U.S. flag

An official website of the United States government

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Cataracts in adults: management. London: National Institute for Health and Care Excellence (NICE); 2017 Oct. (NICE Guideline, No. 77.)

Cover of Cataracts in adults: management

Cataracts in adults: management.

Show details

Appendix HMeta-analysis and Network meta-analysis Results

H.1. Patient information

  • What information do people with cataracts and their carers find useful, and what format (for example written or verbal) do they prefer it to be provided in?
  • What information on cataract surgery do people and their carers find useful when deciding whether surgery is appropriate for them, and before, during and after any operation(s) they elect to undergo? What format (for example written or verbal) do they prefer it to be provided in?

There were no meta-analyses conducted for these questions.

H.2. Indicators for referral

  • What are the indicators for referral for cataract surgery?
  • What are the optimal clinical thresholds in terms of severity and impairment for referral for cataract surgery?

There were no meta-analyses conducted for these questions.

H.3. Pre-operative assessment and biometry

  • What is the effectiveness of different techniques for undertaking biometry?
  • What are the most appropriate formulae to optimise intraocular lens biometry calculation?
  • What is the effectiveness of strategies used to select intraocular lens constants in order to optimise biometry calculation?
  • What other factors should be considered such as, who should undertake biometry and when should preoperative biometry be assessed?
  • What is the effectiveness of risk stratification techniques to reduce surgical complications?
  • What are the risk factors associated with increased surgical complications in cataract surgery?

H.3.1. Biometry techniques - Forest plots of outcomes

H.3.1.1. Ultrasound (immersion and contact) and optical biometry to measure axial length

Mean absolute prediction errors.

Mean absolute prediction errors

Cumulative proportion of eyes within various ranges of absolute prediction errors.

Cumulative proportion of eyes within various ranges of absolute prediction errors

H.3.1.2. Keratometry (manual and automated) and topography to measure corneal curvature

Mean absolute prediction errors.

Mean absolute prediction errors

Cumulative proportion of eyes within various ranges of absolute prediction errors.

Cumulative proportion of eyes within various ranges of absolute prediction errors

H.3.1.3. Observational studies in people undergoing phacoemulsification cataract surgery with a history of corneal refractive surgery

Studies including mixed populations of individuals with a history of different types of refractive surgery (laser-assisted in situ keratomileusis, photorefractive keratectomy and radial keratotomy) for various indications (myopia, hyperopia)

Mean prediction errors.

Mean prediction errors

Mean absolute prediction errors.

Mean absolute prediction errors

Studies including individuals with a history of laser-assisted in situ keratomileusis and photorefractive keratectomy for myopia
Mean prediction errors: keratometry (SRK-T formula) vs topography (SRK-T formula).

Mean prediction errors: keratometry (SRK-T formula) vs topography (SRK-T formula)

Mean prediction errors: keratometry (Haigis-L formula) vs topography (SRK-T formula).

Mean prediction errors: keratometry (Haigis-L formula) vs topography (SRK-T formula)

Cumulative proportion of eyes within various ranges of absolute prediction errors: keratometry (SRK-T formula) vs topography (SRK-T formula).

Cumulative proportion of eyes within various ranges of absolute prediction errors: keratometry (SRK-T formula) vs topography (SRK-T formula)

Cumulative proportion of eyes within various ranges of absolute prediction errors: keratometry (Haigis-L formula) vs topography (SRK-T formula).

Cumulative proportion of eyes within various ranges of absolute prediction errors: keratometry (Haigis-L formula) vs topography (SRK-T formula)

H.3.2. Intraocular lens formulas: Network meta-analyses results: Virgin eyes without a history of corneal refractive surgery

H.3.2.1. Model fit statistics for all outcomes

Table 1Model fit statistics used to select fixed or random effect models for all comparisons and outcomes

StudiesOutcomeModelTotal model DICTotal model DIC (FE – RE)Total residual devianceNo. of data-pointsBetween-study SD (95% CrI)Preferred model
VIRGIN EYES WITHOUT A HISTORY OF CORNEAL REFRACTIVE SURGERY
Axial length less than 22.00mm
7 (Carifi, Cooke, Day, Doshi, Moschos, Ozcura, Srivannaboon)Mean absolute errorFE43.02487.582137.837-RE
RE−44.55836.720.267 (0.174, 0.424)
5 (Aristodemou, Day, Eom, Kane, Srivannaboon)Within 0.25DFE157.592−1.14322.4728-FE
RE158.73521.670.124 (0.004, 0.439)
11 (Aristodemou, Carifi, Cooke, Day, Doshi, Eom, Kane, Moschos, Ozcura, Percival, Srivannaboon)Within 0.5DFE344.2128.65897.8252-RE
RE315.55252.320.589 (0.345, 0.920)
11 (Aristodemou, Carifi, Cooke, Day, Doshi, Eom, Kane, Moschos, Ozcura, Percival, Srivannaboon)Within 1.0DFE295.16418.29583.6652-RE
RE276.86950.870.653 (0.367, 1.035)
3 (Carifi, Kane, Ozcura)Within 2.0DFE43.015−1.0919.17412-FE
RE44.1069.7310.593 (0.024, 1.834)
Axial length 22.00 to 24.50mm
2 (Ozcura, Srivannaboon)Mean absolute errorFE−20.877−0.0276.0156-FE
RE−20.856.0281004 (0.051, 1.948)
3 (Aristodemou, Kane, Srivannaboon)Within 0.25DFE126.752−1.49112.9814-FE
RE128.24312.620.069 (0.005, 0.360)
4 (Aristodemou, Kane, Ozcura, Srivannaboon)Within 0.5DFE138.89−2.35712.816-FE
RE141.24713.650.046 (0.002, 0.242)
4 (Aristodemou, Kane, Ozcura, Srivannaboon)Within 1.0DFE119.799−1.35813.6816-FE
RE121.15713.70.090 (0.003, 0.465)
2 (Kane, Ozcura)Within 2.0DFE43.439−1.0117.3678-FE
RE44.457.7820.745 (0.044, 1.918)
Axial length 24.50 to 26.00mm
1 (Srivannaboon)Mean absolute errorFE−6.1330.0093.9914-FE
RE−6.1423.9860.964 (0.040, 1.949)
3 (Aristodemou, Kane, Srivannaboon)Within 0.25DFE99.785−1.78511.5514-FE
RE101.5712.010.132 (0.006, 0.726)
4 (Aristodemou, Kane, Percival, Srivannaboon)Within 0.5DFE113.319−2.3381517-FE
RE115.65715.630.129 (0.004, 0.755)
6 (Aristodemou, El-Nafees, Kane, Mitra, Percival, Srivannaboon)Within 1.0DFE113.402−1.720.921-FE
RE115.10220.490.227 (0.012, 1.052)
1 (Kane)Within 2.0DFE26.195−0.0296.6666-FE
RE26.2246.7111.010 (0.059, 1.948)
Axial length greater than 26.00mm
2 (Bang, Cooke)Mean absolute errorFE−19.0621.22716.3113-FE
RE−20.28913.250.294 (0.021, 1.623)
2 (Aristodemou, Kane)Within 0.25DFE73.422−2.14810.0512-FE
RE75.5710.930.260 (0.008, 1.540)
5 (Aristodemou, Bang, Cooke, Kane, Percival)Within 0.5DFE160.51−1.36924.4528-FE
RE161.87924.310.122 (0.007, 0.457)
8 (Aristodemou, Bang, Cooke, El-Nafees, Kane, Mitra, Percival, Petermeier)Within 1.0DFE196.07419.25564.135-RE
RE176.81935.10.974 (0.506, 1724)
2 (Bang, Kane)Within 2.0DFE44.4661.00213.4910-FE
RE43.46410.791.033 (0.095, 1.933)

H.3.2.2. Full dataset: Axial length subgroup – less than 22.00mm

Figure 1. AL <22.0mm: Mean absolute error - random effects model – evidence network.

Figure 1AL <22.0mm: Mean absolute error - random effects model – evidence network

Table 2AL <22.0mm: Mean absolute error - random effects model – input data

SRKTBarrettUniversal IIHaigisHofferQHolladay2Holladay2_NoPreSurgRefHolladay2_PreSurgRefHolladay2-LTLadasSuperFormulaOlsen_standaloneT2
Cooke & (2016)0.40
(0.51)		0.39
(0.48)		0.41
(0.51)		0.48
(0.49)		0.44
(0.47)		0.43
(0.47)		0.40
(0.48)		0.46
(0.57)		0.39
(0.49)
Doshi et al. (2017)0.54
(0.46)		1.36
(0.75)		0.59
(0.36)
Ozcura et al. (2016)0.70
(0.64)		0.76
(0.65)
Carifi et al. (2015)1.34
(1.04)		1.03
(0.87)		0.95
(0.78)		0.82
(0.77)
Srivannaboon et al. (2013)0.44
(0.40)		0.42
(0.33)		0.44
(0.31)		0.45
(0.30)
Day et al. (2012)0.52
(0.42)		0.44
(0.35)		0.46
(0.39)
Day et al. (2012)0.50
(0.37)		0.37
(0.28)		0.50
(0.37)
Day et al. (2012)0.79
(0.56)		0.86
(0.58)		0.74
(0.58)
Day et al. (2012)0.85
(0.56)		0.77
(0.51)		0.83
(0.61)
Moschos et al. (2014)0.97
(0.38)		0.43
(0.22)		0.72
(0.51)

Table 3AL <22.0mm: Mean absolute error random effects model relative effectiveness of all pairwise combinations (MD and 95% credible interval)

SRKTBarrettUniversalIIHaigisHofferQHolladay2Holladay2_NoPreSurgRefHolladay2_PreSurgRefHolladay2-LTLadasSuperFormulaOlsen_standaloneT2
SRKT−0.01
(−0.22, 0.20)		−0.03
(−0.30, 0.24)		−0.05
(−0.14, 0.04)		−0.52
(−1.00, −0.04)		0.04
(−0.17, 0.25)		0.03
(−0.19, 0.24)		-0.00
(−0.21, 0.21)		0.06
(−0.18, 0.29)		−0.01
(−0.22, 0.21)
BarrettUniversalII−0.07
(−0.57, 0.44)		0.02
(−0.20, 0.23)		0.09
(−0.12, 0.30)		-0.05
(−0.15, 0.26)		0.04
(−0.17, 0.24)		-0.01
(−0.20, 0.22)		0.07
(−0.16, 0.29)		0.00
(−0.21, 0.21)
Haigis−0.04
(−0.25, 0.18)		0.03
(−0.48, 0.54)		−0.03
(−0.22, 0.15)		−0.05
(−0.27, 0.17)		0.04
(−0.18, 0.25)		0.02
(−0.19, 0.23)		0.01
(−0.24, 0.26)		−0.01
(−0.22, 0.21)		0.05
(−0.18, 0.28)		−0.01
(−0.23, 0.20)
HofferQ−0.06
(−0.27, 0.14)		0.01
(−0.49, 0.51)		−0.02
(−0.23, 0.18)		−0.02
(−0.22, 0.18)		−0.04
(−0.25, 0.17)		−0.06
(−0.26, 0.15)		0.03
(−0.20, 0.26)		−0.08
(−0.29, 0.13)		−0.03
(−0.25, 0.20)		−0.09
(−0.30, 0.12)
Holladay2−0.16
(−0.58, 0.27)		−0.09
(−0.72, 0.54)		−0.12
(−0.53, 0.28)		−0.09
(−0.50, 0.30)		--0.01
(−0.21, 0.23)		---
Holladay2_NoPreSurgRef−0.02
(−0.51, 0.48)		0.05
(−0.54, 0.65)		0.02
(−0.48, 0.52)		0.04
(−0.46, 0.54)		0.14
(−0.49, 0.77)		−0.02
(−0.22, 0.19)		-−0.04
(−0.25, 0.16)		0.02
(−0.21, 0.24)		−0.05
(−0.26, 0.16)
Holladay2_PreSurgRef−0.04
(−0.54, 0.47)		0.03
(−0.57, 0.64)		0.00
(−0.51, 0.50)		0.02
(−0.48, 0.53)		0.12
(−0.50, 0.75)		−0.02
(−0.62, 0.57)		-−0.03
(−0.23, 0.18)		0.03
(−0.19, 0.26)		−0.03
(−0.24, 0.18)
Holladay2-LT−0.07
(−0.61, 0.47)		0.00
(−0.72, 0.72)		−0.03
(−0.56, 0.49)		−0.01
(−0.53, 0.52)		0.08
(−0.46, 0.65)		−0.05
(−0.76, 0.66)		−0.03
(−0.75, 0.68)		---
LadasSuperFormula−0.06
(−0.56, 0.45)		0.01
(−0.60, 0.61)		−0.02
(−0.53, 0.48)		0.00
(−0.50, 0.51)		0.10
(−0.53, 0.72)		−0.04
(−0.63, 0.54)		−0.02
(−0.62, 0.58)		0.01
(−0.70, 0.72)		0.06
(−0.17, 0.28)		−0.01
(−0.22, 0.20)
Olsen_standalone0.00
(−0.52, 0.50)		0.07
(−0.54, 0.67)		0.04
(−0.49, 0.54)		0.06
(−0.46, 0.56)		0.16
(−0.48, 0.79)		0.02
(−0.59, 0.62)		0.04
(−0.57, 0.64)		0.07
(−0.65, 0.78)		0.06
(−0.55, 0.67)		−0.06
(−0.29, 0.17)
T2−0.07
(−0.57, 0.43)		0.00
(−0.60, 0.60)		−0.03
(−0.54, 0.47)		−0.01
(−0.51, 0.49)		0.09
(−0.54, 0.71)		−0.05
(−0.65, 0.55)		−0.03
(−0.63, 0.56)		0.00
(−0.71, 0.71)		−0.01
(−0.61, 0.59)		−0.07
(−0.67, 0.53)
Figure 2. AL <22.0mm: Mean absolute error - random effects model – relative effect of all options versus common comparator.

Figure 2AL <22.0mm: Mean absolute error - random effects model – relative effect of all options versus common comparator

Table 4AL <22.0mm: Mean absolute error - random effects model – rankings for each comparator

Probability bestMedian rank (95%CI)
SRKT0.0057 (2, 11)
BarrettUniversalII0.1225 (1, 11)
Haigis0.0126 (2, 10)
HofferQ0.0176 (2, 10)
Holladay20.2143 (1, 11)
Holladay2_NoPreSurgRef0.0787 (1, 11)
Holladay2_PreSurgRef0.0906 (1, 11)
Holladay2-LT0.1555 (1, 11)
LadasSuperFormula0.1126 (1, 11)
Olsen_standalone0.0737 (1, 11)
T20.1225 (1, 11)
Figure 3. AL <22.0mm: Mean absolute error - random effects model – rank probability histograms.

Figure 3AL <22.0mm: Mean absolute error - random effects model – rank probability histograms

Table 5AL <22.0mm: Mean absolute error - random effects model – model fit statistics

Residual devianceDbarDhatpDDICtau
36.72
(compared to 37 datapoints)		−78.017−111.47633.459−44.5580.267 (95%CI: 0.174, 0.424)

Table 6AL <22.0mm: Mean absolute error - random effects model – notes

  • Continuous (normal; identity link); random effects
  • Prior distribution for between-study heterogeneity: uniform (Min=0; Max=2)
  • 50000 burn-ins; 10000 recorded iterations
Figure 4. AL <22.0mm: Within 0.25D - fixed effects model – evidence network.

Figure 4AL <22.0mm: Within 0.25D - fixed effects model – evidence network

Table 7AL <22.0mm: Within 0.25D - fixed effects model – input data

SRKTBarrettUniversalIIHaigisHofferQHolladay2Holladay2-LTT2
Kane, J. et al. (2016)51/15648/15657/15644/15649/15652/156
Eom et al. (2014)28/7522/75
Srivannaboon et al. (2013)5/156/155/155/15
Day et al. (2012)11/3212/3210/32
Day et al. (2012)32/10035/10039/100
Day et al. (2012)2/193/193/19
Day et al. (2012)3/122/124/12
Aristodemou et al. (2011)50/15144/151
Aristodemou et al. (2011)145/457168/457

Table 8AL <22.0mm: Within 0.25D - fixed effects model – relative effectiveness of all pairwise combinations (RR and 95% credible interval)

SRKTBarrettUniversalIIHaigisHofferQHolladay2Holladay2-LTT2
SRKT0.92
(0.57, 1.47)		1.16
(0.83, 1.61)		1.09
(0.90, 1.33)		0.94
(0.59, 1.52)		-1.03
(0.64, 1.65)
BarrettUniversalII1.01
(0.67, 1.53)		1.30
(0.81, 2.07)		0.88
(0.54, 1.44)		1.03
(0.64, 1.66)		-1.13
(0.70, 1.81)
Haigis1.23
(0.92, 1.64)		1.21
(0.79, 1.85)		0.86
(0.64, 1.15)		0.81
(0.52, 1.27)		1.00
(0.22, 4.56)		0.87
(0.55, 1.38)
HofferQ1.09
(0.90, 1.32)		1.07
(0.71, 1.62)		0.88
(0.67, 1.16)		1.12
(0.70, 1.77)		0.75
(0.17, 3.33)		1.27
(0.79, 2.06)
Holladay21.04
(0.70, 1.55)		1.03
(0.64, 1.65)		0.85
(0.56, 1.27)		0.96
(0.64, 1.41)		1.00
(0.22, 4.56)		1.09
(0.68, 1.75)
Holladay2-LT0.99
(0.26, 3.53)		0.98
(0.24, 3.57)		0.81
(0.21, 2.82)		0.92
(0.24, 3.20)		0.96
(0.25, 3.38)		-
T21.14
(0.76, 1.71)		1.13
(0.70, 1.83)		0.93
(0.61, 1.42)		1.05
(0.70, 1.57)		1.10
(0.69, 1.76)		1.15
(0.31, 4.64)
Figure 5. AL <22.0mm: Within 0.25D - fixed effects model – relative effect of all options versus common comparator.

Figure 5AL <22.0mm: Within 0.25D - fixed effects model – relative effect of all options versus common comparator

Table 9AL <22.0mm: Within 0.25D - fixed effects model – rankings for each comparator

Probability bestMedian rank (95%CI)
SRKT0.0075 (2, 7)
BarrettUniversalII0.0665 (1, 7)
Haigis0.2862 (1, 6)
HofferQ0.0444 (1, 7)
Holladay20.0745 (1, 7)
Holladay2-LT0.3275 (1, 7)
T20.1963 (1, 7)
Figure 6. AL <22.0mm: Within 0.25D - fixed effects model – rank probability histograms.

Figure 6AL <22.0mm: Within 0.25D - fixed effects model – rank probability histograms

Table 10AL <22.0mm: Within 0.25D - fixed effects model – model fit statistics

Residual devianceDbarDhatpDDIC
22.47
(compared to 28 datapoints)		142.482127.37315.109157.592

Table 11AL <22.0mm: Within 0.25D - fixed effects model – notes

  • Dichotomous synchronic (binomial; logit link); fixed effects
  • 50000 burn-ins; 10000 recorded iterations
Figure 7. AL <22.0mm: Within 0.50D - random effects model – evidence network.

Figure 7AL <22.0mm: Within 0.50D - random effects model – evidence network

Table 12AL <22.0mm: Within 0.50D - random effects model – input data

SRKTAvgHofferQSRKTBarrettUniversalIIHaigisHofferQHolladay2Holladay2_NoPreSurgRefHolladay2_PreSurgRefHolladay2-LTLadasSuperFormulaOlsen_standaloneT2
Cooke & (2016)28/4132/4128/4126/4130/4127/4133/4125/4130/41
Doshi et al. (2017)22/407/4017/40
Kane, J. et al. (2016)93/15697/15698/15687/15696/15694/156
Ozcura et al. (2016)14/3215/32
Carifi et al. (2015)6/2812/2811/2812/28
Eom et al. (2014)50/7547/75
Srivannaboon et al. (2013)6/159/157/157/15
Day et al. (2012)20/3224/3218/32
Day et al. (2012)54/10068/10060/100
Day et al. (2012)6/194/199/19
Day et al. (2012)4/124/124/12
Aristodemou et al. (2011)91/15185/151
Aristodemou et al. (2011)276/457293/457
Percival et al. (2002)25/5436/5435/54
Moschos et al. (2014)13/6950/6941/69

Table 13AL <22.0mm: Within 0.50D - random effects model – relative effectiveness of all pairwise combinations (RR and 95% credible interval)

SRKTAvgHofferQSRKTBarrettUniversalIIHaigisHofferQHolladay2Holladay2_NoPreSurgRefHolladay2_PreSurgRefHolladay2-LTLadasSuperFormulaOlsen_standaloneT2
SRKT2.32
(1.06, 5.05)		1.19
(0.79, 1.80)		1.38
(0.68, 2.80)		1.25
(0.92, 1.71)		1.47
(0.62, 3.45)		1.27
(0.49, 3.29)		0.90
(0.36, 2.25)		-1.92
(0.69, 5.28)		0.73
(0.29, 1.80)		1.07
(0.71, 1.61)
AvgHofferQSRKT1.82
(0.51, 6.65)		--0.92
(0.42, 2.04)		-------
BarrettUniversalII1.65
(0.70, 3.92)		0.91
(0.20, 4.12)		0.94
(0.62, 1.42)		0.71
(0.47, 1.07)		0.97
(0.62, 1.54)		0.77
(0.28, 2.11)		0.54
(0.20, 1.45)		-1.16
(0.40, 3.38)		0.44
(0.17, 1.16)		0.89
(0.59, 1.35)
Haigis1.38
(0.84, 2.21)		0.76
(0.20, 2.80)		0.84
(0.34, 1.96)		0.90
(0.65, 1.26)		0.98
(0.65, 1.47)		1.27
(0.49, 3.29)		0.90
(0.36, 2.25)		1.31
(0.31, 5.58)		1.92
(0.69, 5.28)		0.73
(0.29, 1.80)		0.96
(0.63, 1.44)
HofferQ1.30
(0.86, 1.96)		0.71
(0.20, 2.53)		0.79
(0.34, 1.85)		0.94
(0.61, 1.50)		1.18
(0.79, 1.76)		1.57
(0.62, 4.02)		1.11
(0.45, 2.75)		0.58
(0.14, 2.48)		2.38
(0.88, 6.47)		0.90
(0.37, 2.20)		1.26
(0.84, 1.90)
Holladay21.45
(0.66, 3.17)		0.80
(0.18, 3.42)		0.88
(0.31, 2.43)		1.05
(0.49, 2.31)		1.11
(0.52, 2.41)		--1.00
(0.24, 4.20)		--0.95
(0.60, 1.49)
Holladay2_NoPreSurgRef1.53
(0.43, 5.53)		0.84
(0.14, 4.98)		0.93
(0.23, 3.75)		1.11
(0.32, 4.08)		1.17
(0.34, 4.21)		1.06
(0.26, 4.40)		0.71
(0.27, 1.82)		-1.51
(0.54, 4.26)		0.57
(0.23, 1.46)		1.00
(0.38, 2.66)
Holladay2_PreSurgRef1.07
(0.31, 3.67)		0.59
(0.10, 3.36)		0.65
(0.17, 2.50)		0.78
(0.23, 2.69)		0.83
(0.24, 2.81)		0.74
(0.18, 2.95)		0.70
(0.15, 3.30)		-2.14
(0.78, 5.85)		0.81
(0.33, 1.99)		1.41
(0.55, 3.64)
Holladay2-LT1.23
(0.24, 6.19)		0.67
(0.09, 5.14)		0.75
(0.13, 4.28)		0.90
(0.18, 4.45)		0.95
(0.19, 4.60)		0.85
(0.16, 4.39)		0.80
(0.11, 5.95)		1.14
(0.16, 8.35)		---
LadasSuperFormula2.36
(0.64, 8.96)		1.30
(0.21, 7.91)		1.44
(0.35, 6.04)		1.73
(0.47, 6.49)		1.82
(0.49, 6.79)		1.64
(0.39, 7.07)		1.55
(0.31, 7.85)		2.21
(0.46, 10.94)		1.94
(0.26, 14.95)		0.38
(0.14, 1.02)		0.66
(0.23, 1.86)
Olsen_standalone0.86
(0.25, 2.98)		0.47
(0.08, 2.72)		0.52
(0.14, 2.01)		0.62
(0.18, 2.17)		0.66
(0.19, 2.25)		0.59
(0.15, 2.41)		0.56
(0.12, 2.64)		0.80
(0.18, 3.69)		0.70
(0.10, 5.04)		0.36
(0.07, 1.73)		1.75
(0.69, 4.44)
T21.42
(0.60, 3.35)		0.78
(0.17, 3.48)		0.86
(0.31, 2.36)		1.03
(0.44, 2.47)		1.09
(0.47, 2.56)		0.98
(0.35, 2.72)		0.93
(0.23, 3.64)		1.32
(0.35, 4.97)		1.15
(0.20, 6.75)		0.60
(0.14, 2.43)		1.65
(0.43, 6.43)
Figure 8. AL <22.0mm: Within 0.50D - random effects model – relative effect of all options versus common comparator.

Figure 8AL <22.0mm: Within 0.50D - random effects model – relative effect of all options versus common comparator

Table 14AL <22.0mm: Within 0.50D - random effects model – rankings for each comparator

Probability bestMedian rank (95%CI)
SRKT0.00010 (5, 12)
AvgHofferQSRKT0.2174 (1, 12)
BarrettUniversalII0.0744 (1, 11)
Haigis0.0066 (2, 10)
HofferQ0.0027 (3, 11)
Holladay20.0346 (1, 11)
Holladay2_NoPreSurgRef0.1135 (1, 12)
Holladay2_PreSurgRef0.0329 (1, 12)
Holladay2-LT0.1228 (1, 12)
LadasSuperFormula0.3572 (1, 11)
Olsen_standalone0.01411 (2, 12)
T20.0316 (1, 12)
Figure 9. AL <22.0mm: Within 0.50D - random effects model – rank probability histograms.

Figure 9AL <22.0mm: Within 0.50D - random effects model – rank probability histograms

Table 15AL <22.0mm: Within 0.50D - random effects model – model fit statistics

Residual devianceDbarDhatpDDICtau
52.32
(compared to 52 datapoints)		272.542229.53243.01315.5520.589 (95%CI: 0.345, 0.920)

Table 16AL <22.0mm: Within 0.50D - random effects model – notes

  • Dichotomous synchronic (binomial; logit link); random effects
  • Prior distribution for between-study heterogeneity: uniform (Min=0; Max=2)
  • 50000 burn-ins; 10000 recorded iterations
Figure 10. AL <22.0mm: Within 1.0D - random effects model – evidence network.

Figure 10AL <22.0mm: Within 1.0D - random effects model – evidence network

Table 17AL <22.0mm: Within 1.0D - random effects model – input data

SRKTAvgHofferQSRKTBarrettUniversalIIHaigisHofferQHolladay2Holladay2_NoPreSurgRefHolladay2_PreSurgRefHolladay2-LTLadasSuperFormulaOlsen_standaloneT2
Cooke & (2016)39/4138/4139/4136/4136/4138/4138/4139/4139/41
Doshi et al. (2017)33/4014/4036/40
Kane, J. et al. (2016)144/156144/156142/156142/156143/156145/156
Ozcura et al. (2016)24/3226/32
Carifi et al. (2015)12/2813/2817/2818/28
Eom et al. (2014)66/7566/75
Srivannaboon et al. (2013)11/1513/1513/1513/15
Day et al. (2012)28/3231/3228/32
Day et al. (2012)89/10093/10092/100
Day et al. (2012)14/1912/1914/19
Day et al. (2012)6/127/126/12
Aristodemou et al. (2011)130/151131/151
Aristodemou et al. (2011)399/457408/457
Percival et al. (2002)43/5445/5448/54
Moschos et al. (2014)47/6964/6959/69

Table 18AL <22.0mm: Within 1.0D - random effects model – relative effectiveness of all pairwise combinations (RR and 95% credible interval)

SRKTAvgHofferQSRKTBarrettUniversalIIHaigisHofferQHolladay2Holladay2_NoPreSurgRefHolladay2_PreSurgRefHolladay2-LTLadasSuperFormulaOlsen_standaloneT2
SRKT1.28
(0.48, 3.39)		0.93
(0.44, 1.99)		1.10
(0.49, 2.48)		1.29
(1.02, 1.65)		1.39
(0.55, 3.53)		0.37
(0.07, 2.02)		0.65
(0.10, 4.11)		-0.65
(0.10, 4.11)		1.00
(0.13, 7.46)		1.08
(0.50, 2.37)
AvgHofferQSRKT1.05
(0.23, 4.61)		--1.60
(0.53, 4.86)		-------
BarrettUniversalII1.27
(0.43, 3.83)		1.21
(0.20, 7.73)		0.93
(0.44, 1.95)		0.77
(0.38, 1.57)		0.92
(0.40, 2.08)		0.57
(0.13, 2.55)		1.00
(0.19, 5.27)		-1.00
(0.19, 5.27)		1.54
(0.24, 9.73)		1.17
(0.54, 2.52)
Haigis1.07
(0.62, 1.88)		1.02
(0.22, 4.95)		0.84
(0.28, 2.50)		1.18
(0.63, 2.20)		1.44
(0.79, 2.63)		0.37
(0.07, 2.02)		0.65
(0.10, 4.11)		2.36
(0.36, 15.45)		0.65
(0.10, 4.11)		1.00
(0.13, 7.46)		1.25
(0.58, 2.68)
HofferQ1.32
(0.81, 2.11)		1.26
(0.28, 5.75)		1.04
(0.34, 3.00)		1.23
(0.71, 2.10)		1.10
(0.60, 2.03)		1.00
(0.27, 3.75)		1.76
(0.39, 7.90)		1.00
(0.12, 8.21)		1.76
(0.39, 7.90)		2.71
(0.49, 14.84)		1.49
(0.71, 3.13)
Holladay21.56
(0.63, 4.02)		1.48
(0.27, 8.68)		1.23
(0.35, 4.38)		1.45
(0.59, 3.70)		1.18
(0.48, 3.04)		--1.00
(0.12, 8.21)		--1.20
(0.52, 2.76)
Holladay2_NoPreSurgRef0.63
(0.13, 3.23)		0.60
(0.07, 5.29)		0.50
(0.09, 2.81)		0.59
(0.12, 2.94)		0.48
(0.10, 2.41)		0.41
(0.07, 2.40)		1.76
(0.39, 7.90)		-1.76
(0.39, 7.90)		2.71
(0.49, 14.84)		2.71
(0.49, 14.84)
Holladay2_PreSurgRef1.20
(0.22, 8.15)		1.16
(0.12, 13.02)		0.95
(0.15, 7.30)		1.12
(0.20, 7.69)		0.91
(0.16, 6.23)		0.77
(0.12, 6.00)		1.90
(0.25, 16.72)		-1.00
(0.19, 5.27)		1.54
(0.24, 9.73)		1.54
(0.24, 9.73)
Holladay2-LT2.00
(0.26, 22.38)		1.94
(0.16, 32.60)		1.58
(0.17, 21.34)		1.87
(0.25, 21.28)		1.52
(0.20, 17.15)		1.29
(0.16, 14.88)		3.20
(0.24, 55.29)		1.70
(0.10, 30.39)		---
LadasSuperFormula1.18
(0.21, 7.80)		1.14
(0.12, 12.03)		0.94
(0.14, 6.75)		1.11
(0.20, 7.26)		0.90
(0.16, 5.99)		0.76
(0.11, 5.82)		1.87
(0.24, 16.44)		0.99
(0.10, 9.58)		0.59
(0.03, 8.93)		1.54
(0.24, 9.73)		1.54
(0.24, 9.73)
Olsen_standalone2.00
(0.31, 20.29)		1.96
(0.17, 27.07)		1.58
(0.21, 16.97)		1.86
(0.28, 18.71)		1.51
(0.24, 15.13)		1.29
(0.17, 14.30)		3.21
(0.35, 38.30)		1.66
(0.15, 22.15)		1.01
(0.05, 19.87)		1.69
(0.16, 21.88)		1.00
(0.13, 7.46)
T21.57
(0.53, 4.94)		1.50
(0.24, 9.77)		1.23
(0.34, 4.73)		1.46
(0.48, 4.64)		1.19
(0.40, 3.73)		1.01
(0.28, 3.68)		2.46
(0.43, 14.60)		1.31
(0.17, 8.54)		0.79
(0.06, 7.53)		1.32
(0.17, 8.98)		0.78
(0.07, 6.03)
Figure 11. AL <22.0mm: Within 1.0D - random effects model – relative effect of all options versus common comparator.

Figure 11AL <22.0mm: Within 1.0D - random effects model – relative effect of all options versus common comparator

Table 19AL <22.0mm: Within 1.0D - random effects model – rankings for each comparator

Probability bestMedian rank (95%CI)
SRKT0.0018 (4, 12)
AvgHofferQSRKT0.0588 (1, 12)
BarrettUniversalII0.0326 (1, 12)
Haigis0.0028 (3, 11)
HofferQ0.0066 (2, 10)
Holladay20.0565 (1, 11)
Holladay2_NoPreSurgRef0.01211 (2, 12)
Holladay2_PreSurgRef0.0957 (1, 12)
Holladay2-LT0.2983 (1, 12)
LadasSuperFormula0.0937 (1, 12)
Olsen_standalone0.2733 (1, 12)
T20.0755 (1, 11)
Figure 12. AL <22.0mm: Within 1.0D - random effects model – rank probability histograms.

Figure 12AL <22.0mm: Within 1.0D - random effects model – rank probability histograms

Table 20AL <22.0mm: Within 1.0D - random effects model – model fit statistics

Residual devianceDbarDhatpDDICtau
50.87
(compared to 52 datapoints)		236.411195.95440.458276.8690.653 (95%CI: 0.367, 1.035)

Table 21AL <22.0mm: Within 1.0D - random effects model – notes

  • Dichotomous synchronic (binomial; logit link); random effects
  • Prior distribution for between-study heterogeneity: uniform (Min=0; Max=2)
  • 50000 burn-ins; 10000 recorded iterations
Figure 13. AL <22.0mm: Within 2.0D - fixed effects model – evidence network.

Figure 13AL <22.0mm: Within 2.0D - fixed effects model – evidence network

Table 22AL <22.0mm: Within 2.0D - fixed effects model – input data

SRKTBarrettUniversalIIHaigisHofferQHolladay2T2
Kane, J. et al. (2016)155/156156/156156/156156/156156/156155/156
Ozcura et al. (2016)31/3231/32
Carifi et al. (2015)22/2824/2825/2826/28

Table 23AL <22.0mm: Within 2.0D - fixed effects model – relative effectiveness of all pairwise combinations (RR and 95% credible interval)

SRKTBarrettUniversalIIHaigisHofferQHolladay2T2
SRKT3.02
(0.12, 74.69)		1.83
(0.51, 6.48)		2.05
(0.61, 6.86)		3.42
(0.76, 15.35)		1.00
(0.06, 16.13)
BarrettUniversalII4.65
(0.25, 2033.00)		1.00
(0.02, 50.71)		1.00
(0.02, 50.71)		1.00
(0.02, 50.71)		0.33
(0.01, 8.19)
Haigis1.77
(0.51, 7.04)		0.38
(0.00, 8.10)		1.33
(0.30, 5.81)		1.90
(0.38, 9.54)		0.33
(0.01, 8.19)
HofferQ2.14
(0.63, 8.20)		0.46
(0.00, 9.67)		1.21
(0.28, 5.31)		1.44
(0.27, 7.74)		0.33
(0.01, 8.19)
Holladay23.61
(0.85, 21.95)		0.77
(0.00, 20.64)		2.03
(0.40, 13.61)		1.67
(0.32, 11.15)		0.33
(0.01, 8.19)
T20.91
(0.10, 12.74)		0.20
(0.00, 6.00)		0.51
(0.05, 7.38)		0.42
(0.04, 6.24)		0.25
(0.02, 4.14)
Figure 14. AL <22.0mm: Within 2.0D - fixed effects model – relative effect of all options versus common comparator.

Figure 14AL <22.0mm: Within 2.0D - fixed effects model – relative effect of all options versus common comparator

Table 24AL <22.0mm: Within 2.0D - fixed effects model – rankings for each comparator

Probability bestMedian rank (95%CI)
SRKT0.0035 (3, 6)
BarrettUniversalII0.4922 (1, 6)
Haigis0.0524 (1, 6)
HofferQ0.0893 (1, 6)
Holladay20.3092 (1, 5)
T20.0555 (1, 6)
Figure 15. AL <22.0mm: Within 2.0D - fixed effects model – rank probability histograms.

Figure 15AL <22.0mm: Within 2.0D - fixed effects model – rank probability histograms

Table 25AL <22.0mm: Within 2.0D - fixed effects model – model fit statistics

Residual devianceDbarDhatpDDIC
9.174
(compared to 12 datapoints)		35.54128.0667.47443.015

Table 26AL <22.0mm: Within 2.0D - fixed effects model – notes

  • Dichotomous synchronic (binomial; logit link); fixed effects
  • 50000 burn-ins; 10000 recorded iterations

H.3.2.3. Full dataset: Axial length subgroup – 22.00 to 24.50mm

Figure 16. AL 22.0–24.5mm: Mean absolute error - fixed effects model – evidence network.

Figure 16AL 22.0–24.5mm: Mean absolute error - fixed effects model – evidence network

Table 27AL 22.0−24.5mm: Mean absolute error - fixed effects model – input data

SRKTHaigisHofferQHolladay2Holladay2-LT
Ozcura et al. (2016)0.51 (0.42)0.55 (0.44)
Srivannaboon et al. (2013)0.40 (0.33)0.39 (0.33)0.41 (0.31)0.42 (0.30)

Table 28AL 22.0−24.5mm: Mean absolute error - fixed effects model – relative effectiveness of all pairwise combinations (MD and 95% credible interval)

SRKTHaigisHofferQHolladay2Holladay2-LT
SRKT-0.04
(−0.02, 0.10)		--
Haigis0.05
(−0.05, 0.15)		−0.01
(−0.09, 0.07)		0.01
(−0.07, 0.09)		0.02
(−0.06, 0.10)
HofferQ0.04
(−0.02, 0.10)		−0.01
(−0.09, 0.07)		0.02
(−0.06, 0.10)		0.03
(−0.05, 0.11)
Holladay20.06
(−0.04, 0.16)		0.01
(−0.07, 0.09)		0.02
(−0.06, 0.10)		0.01
(−0.07, 0.09)
Holladay2-LT0.07
(−0.03, 0.17)		0.02
(−0.06, 0.10)		0.03
(−0.05, 0.11)		0.01
(−0.07, 0.09)
Figure 17. AL 22.0–24.5mm: Mean absolute error - fixed effects model – relative effect of all options versus common comparator.

Figure 17AL 22.0–24.5mm: Mean absolute error - fixed effects model – relative effect of all options versus common comparator

Table 29AL 22.0−24.5mm: Mean absolute error - fixed effects model – rankings for each comparator

Probability bestMedian rank (95%CI)
SRKT0.7391 (1, 4)
Haigis0.1233 (1, 5)
HofferQ0.0373 (1, 5)
Holladay20.0684 (1, 5)
Holladay2-LT0.0324 (1, 5)
Figure 18. AL 22.0–24.5mm: Mean absolute error - fixed effects model – rank probability histograms.

Figure 18AL 22.0–24.5mm: Mean absolute error - fixed effects model – rank probability histograms

Table 30AL 22.0−24.5mm: Mean absolute error - fixed effects model – model fit statistics

Residual devianceDbarDhatpDDIC
6.015
(compared to 6 datapoints)		−26.891−32.9056.014−20.877

Table 31AL 22.0−24.5mm: Mean absolute error - fixed effects model – notes

  • Continuous (normal; identity link); fixed effects
  • 50000 burn-ins; 10000 recorded iterations
Figure 19. AL 22.0–24.5mm: Within 0.25D - fixed effects model – evidence network.

Figure 19AL 22.0–24.5mm: Within 0.25D - fixed effects model – evidence network

Table 32AL 22.0−24.5mm: Within 0.25D - fixed effects model – input data

SRKTBarrettUniversalIIHaigisHofferQHolladay2Holladay2-LTT2
Kane, J. et al. (2016)1018/26381126/26381029/26381029/26381000/26381029/2638
Srivannaboon et al. (2013)52/12450/12445/12446/124
Aristodemou et al. (2011)599/1508609/1508
Aristodemou et al. (2011)1985/46991900/4699

Table 33AL 22.0−24.5mm: Within 0.25D - fixed effects model – relative effectiveness of all pairwise combinations (RR and 95% credible interval)

SRKTBarrettUniversalIIHaigisHofferQHolladay2Holladay2-LTT2
SRKT1.19
(1.06, 1.32)		1.02
(0.91, 1.14)		0.97
(0.91, 1.03)		0.97
(0.87, 1.09)		-1.02
(0.91, 1.14)
BarrettUniversalII1.16
(1.05, 1.28)		0.86
(0.77, 0.96)		0.86
(0.77, 0.96)		0.82
(0.73, 0.92)		-0.86
(0.77, 0.96)
Haigis1.00
(0.90, 1.10)		0.86
(0.77, 0.96)		1.00
(0.89, 1.11)		0.95
(0.85, 1.05)		0.82
(0.49, 1.36)		1.00
(0.90, 1.12)
HofferQ0.97
(0.91, 1.03)		0.84
(0.76, 0.93)		0.97
(0.88, 1.08)		0.95
(0.85, 1.06)		0.87
(0.52, 1.46)		1.00
(0.90, 1.12)
Holladay20.94
(0.85, 1.04)		0.82
(0.73, 0.91)		0.95
(0.85, 1.06)		0.97
(0.88, 1.07)		1.04
(0.62, 1.74)		1.05
(0.94, 1.17)
Holladay2-LT0.87
(0.57, 1.33)		0.76
(0.49, 1.15)		0.88
(0.57, 1.33)		0.90
(0.59, 1.36)		0.93
(0.60, 1.41)		-
T20.99
(0.90, 1.10)		0.86
(0.77, 0.96)		1.00
(0.89, 1.11)		1.02
(0.93, 1.13)		1.05
(0.94, 1.18)		1.14
(0.75, 1.76)
Figure 20. AL 22.0–24.5mm: Within 0.25D - fixed effects model – relative effect of all options versus common comparator.

Figure 20AL 22.0–24.5mm: Within 0.25D - fixed effects model – relative effect of all options versus common comparator

Table 34AL 22.0−24.5mm: Within 0.25D - fixed effects model – rankings for each comparator

Probability bestMedian rank (95%CI)
SRKT0.0013 (2, 6)
BarrettUniversalII0.8951 (1, 2)
Haigis0.0034 (2, 7)
HofferQ0.0005 (2, 7)
Holladay20.0006 (3, 7)
Holladay2-LT0.0987 (1, 7)
T20.0034 (2, 7)
Figure 21. AL 22.0–24.5mm: Within 0.25D - fixed effects model – rank probability histograms.

Figure 21AL 22.0–24.5mm: Within 0.25D - fixed effects model – rank probability histograms

Table 35AL 22.0−24.5mm: Within 0.25D - fixed effects model – model fit statistics

Residual devianceDbarDhatpDDIC
12.98
(compared to 14 datapoints)		116.779106.8059.974126.752

Table 36AL 22.0−24.5mm: Within 0.25D - fixed effects model – notes

  • Dichotomous synchronic (binomial; logit link); fixed effects
  • 50000 burn-ins; 10000 recorded iterations
Figure 22. AL 22.0–24.5mm: Within 0.5D - fixed effects model – evidence network.

Figure 22AL 22.0–24.5mm: Within 0.5D - fixed effects model – evidence network

Table 37AL 22.0−24.5mm: Within 0.5D - fixed effects model – input data

SRKTBarrettUniversalIIHaigisHofferQHolladay2Holladay2-LTT2
Kane, J. et al. (2016)1868/26381881/26381820/26381796/26381796/26381833/2638
Ozcura et al. (2016)245/422221/422
Srivannaboon et al. (2013)82/12484/12487/12489/124
Aristodemou et al. (2011)1062/15081033/1508
Aristodemou et al. (2011)3353/46993266/4699

Table 38AL 22.0−24.5mm: Within 0.5D - fixed effects model – relative effectiveness of all pairwise combinations (RR and 95% credible interval)

SRKTBarrettUniversalIIHaigisHofferQHolladay2Holladay2-LTT2
SRKT1.02
(0.91, 1.15)		0.92
(0.82, 1.03)		0.90
(0.84, 0.96)		0.88
(0.78, 0.99)		-0.94
(0.83, 1.06)
BarrettUniversalII1.04
(0.93, 1.15)		0.90
(0.80, 1.01)		0.86
(0.76, 0.97)		0.86
(0.76, 0.97)		-0.92
(0.81, 1.03)
Haigis0.92
(0.83, 1.03)		0.89
(0.79, 1.00)		0.96
(0.86, 1.08)		0.97
(0.86, 1.09)		1.30
(0.76, 2.23)		1.02
(0.91, 1.15)
HofferQ0.90
(0.84, 0.96)		0.87
(0.78, 0.96)		0.97
(0.88, 1.08)		1.01
(0.90, 1.13)		1.21
(0.70, 2.08)		1.07
(0.95, 1.20)
Holladay20.89
(0.81, 0.99)		0.86
(0.77, 0.97)		0.97
(0.86, 1.09)		1.00
(0.90, 1.11)		1.08
(0.62, 1.87)		1.07
(0.95, 1.20)
Holladay2-LT1.09
(0.70, 1.73)		1.05
(0.67, 1.68)		1.18
(0.75, 1.87)		1.21
(0.78, 1.92)		1.22
(0.78, 1.93)		-
T20.95
(0.85, 1.06)		0.92
(0.81, 1.03)		1.03
(0.92, 1.16)		1.06
(0.95, 1.17)		1.06
(0.95, 1.19)		0.87
(0.55, 1.37)
Figure 23. AL 22.0–24.5mm: Within 0.5D - fixed effects model – relative effect of all options versus common comparator.

Figure 23AL 22.0–24.5mm: Within 0.5D - fixed effects model – relative effect of all options versus common comparator

Table 39AL 22.0−24.5mm: Within 0.5D - fixed effects model – rankings for each comparator

Probability bestMedian rank (95%CI)
SRKT0.0943 (1, 5)
BarrettUniversalII0.3182 (1, 4)
Haigis0.0045 (2, 7)
HofferQ0.0006 (4, 7)
Holladay20.0006 (3, 7)
Holladay2-LT0.5691 (1, 7)
T20.0164 (2, 7)
Figure 24. AL 22.0–24.5mm: Within 0.5D - fixed effects model – rank probability histograms.

Figure 24AL 22.0–24.5mm: Within 0.5D - fixed effects model – rank probability histograms

Table 40AL 22.0−24.5mm: Within 0.5D - fixed effects model – model fit statistics

Residual devianceDbarDhatpDDIC
12.8
(compared to 16 datapoints)		127.886116.88211.004138.89

Table 41AL 22.0−24.5mm: Within 0.5D - fixed effects model – notes

  • Dichotomous synchronic (binomial; logit link); fixed effects
  • 50000 burn-ins; 10000 recorded iterations
Figure 25. AL 22.0–24.5mm: Within 1.0D - fixed effects model – evidence network.

Figure 25AL 22.0–24.5mm: Within 1.0D - fixed effects model – evidence network

Table 42AL 22.0−24.5mm: Within 1.0D - fixed effects model – input data

SRKTBarrettUniversalIIHaigisHofferQHolladay2Holladay2-LTT2
Kane, J. et al. (2016)2477/26382485/26382453/26382451/26382480/26382467/2638
Ozcura et al. (2016)374/422374/422
Srivannaboon et al. (2013)114/124118/124118/124118/124
Aristodemou et al. (2011)1398/15081400/1508
Aristodemou et al. (2011)4430/46994432/4699

Table 43AL 22.0−24.5mm: Within 1.0D - fixed effects model – relative effectiveness of all pairwise combinations (RR and 95% credible interval)

RKTBarrettUniversalIIHaigisHofferQHolladay2Holladay2-LTT2
SRKT1.06
(0.84, 1.33)		0.86
(0.69, 1.07)		0.96
(0.86, 1.08)		1.02
(0.81, 1.28)		-0.94
(0.75, 1.17)
BarrettUniversalII1.12
(0.91, 1.38)		0.82
(0.65, 1.02)		0.81
(0.65, 1.01)		0.97
(0.77, 1.22)		-0.89
(0.71, 1.11)
Haigis0.90
(0.75, 1.10)		0.81
(0.65, 1.00)		1.01
(0.82, 1.24)		1.20
(0.97, 1.49)		1.73
(0.61, 4.90)		1.09
(0.88, 1.35)
HofferQ0.96
(0.86, 1.08)		0.86
(0.70, 1.05)		1.07
(0.88, 1.29)		1.19
(0.96, 1.48)		1.00
(0.31, 3.19)		1.10
(0.89, 1.36)
Holladay21.09
(0.89, 1.33)		0.97
(0.77, 1.22)		1.21
(0.97, 1.49)		1.13
(0.92, 1.38)		1.00
(0.31, 3.19)		0.92
(0.74, 1.15)
Holladay2-LT1.25
(0.51, 3.57)		1.12
(0.45, 3.25)		1.39
(0.57, 3.99)		1.30
(0.54, 3.70)		1.15
(0.47, 3.29)		-
T21.00
(0.82, 1.22)		0.89
(0.71, 1.11)		1.10
(0.89, 1.37)		1.03
(0.85, 1.27)		0.92
(0.73, 1.14)		0.79
(0.28, 1.97)
Figure 26. AL 22.0–24.5mm: Within 1.0D - fixed effects model – relative effect of all options versus common comparator.

Figure 26AL 22.0–24.5mm: Within 1.0D - fixed effects model – relative effect of all options versus common comparator

Table 44AL 22.0−24.5mm: Within 1.0D - fixed effects model – rankings for each comparator

Probability bestMedian rank (95%CI)
SRKT0.0174 (2, 7)
BarrettUniversalII0.2452 (1, 6)
Haigis0.0016 (3, 7)
HofferQ0.0035 (2, 7)
Holladay20.1413 (1, 6)
Holladay2-LT0.5621 (1, 7)
T20.0294 (1, 7)
Figure 27. AL 22.0–24.5mm: Within 1.0D - fixed effects model – rank probability histograms.

Figure 27AL 22.0–24.5mm: Within 1.0D - fixed effects model – rank probability histograms

Table 45AL 22.0−24.5mm: Within 1.0D - fixed effects model – model fit statistics

Residual devianceDbarDhatpDDIC
13.68
(compared to 16 datapoints)		108.8297.84110.979119.799

Table 46AL 22.0−24.5mm: Within 1.0D - fixed effects model – notes

  • Dichotomous synchronic (binomial; logit link); fixed effects
  • 50000 burn-ins; 10000 recorded iterations
Figure 28. AL 22.0–24.5mm: Within 2.0D - fixed effects model – evidence network.

Figure 28AL 22.0–24.5mm: Within 2.0D - fixed effects model – evidence network

Table 47AL 22.0−24.5mm: Within 2.0D - fixed effects model – input data

SRKTBarrettUniversalIIHaigisHofferQHolladay2T2
Kane, J. et al. (2016)2633/26382635/26382627/26382627/26382630/26382630/2638
Ozcura et al. (2016)420/422415/422

Table 48AL 22.0−24.5mm: Within 2.0D - fixed effects model – relative effectiveness of all pairwise combinations (RR and 95% credible interval)

SRKTBarrettUniversalIIHaigisHofferQHolladay2T2
SRKT1.67
(0.40, 6.99)		0.45
(0.16, 1.31)		0.39
(0.16, 0.93)		0.62
(0.20, 1.91)		0.62
(0.20, 1.91)
BarrettUniversalII1.57
(0.41, 7.93)		0.27
(0.08, 0.98)		0.27
(0.08, 0.98)		0.37
(0.10, 1.41)		0.37
(0.10, 1.41)
Haigis0.39
(0.14, 1.05)		0.25
(0.05, 0.83)		1.00
(0.43, 2.31)		1.38
(0.55, 3.43)		1.38
(0.55, 3.43)
HofferQ0.37
(0.14, 0.87)		0.24
(0.05, 0.77)		0.95
(0.41, 2.14)		1.38
(0.55, 3.43)		1.38
(0.55, 3.43)
Holladay20.54
(0.18, 1.59)		0.35
(0.07, 1.23)		1.39
(0.56, 3.66)		1.46
(0.61, 3.71)		1.00
(0.37, 2.67)
T20.55
(0.18, 1.59)		0.35
(0.07, 1.25)		1.39
(0.56, 3.70)		1.46
(0.61, 3.75)		1.00
(0.36, 2.75)
Figure 29. AL 22.0–24.5mm: Within 2.0D - fixed effects model – relative effect of all options versus common comparator.

Figure 29AL 22.0–24.5mm: Within 2.0D - fixed effects model – relative effect of all options versus common comparator

Table 49AL 22.0−24.5mm: Within 2.0D - fixed effects model – rankings for each comparator

Probability bestMedian rank (95%CI)
SRKT0.2422 (1, 4)
BarrettUniversalII0.7141 (1, 3)
Haigis0.0025 (3, 6)
HofferQ0.0015 (3, 6)
Holladay20.0204 (2, 6)
T20.0224 (2, 6)
Figure 30. AL 22.0–24.5mm: Within 2.0D - fixed effects model – rank probability histograms.

Figure 30AL 22.0–24.5mm: Within 2.0D - fixed effects model – rank probability histograms

Table 50AL 22.0−24.5mm: Within 2.0D - fixed effects model – model fit statistics

Residual devianceDbarDhatpDDIC
7.367
(compared to 8 datapoints)		36.629.7616.83943.439

Table 51AL 22.0−24.5mm: Within 2.0D - fixed effects model – notes

  • Dichotomous synchronic (binomial; logit link); fixed effects
  • 50000 burn-ins; 10000 recorded iterations

H.3.2.4. Full dataset: Axial length subgroup – 24.50 to 26.00mm

Figure 31. AL 24.5–26.0mm: Mean absolute error - fixed effects model – evidence network.

Figure 31AL 24.5–26.0mm: Mean absolute error - fixed effects model – evidence network

Table 52AL 24.5−26.0mm: Mean absolute error - fixed effects model – input data

HaigisHofferQHolladay2Holladay2-LT
Srivannaboon et al. (2013)0.39 (0.32)0.45 (0.35)0.38 (0.34)0.39 (0.33)

Table 53AL 24.5−26.0mm: Mean absolute error - fixed effects model – relative effectiveness of all pairwise combinations (MD and 95% credible interval)

HaigisHofferQHolladay2Holladay2-LT
Haigis0.06
(−0.13, 0.25)		−0.01
(−0.20, 0.18)		0.00
(−0.18, 0.18)
HofferQ0.06
(−0.13, 0.25)		−0.07
(−0.27, 0.13)		−0.06
(−0.25, 0.13)
Holladay2−0.01
(−0.20, 0.18)		−0.07
(−0.27, 0.13)		0.01
(−0.18, 0.20)
Holladay2-LT0.00
(−0.18, 0.18)		−0.06
(−0.25, 0.13)		0.01
(−0.18, 0.20)
Figure 32. AL 24.5–26.0mm: Mean absolute error - fixed effects model – relative effect of all options versus common comparator.

Figure 32AL 24.5–26.0mm: Mean absolute error - fixed effects model – relative effect of all options versus common comparator

Table 54AL 24.5−26.0mm: Mean absolute error - fixed effects model – rankings for each comparator

Probability bestMedian rank (95%CI)
Haigis0.2802 (1, 4)
HofferQ0.0854 (1, 4)
Holladay20.3492 (1, 4)
Holladay2-LT0.2862 (1, 4)
Figure 33. AL 24.5–26.0mm: Mean absolute error - fixed effects model – rank probability histograms.

Figure 33AL 24.5–26.0mm: Mean absolute error - fixed effects model – rank probability histograms

Table 55AL 24.5−26.0mm: Mean absolute error - fixed effects model – model fit statistics

Residual devianceDbarDhatpDDIC
3.991
(compared to 4 datapoints)		−10.123−14.1143.991−6.133

Table 56AL 24.5−26.0mm: Mean absolute error - fixed effects model – notes

  • Continuous (normal; identity link); fixed effects
  • 50000 burn-ins; 10000 recorded iterations
Figure 34. AL 24.5–26.0mm: Within 0.25D - fixed effects model – evidence network.

Figure 34AL 24.5–26.0mm: Within 0.25D - fixed effects model – evidence network

Table 57AL 24.5−26.0mm: Within 0.25D - fixed effects model – input data

SRKTBarrettUniversalIIHaigisHofferQHolladay2Holladay2-LTT2
Kane, J. et al. (2016)144/372172/372143/372130/372154/372147/372
Srivannaboon et al. (2013)12/2410/2414/2412/24
Aristodemou et al. (2011)105/234104/234
Aristodemou et al. (2011)272/712275/712

Table 58AL 24.5−26.0mm: Within 0.25D - fixed effects model – relative effectiveness of all pairwise combinations (RR and 95% credible interval)

SRKTBarrettUniversalIIHaigisHofferQHolladay2Holladay2-LTT2
SRKT1.36
(1.02, 1.82)		0.99
(0.74, 1.33)		0.96
(0.82, 1.13)		1.12
(0.83, 1.50)		-1.03
(0.77, 1.39)
BarrettUniversalII1.45
(1.12, 1.89)		0.73
(0.54, 0.97)		0.62
(0.47, 0.84)		0.82
(0.61, 1.10)		-0.76
(0.57, 1.02)
Haigis1.06
(0.81, 1.38)		0.73
(0.54, 0.97)		0.85
(0.64, 1.13)		1.15
(0.86, 1.52)		1.00
(0.32, 3.10)		1.05
(0.78, 1.40)
HofferQ0.96
(0.82, 1.12)		0.66
(0.51, 0.86)		0.91
(0.70, 1.18)		1.35
(1.01, 1.80)		1.40
(0.45, 4.38)		1.22
(0.90, 1.64)
Holladay21.21
(0.93, 1.57)		0.83
(0.63, 1.11)		1.15
(0.86, 1.52)		1.27
(0.98, 1.64)		0.71
(0.23, 2.23)		0.92
(0.69, 1.24)
Holladay2-LT1.07
(0.41, 2.76)		0.74
(0.28, 1.94)		1.01
(0.39, 2.61)		1.12
(0.43, 2.89)		0.88
(0.34, 2.30)		-
T21.10
(0.85, 1.44)		0.76
(0.57, 1.01)		1.05
(0.78, 1.40)		1.15
(0.88, 1.51)		0.91
(0.68, 1.22)		1.04
(0.39, 2.73)
Figure 35. AL 24.5–26.0mm: Within 0.25D - fixed effects model – relative effect of all options versus common comparator.

Figure 35AL 24.5–26.0mm: Within 0.25D - fixed effects model – relative effect of all options versus common comparator

Table 59AL 24.5−26.0mm: Within 0.25D - fixed effects model – rankings for each comparator

Probability bestMedian rank (95%CI)
SRKT0.0015 (3, 7)
BarrettUniversalII0.6581 (1, 3)
Haigis0.0054 (2, 7)
HofferQ0.0006 (3, 7)
Holladay20.0653 (1, 6)
Holladay2-LT0.2564 (1, 7)
T20.0154 (2, 7)
Figure 36. AL 24.5–26.0mm: Within 0.25D - fixed effects model – rank probability histograms.

Figure 36AL 24.5–26.0mm: Within 0.25D - fixed effects model – rank probability histograms

Table 60AL 24.5−26.0mm: Within 0.25D - fixed effects model – model fit statistics

Residual devianceDbarDhatpDDIC
11.55
(compared to 14 datapoints)		89.77179.75710.01499.785

Table 61AL 24.5−26.0mm: Within 0.25D - fixed effects model – notes

  • Dichotomous synchronic (binomial; logit link); fixed effects
  • 50000 burn-ins; 10000 recorded iterations
Figure 37. AL 24.5–26.0mm: Within 0.5D - fixed effects model – evidence network.

Figure 37AL 24.5–26.0mm: Within 0.5D - fixed effects model – evidence network

Table 62AL 24.5−26.0mm: Within 0.5D - fixed effects model – input data

SRKTAvgHofferQSRKTBarrettUniversalIIHaigisHofferQHolladay2Holladay2-LTT2
Kane, J. et al. (2016)248/372285/372255/372256/372250/372265/372
Srivannaboon et al. (2013)19/2414/2417/2414/24
Aristodemou et al. (2011)170/234173/234
Aristodemou et al. (2011)481/712473/712
Percival et al. (2002)20/2621/2620/26

Table 63AL 24.5−26.0mm: Within 0.5D - fixed effects model – relative effectiveness of all pairwise combinations (RR and 95% credible interval)

SRKTAvgHofferQSRKTBarrettUniversalIIHaigisHofferQHolladay2Holladay2-LTT2
SRKT1.26
(0.33, 4.79)		1.64
(1.19, 2.26)		1.09
(0.80, 1.48)		1.01
(0.86, 1.19)		1.02
(0.75, 1.39)		-1.24
(0.91, 1.69)
AvgHofferQSRKT1.30
(0.41, 4.73)		--0.79
(0.21, 3.02)		---
BarrettUniversalII1.59
(1.19, 2.15)		1.21
(0.33, 4.02)		0.67
(0.48, 0.92)		0.67
(0.49, 0.93)		0.63
(0.45, 0.86)		-0.76
(0.54, 1.05)
Haigis1.08
(0.82, 1.43)		0.83
(0.22, 2.75)		0.68
(0.49, 0.94)		0.95
(0.71, 1.29)		0.92
(0.68, 1.24)		0.37
(0.10, 1.32)		1.14
(0.83, 1.55)
HofferQ1.00
(0.85, 1.17)		0.77
(0.21, 2.45)		0.63
(0.47, 0.84)		0.92
(0.70, 1.22)		0.97
(0.72, 1.30)		1.00
(0.32, 3.15)		1.12
(0.82, 1.54)
Holladay21.00
(0.76, 1.31)		0.76
(0.21, 2.53)		0.63
(0.46, 0.86)		0.92
(0.68, 1.24)		1.00
(0.76, 1.31)		0.58
(0.17, 1.91)		1.21
(0.88, 1.65)
Holladay2-LT0.63
(0.24, 1.72)		0.48
(0.10, 2.21)		0.40
(0.15, 1.10)		0.58
(0.22, 1.58)		0.63
(0.24, 1.70)		0.63
(0.24, 1.71)		-
T21.20
(0.91, 1.60)		0.92
(0.25, 3.07)		0.76
(0.54, 1.05)		1.11
(0.81, 1.52)		1.20
(0.90, 1.60)		1.20
(0.88, 1.63)		1.90
(0.69, 5.16)
Figure 38. AL 24.5–26.0mm: Within 0.5D - fixed effects model – relative effect of all options versus common comparator.

Figure 38AL 24.5–26.0mm: Within 0.5D - fixed effects model – relative effect of all options versus common comparator

Table 64AL 24.5−26.0mm: Within 0.5D - fixed effects model – rankings for each comparator

Probability bestMedian rank (95%CI)
SRKT0.0006 (3, 8)
AvgHofferQSRKT0.3732 (1, 8)
BarrettUniversalII0.5741 (1, 3)
Haigis0.0034 (2, 7)
HofferQ0.0006 (3, 8)
Holladay20.0016 (2, 8)
Holladay2-LT0.0258 (2, 8)
T20.0253 (2, 7)
Figure 39. AL 24.5–26.0mm: Within 0.5D - fixed effects model – rank probability histograms.

Figure 39AL 24.5–26.0mm: Within 0.5D - fixed effects model – rank probability histograms

Table 65AL 24.5−26.0mm: Within 0.5D - fixed effects model – model fit statistics

Residual devianceDbarDhatpDDIC
15
(compared to 17 datapoints)		101.30389.28812.015113.319

Table 66AL 24.5−26.0mm: Within 0.5D - fixed effects model – notes

  • Dichotomous synchronic (binomial; logit link); fixed effects
  • 50000 burn-ins; 10000 recorded iterations
Figure 40. AL 24.5–26.0mm: Within 1.0D - fixed effects model – evidence network.

Figure 40AL 24.5–26.0mm: Within 1.0D - fixed effects model – evidence network

Table 67AL 24.5−26.0mm: Within 1.0D - fixed effects model – input data

SRKTAvgHofferQSRKTBarrettUniversalIIHaigisHofferQHolladay2Holladay2-LTT2
Kane, J. et al. (2016)351/372364/372349/372350/372348/372353/372
Srivannaboon et al. (2013)24/2422/2420/2420/24
Aristodemou et al. (2011)215/234224/234
Aristodemou et al. (2011)673/712672/712
Percival et al. (2002)26/2626/2626/26
Mitra et al. (2014)17/4319/43
El-Nafees et al. (2010)44/5344/53

Table 68AL 24.5−26.0mm: Within 1.0D - fixed effects model – relative effectiveness of all pairwise combinations (RR and 95% credible interval)

SRKTAvgHofferQSRKTBarrettUniversalIIHaigisHofferQHolladay2Holladay2-LTT2
SRKT1.00
(0.02, 52.29)		2.72
(1.19, 6.23)		0.93
(0.55, 1.57)		1.12
(0.82, 1.52)		0.87
(0.47, 1.59)		-1.11
(0.59, 2.10)
AvgHofferQSRKT1.58
(0.03, 769.90)		--1.00
(0.02, 52.29)		---
BarrettUniversalII3.03
(1.46, 7.12)		1.96
(0.00, 130.80)		0.33
(0.15, 0.76)		0.35
(0.15, 0.80)		0.32
(0.14, 0.72)		-0.41
(0.18, 0.94)
Haigis1.07
(0.67, 1.73)		0.67
(0.00, 42.56)		0.35
(0.15, 0.75)		0.95
(0.53, 1.71)		0.81
(0.46, 1.42)		0.09
(0.00, 1.83)		1.22
(0.66, 2.29)
HofferQ1.11
(0.82, 1.51)		0.71
(0.00, 43.61)		0.37
(0.16, 0.76)		1.04
(0.64, 1.71)		0.85
(0.48, 1.49)		0.45
(0.07, 2.76)		1.17
(0.62, 2.20)
Holladay20.88
(0.53, 1.46)		0.55
(0.00, 34.21)		0.29
(0.12, 0.61)		0.82
(0.48, 1.42)		0.79
(0.48, 1.32)		1.00
(0.22, 4.56)		1.28
(0.69, 2.38)
Holladay2-LT0.51
(0.14, 2.12)		0.32
(0.00, 24.96)		0.17
(0.04, 0.77)		0.48
(0.13, 1.94)		0.46
(0.13, 1.87)		0.59
(0.16, 2.35)		-
T21.21
(0.70, 2.18)		0.77
(0.00, 48.51)		0.40
(0.16, 0.90)		1.14
(0.63, 2.14)		1.09
(0.63, 1.99)		1.39
(0.76, 2.60)		2.37
(0.55, 9.43)
Figure 41. AL 24.5–26.0mm: Within 1.0D - fixed effects model – relative effect of all options versus common comparator.

Figure 41AL 24.5–26.0mm: Within 1.0D - fixed effects model – relative effect of all options versus common comparator

Table 69AL 24.5−26.0mm: Within 1.0D - fixed effects model – rankings for each comparator

Probability bestMedian rank (95%CI)
SRKT0.0005 (3, 8)
AvgHofferQSRKT0.3862 (1, 8)
BarrettUniversalII0.5981 (1, 2)
Haigis0.0015 (2, 7)
HofferQ0.0014 (2, 7)
Holladay20.0006 (3, 8)
Holladay2-LT0.0078 (2, 8)
T20.0073 (2, 7)
Figure 42. AL 24.5–26.0mm: Within 1.0D - fixed effects model – rank probability histograms.

Figure 42AL 24.5–26.0mm: Within 1.0D - fixed effects model – rank probability histograms

Table 70AL 24.5−26.0mm: Within 1.0D - fixed effects model – model fit statistics

Residual devianceDbarDhatpDDIC
20.9
(compared to 21 datapoints)		99.81386.22413.589113.402

Table 71AL 24.5−26.0mm: Within 1.0D - fixed effects model – notes

  • Dichotomous synchronic (binomial; logit link); fixed effects
  • 50000 burn-ins; 10000 recorded iterations
Figure 43. AL 24.5–26.0mm: Within 2.0D - fixed effects model – evidence network.

Figure 43AL 24.5–26.0mm: Within 2.0D - fixed effects model – evidence network

Table 72AL 24.5−26.0mm: Within 2.0D - fixed effects model – input data

SRKTBarrettUniversalIIHaigisHofferQHolladay2T2
Kane, J. et al. (2016)371/372372/372370/372370/372371/372371/372

Table 73AL 24.5−26.0mm: Within 2.0D - fixed effects model – relative effectiveness of all pairwise combinations (RR and 95% credible interval)

SRKTBarrettUniversalIIHaigisHofferQHolladay2T2
SRKT3.01
(0.12, 74.08)		0.50
(0.05, 5.52)		0.50
(0.05, 5.52)		1.00
(0.06, 16.05)		1.00
(0.06, 16.05)
BarrettUniversalII5.33
(0.18, 2510.00)		0.20
(0.01, 4.16)		0.20
(0.01, 4.16)		0.33
(0.01, 8.19)		0.33
(0.01, 8.19)
Haigis0.56
(0.04, 4.75)		0.11
(0.00, 2.01)		1.00
(0.14, 7.14)		2.01
(0.18, 22.21)		2.01
(0.18, 22.21)
HofferQ0.56
(0.04, 4.76)		0.11
(0.00, 1.99)		1.01
(0.14, 7.20)		2.01
(0.18, 22.21)		2.01
(0.18, 22.21)
Holladay21.04
(0.07, 15.34)		0.19
(0.00, 5.79)		1.83
(0.22, 24.22)		1.83
(0.22, 25.09)		1.00
(0.06, 16.05)
T21.03
(0.07, 15.35)		0.20
(0.00, 5.62)		1.83
(0.21, 24.40)		1.82
(0.21, 24.37)		1.00
(0.07, 14.99)
Figure 44. AL 24.5–26.0mm: Within 2.0D - fixed effects model – relative effect of all options versus common comparator.

Figure 44AL 24.5–26.0mm: Within 2.0D - fixed effects model – relative effect of all options versus common comparator

Table 74AL 24.5−26.0mm: Within 2.0D - fixed effects model – rankings for each comparator

Probability bestMedian rank (95%CI)
SRKT0.0973 (1, 6)
BarrettUniversalII0.6511 (1, 5)
Haigis0.0215 (2, 6)
HofferQ0.0215 (2, 6)
Holladay20.1043 (1, 6)
T20.1063 (1, 6)
Figure 45. AL 24.5–26.0mm: Within 2.0D - fixed effects model – rank probability histograms.

Figure 45AL 24.5–26.0mm: Within 2.0D - fixed effects model – rank probability histograms

Table 75AL 24.5−26.0mm: Within 2.0D - fixed effects model – model fit statistics

Residual devianceDbarDhatpDDIC
6.666
(compared to 6 datapoints)		20.79115.3875.40426.195

Table 76AL 24.5−26.0mm: Within 2.0D - fixed effects model – notes

  • Dichotomous synchronic (binomial; logit link); fixed effects
  • 50000 burn-ins; 10000 recorded iterations

H.3.2.5. Full dataset: Axial length subgroup – Greater than 26.00mm

Figure 46. AL >26.0mm: Mean absolute error - fixed effects model – evidence network.

Figure 46AL >26.0mm: Mean absolute error - fixed effects model – evidence network

Table 77AL >26.0mm: Mean absolute error - fixed effects model – input data

SRKTBarrettUniversalIIHaigisHofferQHolladay2Holladay2_NoPreSurgRefHolladay2_PreSurgRefLadasSuperFormulaOlsen_standaloneT2
Cooke & (2016)0.40 (0.45)0.30 (0.38)0.28 (0.37)0.43 (0.45)0.39 (0.41)0.41 (0.43)0.35 (0.40)0.29 (0.35)0.32 (0.40)
Bang et al. (2011)0.62 (0.77)0.52 (0.63)1.02 (0.88)0.81 (0.81)

Table 78AL >26.0mm: Mean absolute error - fixed effects model – relative effectiveness of all pairwise combinations (MD and 95% credible interval)

SRKTBarrettUniversalIIHaigisHofferQHolladay2Holladay2_NoPreSurgRefHolladay2_PreSurgRefLadasSuperFormulaOlsen_standaloneT2
SRKT−0.10
(−0.25, 0.06)		−0.11
(−0.25, 0.02)		0.11
(−0.04, 0.26)		0.19
(−0.11, 0.49)		−0.01
(−0.17, 0.15)		0.01
(−0.16, 0.17)		−0.05
(−0.21, 0.11)		−0.11
(−0.26, 0.04)		−0.08
(−0.24, 0.08)
BarrettUniversalII−0.07
(−0.22, 0.08)		−0.02
(−0.16, 0.12)		0.13
(−0.03, 0.28)		-0.09
(−0.06, 0.24)		0.10
(−0.05, 0.26)		0.05
(−0.10, 0.19)		−0.01
(−0.15, 0.12)		0.02
(−0.13, 0.16)
Haigis−0.11
(−0.25, 0.02)		−0.04
(−0.18, 0.09)		0.23
(0.09, 0.36)		0.29
(0.01, 0.57)		0.11
(−0.04, 0.26)		0.13
(−0.02, 0.28)		0.07
(−0.08, 0.21)		0.01
(−0.13, 0.15)		0.04
(−0.11, 0.18)
HofferQ0.11
(−0.04, 0.26)		0.18
(0.03, 0.33)		0.22
(0.09, 0.36)		−0.21
(−0.53, 0.11)		−0.04
(−0.20, 0.12)		−0.02
(−0.19, 0.14)		−0.08
(−0.24, 0.08)		−0.14
(−0.29, 0.01)		−0.11
(−0.27, 0.05)
Holladay20.12
(−0.15, 0.38)		0.19
(−0.09, 0.46)		0.23
(−0.02, 0.48)		0.01
(−0.26, 0.27)		-----
Holladay2_NoPreSurgRef0.02
(−0.14, 0.17)		0.09
(−0.06, 0.24)		0.13
(−0.01, 0.27)		−0.09
(−0.25, 0.06)		−0.10
(−0.38, 0.18)		0.02
(−0.14, 0.17)		−0.04
(−0.19, 0.11)		−0.10
(−0.24, 0.04)		−0.07
(−0.22, 0.08)
Holladay2_PreSurgRef0.03
(−0.13, 0.19)		0.10
(−0.05, 0.26)		0.15
(0.00, 0.29)		−0.08
(−0.24, 0.08)		−0.08
(−0.36, 0.20)		0.02
(−0.14, 0.17)		−0.06
(−0.22, 0.10)		−0.12
(−0.26, 0.03)		−0.09
(−0.24, 0.07)
LadasSuperFormula−0.02
(−0.18, 0.13)		0.05
(−0.10, 0.19)		0.09
(−0.05, 0.23)		−0.14
(−0.29, 0.02)		−0.14
(−0.42, 0.14)		−0.04
(−0.19, 0.11)		−0.06
(−0.21, 0.10)		−0.06
(−0.20, 0.08)		−0.03
(−0.18, 0.12)
Olsen_standalone−0.08
(−0.23, 0.06)		−0.01
(−0.15, 0.13)		0.03
(−0.10, 0.16)		−0.19
(−0.34, −0.05)		−0.20
(−0.47, 0.07)		−0.10
(−0.24, 0.04)		−0.12
(−0.26, 0.03)		−0.06
(−0.20, 0.09)		0.03
(−0.11, 0.17)
T2−0.05
(−0.21, 0.10)		0.02
(−0.13, 0.16)		0.06
(−0.08, 0.20)		−0.17
(−0.32, −0.01)		−0.17
(−0.45, 0.11)		−0.07
(−0.22, 0.08)		−0.09
(−0.25, 0.07)		−0.03
(−0.18, 0.12)		0.03
(−0.11, 0.17)
Figure 47. AL >26.0mm: Mean absolute error - fixed effects model – relative effect of all options versus common comparator.

Figure 47AL >26.0mm: Mean absolute error - fixed effects model – relative effect of all options versus common comparator

Table 79AL >26.0mm: Mean absolute error - fixed effects model – rankings for each comparator

Probability bestMedian rank (95%CI)
SRKT0.0126 (2, 9)
BarrettUniversalII0.1533 (1, 8)
Haigis0.4502 (1, 5)
HofferQ0.0009 (6, 10)
Holladay20.0259 (1, 10)
Holladay2_NoPreSurgRef0.0087 (2, 10)
Holladay2_PreSurgRef0.0047 (3, 10)
LadasSuperFormula0.0385 (1, 9)
Olsen_standalone0.2043 (1, 7)
T20.1054 (1, 8)
Figure 48. AL >26.0mm: Mean absolute error - fixed effects model – rank probability histograms.

Figure 48AL >26.0mm: Mean absolute error - fixed effects model – rank probability histograms

Table 80AL >26.0mm: Mean absolute error - fixed effects model – model fit statistics

Residual devianceDbarDhatpDDIC
16.31
(compared to 13 datapoints)		−30.087−41.11211.025−19.062

Table 81AL >26.0mm: Mean absolute error - fixed effects model – notes

  • Continuous (normal; identity link); fixed effects
  • 50000 burn-ins; 10000 recorded iterations
Figure 49. AL >26.0mm: Within 0.25D - fixed effects model – evidence network.

Figure 49AL >26.0mm: Within 0.25D - fixed effects model – evidence network

Table 82AL >26.0mm: Within 0.25D - fixed effects model – input data

SRKTBarrettUniversalIIHaigisHofferQHolladay2T2
Kane, J. et al. (2016)30/7726/7728/7726/7725/7724/77
Aristodemou et al. (2011)21/4718/47
Aristodemou et al. (2011)111/27196/271
Aristodemou et al. (2011)5/172/17

Table 83AL >26.0mm: Within 0.25D - fixed effects model – relative effectiveness of all pairwise combinations (RR and 95% credible interval)

SRKTBarrettUniversalIIHaigisHofferQHolladay2T2
SRKT0.80
(0.41, 1.54)		0.90
(0.47, 1.72)		0.77
(0.58, 1.02)		0.75
(0.39, 1.46)		0.71
(0.36, 1.38)
BarrettUniversalII0.78
(0.43, 1.41)		1.12
(0.58, 2.17)		1.00
(0.51, 1.95)		0.94
(0.48, 1.85)		0.89
(0.45, 1.74)
Haigis0.88
(0.48, 1.57)		1.12
(0.58, 2.21)		0.89
(0.46, 1.73)		0.84
(0.43, 1.64)		0.79
(0.41, 1.55)
HofferQ0.77
(0.58, 1.02)		0.98
(0.55, 1.80)		0.88
(0.49, 1.61)		0.94
(0.48, 1.85)		0.89
(0.45, 1.74)
Holladay20.74
(0.40, 1.32)		0.94
(0.48, 1.84)		0.84
(0.43, 1.63)		0.96
(0.52, 1.71)		0.94
(0.48, 1.86)
T20.69
(0.38, 1.25)		0.88
(0.45, 1.74)		0.79
(0.40, 1.55)		0.90
(0.49, 1.63)		0.94
(0.48, 1.85)
Figure 50. AL >26.0mm: Within 0.25D - fixed effects model – relative effect of all options versus common comparator.

Figure 50AL >26.0mm: Within 0.25D - fixed effects model – relative effect of all options versus common comparator

Table 84AL >26.0mm: Within 0.25D - fixed effects model – rankings for each comparator

Probability bestMedian rank (95%CI)
SRKT0.4902 (1, 4)
BarrettUniversalII0.1194 (1, 6)
Haigis0.2473 (1, 6)
HofferQ0.0114 (2, 6)
Holladay20.0794 (1, 6)
T20.0535 (1, 6)
Figure 51. AL >26.0mm: Within 0.25D - fixed effects model – rank probability histograms.

Figure 51AL >26.0mm: Within 0.25D - fixed effects model – rank probability histograms

Table 85AL >26.0mm: Within 0.25D - fixed effects model – model fit statistics

Residual devianceDbarDhatpDDIC
10.05
(compared to 12 datapoints)		64.40555.3889.01773.422

Table 86AL >26.0mm: Within 0.25D - fixed effects model – notes

  • Dichotomous synchronic (binomial; logit link); fixed effects
  • 50000 burn-ins; 10000 recorded iterations
Figure 52. AL >26.0mm: Within 0.5D - fixed effects model – evidence network.

Figure 52AL >26.0mm: Within 0.5D - fixed effects model – evidence network

Table 87AL >26.0mm: Within 0.5D - fixed effects model – input data

SRKTAvgHofferQSRKTBarrettUniversalIIHaigisHofferQHolladay2Holladay2_NoPreSurgRefHolladay2_PreSurgRefLadasSuperFormulaOlsen_standaloneT2
Cooke & (2016)41/5441/5444/5434/5437/5437/5441/5445/5444/54
Kane, J. et al. (2016)48/7748/7744/7741/7744/7749/77
Bang et al. (2011)27/5330/5318/5322/53
Aristodemou et al. (2011)37/4733/47
Aristodemou et al. (2011)197/271167/271
Aristodemou et al. (2011)10/173/17
Percival et al. (2002)16/2015/2012/20

Table 88AL >26.0mm: Within 0.5D - fixed effects model – relative effectiveness of all pairwise combinations (RR and 95% credible interval)

SRKTAvgHofferQSRKTBarrettUniversalIIHaigisHofferQHolladay2Holladay2_NoPreSurgRefHolladay2_PreSurgRefLadasSuperFormulaOlsen_standaloneT2
SRKT0.75
(0.17, 3.33)		1.00
(0.59, 1.69)		1.05
(0.68, 1.62)		0.57
(0.44, 0.73)		0.75
(0.46, 1.23)		0.69
(0.30, 1.61)		0.69
(0.30, 1.61)		1.00
(0.41, 2.42)		1.59
(0.61, 4.10)		1.16
(0.68, 1.98)
AvgHofferQSRKT0.98
(0.29, 3.72)		--0.50
(0.13, 1.93)		------
BarrettUniversalII0.97
(0.61, 1.54)		0.98
(0.25, 3.65)		0.96
(0.57, 1.63)		0.63
(0.38, 1.04)		0.81
(0.42, 1.54)		0.69
(0.30, 1.61)		0.69
(0.30, 1.61)		1.00
(0.41, 2.42)		1.59
(0.61, 4.10)		1.16
(0.68, 1.98)
Haigis1.03
(0.69, 1.53)		1.05
(0.26, 3.74)		1.07
(0.64, 1.74)		0.56
(0.36, 0.86)		0.78
(0.48, 1.27)		0.49
(0.20, 1.21)		0.49
(0.20, 1.21)		0.72
(0.28, 1.81)		1.14
(0.42, 3.06)		1.21
(0.70, 2.07)
HofferQ0.56
(0.43, 0.72)		0.57
(0.15, 1.95)		0.58
(0.37, 0.91)		0.54
(0.37, 0.80)		1.25
(0.76, 2.05)		1.28
(0.58, 2.84)		1.28
(0.58, 2.84)		1.86
(0.81, 4.27)		2.94
(1.19, 7.26)		1.85
(1.10, 3.10)
Holladay20.75
(0.49, 1.17)		0.76
(0.19, 2.79)		0.77
(0.45, 1.32)		0.73
(0.45, 1.17)		1.34
(0.87, 2.08)		----1.31
(0.69, 2.51)
Holladay2_NoPreSurgRef0.63
(0.33, 1.27)		0.65
(0.15, 2.62)		0.65
(0.32, 1.37)		0.61
(0.30, 1.27)		1.13
(0.58, 2.26)		0.84
(0.40, 1.82)		1.00
(0.44, 2.25)		1.45
(0.62, 3.38)		2.30
(0.92, 5.75)		2.02
(0.83, 4.95)
Holladay2_PreSurgRef0.63
(0.33, 1.28)		0.64
(0.15, 2.63)		0.65
(0.32, 1.38)		0.61
(0.31, 1.27)		1.13
(0.58, 2.26)		0.84
(0.40, 1.82)		1.00
(0.44, 2.29)		1.45
(0.62, 3.38)		2.30
(0.92, 5.75)		2.02
(0.83, 4.95)
LadasSuperFormula0.93
(0.46, 1.97)		0.95
(0.21, 3.93)		0.96
(0.45, 2.12)		0.90
(0.44, 1.95)		1.65
(0.82, 3.48)		1.23
(0.57, 2.80)		1.46
(0.62, 3.52)		1.46
(0.63, 3.50)		1.59
(0.61, 4.10)		1.40
(0.55, 3.53)
Olsen_standalone1.50
(0.69, 3.52)		1.53
(0.32, 6.72)		1.54
(0.68, 3.79)		1.45
(0.65, 3.49)		2.66
(1.23, 6.27)		1.99
(0.86, 4.99)		2.37
(0.94, 6.21)		2.36
(0.95, 6.21)		1.62
(0.62, 4.35)		0.88
(0.33, 2.37)
T21.13
(0.71, 1.81)		1.14
(0.28, 4.25)		1.16
(0.68, 1.99)		1.09
(0.66, 1.81)		2.01
(1.26, 3.22)		1.50
(0.87, 2.58)		1.78
(0.85, 3.66)		1.78
(0.84, 3.65)		1.22
(0.55, 2.58)		0.75
(0.31, 1.72)
Figure 53. AL >26.0mm: Within 0.5D - fixed effects model – relative effect of all options versus common comparator.

Figure 53AL >26.0mm: Within 0.5D - fixed effects model – relative effect of all options versus common comparator

Table 89AL >26.0mm: Within 0.5D - fixed effects model – rankings for each comparator

Probability bestMedian rank (95%CI)
SRKT0.0175 (2, 8)
AvgHofferQSRKT0.2315 (1, 11)
BarrettUniversalII0.0295 (1, 9)
Haigis0.0354 (1, 8)
HofferQ0.00010 (8, 11)
Holladay20.0028 (3, 11)
Holladay2_NoPreSurgRef0.0059 (3, 11)
Holladay2_PreSurgRef0.0049 (3, 11)
LadasSuperFormula0.0696 (1, 11)
Olsen_standalone0.5081 (1, 8)
T20.1003 (1, 8)
Figure 54. AL >26.0mm: Within 0.5D - fixed effects model – rank probability histograms.

Figure 54AL >26.0mm: Within 0.5D - fixed effects model – rank probability histograms

Table 90AL >26.0mm: Within 0.5D - fixed effects model – model fit statistics

Residual devianceDbarDhatpDDIC
24.45
(compared to 28 datapoints)		143.413126.31617.097160.51

Table 91AL >26.0mm: Within 0.5D - fixed effects model – notes

  • Dichotomous synchronic (binomial; logit link); fixed effects
  • 50000 burn-ins; 10000 recorded iterations
Figure 55. AL >26.0mm: Within 1.0D - random effects model – evidence network.

Figure 55AL >26.0mm: Within 1.0D - random effects model – evidence network

Table 92AL >26.0mm: Within 1.0D - random effects model – input data

SRKTAvgHofferQSRKTBarrettUniversalIIHaigisHofferQHolladay2Holladay2_NoPreSurgRefHolladay2_PreSurgRefLadasSuperFormulaOlsen_standaloneT2
Cooke & (2016)53/5453/5453/5452/5453/5453/5452/5453/5453/54
Kane, J. et al. (2016)71/7771/7768/7764/7768/7767/77
Bang et al. (2011)35/5339/5332/5333/53
Aristodemou et al. (2011)44/4743/47
Aristodemou et al. (2011)253/271239/271
Aristodemou et al. (2011)14/1712/17
Petermeier et al. (2009)50/5032/5050/50
Percival et al. (2002)19/2017/2017/20
Mitra et al. (2014)17/4319/43
El-Nafees et al. (2010)44/5344/53

Table 93AL >26.0mm: Within 1.0D - random effects model – relative effectiveness of all pairwise combinations (RR and 95% credible interval)

SRKTAvgHofferQSRKTBarrettUniversalIIHaigisHofferQHolladay2Holladay2_NoPreSurgRefHolladay2_PreSurgRefLadasSuperFormulaOlsen_standaloneT2
SRKT0.30
(0.03, 3.15)		1.00
(0.34, 2.96)		0.71
(0.28, 1.78)		0.65
(0.45, 0.92)		0.77
(0.40, 1.46)		1.00
(0.06, 16.41)		1.00
(0.06, 16.41)		0.49
(0.04, 5.58)		1.00
(0.06, 16.41)		0.61
(0.22, 1.65)
AvgHofferQSRKT0.46
(0.04, 5.92)		--1.00------
BarrettUniversalII1.05
(0.19, 6.08)		2.28
(0.12, 47.00)		0.68
(0.25, 1.86)		0.43
(0.17, 1.10)		0.64
(0.22, 1.89)		1.00
(0.06, 16.41)		1.00
(0.06, 16.41)		0.49
(0.04, 5.58)		1.00
(0.06, 16.41)		0.61
(0.22, 1.65)
Haigis0.48
(0.16, 1.36)		1.04
(0.07, 15.16)		0.46
(0.07, 2.53)		1.07
(0.30, 3.81)		0.74
(0.39, 1.38)		1.00
(0.06, 16.41)		1.00
(0.06, 16.41)		0.49
(0.04, 5.58)		1.00
(0.06, 16.41)		0.90
(0.36, 2.23)
HofferQ0.65
(0.27, 1.49)		1.40
(0.11, 17.29)		0.62
(0.11, 3.36)		1.35
(0.46, 4.25)		1.25
(0.69, 2.27)		2.04
(0.18, 23.17)		2.04
(0.18, 23.17)		1.00
(0.14, 7.37)		2.04
(0.18, 23.17)		1.43
(0.62, 3.30)
Holladay20.63
(0.15, 2.59)		1.35
(0.08, 23.34)		0.59
(0.08, 4.24)		1.31
(0.30, 6.02)		0.97
(0.23, 4.04)		----0.89
(0.34, 2.32)
Holladay2_NoPreSurgRef1.07
(0.07, 41.13)		2.48
(0.06, 174.90)		1.03
(0.05, 46.32)		2.24
(0.14, 88.12)		1.64
(0.11, 63.46)		1.75
(0.09, 77.21)		1.00
(0.06, 16.41)		0.49
(0.04, 5.58)		1.00
(0.06, 16.41)		1.00
(0.06, 16.41)
Holladay2_PreSurgRef1.08
(0.07, 46.67)		2.46
(0.06, 215.20)		1.03
(0.05, 49.49)		2.22
(0.14, 104.20)		1.66
(0.11, 72.32)		1.73
(0.09, 89.57)		1.01
(0.01, 77.45)		0.49
(0.04, 5.58)		1.00
(0.06, 16.41)		1.00
(0.06, 16.41)
LadasSuperFormula0.42
(0.03, 5.61)		0.91
(0.03, 32.60)		0.40
(0.03, 6.94)		0.87
(0.07, 12.61)		0.64
(0.06, 9.08)		0.68
(0.04, 11.28)		0.38
(0.01, 10.65)		0.38
(0.01, 11.08)		2.04
(0.18, 23.17)		2.04
(0.18, 23.17)
Olsen_standalone1.10
(0.07, 45.60)		2.43
(0.06, 207.10)		1.05
(0.05, 49.59)		2.28
(0.14, 97.44)		1.71
(0.10, 69.48)		1.79
(0.09, 83.45)		1.01
(0.02, 70.36)		0.98
(0.01, 70.90)		2.64
(0.09, 158.90)		1.00
(0.06, 16.41)
T20.68
(0.13, 3.61)		1.48
(0.08, 30.72)		0.64
(0.09, 4.94)		1.41
(0.28, 8.25)		1.05
(0.21, 5.74)		1.10
(0.17, 7.52)		0.63
(0.01, 12.17)		0.63
(0.01, 12.54)		1.62
(0.10, 23.86)		0.62
(0.01, 12.23)
Figure 56. AL >26.0mm: Within 1.0D - random effects model – relative effect of all options versus common comparator.

Figure 56AL >26.0mm: Within 1.0D - random effects model – relative effect of all options versus common comparator

Table 94AL >26.0mm: Within 1.0D - random effects model – rankings for each comparator

Probability bestMedian rank (95%CI)
SRKT0.0454 (1, 8)
AvgHofferQSRKT0.0668 (1, 11)
BarrettUniversalII0.1024 (1, 10)
Haigis0.0028 (3, 11)
HofferQ0.0047 (3, 10)
Holladay20.0227 (2, 11)
Holladay2_NoPreSurgRef0.2274 (1, 11)
Holladay2_PreSurgRef0.2294 (1, 11)
LadasSuperFormula0.0429 (1, 11)
Olsen_standalone0.2264 (1, 11)
T20.0356 (1, 11)
Figure 57. AL >26.0mm: Within 1.0D - random effects model – rank probability histograms.

Figure 57AL >26.0mm: Within 1.0D - random effects model – rank probability histograms

Table 95AL >26.0mm: Within 1.0D - random effects model – model fit statistics

Residual devianceDbarDhatpDDICtau
35.1
(compared to 35 datapoints)		147.479118.13929.34176.8190.974 (95%CI: 0.506, 1.724)

Table 96AL >26.0mm: Within 1.0D - random effects model – notes

  • Dichotomous synchronic (binomial; logit link); random effects
  • Prior distribution for between-study heterogeneity: uniform (Min=0; Max=2)
  • 50000 burn-ins; 10000 recorded iterations
Figure 58. AL >26.0mm: Within 2.0D - fixed effects model – evidence network.

Figure 58AL >26.0mm: Within 2.0D - fixed effects model – evidence network

Table 97AL >26.0mm: Within 2.0D - fixed effects model – input data

SRKTBarrettUniversalIIHaigisHofferQHolladay2T2
Kane, J. et al. (2016)75/7777/7776/7776/7775/7777/77
Bang et al. (2011)51/5352/5342/5350/53

Table 98AL >26.0mm: Within 2.0D - fixed effects model – relative effectiveness of all pairwise combinations (RR and 95% credible interval)

SRKTBarrettUniversalIIHaigisHofferQHolladay2T2
SRKT5.13
(0.24, 108.68)		2.03
(0.37, 11.30)		0.31
(0.10, 0.97)		0.79
(0.21, 3.03)		5.13
(0.24, 108.68)
BarrettUniversalII5.53
(0.33, 2725.00)		0.33
(0.01, 8.20)		0.33
(0.01, 8.20)		0.19
(0.01, 4.13)		1.00
(0.02, 51.04)
Haigis1.93
(0.39, 12.36)		0.36
(0.00, 7.38)		0.15
(0.03, 0.68)		0.39
(0.07, 2.05)		3.04
(0.12, 75.77)
HofferQ0.31
(0.09, 0.90)		0.06
(0.00, 0.76)		0.16
(0.03, 0.58)		2.61
(0.89, 7.67)		3.04
(0.12, 75.77)
Holladay20.79
(0.20, 3.02)		0.14
(0.00, 2.21)		0.41
(0.07, 1.85)		2.55
(0.92, 8.15)		5.13
(0.24, 108.68)
T25.32
(0.34, 2364.00)		0.98
(0.00, 635.90)		2.75
(0.13, 1348.00)		17.49
(1.34, 7933.00)		6.84
(0.44, 3162.00)
Figure 59. AL >26.0mm: Within 2.0D - fixed effects model – relative effect of all options versus common comparator.

Figure 59AL >26.0mm: Within 2.0D - fixed effects model – relative effect of all options versus common comparator

Table 99AL >26.0mm: Within 2.0D - fixed effects model – rankings for each comparator

Probability bestMedian rank (95%CI)
SRKT0.0144 (2, 5)
BarrettUniversalII0.4452 (1, 5)
Haigis0.0963 (1, 5)
HofferQ0.0006 (5, 6)
Holladay20.0055 (2, 6)
T20.4402 (1, 5)
Figure 60. AL >26.0mm: Within 2.0D - fixed effects model – rank probability histograms.

Figure 60AL >26.0mm: Within 2.0D - fixed effects model – rank probability histograms

Table 100AL >26.0mm: Within 2.0D - fixed effects model – model fit statistics

Residual devianceDbarDhatpDDIC
13.49
(compared to 10 datapoints)		38.13231.7976.33544.466

Table 101AL >26.0mm: Within 2.0D - fixed effects model – notes

  • Dichotomous synchronic (binomial; logit link); fixed effects
  • 50000 burn-ins; 10000 recorded iterations

H.3.3. Intraocular lens formulas: Network meta-analyses results: Eyes with a history of myopic LASIK/LASEK/PRK

H.3.3.1. Model fit statistics for all outcomes

Table 102Model fit statistics used to select fixed or random effect models for all comparisons and outcomes

StudiesOutcomeModelTotal model DICTotal model DIC (FE – RE)Total residual devianceNo. of data-pointsBetween-study SD (95% CrI)Preferred model
Full dataset: historical and no historical data methods
2 (Fam, Savini)Mean absolute errorFE9.83338.5427-RE
RE0.227.10.81 (0.24, 1.89)
5 (Fam, Huang, Kim, Savini, Xu)Prediction errorFE62.146.779.731-RE
RE15.3311.42 (0.72, 1.97)
5 (Fam, Huang, Kim, Saiki, Xu)Within 0.5DFE144.00.729.826-RE
RE143.327.40.96 (0.07, 1.93)
5 (Fam, Huang, Kim, Saiki, Xu)Within 1.0DFE151.75.834.726-RE
RE145.927.01.21 (0.29, 1.95)
1 (Kim – pairwise comparison)Within 1.5DFE-----FE
1 (Fam)Within 2.0DFE29.4-6.46-FE
1 (Kim – pairwise comparison)FE-----FE
No historical data methods only
4 (Huang, Kim, Saiki, Xu)Within 0.5DFE78.71.117.514-RE
RE77.614.90.94 (0.07, 1.93)
4 (Huang, Kim, Saiki, Xu)Within 1.0DFE86.06.122.414-RE
RE79.914.51.20 (0.30, 1.95)
Historical data methods only
2 (Fam, Savini)Mean absolute errorFE9.89.732.521-RE
RE0.1321.00.82 (0.24, 1.89)
2 (Fam, Saiki) – NB: network connector (SRKT DK) uses historical data in Fam but no historical data in SaikiWithin 0.5DFE60.0-11.411-FE
2 (Fam, Saiki) – NB: network connector (SRKT DK) uses historical data in Fam but no historical data in SaikiWithin 1.0DFE60.5-11.311-FE
1 (Fam)Within 2.0DFE29.4-6.46-FE

H.3.3.2. Full dataset: historical and no historical methods

Figure 61. Myopic corneal refractive surgery: Mean absolute error – evidence network.

Figure 61Myopic corneal refractive surgery: Mean absolute error – evidence network

Table 103Myopic corneal refractive surgery: Mean absolute error – input data

Clinical historyCamellin-CalossiHofferQ DKHolladay2 DKShammas-PLSRKT Clinical historySRKT DiehlSRKT DKSRKT DK AwwadSRKT DK SaviniSRKT DK Seitz/SpeicherSRKT DK Seitz/Speicher/SaviniSRKT DK ShammasSRKT Feiz-MannisSRKT Feiz-Mannis nomogramSRKT Ladas-StarkSRKT LatkanySRKT MasketSRKT SK FerraraSRKT SK Rosa
Savini et al. (2010)1.62
(1.25)		1.41
(0.76)		0.93
(0.48)		1.33
(1.03)		1.00
(0.57)		1.79
(1.13)		0.62
(0.52)		0.56
(0.45)		0.53
(0.46)		1.60
(0.98)		1.87
(1.44)		2.04
(1.48)		2.18
(1.52)		1.08
(0.86)		0.76
(0.49)		3.64
(1.45)		1.94
(1.01)
Fam & (2008)0.75
(0.52)		0.75
(0.62)		1.32
(0.73)		0.76
(0.60)		0.93
(0.83)		0.80
(0.63)

Table 104Myopic corneal refractive surgery: Mean absolute error – relative effectiveness of all pairwise combinations (MD and 95% credible interval)

Clinical historyCamellin-CalossiHofferQ DKHolladay2 DKShammas-PLSRKT Clinical historySRKT DiehlSRKT DKSRKT DK AwwadSRKT DK SaviniSRKT DK Seitz/SpeicherSRKT DK Seitz/Speicher/SaviniSRKT DK ShammasSRKT Feiz-MannisSRKT Feiz-Mannis nomogramSRKT Ladas-StarkSRKT LatkanySRKT MasketSRKT SK FerraraSRKT SK Rosa
Clinical history−0.21
(−0.75, 0.33)		--−0.69
(−1.19, −0.19)		-−0.29
(−0.89, 0.31)		−0.62
(−1.13, −0.11)		0.17
(−0.45, 0.79)		−1.00
(−1.50, −0.50)		−1.06
(−1.55, −0.57)		−1.09
(−1.58, −0.60)		−0.02
(−0.61, 0.57)		0.25
(−0.46, 0.96)		0.42
(−0.30, 1.14)		0.56
(−0.17, 1.29)		−0.54
(−1.10, 0.02)		−0.86
(−1.36, −0.36)		2.02
(1.31, 2.73)		0.32
(−0.28, 0.92)
Camellin-Calossi−0.20
(−5.27, 4.82)		--−0.48
(−0.81, −0.15)		-−0.08
(−0.55, 0.39)		−0.41
(−0.76, −0.06)		0.38
(−0.12, 0.88)		−0.79
(−1.13, −0.45)		−0.85
(−1.18, −0.52)		−0.88
(−1.21, −0.55)		0.19
(−0.27, 0.65)		0.46
(−0.14, 1.06)		0.63
(0.01, 1.25)		0.77
(0.14, 1.40)		−0.33
(−0.76, 0.10)		−0.65
(−0.98, −0.32)		2.23
(1.62, 2.84)		0.53
(0.06, 1.00)
HofferQ DK−0.06
(−5.75, 6.04)		0.14
(−5.72, 6.24)		0.00
(−0.26, 0.26)		-0.57
(0.28, 0.86)		-0.01
(−0.25, 0.27)		-----0.18
(−0.14, 0.50)		-0.05
(−0.21, 0.31)		----
Holladay2 DK−0.07
(−5.66, 5.96)		0.13
(−5.52, 6.08)		0.00
(−5.08, 5.04)		-0.57
(0.26, 0.88)		-0.01
(−0.27, 0.29)		-----0.18
(−0.15, 0.51)		-0.05
(−0.23, 0.33)		----
Shammas-PL−0.69
(−5.55, 4.63)		−0.48
(−5.36, 4.81)		−0.62
(−6.50, 5.44)		−0.62
(−6.57, 5.45)		-0.40
(−0.02, 0.82)		0.07
(−0.21, 0.35)		0.86
(0.41, 1.31)		−0.31
(−0.57, −0.05)		−0.37
(−0.61, −0.13)		−0.40
(−0.65, −0.15)		0.67
(0.27, 1.07)		0.94
(0.38, 1.50)		1.11
(0.53, 1.69)		1.25
(0.66, 1.84)		0.15
(−0.21, 0.51)		−0.17
(−0.42, 0.08)		2.71
(2.14, 3.28)		1.01
(0.60, 1.42)
SRKT Clinical history0.49
(−5.24, 6.52)		0.69
(−5.09, 6.67)		0.56
(−4.64, 5.56)		0.57
(−4.59, 5.72)		1.18
(−4.75, 7.02)		-−0.56
(−0.86, −0.26)		-----−0.39
(−0.75, −0.03)		-−0.52
(−0.83, −0.21)		----
SRKT Diehl−0.28
(−5.11, 4.98)		−0.08
(−5.10, 5.04)		−0.22
(−6.08, 5.43)		−0.21
(−6.11, 5.52)		0.41
(−4.81, 5.39)		−0.79
(−6.63, 4.93)		−0.33
(−0.77, 0.11)		0.46
(−0.11, 1.03)		−0.71
(−1.14, −0.28)		−0.77
(−1.19, −0.35)		−0.80
(−1.22, −0.38)		0.27
(−0.26, 0.80)		0.54
(−0.12, 1.20)		0.71
(0.04, 1.38)		0.85
(0.17, 1.53)		−0.25
(−0.75, 0.25)		−0.57
(−0.99, −0.15)		2.31
(1.65, 2.97)		0.61
(0.08, 1.14)
SRKT DK−0.34
(−4.78, 4.35)		−0.13
(−4.75, 4.46)		−0.27
(−4.96, 4.26)		−0.26
(−4.94, 4.28)		0.35
(−4.45, 4.87)		−0.84
(−5.45, 3.67)		−0.06
(−4.72, 4.51)		0.79
(0.32, 1.26)		−0.38
(−0.67, −0.09)		−0.44
(−0.71, −0.17)		−0.47
(−0.74, −0.20)		0.60
(0.18, 1.02)		0.48
(−0.20, 1.16)		1.04
(0.45, 1.63)		0.58
(−0.54, 1.69)		0.08
(−0.30, 0.46)		−0.24
(−0.52, 0.04)		2.64
(2.06, 3.22)		0.94
(0.51, 1.37)
SRKT DK Awwad0.16
(−4.64, 5.51)		0.37
(−4.53, 5.70)		0.23
(−5.66, 6.36)		0.23
(−5.65, 6.23)		0.85
(−4.26, 6.06)		−0.34
(−6.03, 5.68)		0.45
(−4.63, 5.68)		0.50
(−4.03, 5.37)		−1.17
(−1.63, −0.71)		−1.23
(−1.68, −0.78)		−1.26
(−1.71, −0.81)		−0.19
(−0.74, 0.36)		0.08
(−0.60, 0.76)		0.25
(−0.44, 0.94)		0.39
(−0.31, 1.09)		−0.71
(−1.24, −0.18)		−1.03
(−1.49, −0.57)		1.85
(1.17, 2.53)		0.15
(−0.41, 0.71)
SRKT DK Savini−1.01
(−5.76, 4.24)		−0.79
(−5.77, 4.31)		−0.93
(−6.95, 4.98)		−0.93
(−6.75, 4.98)		−0.31
(−5.57, 4.78)		−1.51
(−7.46, 4.35)		−0.72
(−5.93, 4.37)		−0.66
(−5.29, 4.06)		−1.17
(−6.33, 3.91)		−0.06
(−0.31, 0.19)		−0.09
(−0.35, 0.17)		0.98
(0.57, 1.39)		1.25
(0.68, 1.82)		1.42
(0.84, 2.00)		1.56
(0.96, 2.16)		0.46
(0.09, 0.83)		0.14
(−0.12, 0.40)		3.02
(2.45, 3.59)		1.32
(0.90, 1.74)
SRKT DK Seitz/Speicher−1.07
(−6.07, 4.11)		−0.86
(−5.94, 4.32)		−0.99
(−6.94, 4.78)		−0.99
(−6.95, 4.83)		−0.37
(−5.54, 4.71)		−1.56
(−7.36, 4.29)		−0.78
(−5.86, 4.31)		−0.72
(−5.41, 3.94)		−1.22
(−6.50, 3.77)		−0.06
(−5.23, 5.07)		−0.03
(−0.27, 0.21)		1.04
(0.64, 1.44)		1.31
(0.75, 1.87)		1.48
(0.91, 2.05)		1.62
(1.03, 2.21)		0.52
(0.16, 0.88)		0.20
(−0.05, 0.45)		3.08
(2.52, 3.64)		1.38
(0.97, 1.79)
SRKT DK Seitz/Speicher/Savini−1.09
(−5.98, 4.07)		−0.88
(−5.81, 4.25)		−1.01
(−6.86, 4.76)		−1.02
(−6.86, 4.77)		−0.40
(−5.64, 4.62)		−1.58
(−7.32, 4.29)		−0.81
(−5.89, 4.32)		−0.74
(−5.33, 3.95)		−1.25
(−6.41, 3.64)		−0.08
(−5.09, 5.02)		−0.02
(−4.99, 5.09)		1.07
(0.67, 1.47)		1.34
(0.78, 1.90)		1.51
(0.94, 2.08)		1.65
(1.06, 2.24)		0.55
(0.19, 0.91)		0.23
(−0.02, 0.48)		3.11
(2.55, 3.67)		1.41
(1.00, 1.82)
SRKT DK Shammas−0.02
(−4.87, 5.06)		0.18
(−4.81, 5.29)		0.05
(−5.84, 5.89)		0.05
(−5.80, 5.73)		0.66
(−4.41, 5.53)		−0.52
(−6.31, 5.21)		0.26
(−4.80, 5.28)		0.32
(−4.19, 4.86)		−0.18
(−5.41, 4.62)		0.97
(−4.14, 5.94)		1.05
(−4.10, 6.06)		1.07
(−4.03, 6.04)		0.27
(−0.38, 0.92)		0.44
(−0.22, 1.10)		0.58
(−0.09, 1.25)		−0.52
(−1.00, −0.04)		−0.84
(−1.25, −0.43)		2.04
(1.39, 2.69)		0.34
(−0.18, 0.86)
SRKT Feiz-Mannis0.15
(−4.37, 5.01)		0.37
(−4.18, 5.07)		0.23
(−4.41, 4.92)		0.25
(−4.42, 4.84)		0.86
(−3.89, 5.49)		−0.33
(−4.88, 4.30)		0.46
(−4.12, 5.00)		0.50
(−3.03, 4.17)		0.00
(−4.84, 4.52)		1.16
(−3.53, 5.86)		1.23
(−3.40, 5.81)		1.26
(−3.32, 5.86)		0.19
(−4.39, 4.81)		0.17
(−0.59, 0.93)		−0.05
(−0.38, 0.27)		−0.79
(−1.41, −0.17)		−1.11
(−1.67, −0.55)		1.77
(1.01, 2.53)		0.07
(−0.58, 0.72)
SRKT Feiz-Mannis nomogram0.42
(−4.33, 5.70)		0.63
(−4.36, 5.80)		0.49
(−5.33, 6.26)		0.49
(−5.33, 6.27)		1.11
(−4.03, 6.21)		−0.09
(−5.86, 5.73)		0.71
(−4.41, 5.78)		0.76
(−3.74, 5.37)		0.26
(−4.95, 5.27)		1.42
(−3.73, 6.61)		1.49
(−3.51, 6.52)		1.51
(−3.51, 6.53)		0.44
(−4.67, 5.70)		0.25
(−4.34, 4.97)		0.14
(−0.65, 0.93)		−0.96
(−1.59, −0.33)		−1.28
(−1.86, −0.70)		1.60
(0.83, 2.37)		−0.10
(−0.76, 0.56)
SRKT Ladas-Stark0.22
(−4.17, 5.10)		0.42
(−4.00, 5.17)		0.28
(−4.35, 4.96)		0.28
(−4.22, 4.92)		0.91
(−3.77, 5.64)		−0.29
(−4.81, 4.42)		0.50
(−4.14, 5.12)		0.56
(−3.01, 4.22)		0.05
(−4.68, 4.62)		1.22
(−3.35, 6.03)		1.28
(−3.20, 5.94)		1.30
(−3.29, 5.90)		0.23
(−4.35, 4.91)		0.05
(−3.43, 3.61)		−0.20
(−4.80, 4.46)		−1.10
(−1.75, −0.45)		−1.42
(−2.01, −0.83)		1.46
(0.68, 2.24)		−0.24
(−0.92, 0.44)
SRKT Latkany−0.54
(−5.45, 4.70)		−0.33
(−5.39, 4.96)		−0.48
(−6.28, 5.38)		−0.47
(−6.36, 5.33)		0.15
(−5.04, 5.26)		−1.05
(−6.79, 4.90)		−0.26
(−5.31, 4.84)		−0.20
(−4.78, 4.50)		−0.71
(−5.95, 4.30)		0.46
(−4.64, 5.58)		0.52
(−4.45, 5.72)		0.55
(−4.36, 5.59)		−0.52
(−5.56, 4.62)		−0.70
(−5.33, 3.98)		−0.96
(−6.06, 4.03)		−0.76
(−5.48, 3.98)		−0.32
(−0.69, 0.05)		2.56
(1.94, 3.18)		0.86
(0.37, 1.35)
SRKT Masket−0.85
(−5.63, 4.47)		−0.64
(−5.63, 4.64)		−0.79
(−6.55, 5.18)		−0.78
(−6.73, 5.25)		−0.17
(−5.32, 5.06)		−1.35
(−7.20, 4.68)		−0.57
(−5.48, 4.63)		−0.51
(−4.99, 4.31)		−1.01
(−6.22, 4.22)		0.15
(−4.70, 5.27)		0.21
(−4.79, 5.33)		0.25
(−4.75, 5.35)		−0.83
(−5.79, 4.36)		−1.02
(−5.63, 3.79)		−1.27
(−6.28, 3.94)		−1.06
(−5.65, 3.67)		−0.32
(−5.36, 4.90)		2.88
(2.31, 3.45)		1.18
(0.76, 1.60)
SRKT SK Ferrara2.01
(−2.79, 7.30)		2.21
(−2.73, 7.42)		2.08
(−3.74, 7.90)		2.09
(−3.64, 7.93)		2.71
(−2.43, 7.68)		1.51
(−4.14, 7.46)		2.30
(−2.82, 7.50)		2.35
(−2.19, 7.18)		1.86
(−3.29, 6.81)		3.01
(−1.95, 8.27)		3.07
(−1.90, 8.20)		3.10
(−1.84, 8.09)		2.03
(−2.83, 7.19)		1.85
(−2.62, 6.53)		1.59
(−3.48, 6.71)		1.80
(−2.85, 6.53)		2.55
(−2.38, 7.80)		2.86
(−2.16, 7.84)		−1.70
(−2.35, −1.05)
SRKT SK Rosa0.32
(−4.47, 5.49)		0.53
(−4.42, 5.75)		0.40
(−5.53, 6.31)		0.40
(−5.49, 6.19)		1.01
(−4.04, 6.02)		−0.17
(−5.93, 5.83)		0.61
(−4.44, 5.66)		0.67
(−3.90, 5.44)		0.16
(−5.09, 5.32)		1.32
(−3.66, 6.33)		1.39
(−3.69, 6.51)		1.41
(−3.58, 6.56)		0.34
(−4.64, 5.47)		0.16
(−4.43, 4.86)		−0.09
(−5.13, 5.03)		0.11
(−4.50, 4.68)		0.86
(−4.16, 5.96)		1.18
(−4.00, 6.21)		−1.69
(−6.78, 3.31)
Figure 62. Myopic corneal refractive surgery: Mean absolute error – relative effect of all options versus common comparator.

Figure 62Myopic corneal refractive surgery: Mean absolute error – relative effect of all options versus common comparator

Table 105Myopic corneal refractive surgery: Mean absolute error – rankings for each comparator

Probability bestMedian rank (95%CI)
Clinical history0.01912 (2, 19)
Camellin-Calossi0.02410 (2, 19)
HofferQ DK0.03511 (1, 20)
Holladay2 DK0.03511 (1, 20)
Shammas-PL0.0676 (1, 18)
SRKT Clinical history0.01716 (2, 20)
SRKT Diehl0.0309 (1, 19)
SRKT DK0.0119 (2, 17)
SRKT DK Awwad0.01413 (2, 20)
SRKT DK Savini0.1574 (1, 18)
SRKT DK Seitz/Speicher0.1864 (1, 17)
SRKT DK Seitz/Speicher/Savini0.2004 (1, 17)
SRKT DK Shammas0.01812 (2, 19)
SRKT Feiz-Mannis0.00413 (4, 19)
SRKT Feiz-Mannis nomogram0.01015 (3, 20)
SRKT Ladas-Stark0.00414 (4, 19)
SRKT Latkany0.0517 (1, 19)
SRKT Masket0.1065 (1, 18)
SRKT SK Ferrara0.00220 (7, 20)
SRKT SK Rosa0.01215 (2, 20)
Figure 63. Myopic corneal refractive surgery: Mean absolute error – rank probability histograms.

Figure 63Myopic corneal refractive surgery: Mean absolute error – rank probability histograms

Table 106Myopic corneal refractive surgery: Mean absolute error – model fit statistics

Residual devianceDbarDhatpDDICtau
23.18
(compared to 23 datapoints)		−23.582−46.6523.068−0.5131.053 (95%CI: 0.250, 7.282)

Table 107Myopic corneal refractive surgery: Mean absolute error – notes

  • Continuous (normal; identity link); random effects
  • Prior distribution for between-study heterogeneity: uniform (Min=0; Max=10)
  • 50000 burn-ins; 10000 recorded iterations
Figure 64. Myopic corneal refractive surgery: prediction error – evidence network.

Figure 64Myopic corneal refractive surgery: prediction error – evidence network

Table 108Myopic corneal refractive surgery: prediction error – input data

Clinical historyCamellin-CalossiHaigis-LHofferQ DKHofferQ KHofferQ TNPHolladay2 DKShammas-PLSRKT Clinical historySRKT DiehlSRKT DKSRKT DK AwwadSRKT DK SaviniSRKT DK Seitz/SpeicherSRKT DK Seitz/Speicher/SaviniSRKT DK ShammasSRKT Feiz-MannisSRKT Feiz-Mannis nomogramSRKT KSRKT Ladas-StarkSRKT LatkanySRKT MasketSRKT SK FerraraSRKT SK RosaSRKT TNP
Xu et al. (2014)1.58
(1.20)		−2.30
(1.25)		1.64
(0.93)		−1.79
(1.11)
Huang et al. (2013)0.14
(0.83)		0.24
(0.82)
Kim et al. (2013)0.03
(1.06)		1.68
(1.34)
Savini et al. (2010)1.08
(1.75)		1.37
(0.83)		0.50
(0.94)		0.83
(1.48)		−0.88
(0.75)		1.73
(1.23)		0.21
(0.79)		0.05
(0.73)		0.09
(0.70)		1.60
(0.98)		1.37
(1.94)		2.00
(1.53)		1.83
(1.95)		0.80
(1.13)		−0.27
(0.88)		3.64
(1.45)		1.90
(1.10)
Fam & (2008)0.19
(0.90)		−0.04
(0.98)		1.15
(0.99)		−0.19
(0.95)		−0.51
(1.15)		−0.01
(1.02)

Table 109Myopic corneal refractive surgery: prediction error – relative effectiveness of all pairwise combinations (MD and 95% credible interval)

Clinical historyCamellin-CalossiHaigis-LHofferQ DKHofferQ KHofferQ TNPHolladay2 DKShammas-PLSRKT Clinical historySRKT DiehlSRKT DKSRKT DK AwwadSRKT DK SaviniSRKT DK Seitz/SpeicherSRKT DK Seitz/Speicher/SaviniSRKT DK ShammasSRKT Feiz-MannisSRKT Feiz-Mannis nomogramSRKT KSRKT Ladas-StarkSRKT LatkanySRKT MasketSRKT SK FerraraSRKT SK RosaSRKT TNP
Clinical history0.29
(−0.43, 1.01)		-----−0.58
(−1.32, 0.16)		-−0.25
(−1.10, 0.60)		−1.96
(−2.67, −1.25)		0.65
(−0.14, 1.44)		−0.87
(−1.58, −0.16)		−1.03
(−1.73, −0.33)		−0.99
(−1.69, −0.29)		0.52
(−0.22, 1.26)		0.29
(−0.68, 1.26)		0.92
(0.06, 1.78)		-0.75
(−0.22, 1.72)		−0.28
(−1.05, 0.49)		−1.35
(−2.08, −0.62)		2.56
(1.72, 3.40)		0.82
(0.05, 1.59)		-
Camellin-Calossi0.30
(−2.67, 3.28)		-----−0.87
(−1.33, −0.41)		-−0.54
(−1.17, 0.09)		−2.25
(−2.66, −1.84)		0.36
(−0.19, 0.91)		−1.16
(−1.58, −0.74)		−1.32
(−1.73, −0.91)		−1.28
(−1.68, −0.88)		0.23
(−0.25, 0.71)		0.00
(−0.78, 0.78)		0.63
(−0.01, 1.27)		-0.46
(−0.32, 1.24)		−0.57
(−1.09, −0.05)		−1.64
(−2.09, −1.19)		2.27
(1.65, 2.89)		0.53
(0.02, 1.04)		-
Haigis-L−0.66
(−4.85, 3.58)		−0.94
(−5.14, 3.22)		----0.10
(−0.24, 0.44)		----------1.65
(1.16, 2.14)		------
HofferQ DK0.06
(−3.35, 3.57)		−0.24
(−3.64, 3.25)		0.71
(−3.80, 5.18)		--−0.23
(−0.66, 0.20)		-0.96
(0.53, 1.39)		-−0.38
(−0.80, 0.04)		-----−0.70
(−1.17, −0.23)		--−0.20
(−0.64, 0.24)		-----
HofferQ K0.95
(−5.07, 6.96)		0.65
(−5.28, 6.55)		1.59
(−2.61, 5.80)		0.87
(−5.31, 7.03)		−3.88
(−4.44, −3.32)		------------0.06
(−0.43, 0.55)		-----−3.37
(−3.90, −2.84)
HofferQ TNP−2.93
(−8.90, 3.01)		−3.24
(−9.18, 2.68)		−2.29
(−6.49, 1.92)		−3.01
(−9.14, 3.14)		−3.88
(−6.82, −0.93)		------------3.94
(3.44, 4.44)		-----0.51
(−0.03, 1.05)
Holladay2 DK−0.16
(−3.60, 3.33)		−0.45
(−3.84, 3.05)		0.50
(−4.03, 5.02)		−0.21
(−3.15, 2.76)		−1.12
(−7.27, 5.15)		2.79
(−3.37, 8.96)		-1.19
(0.74, 1.64)		-−0.15
(−0.59, 0.29)		-----−0.47
(−0.96, 0.02)		--0.03
(−0.43, 0.49)		-----
Shammas-PL−0.56
(−3.54, 2.49)		−0.86
(−3.82, 2.11)		0.08
(−2.88, 3.02)		−0.63
(−4.01, 2.77)		−1.51
(−6.65, 3.68)		2.36
(−2.79, 7.54)		−0.41
(−3.83, 3.01)		-0.33
(−0.32, 0.98)		−1.38
(−1.83, −0.93)		1.23
(0.66, 1.80)		−0.29
(−0.74, 0.16)		−0.45
(−0.89, −0.01)		−0.41
(−0.84, 0.02)		1.10
(0.60, 1.60)		0.87
(0.07, 1.67)		1.50
(0.83, 2.17)		-1.33
(0.53, 2.13)		0.30
(−0.24, 0.84)		−0.77
(−1.25, −0.29)		3.14
(2.50, 3.78)		1.40
(0.86, 1.94)		-
SRKT Clinical history1.03
(−2.38, 4.57)		0.74
(−2.63, 4.18)		1.68
(−2.86, 6.23)		0.97
(−1.94, 3.88)		0.10
(−6.02, 6.33)		3.96
(−2.18, 10.18)		1.19
(−1.77, 4.13)		1.60
(−1.81, 5.02)		-−1.34
(−1.78, −0.90)		-----−1.66
(−2.15, −1.17)		--−1.16
(−1.62, −0.70)		-----
SRKT Diehl−0.24
(−3.26, 2.79)		−0.54
(−3.53, 2.45)		0.41
(−3.77, 4.64)		−0.32
(−3.78, 3.12)		−1.18
(−7.09, 4.82)		2.67
(−3.21, 8.67)		−0.10
(−3.56, 3.30)		0.33
(−2.69, 3.26)		−1.27
(−4.76, 2.10)		−1.71
(−2.32, −1.10)		0.90
(0.19, 1.61)		−0.62
(−1.24, 0.00)		−0.78
(−1.39, −0.17)		−0.74
(−1.35, −0.13)		0.77
(0.11, 1.43)		0.54
(−0.36, 1.44)		1.17
(0.38, 1.96)		-1.00
(0.09, 1.91)		−0.03
(−0.72, 0.66)		−1.10
(−1.74, −0.46)		2.81
(2.04, 3.58)		1.07
(0.39, 1.75)		-
SRKT DK−1.13
(−3.84, 1.61)		−1.43
(−4.14, 1.31)		−0.47
(−4.53, 3.52)		−1.19
(−3.91, 1.49)		−2.06
(−7.87, 3.85)		1.80
(−4.00, 7.71)		−0.98
(−3.69, 1.67)		−0.56
(−3.28, 2.12)		−2.16
(−4.87, 0.52)		−0.89
(−3.61, 1.84)		2.61
(2.08, 3.14)		1.09
(0.69, 1.49)		0.93
(0.54, 1.32)		0.97
(0.59, 1.35)		2.48
(2.02, 2.94)		0.95
(−1.57, 3.46)		2.88
(2.25, 3.51)		-1.42
(−1.05, 3.90)		1.68
(1.18, 2.18)		0.61
(0.18, 1.04)		4.52
(3.92, 5.12)		2.78
(2.29, 3.27)		-
SRKT DK Awwad0.66
(−2.35, 3.71)		0.36
(−2.63, 3.36)		1.30
(−2.90, 5.50)		0.59
(−2.87, 4.04)		−0.29
(−6.28, 5.70)		3.56
(−2.36, 9.53)		0.81
(−2.63, 4.22)		1.24
(−1.76, 4.18)		−0.39
(−3.85, 3.08)		0.90
(−2.12, 3.90)		1.79
(−0.93, 4.50)		−1.52
(−2.06, −0.98)		−1.68
(−2.21, −1.15)		−1.64
(−2.16, −1.12)		−0.13
(−0.71, 0.45)		−0.36
(−1.21, 0.49)		0.27
(−0.46, 1.00)		-0.10
(−0.75, 0.95)		−0.93
(−1.55, −0.31)		−2.00
(−2.56, −1.44)		1.91
(1.21, 2.61)		0.17
(−0.44, 0.78)		-
SRKT DK Savini−0.85
(−3.87, 2.19)		−1.16
(−4.14, 1.79)		−0.21
(−4.46, 3.98)		−0.93
(−4.39, 2.47)		−1.81
(−7.78, 4.14)		2.06
(−3.86, 8.10)		−0.70
(−4.20, 2.68)		−0.29
(−3.28, 2.65)		−1.89
(−5.35, 1.49)		−0.62
(−3.63, 2.39)		0.26
(−2.46, 2.94)		−1.51
(−4.48, 1.49)		−0.16
(−0.56, 0.24)		−0.12
(−0.51, 0.27)		1.39
(0.92, 1.86)		1.16
(0.38, 1.94)		1.79
(1.15, 2.43)		-1.62
(0.84, 2.40)		0.59
(0.08, 1.10)		−0.48
(−0.92, −0.04)		3.43
(2.82, 4.04)		1.69
(1.19, 2.19)		-
SRKT DK Seitz/Speicher−1.02
(−4.00, 1.97)		−1.32
(−4.31, 1.64)		−0.38
(−4.59, 3.81)		−1.08
(−4.56, 2.28)		−1.97
(−7.93, 4.00)		1.92
(−4.05, 7.89)		−0.87
(−4.36, 2.51)		−0.46
(−3.43, 2.47)		−2.05
(−5.51, 1.27)		−0.77
(−3.76, 2.20)		0.11
(−2.63, 2.76)		−1.68
(−4.67, 1.28)		−0.15
(−3.08, 2.80)		0.04
(−0.33, 0.41)		1.55
(1.10, 2.00)		1.32
(0.55, 2.09)		1.95
(1.32, 2.58)		-1.78
(1.01, 2.55)		0.75
(0.25, 1.25)		−0.32
(−0.74, 0.10)		3.59
(2.99, 4.19)		1.85
(1.36, 2.34)		-
SRKT DK Seitz/Speicher/Savini−0.99
(−3.98, 2.02)		−1.28
(−4.26, 1.70)		−0.35
(−4.53, 3.84)		−1.04
(−4.57, 2.36)		−1.93
(−7.83, 4.04)		1.95
(−3.99, 7.96)		−0.83
(−4.28, 2.53)		−0.42
(−3.38, 2.54)		−2.01
(−5.47, 1.39)		−0.74
(−3.73, 2.24)		0.14
(−2.60, 2.84)		−1.64
(−4.67, 1.32)		−0.12
(−3.08, 2.85)		0.04
(−2.91, 3.02)		1.51
(1.06, 1.96)		1.28
(0.52, 2.04)		1.91
(1.29, 2.53)		-1.74
(0.97, 2.51)		0.71
(0.22, 1.20)		−0.36
(−0.78, 0.06)		3.55
(2.95, 4.15)		1.81
(1.33, 2.29)		-
SRKT DK Shammas0.53
(−2.44, 3.54)		0.22
(−2.71, 3.18)		1.16
(−3.01, 5.37)		0.46
(−3.04, 3.81)		−0.44
(−6.39, 5.58)		3.43
(−2.46, 9.42)		0.68
(−2.79, 4.08)		1.08
(−1.87, 4.05)		−0.51
(−3.98, 2.89)		0.77
(−2.20, 3.73)		1.66
(−1.05, 4.37)		−0.13
(−3.13, 2.83)		1.39
(−1.54, 4.33)		1.55
(−1.39, 4.50)		1.50
(−1.48, 4.51)		−0.23
(−1.04, 0.58)		0.40
(−0.27, 1.07)		-0.23
(−0.58, 1.04)		−0.80
(−1.35, −0.25)		−1.87
(−2.36, −1.38)		2.04
(1.39, 2.69)		0.30
(−0.25, 0.85)		-
SRKT Feiz-Mannis−0.20
(−2.89, 2.58)		−0.49
(−3.20, 2.25)		0.47
(−3.62, 4.48)		−0.25
(−2.97, 2.42)		−1.14
(−6.93, 4.77)		2.74
(−3.06, 8.65)		−0.04
(−2.75, 2.65)		0.37
(−2.36, 3.10)		−1.22
(−3.96, 1.45)		0.06
(−2.68, 2.82)		0.93
(−1.16, 3.05)		−0.85
(−3.58, 1.90)		0.67
(−2.04, 3.43)		0.83
(−1.87, 3.60)		0.80
(−1.92, 3.51)		−0.71
(−3.42, 2.02)		0.63
(−0.29, 1.55)		-0.49
(0.05, 0.94)		−0.57
(−1.40, 0.26)		−1.64
(−2.43, −0.85)		2.27
(1.37, 3.17)		0.53
(−0.30, 1.36)		-
SRKT Feiz-Mannis nomogram0.93
(−2.11, 3.98)		0.64
(−2.37, 3.63)		1.59
(−2.66, 5.82)		0.87
(−2.58, 4.27)		−0.01
(−6.01, 5.96)		3.86
(−2.11, 9.87)		1.10
(−2.45, 4.49)		1.51
(−1.52, 4.49)		−0.09
(−3.57, 3.35)		1.18
(−1.84, 4.22)		2.08
(−0.65, 4.80)		0.28
(−2.77, 3.28)		1.80
(−1.19, 4.81)		1.96
(−1.01, 4.94)		1.92
(−1.04, 4.90)		0.42
(−2.58, 3.41)		1.13
(−1.66, 3.87)		-−0.17
(−1.09, 0.75)		−1.20
(−1.90, −0.50)		−2.27
(−2.92, −1.62)		1.64
(0.86, 2.42)		−0.10
(−0.80, 0.60)		-
SRKT K1.00
(−4.17, 6.16)		0.71
(−4.40, 5.83)		1.65
(−1.29, 4.59)		0.93
(−4.43, 6.35)		0.06
(−2.87, 3.01)		3.94
(0.95, 6.94)		1.13
(−4.30, 6.54)		1.59
(−2.59, 5.75)		−0.03
(−5.41, 5.42)		1.25
(−3.89, 6.40)		2.14
(−2.89, 7.10)		0.33
(−4.73, 5.52)		1.87
(−3.27, 7.06)		2.03
(−3.12, 7.17)		1.99
(−3.15, 7.08)		0.49
(−4.72, 5.65)		1.19
(−3.79, 6.22)		0.08
(−5.15, 5.21)		-----−3.43
(−3.90, −2.96)
SRKT Ladas-Stark0.28
(−2.47, 3.08)		−0.02
(−2.72, 2.76)		0.92
(−3.09, 4.96)		0.23
(−2.51, 2.94)		−0.67
(−6.47, 5.18)		3.20
(−2.56, 9.13)		0.44
(−2.28, 3.14)		0.86
(−1.86, 3.58)		−0.76
(−3.46, 1.94)		0.53
(−2.23, 3.29)		1.42
(−0.70, 3.51)		−0.37
(−3.15, 2.38)		1.13
(−1.57, 3.90)		1.30
(−1.39, 4.07)		1.27
(−1.41, 4.04)		−0.23
(−2.95, 2.51)		0.47
(−1.67, 2.61)		−0.66
(−3.39, 2.15)		−0.71
(−5.72, 4.26)		−1.03
(−1.86, −0.20)		−2.10
(−2.89, −1.31)		1.81
(0.91, 2.71)		0.07
(−0.76, 0.90)		-
SRKT Latkany−0.26
(−3.27, 2.69)		−0.57
(−3.51, 2.40)		0.38
(−3.88, 4.57)		−0.34
(−3.72, 3.05)		−1.21
(−7.16, 4.82)		2.66
(−3.28, 8.69)		−0.12
(−3.55, 3.25)		0.31
(−2.66, 3.28)		−1.31
(−4.72, 2.07)		−0.03
(−3.01, 2.96)		0.87
(−1.85, 3.54)		−0.93
(−3.91, 2.05)		0.59
(−2.39, 3.57)		0.75
(−2.17, 3.72)		0.73
(−2.23, 3.67)		−0.78
(−3.74, 2.16)		−0.08
(−2.83, 2.63)		−1.21
(−4.19, 1.74)		−1.27
(−6.37, 3.91)		−0.54
(−3.31, 2.14)		−1.07
(−1.60, −0.54)		2.84
(2.16, 3.52)		1.10
(0.52, 1.68)		-
SRKT Masket−1.35
(−4.33, 1.66)		−1.64
(−4.59, 1.27)		−0.71
(−4.88, 3.49)		−1.41
(−4.87, 2.01)		−2.29
(−8.23, 3.68)		1.58
(−4.37, 7.59)		−1.20
(−4.63, 2.14)		−0.77
(−3.79, 2.12)		−2.37
(−5.82, 1.02)		−1.09
(−4.05, 1.86)		−0.22
(−2.94, 2.47)		−2.00
(−5.03, 0.98)		−0.48
(−3.47, 2.48)		−0.33
(−3.26, 2.65)		−0.36
(−3.32, 2.58)		−1.87
(−4.84, 1.08)		−1.16
(−3.88, 1.53)		−2.28
(−5.29, 0.65)		−2.35
(−7.42, 2.75)		−1.63
(−4.38, 1.06)		−1.08
(−4.01, 1.84)		3.91
(3.28, 4.54)		2.17
(1.65, 2.69)		-
SRKT SK Ferrara2.56
(−0.48, 5.57)		2.27
(−0.74, 5.29)		3.20
(−1.05, 7.41)		2.49
(−0.99, 5.92)		1.61
(−4.35, 7.61)		5.49
(−0.50, 11.47)		2.71
(−0.78, 6.13)		3.13
(0.08, 6.12)		1.52
(−2.02, 5.01)		2.80
(−0.20, 5.87)		3.69
(0.92, 6.42)		1.91
(−1.14, 4.93)		3.43
(0.46, 6.42)		3.59
(0.57, 6.62)		3.55
(0.50, 6.52)		2.04
(−0.95, 5.04)		2.76
(−0.02, 5.46)		1.62
(−1.39, 4.65)		1.56
(−3.64, 6.71)		2.28
(−0.56, 5.04)		2.83
(−0.17, 5.80)		3.91
(0.93, 6.89)		−1.74
(−2.41, −1.07)		-
SRKT SK Rosa0.82
(−2.17, 3.82)		0.53
(−2.43, 3.48)		1.46
(−2.68, 5.67)		0.77
(−2.71, 4.16)		−0.12
(−5.99, 5.79)		3.76
(−2.19, 9.77)		0.98
(−2.51, 4.39)		1.39
(−1.59, 4.35)		−0.22
(−3.65, 3.22)		1.07
(−1.92, 4.08)		1.96
(−0.78, 4.65)		0.16
(−2.83, 3.20)		1.68
(−1.28, 4.66)		1.84
(−1.12, 4.87)		1.82
(−1.14, 4.78)		0.30
(−2.66, 3.27)		1.02
(−1.72, 3.75)		−0.11
(−3.07, 2.91)		−0.17
(−5.26, 4.97)		0.54
(−2.22, 3.29)		1.09
(−1.89, 4.09)		2.18
(−0.78, 5.13)		−1.74
(−4.71, 1.27)		-
SRKT TNP−2.43
(−8.33, 3.58)		−2.73
(−8.57, 3.24)		−1.78
(−5.94, 2.38)		−2.51
(−8.63, 3.68)		−3.37
(−6.34, −0.37)		0.49
(−2.50, 3.50)		−2.30
(−8.45, 3.81)		−1.87
(−6.94, 3.31)		−3.45
(−9.62, 2.67)		−2.20
(−8.18, 3.74)		−1.28
(−7.14, 4.48)		−3.10
(−9.01, 2.83)		−1.56
(−7.54, 4.42)		−1.40
(−7.32, 4.50)		−1.43
(−7.37, 4.45)		−2.94
(−8.88, 2.98)		−2.25
(−8.09, 3.59)		−3.36
(−9.26, 2.55)		−3.44
(−6.38, −0.45)		−2.71
(−8.52, 3.09)		−2.15
(−8.04, 3.79)		−1.09
(−6.95, 4.84)		−4.98
(−10.99, 1.03)		−3.25
(−9.16, 2.62)
Figure 65. Myopic corneal refractive surgery: prediction error – relative effect of all options versus common comparator.

Figure 65Myopic corneal refractive surgery: prediction error – relative effect of all options versus common comparator

Table 110Myopic corneal refractive surgery: prediction error – rankings for each comparator

Probability bestMedian rank (95%CI)
Clinical history0.00914 (3, 24)
Camellin-Calossi0.00416 (4, 24)
Haigis-L0.0149 (2, 23)
HofferQ DK0.01314 (2, 24)
HofferQ K0.00120 (3, 25)
HofferQ TNP0.4232 (1, 22)
Holladay2 DK0.01813 (2, 24)
Shammas-PL0.00710 (3, 21)
SRKT Clinical history0.00220 (5, 25)
SRKT Diehl0.01312 (2, 23)
SRKT DK0.0397 (1, 17)
SRKT DK Awwad0.00218 (5, 25)
SRKT DK Savini0.0388 (1, 21)
SRKT DK Seitz/Speicher0.0487 (1, 20)
SRKT DK Seitz/Speicher/Savini0.0467 (1, 20)
SRKT DK Shammas0.00217 (5, 24)
SRKT Feiz-Mannis0.00313 (4, 22)
SRKT Feiz-Mannis nomogram0.00120 (6, 25)
SRKT K0.00020 (4, 25)
SRKT Ladas-Stark0.00116 (5, 23)
SRKT Latkany0.01212 (2, 23)
SRKT Masket0.0846 (1, 19)
SRKT SK Ferrara0.00024 (15, 25)
SRKT SK Rosa0.00119 (6, 25)
SRKT TNP0.2182 (1, 22)
Figure 66. Myopic corneal refractive surgery: prediction error – rank probability histograms.

Figure 66Myopic corneal refractive surgery: prediction error – rank probability histograms

Table 111Myopic corneal refractive surgery: prediction error – model fit statistics

Residual devianceDbarDhatpDDICtau
31
(compared to 31 datapoints)		−15.563−46.42430.8615.2971.416 (95%CI: 0.720, 1.966)

Table 112Myopic corneal refractive surgery: prediction error – notes

  • Continuous (normal; identity link); random effects
  • Prior distribution for between-study heterogeneity: uniform (Min=0; Max=2)
  • 50000 burn-ins; 10000 recorded iterations (thinned from 500000)
Figure 67. Myopic corneal refractive surgery: within 0.5D – evidence network.

Figure 67Myopic corneal refractive surgery: within 0.5D – evidence network

Table 113Myopic corneal refractive surgery: within 0.5D – input data

SRKT Clinical historyBESStCamellin-CalossiDouble-KFeiz-MannisHaigis-LHofferQ DKHofferQ KHofferQ TNPHolladay2 DKMasketModified MasketShammas-PLSRKT A-PSRKT C-PSRKT DKSRKT Feiz-MannisSRKT KSRKT Ladas-StarkSRKT TNP
Xu et al. (2014)6/373/374/373/37
Huang et al. (2013)21/4621/46
Kim et al. (2013)30/475/47
Saiki et al. (2013)3/289/194/121/126/254/125/127/2813/2812/255/285/28
Fam & (2008)5/3713/3717/3719/3715/3717/37

Table 114Myopic corneal refractive surgery: within 0.5D – relative effectiveness of all pairwise combinations (RR and 95% credible interval)

SRKT Clinical historyBESStCamellin-CalossiDouble-KFeiz-MannisHaigis-LHofferQ DKHofferQ KHofferQ TNPHolladay2 DKMasketModified MasketShammas-PLSRKT A-PSRKT C-PSRKT DKSRKT Feiz-MannisSRKT KSRKT Ladas-StarkSRKT TNP
SRKT Clinical history-----3.47
(1.09, 11.05)		--5.44
(1.73, 17.06)		-----6.76
(2.16, 21.16)		4.36
(1.38, 13.76)		-5.44
(1.73, 17.06)		-
BESSt3.88
(0.08, 180.60)		7.50
(1.68, 33.56)		4.17
(0.76, 22.71)		0.76
(0.07, 8.12)		2.63
(0.58, 11.90)		----4.17
(0.76, 22.71)		5.95
(1.13, 31.26)		2.78
(0.64, 12.10)		7.22
(1.76, 29.56)		7.69
(1.84, 32.20)		1.81
(0.39, 8.44)		---1.81
(0.39, 8.44)
CamellinCalossi32.89
(0.76, 1490.00)		8.63
(0.54, 149.10)		0.56
(0.12, 2.49)		0.10
(0.01, 0.95)		0.35
(0.10, 1.27)		----0.56
(0.12, 2.49)		0.79
(0.18, 3.41)		0.37
(0.11, 1.28)		0.96
(0.30, 3.09)		1.03
(0.31, 3.39)		0.24
(0.06, 0.91)		---0.24
(0.06, 0.91)
Double-K17.55
(0.36, 820.50)		4.53
(0.25, 90.35)		0.53
(0.03, 8.48)		0.18
(0.02, 1.95)		0.63
(0.14, 2.86)		----1.00
(0.18, 5.46)		1.43
(0.27, 7.52)		0.67
(0.15, 2.91)		1.73
(0.42, 7.11)		1.85
(0.44, 7.74)		0.43
(0.09, 2.03)		---0.43
(0.09, 2.03)
Feiz-Mannis2.20
(0.02, 145.80)		0.59
(0.01, 15.97)		0.07
(0.00, 1.64)		0.13
(0.00, 3.46)		3.47
(0.37, 32.74)		----5.50
(0.51, 59.01)		7.86
(0.75, 82.13)		3.67
(0.40, 33.71)		9.53
(1.08, 84.14)		10.15
(1.13, 90.94)		2.39
(0.25, 23.01)		---2.39
(0.25, 23.01)
Haigis-L15.45
(0.44, 630.20)		4.00
(0.33, 61.03)		0.47
(0.04, 5.51)		0.88
(0.07, 11.93)		6.75
(0.37, 328.20)		----1.58
(0.35, 7.17)		2.26
(0.52, 9.83)		1.02
(0.51, 2.02)		2.74
(0.84, 8.94)		2.92
(0.87, 9.78)		0.69
(0.18, 2.61)		-0.07
(0.02, 0.20)		-0.69
(0.18, 2.61)
HofferQ DK3.68
(0.26, 50.80)		0.97
(0.02, 44.53)		0.11
(0.00, 4.34)		0.22
(0.00, 10.12)		1.68
(0.03, 184.60)		0.24
(0.01, 7.79)		--1.57
(0.62, 4.00)		-----1.95
(0.77, 4.96)		1.26
(0.49, 3.23)		-1.57
(0.62, 4.00)		-
HofferQ K6.00
(0.08, 518.00)		1.56
(0.05, 63.77)		0.18
(0.01, 6.20)		0.34
(0.01, 13.15)		2.78
(0.06, 261.40)		0.39
(0.02, 7.74)		1.60
(0.02, 125.70)		0.46
(0.10, 1.98)		--------0.63
(0.16, 2.43)		-0.46
(0.10, 1.98)
HofferQ TNP2.53
(0.03, 198.90)		0.66
(0.02, 27.58)		0.08
(0.00, 2.82)		0.14
(0.00, 6.06)		1.15
(0.02, 119.30)		0.16
(0.01, 3.52)		0.67
(0.01, 57.33)		0.42
(0.03, 6.24)		--------1.37
(0.29, 6.61)		-1.00
(0.19, 5.31)
Holladay2 DK5.83
(0.42, 82.42)		1.56
(0.04, 72.66)		0.18
(0.00, 6.71)		0.34
(0.01, 15.19)		2.71
(0.04, 294.60)		0.38
(0.01, 12.49)		1.58
(0.13, 20.93)		0.99
(0.01, 69.46)		2.34
(0.03, 182.90)		-----1.24
(0.50, 3.09)		0.80
(0.32, 2.02)		-1.00
(0.40, 2.50)		-
Masket17.47
(0.38, 814.40)		4.51
(0.26, 87.62)		0.52
(0.03, 8.83)		0.99
(0.06, 18.32)		7.79
(0.31, 467.40)		1.12
(0.08, 13.95)		4.64
(0.11, 208.30)		2.88
(0.08, 95.93)		6.88
(0.17, 270.30)		2.94
(0.07, 128.20)		1.43
(0.27, 7.52)		0.67
(0.15, 2.91)		1.73
(0.42, 7.11)		1.85
(0.44, 7.74)		0.43
(0.09, 2.03)		---0.43
(0.09, 2.03)
Modified Masket25.83
(0.58, 1224.00)		6.73
(0.37, 128.50)		0.78
(0.05, 12.25)		1.47
(0.09, 26.82)		11.47
(0.46, 681.20)		1.67
(0.13, 20.40)		6.85
(0.17, 301.80)		4.29
(0.12, 134.10)		10.23
(0.27, 362.90)		4.31
(0.11, 201.20)		1.49
(0.08, 25.81)		0.47
(0.11, 1.95)		1.21
(0.31, 4.76)		1.29
(0.32, 5.19)		0.30
(0.07, 1.37)		---0.30
(0.07, 1.37)
Shammas-PL13.19
(0.36, 515.30)		3.45
(0.26, 50.02)		0.40
(0.03, 4.76)		0.76
(0.06, 10.23)		5.86
(0.31, 286.60)		0.86
(0.14, 4.99)		3.56
(0.10, 133.40)		2.20
(0.09, 52.58)		5.28
(0.18, 150.60)		2.24
(0.07, 83.46)		0.76
(0.06, 10.92)		0.51
(0.04, 7.03)		2.60
(0.84, 8.07)		2.77
(0.87, 8.84)		0.65
(0.18, 2.37)		---0.65
(0.18, 2.37)
SRKT A-P32.41
(0.79, 1340.00)		8.36
(0.54, 141.30)		0.96
(0.07, 13.33)		1.84
(0.12, 30.37)		14.38
(0.66, 757.50)		2.08
(0.19, 22.65)		8.58
(0.22, 341.20)		5.34
(0.16, 159.30)		12.68
(0.37, 466.30)		5.39
(0.14, 214.50)		1.84
(0.12, 30.10)		1.24
(0.08, 19.41)		2.39
(0.21, 26.76)		1.07
(0.36, 3.14)		0.25
(0.07, 0.85)		---0.25
(0.07, 0.85)
SRKT C-P34.02
(0.83, 1408.00)		8.83
(0.58, 145.40)		1.03
(0.08, 14.14)		1.95
(0.13, 30.74)		15.09
(0.68, 831.20)		2.21
(0.19, 23.11)		9.07
(0.24, 373.40)		5.70
(0.18, 163.20)		13.73
(0.38, 458.20)		5.69
(0.16, 230.60)		1.97
(0.13, 30.91)		1.32
(0.09, 19.92)		2.55
(0.23, 29.17)		1.06
(0.08, 14.37)		0.24
(0.07, 0.82)		---0.24
(0.07, 0.82)
SRKT DK7.35
(0.53, 104.30)		1.94
(0.12, 35.33)		0.23
(0.01, 3.24)		0.42
(0.03, 7.05)		3.31
(0.14, 190.00)		0.48
(0.04, 5.34)		1.97
(0.16, 26.69)		1.23
(0.04, 37.44)		2.94
(0.08, 104.20)		1.25
(0.10, 16.39)		0.43
(0.03, 6.89)		0.29
(0.02, 4.42)		0.56
(0.04, 6.64)		0.23
(0.02, 3.10)		0.22
(0.02, 2.97)		0.65
(0.26, 1.62)		-0.81
(0.32, 2.01)		1.00
(0.25, 3.93)
SRKT FeizMannis4.67
(0.34, 67.63)		1.22
(0.03, 56.19)		0.14
(0.00, 5.68)		0.27
(0.01, 11.82)		2.11
(0.04, 236.80)		0.30
(0.01, 10.31)		1.27
(0.10, 16.86)		0.79
(0.01, 54.08)		1.86
(0.02, 151.40)		0.80
(0.06, 10.32)		0.27
(0.01, 11.89)		0.18
(0.00, 7.91)		0.35
(0.01, 12.94)		0.15
(0.00, 5.68)		0.14
(0.00, 5.32)		0.63
(0.05, 8.18)		-1.25
(0.50, 3.13)		-
SRKT K2.04
(0.04, 124.20)		0.53
(0.03, 14.55)		0.06
(0.00, 1.34)		0.12
(0.01, 2.90)		0.94
(0.03, 66.34)		0.13
(0.02, 1.27)		0.54
(0.01, 34.83)		0.34
(0.03, 5.01)		0.81
(0.06, 14.05)		0.34
(0.01, 20.26)		0.12
(0.01, 2.82)		0.08
(0.00, 1.86)		0.15
(0.01, 2.31)		0.06
(0.00, 1.34)		0.06
(0.00, 1.32)		0.27
(0.02, 6.20)		0.43
(0.01, 25.76)		-0.73
(0.15, 3.50)
SRKT Ladas-Stark5.86
(0.42, 84.94)		1.57
(0.03, 69.48)		0.18
(0.00, 6.77)		0.34
(0.01, 14.92)		2.68
(0.05, 284.60)		0.39
(0.01, 12.06)		1.59
(0.13, 20.83)		0.98
(0.01, 65.53)		2.35
(0.03, 186.80)		1.00
(0.08, 12.46)		0.34
(0.01, 14.34)		0.23
(0.01, 9.70)		0.45
(0.01, 15.39)		0.19
(0.00, 6.86)		0.17
(0.00, 6.28)		0.80
(0.06, 9.94)		1.26
(0.10, 15.77)		2.97
(0.05, 124.20)		-
SRKT TNP4.50
(0.11, 187.50)		1.17
(0.08, 19.22)		0.14
(0.01, 1.87)		0.26
(0.02, 3.98)		2.01
(0.09, 107.30)		0.29
(0.04, 2.33)		1.21
(0.03, 49.20)		0.76
(0.05, 9.75)		1.77
(0.11, 28.02)		0.76
(0.02, 30.60)		0.26
(0.02, 3.90)		0.17
(0.01, 2.59)		0.34
(0.03, 3.39)		0.14
(0.01, 1.76)		0.13
(0.01, 1.74)		0.61
(0.04, 8.43)		0.96
(0.03, 38.36)		2.19
(0.22, 18.56)		0.75
(0.02, 29.15)
Figure 68. Myopic corneal refractive surgery: within 0.5D – relative effect of all options versus common comparator.

Figure 68Myopic corneal refractive surgery: within 0.5D – relative effect of all options versus common comparator

Table 115Myopic corneal refractive surgery: within 0.5D rankings for each comparator

Probability bestMedian rank (95%CI)
SRKT Clinical history0.00119 (8, 20)
BESSt0.00214 (5, 20)
Camellin-Calossi0.1993 (1, 14)
Double-K0.0696 (1, 17)
Feiz-Mannis0.00216 (5, 20)
Haigis-L0.0177 (2, 15)
HofferQ DK0.01114 (3, 20)
HofferQ K0.02812 (1, 19)
HofferQ TNP0.00816 (3, 20)
Holladay2 DK0.02412 (2, 19)
Masket0.0706 (1, 17)
Modified Masket0.1425 (1, 15)
Shammas-PL0.0138 (2, 16)
SRKT A-P0.1714 (1, 13)
SRKT C-P0.1993 (1, 13)
SRKT DK0.00211 (4, 17)
SRKT Feiz-Mannis0.01813 (2, 19)
SRKT K0.00117 (7, 20)
SRKT Ladas-Stark0.02412 (2, 19)
SRKT TNP0.00113 (5, 19)
Figure 69. Myopic corneal refractive surgery: within 0.5D – rank probability histograms.

Figure 69Myopic corneal refractive surgery: within 0.5D – rank probability histograms

Table 116Myopic corneal refractive surgery: within 0.5D – model fit statistics

Residual devianceDbarDhatpDDICtau
27.42
(compared to 26 datapoints)		117.47291.60725.865143.3370.955 (95%CI: 0.065, 1.925)

Table 117Myopic corneal refractive surgery: within 0.5D – notes

  • Dichotomous synchronic (binomial; logit link); random effects
  • Prior distribution for between-study heterogeneity: uniform (Min=0; Max=2)
  • 50000 burn-ins; 10000 recorded iterations (thinned from 500000)
Figure 70. Myopic corneal refractive surgery: within 1.0D – evidence network.

Figure 70Myopic corneal refractive surgery: within 1.0D – evidence network

Table 118Myopic corneal refractive surgery: within 1.0D – input data

SRKT Clinical historyBESStCamellin-CalossiDouble-KFeiz-MannisHaigis-LHofferQ DKHofferQ KHofferQ TNPHolladay2 DKMasketModified MasketShammas-PLSRKT A-PSRKT C-PSRKT DKSRKT Feiz-MannisSRKT KSRKT Ladas-StarkSRKT TNP
Xu et al. (2014)14/373/378/375/37
Huang et al. (2013)36/4639/46
Kim et al. (2013)38/4716/47
Saiki et al. (2013)12/2814/198/126/1213/2510/129/1220/2821/2817/2514/2817/28
Fam & (2008)11/3728/3730/3725/3723/3723/37

Table 119Myopic corneal refractive surgery: within 1.0D – relative effectiveness of all pairwise combinations (RR and 95% credible interval)

SRKT Clinical historyBESStCamellin-CalossiDouble-KFeiz-MannisHaigis-LHofferQ DKHofferQ KHofferQ TNPHolladay2 DKMasketModified MasketShammas-PLSRKT A-PSRKT C-PSRKT DKSRKT Feiz-MannisSRKT KSRKT Ladas-StarkSRKT TNP
SRKT Clinical history-----7.35
(2.63, 20.60)		--10.13
(3.43, 29.93)		-----4.92
(1.84, 13.19)		3.88
(1.47, 10.23)		-3.88
(1.47, 10.23)		-
BESSt3.81
(0.06, 227.60)		3.73
(1.05, 13.24)		2.67
(0.65, 10.97)		1.33
(0.34, 5.18)		1.44
(0.49, 4.27)		----6.67
(1.23, 36.23)		4.00
(0.89, 18.03)		3.33
(1.10, 10.12)		4.00
(1.28, 12.46)		2.83
(0.92, 8.73)		1.33
(0.47, 3.82)		---2.06
(0.71, 5.98)
Camellin-Calossi15.54
(0.26, 888.10)		4.05
(0.21, 78.19)		0.71
(0.15, 3.45)		0.36
(0.08, 1.64)		0.39
(0.11, 1.40)		----1.79
(0.29, 11.13)		1.07
(0.20, 5.63)		0.89
(0.24, 3.31)		1.07
(0.28, 4.06)		0.76
(0.20, 2.85)		0.36
(0.10, 1.26)		---0.55
(0.15, 1.97)
Double-K10.98
(0.17, 731.40)		2.87
(0.15, 59.91)		0.70
(0.03, 15.96)		0.50
(0.10, 2.60)		0.54
(0.13, 2.27)		----2.50
(0.36, 17.32)		1.50
(0.25, 8.84)		1.25
(0.29, 5.35)		1.50
(0.34, 6.55)		1.06
(0.25, 4.60)		0.50
(0.12, 2.05)		---0.77
(0.19, 3.20)
Feiz-Mannis5.22
(0.08, 311.00)		1.37
(0.07, 27.48)		0.34
(0.02, 7.02)		0.48
(0.02, 10.31)		1.08
(0.27, 4.29)		----5.00
(0.75, 33.21)		3.00
(0.53, 16.90)		2.50
(0.62, 10.11)		3.00
(0.73, 12.39)		2.13
(0.52, 8.70)		1.00
(0.26, 3.87)		---1.55
(0.40, 6.03)
Haigis-L9.30
(0.19, 449.10)		2.43
(0.18, 33.65)		0.60
(0.04, 8.86)		0.85
(0.05, 12.83)		1.79
(0.12, 26.69)		----4.62
(0.84, 25.49)		2.77
(0.60, 12.71)		1.87
(0.86, 4.06)		2.77
(0.87, 8.84)		1.96
(0.62, 6.19)		0.92
(0.31, 2.72)		-0.12
(0.05, 0.31)		-1.43
(0.48, 4.25)
HofferQ DK7.79
(0.44, 134.90)		2.06
(0.03, 120.40)		0.51
(0.01, 30.84)		0.71
(0.01, 45.22)		1.50
(0.02, 99.39)		0.84
(0.02, 42.06)		--1.38
(0.45, 4.20)		-----0.67
(0.24, 1.85)		0.53
(0.19, 1.44)		-0.53
(0.19, 1.44)		-
HofferQ K11.10
(0.11, 1221.00)		2.94
(0.08, 119.80)		0.73
(0.02, 30.60)		1.02
(0.02, 42.35)		2.15
(0.05, 94.42)		1.20
(0.05, 27.15)		1.44
(0.01, 159.70)		0.14
(0.04, 0.56)		--------0.45
(0.16, 1.27)		-0.26
(0.08, 0.81)
HofferQ TNP1.41
(0.01, 169.20)		0.36
(0.01, 17.53)		0.09
(0.00, 4.53)		0.13
(0.00, 6.43)		0.27
(0.01, 13.24)		0.15
(0.01, 4.11)		0.18
(0.00, 21.37)		0.13
(0.01, 2.40)		--------3.13
(0.76, 12.89)		-1.77
(0.39, 8.02)
Holladay2 DK10.95
(0.63, 198.80)		2.91
(0.05, 163.90)		0.70
(0.01, 43.02)		1.00
(0.01, 66.08)		2.09
(0.03, 126.40)		1.18
(0.02, 56.26)		1.41
(0.08, 24.39)		0.98
(0.01, 101.70)		7.78
(0.06, 909.80)		-----0.49
(0.17, 1.42)		0.38
(0.13, 1.10)		-0.38
(0.13, 1.10)		-
Masket32.02
(0.46, 2442.00)		8.29
(0.35, 219.80)		2.03
(0.08, 61.95)		2.90
(0.11, 84.96)		6.01
(0.23, 186.40)		3.37
(0.19, 75.67)		4.05
(0.06, 306.10)		2.82
(0.06, 150.30)		22.62
(0.40, 1487.00)		2.95
(0.04, 216.80)		0.60
(0.08, 4.45)		0.50
(0.09, 2.81)		0.60
(0.11, 3.43)		0.43
(0.07, 2.41)		0.20
(0.04, 1.08)		---0.31
(0.06, 1.69)
Modified Masket17.37
(0.27, 1161.00)		4.50
(0.22, 100.40)		1.12
(0.05, 27.63)		1.59
(0.06, 40.92)		3.31
(0.15, 87.14)		1.85
(0.11, 34.73)		2.21
(0.03, 143.70)		1.56
(0.03, 76.63)		12.41
(0.23, 690.50)		1.57
(0.02, 106.70)		0.55
(0.02, 15.91)		0.83
(0.18, 3.90)		1.00
(0.21, 4.77)		0.71
(0.15, 3.35)		0.33
(0.07, 1.50)		---0.52
(0.11, 2.33)
Shammas-PL13.84
(0.28, 686.00)		3.59
(0.25, 53.09)		0.89
(0.05, 14.03)		1.26
(0.07, 21.07)		2.63
(0.16, 44.96)		1.47
(0.20, 11.10)		1.76
(0.03, 85.45)		1.24
(0.04, 37.72)		9.73
(0.28, 351.80)		1.27
(0.03, 61.34)		0.44
(0.02, 8.57)		0.80
(0.04, 13.41)		1.20
(0.37, 3.92)		0.85
(0.26, 2.75)		0.40
(0.13, 1.21)		---0.62
(0.20, 1.89)
SRKT A-P16.35
(0.28, 935.50)		4.26
(0.24, 78.67)		1.06
(0.05, 20.89)		1.51
(0.07, 29.39)		3.10
(0.16, 65.42)		1.74
(0.13, 24.64)		2.08
(0.04, 119.80)		1.46
(0.04, 56.51)		11.52
(0.26, 512.40)		1.49
(0.03, 86.26)		0.52
(0.02, 11.81)		0.95
(0.04, 19.81)		1.18
(0.08, 17.37)		0.71
(0.21, 2.35)		0.33
(0.11, 1.03)		---0.52
(0.16, 1.62)
SRKT C-P11.16
(0.20, 666.50)		2.99
(0.16, 52.55)		0.73
(0.04, 14.10)		1.03
(0.05, 21.10)		2.18
(0.11, 44.03)		1.23
(0.09, 16.46)		1.45
(0.02, 83.72)		1.02
(0.03, 39.19)		8.14
(0.17, 371.30)		1.03
(0.02, 57.93)		0.36
(0.01, 8.30)		0.66
(0.03, 13.48)		0.82
(0.05, 12.32)		0.70
(0.04, 12.77)		0.47
(0.15, 1.44)		---0.73
(0.23, 2.26)
SRKT DK5.13
(0.30, 90.12)		1.35
(0.08, 23.55)		0.33
(0.02, 6.31)		0.47
(0.02, 9.54)		0.99
(0.05, 19.41)		0.55
(0.04, 7.31)		0.66
(0.04, 11.39)		0.46
(0.01, 17.53)		3.68
(0.08, 168.00)		0.47
(0.03, 8.18)		0.16
(0.01, 3.82)		0.30
(0.01, 5.98)		0.37
(0.03, 5.39)		0.31
(0.02, 5.78)		0.46
(0.03, 7.87)		0.79
(0.30, 2.05)		-0.79
(0.30, 2.05)		1.55
(0.54, 4.46)
SRKT Feiz-Mannis4.01
(0.23, 68.91)		1.06
(0.02, 61.32)		0.26
(0.00, 15.85)		0.37
(0.01, 23.36)		0.78
(0.01, 46.62)		0.43
(0.01, 20.86)		0.52
(0.03, 8.69)		0.36
(0.00, 35.39)		2.88
(0.02, 332.60)		0.37
(0.02, 6.33)		0.13
(0.00, 8.85)		0.23
(0.00, 15.16)		0.29
(0.01, 14.21)		0.25
(0.00, 14.64)		0.36
(0.01, 19.83)		0.78
(0.05, 12.83)		-1.00
(0.39, 2.56)		-
SRKT K2.80
(0.04, 216.40)		0.74
(0.03, 19.53)		0.18
(0.01, 4.91)		0.26
(0.01, 7.12)		0.54
(0.02, 14.87)		0.30
(0.03, 3.14)		0.36
(0.01, 28.41)		0.25
(0.02, 3.79)		2.02
(0.11, 38.82)		0.26
(0.00, 20.06)		0.09
(0.00, 2.91)		0.16
(0.01, 4.87)		0.20
(0.01, 3.39)		0.17
(0.01, 4.29)		0.25
(0.01, 6.02)		0.54
(0.02, 12.99)		0.70
(0.01, 53.95)		-0.57
(0.17, 1.93)
SRKT Ladas-Stark4.06
(0.24, 70.52)		1.07
(0.02, 62.02)		0.26
(0.00, 15.31)		0.37
(0.01, 22.91)		0.77
(0.01, 48.40)		0.44
(0.01, 21.22)		0.52
(0.03, 8.77)		0.36
(0.00, 35.32)		2.91
(0.03, 336.20)		0.37
(0.02, 6.26)		0.13
(0.00, 8.54)		0.23
(0.00, 14.58)		0.29
(0.01, 14.89)		0.25
(0.00, 14.05)		0.36
(0.01, 20.29)		0.79
(0.05, 13.65)		1.00
(0.06, 16.65)		1.45
(0.02, 96.99)		-
SRKT TNP4.85
(0.09, 239.20)		1.27
(0.08, 19.62)		0.31
(0.02, 5.20)		0.44
(0.02, 7.52)		0.92
(0.05, 16.07)		0.52
(0.05, 4.81)		0.62
(0.01, 32.52)		0.44
(0.03, 6.58)		3.46
(0.18, 61.45)		0.44
(0.01, 22.90)		0.15
(0.01, 3.15)		0.28
(0.01, 5.18)		0.35
(0.03, 4.09)		0.30
(0.02, 4.59)		0.43
(0.03, 6.30)		0.93
(0.06, 13.63)		1.22
(0.02, 60.79)		1.72
(0.16, 16.40)		1.20
(0.02, 58.13)
Figure 71. Myopic corneal refractive surgery: within 1.0D – relative effect of all options versus common comparator.

Figure 71Myopic corneal refractive surgery: within 1.0D – relative effect of all options versus common comparator

Table 120Myopic corneal refractive surgery: within 1.0D – rankings for each comparator

Probability bestMedian rank (95%CI)
SRKT Clinical history0.00119 (6, 20)
BESSt0.00314 (4, 20)
Camellin-Calossi0.0806 (1, 18)
Double-K0.0538 (1, 19)
Feiz-Mannis0.01013 (2, 20)
Haigis-L0.0089 (2, 17)
HofferQ DK0.05210 (1, 19)
HofferQ K0.0918 (1, 19)
HofferQ TNP0.00418 (4, 20)
Holladay2 DK0.0958 (1, 19)
Masket0.2903 (1, 16)
Modified Masket0.1196 (1, 18)
Shammas-PL0.0387 (1, 17)
SRKT A-P0.0806 (1, 17)
SRKT C-P0.0398 (1, 18)
SRKT DK0.00113 (4, 18)
SRKT Feiz-Mannis0.01614 (2, 20)
SRKT K0.00216 (4, 20)
SRKT Ladas-Stark0.01514 (2, 20)
SRKT TNP0.00213 (4, 19)
Figure 72. Myopic corneal refractive surgery: within 1.0D – rank probability histograms.

Figure 72Myopic corneal refractive surgery: within 1.0D – rank probability histograms

Table 121Myopic corneal refractive surgery: within 1.0D – model fit statistics

Residual devianceDbarDhatpDDICtau
27.01
(compared to 26 datapoints)		119.69993.50426.195145.8941.213 (95%CI: 0.294, 1.954)

Table 122Myopic corneal refractive surgery: within 1.0D – notes

  • Dichotomous synchronic (binomial; logit link); random effects
  • Prior distribution for between-study heterogeneity: uniform (Min=0; Max=2)
  • 50000 burn-ins; 10000 recorded iterations (thinned from 500000)
Pairwise Comparisons: Proportion Within 1.5 Dioptres and 2.0 Dioptres.

Pairwise Comparisons: Proportion Within 1.5 Dioptres and 2.0 Dioptres

Figure 73. Myopic corneal refractive surgery: within 2.0D – evidence network.

Figure 73Myopic corneal refractive surgery: within 2.0D – evidence network

Table 123Myopic corneal refractive surgery: within 2.0D – input data

SRKT Clinical historyHofferQ DKHolladay2 DKSRKT DKSRKT Feiz-MannisSRKT Ladas-Stark
Fam & (2008)33/3735/3734/3735/3732/3735/37

Table 124Myopic corneal refractive surgery: within 2.0D – relative effectiveness of all pairwise combinations (RR and 95% credible interval)

SRKT Clinical historyHofferQ DKHolladay2 DKSRKT DKSRKT Feiz-MannisSRKT Ladas-Stark
SRKT Clinical history2.12
(0.36, 12.36)		1.37
(0.29, 6.61)		2.12
(0.36, 12.36)		0.78
(0.19, 3.15)		2.12
(0.36, 12.36)
HofferQ DK2.32
(0.39, 19.84)		0.65
(0.10, 4.12)		1.00
(0.13, 7.50)		0.37
(0.07, 2.02)		1.00
(0.13, 7.50)
Holladay2 DK1.42
(0.27, 8.24)		0.61
(0.07, 4.32)		1.54
(0.24, 9.82)		0.56
(0.12, 2.56)		1.54
(0.24, 9.82)
SRKT DK2.33
(0.39, 19.87)		1.00
(0.10, 9.99)		1.63
(0.23, 15.74)		0.37
(0.07, 2.02)		1.00
(0.13, 7.50)
SRKT Feiz-Mannis0.76
(0.17, 3.29)		0.33
(0.04, 1.74)		0.53
(0.10, 2.50)		0.33
(0.04, 1.81)		2.73
(0.50, 15.09)
SRKT Ladas-Stark2.30
(0.39, 19.99)		0.99
(0.10, 9.99)		1.62
(0.23, 14.79)		0.99
(0.10, 10.03)		3.02
(0.56, 26.29)
Figure 74. Myopic corneal refractive surgery: within 2.0D – relative effect of all options versus common comparator.

Figure 74Myopic corneal refractive surgery: within 2.0D – relative effect of all options versus common comparator

Table 125Myopic corneal refractive surgery: within 2.0D – rankings for each comparator

Probability bestMedian rank (95%CI)
SRKT Clinical history0.0295 (1, 6)
HofferQ DK0.2922 (1, 6)
Holladay2 DK0.0934 (1, 6)
SRKT DK0.2942 (1, 6)
SRKT Feiz-Mannis0.0086 (2, 6)
SRKT Ladas-Stark0.2842 (1, 6)
Figure 75. Myopic corneal refractive surgery: within 2.0D – rank probability histograms.

Figure 75Myopic corneal refractive surgery: within 2.0D – rank probability histograms

Table 126Myopic corneal refractive surgery: within 2.0D – model fit statistics

Residual devianceDbarDhatpDDIC
6.394
(compared to 6 datapoints)		23.46517.5675.89829.364

Table 127Myopic corneal refractive surgery: within 2.0D – notes

  • Dichotomous synchronic (binomial; logit link); fixed effects
  • 50000 burn-ins; 10000 recorded iterations (thinned from 500000)

H.3.3.3. Sensitivity analyses: no historical data methods only

Figure 76. Myopic CRS No historical data methods: within 0.5D – evidence network.

Figure 76Myopic CRS No historical data methods: within 0.5D – evidence network

Table 128Myopic CRS No historical data methods: within 0.5D – input data

SRKT KBESStHaigis-LHofferQ KHofferQ TNPShammas-PLSRKT A-PSRKT C-PSRKT TNP
Xu et al. (2014)4/376/373/373/37
Huang et al. (2013)21/4621/46
Kim et al. (2013)5/4730/47
Saiki et al. (2013)3/286/257/2813/2812/255/28

Table 129Myopic CRS No historical data methods: within 0.5D – relative effectiveness of all pairwise combinations (RR and 95% credible interval)

SRKT KBESStHaigis-LHofferQ KHofferQ TNPShammas-PLSRKT A-PSRKT C-PSRKT TNP
SRKT K-14.82
(4.93, 44.61)		1.60
(0.41, 6.20)		0.73
(0.15, 3.50)		---0.73
(0.15, 3.50)
BESSt1.91
(0.08, 36.38)		2.63
(0.58, 11.90)		--2.78
(0.64, 12.10)		7.22
(1.76, 29.56)		7.69
(1.84, 32.20)		1.81
(0.39, 8.44)
Haigis-L7.71
(0.83, 59.26)		3.99
(0.33, 55.38)		--1.02
(0.51, 2.02)		2.74
(0.84, 8.94)		2.92
(0.87, 9.78)		0.69
(0.18, 2.61)
HofferQ K2.94
(0.20, 38.58)		1.54
(0.05, 61.70)		0.38
(0.02, 8.18)		0.46
(0.10, 1.98)		---0.46
(0.10, 1.98)
HofferQ TNP1.25
(0.07, 17.11)		0.66
(0.02, 24.84)		0.16
(0.01, 3.45)		0.42
(0.02, 6.30)		---1.00
(0.19, 5.31)
Shammas-PL6.54
(0.44, 72.95)		3.38
(0.27, 48.07)		0.86
(0.14, 4.80)		2.21
(0.08, 50.70)		5.22
(0.21, 135.60)		2.60
(0.84, 8.07)		2.77
(0.87, 8.84)		0.65
(0.18, 2.37)
SRKT A-P15.89
(0.76, 273.70)		8.16
(0.55, 144.60)		2.06
(0.19, 21.26)		5.34
(0.16, 156.30)		12.65
(0.36, 447.50)		2.43
(0.22, 26.60)		1.07
(0.36, 3.14)		0.25
(0.07, 0.85)
SRKT C-P16.96
(0.78, 285.30)		8.84
(0.60, 153.10)		2.23
(0.20, 24.06)		5.74
(0.17, 166.20)		13.68
(0.40, 460.80)		2.58
(0.23, 29.24)		1.08
(0.08, 14.84)		0.24
(0.07, 0.82)
SRKT TNP2.20
(0.21, 18.06)		1.16
(0.08, 19.14)		0.29
(0.04, 2.33)		0.75
(0.05, 9.28)		1.78
(0.12, 27.68)		0.34
(0.04, 3.41)		0.14
(0.01, 1.77)		0.13
(0.01, 1.72)
Figure 77. Myopic CRS No historical data methods: within 0.5D – relative effect of all options versus common comparator.

Figure 77Myopic CRS No historical data methods: within 0.5D – relative effect of all options versus common comparator

Table 130Myopic CRS No historical data methods: within 0.5D – rankings for each comparator

Probability bestMedian rank (95%CI)
SRKT K0.0028 (4, 9)
BESSt0.0087 (2, 9)
Haigis-L0.0613 (1, 7)
HofferQ K0.0576 (1, 9)
HofferQ TNP0.0158 (2, 9)
Shammas-PL0.0434 (1, 8)
SRKT A-P0.3792 (1, 7)
SRKT C-P0.4302 (1, 7)
SRKT TNP0.0046 (3, 9)
Figure 78. Myopic CRS No historical data methods: within 0.5D – rank probability histograms.

Figure 78Myopic CRS No historical data methods: within 0.5D – rank probability histograms

Table 131Myopic CRS No historical data methods: within 0.5D – model fit statistics

Residual devianceDbarDhatpDDICtau
14.86
(compared to 14 datapoints)		64.03150.4313.60177.6320.942 (95%CI: 0.066, 1.925)

Table 132Myopic CRS No historical data methods: within 0.5D – notes

  • Dichotomous synchronic (binomial; logit link); random effects
  • Prior distribution for between-study heterogeneity: uniform (Min=0; Max=2)
  • 50000 burn-ins; 10000 recorded iterations (thinned from 500000)
Figure 79. Myopic CRS No historical data methods: within 1.0D – evidence network.

Figure 79Myopic CRS No historical data methods: within 1.0D – evidence network

Table 133Myopic CRS No historical data methods: within 1.0D – input data

SRKT KBESStHaigis-LHofferQ KHofferQ TNPShammas-PLSRKT A-PSRKT C-PSRKT TNP
Xu et al. (2014)8/3714/373/375/37
Huang et al. (2013)36/4639/46
Kim et al. (2013)16/4738/47
Saiki et al. (2013)12/2813/2520/2821/2817/2517/28

Table 134Myopic CRS No historical data methods: within 1.0D – relative effectiveness of all pairwise combinations (RR and 95% credible interval)

SRKT KBESStHaigis-LHofferQ KHofferQ TNPShammas-PLSRKT A-PSRKT C-PSRKT TNP
SRKT K-8.18
(3.18, 21.03)		2.21
(0.79, 6.16)		0.32
(0.08, 1.32)		---0.57
(0.17, 1.93)
BESSt1.36
(0.06, 29.93)		1.44
(0.49, 4.27)		--3.33
(1.10, 10.12)		4.00
(1.28, 12.46)		2.83
(0.92, 8.73)		2.06
(0.71, 5.98)
Haigis-L3.30
(0.32, 30.30)		2.40
(0.18, 32.35)		--1.87
(0.86, 4.06)		2.77
(0.87, 8.84)		1.96
(0.62, 6.19)		1.43
(0.48, 4.25)
HofferQ K3.93
(0.26, 60.17)		2.89
(0.07, 114.80)		1.20
(0.06, 28.92)		0.14
(0.04, 0.56)		---0.26
(0.08, 0.81)
HofferQ TNP0.50
(0.03, 8.27)		0.37
(0.01, 16.64)		0.15
(0.01, 4.08)		0.13
(0.01, 2.49)		---1.77
(0.39, 8.02)
Shammas-PL4.89
(0.29, 73.62)		3.59
(0.25, 51.59)		1.49
(0.20, 10.83)		1.23
(0.04, 37.75)		9.68
(0.28, 348.20)		1.20
(0.37, 3.92)		0.85
(0.26, 2.75)		0.62
(0.20, 1.89)
SRKT A-P5.77
(0.23, 130.60)		4.24
(0.24, 76.73)		1.76
(0.13, 23.75)		1.46
(0.04, 56.18)		11.34
(0.26, 561.10)		1.19
(0.08, 16.70)		0.71
(0.21, 2.35)		0.52
(0.16, 1.62)
SRKT C-P4.00
(0.16, 91.84)		2.94
(0.16, 53.66)		1.22
(0.09, 16.85)		1.00
(0.02, 39.47)		7.88
(0.18, 378.80)		0.82
(0.05, 12.06)		0.69
(0.04, 12.78)		0.73
(0.23, 2.26)
SRKT TNP1.72
(0.16, 16.54)		1.27
(0.08, 19.13)		0.52
(0.06, 4.77)		0.43
(0.03, 6.28)		3.37
(0.19, 61.78)		0.35
(0.03, 4.05)		0.30
(0.02, 4.49)		0.43
(0.03, 6.83)
Figure 80. Myopic CRS No historical data methods: within 1.0D – relative effect of all options versus common comparator.

Figure 80Myopic CRS No historical data methods: within 1.0D – relative effect of all options versus common comparator

Table 135Myopic CRS No historical data methods: within 1.0D – rankings for each comparator

Probability bestMedian rank (95% CI)
SRKT K0.0087 (2, 9)
BESSt0.0227 (2, 9)
Haigis-L0.0534 (1, 8)
HofferQ K0.2334 (1, 8)
HofferQ TNP0.0109 (2, 9)
Shammas-PL0.1903 (1, 8)
SRKT A-P0.3032 (1, 8)
SRKT C-P0.1694 (1, 9)
SRKT TNP0.0126 (2, 9)
Figure 81. Myopic CRS No historical data methods: within 1.0D – rank probability histograms.

Figure 81Myopic CRS No historical data methods: within 1.0D – rank probability histograms

Table 136Myopic CRS No historical data methods: within 1.0D – model fit statistics

Residual devianceDbarDhatpDDICtau
14.54
(compared to 14 datapoints)		66.00752.11713.8979.8971.204 (95%CI: 0.298, 1.951)

Table 137Myopic CRS No historical data methods: within 1.0D – notes

  • Dichotomous synchronic (binomial; logit link); random effects
  • Prior distribution for between-study heterogeneity: uniform (Min=0; Max=2)
  • 50000 burn-ins; 10000 recorded iterations (thinned from 500000)

H.3.3.4. Sensitivity analysis: historical data methods only

Figure 82. Myopic CRS Historical data methods: mean absolute error – evidence network.

Figure 82Myopic CRS Historical data methods: mean absolute error – evidence network

Table 138Myopic CRS Historical data methods: mean absolute error – input data

Clinical historyHofferQ DKHolladay2 DKSRKT Clinical historySRKT DiehlSRKT DKSRKT DK AwwadSRKT DK SaviniSRKT DK Seitz/SpeicherSRKT DK Seitz/Speicher/SaviniSRKT DK ShammasSRKT Feiz-MannisSRKT Feiz-Mannis nomogramSRKT Ladas-StarkSRKT LatkanySRKT MasketSRKT SK FerraraSRKT SK Rosa
Savini et al. (2010)1.62
(1.25)		1.33
(1.03)		1.00
(0.57)		1.79
(1.13)		0.62
(0.52)		0.56
(0.45)		0.53
(0.46)		1.60
(0.98)		1.87
(1.44)		2.04
(1.48)		2.18
(1.52)		1.08
(0.86)		0.76
(0.49)		3.64
(1.45)		1.94
(1.01)
Fam & (2008)0.75
(0.52)		0.75
(0.62)		1.32
(0.73)		0.76
(0.60)		0.93
(0.83)		0.80
(0.63)

Table 139Myopic CRS Historical data methods: mean absolute error – relative effectiveness of all pairwise combinations (MD and 95% credible interval)

Clinical historyHofferQ DKHolladay2 DKSRKT Clinical historySRKT DiehlSRKT DKSRKT DK AwwadSRKT DK SaviniSRKT DK Seitz/SpeicherSRKT DK Seitz/Speicher/SaviniSRKT DK ShammasSRKT Feiz-MannisSRKT Feiz-Mannis nomogramSRKT Ladas-StarkSRKT LatkanySRKT MasketSRKT SK FerraraSRKT SK Rosa
Clinical history---−0.29
(−0.89, 0.31)		−0.62
(−1.13, −0.11)		0.17
(−0.45, 0.79)		−1.00
(−1.50, −0.50)		−1.06
(−1.55, −0.57)		−1.09
(−1.58, −0.60)		−0.02
(−0.61, 0.57)		0.25
(−0.46, 0.96)		0.42
(−0.30, 1.14)		0.56
(−0.17, 1.29)		−0.54
(−1.10, 0.02)		−0.86
(−1.36, −0.36)		2.02
(1.31, 2.73)		0.32
(−0.28, 0.92)
HofferQ DK−0.07
(−2.66, 2.54)		0.00
(−0.26, 0.26)		0.57
(0.28, 0.86)		-0.01
(−0.25, 0.27)		-----0.18
(−0.14, 0.50)		-0.05
(−0.21, 0.31)		----
Holladay2 DK−0.08
(−2.63, 2.50)		0.00
(−2.16, 2.16)		0.57
(0.26, 0.88)		-0.01
(−0.27, 0.29)		-----0.18
(−0.15, 0.51)		-0.05
(−0.23, 0.33)		----
SRKT Clinical history0.50
(−2.04, 3.08)		0.57
(−1.59, 2.72)		0.57
(−1.63, 2.74)		-−0.56
(−0.86, −0.26)		-----−0.39
(−0.75, −0.03)		-−0.52
(−0.83, −0.21)		----
SRKT Diehl−0.30
(−2.52, 1.93)		−0.22
(−2.80, 2.29)		−0.22
(−2.77, 2.27)		−0.79
(−3.38, 1.74)		−0.33
(−0.77, 0.11)		0.46
(−0.11, 1.03)		−0.71
(−1.14, −0.28)		−0.77
(−1.19, −0.35)		−0.80
(−1.22, −0.38)		0.27
(−0.26, 0.80)		0.54
(−0.12, 1.20)		0.71
(0.04, 1.38)		0.85
(0.17, 1.53)		−0.25
(−0.75, 0.25)		−0.57
(−0.99, −0.15)		2.31
(1.65, 2.97)		0.61
(0.08, 1.14)
SRKT DK−0.35
(−2.38, 1.74)		−0.27
(−2.26, 1.69)		−0.27
(−2.22, 1.68)		−0.84
(−2.84, 1.14)		−0.05
(−2.08, 2.02)		0.79
(0.32, 1.26)		−0.38
(−0.67, −0.09)		−0.44
(−0.71, −0.17)		−0.47
(−0.74, −0.20)		0.60
(0.18, 1.02)		0.48
(−0.20, 1.16)		1.04
(0.45, 1.63)		0.58
(−0.54, 1.69)		0.08
(−0.30, 0.46)		−0.24
(−0.52, 0.04)		2.64
(2.06, 3.22)		0.94
(0.51, 1.37)
SRKT DK Awwad0.16
(−2.08, 2.37)		0.24
(−2.40, 2.74)		0.24
(−2.29, 2.75)		−0.34
(−2.92, 2.19)		0.45
(−1.80, 2.68)		0.51
(−1.57, 2.47)		−1.17
(−1.63, −0.71)		−1.23
(−1.68, −0.78)		−1.26
(−1.71, −0.81)		−0.19
(−0.74, 0.36)		0.08
(−0.60, 0.76)		0.25
(−0.44, 0.94)		0.39
(−0.31, 1.09)		−0.71
(−1.24, −0.18)		−1.03
(−1.49, −0.57)		1.85
(1.17, 2.53)		0.15
(−0.41, 0.71)
SRKT DK Savini−1.00
(−3.26, 1.18)		−0.93
(−3.48, 1.53)		−0.92
(−3.44, 1.56)		−1.50
(−4.07, 0.97)		−0.71
(−2.90, 1.49)		−0.66
(−2.70, 1.34)		−1.17
(−3.33, 1.05)		−0.06
(−0.31, 0.19)		−0.09
(−0.35, 0.17)		0.98
(0.57, 1.39)		1.25
(0.68, 1.82)		1.42
(0.84, 2.00)		1.56
(0.96, 2.16)		0.46
(0.09, 0.83)		0.14
(−0.12, 0.40)		3.02
(2.45, 3.59)		1.32
(0.90, 1.74)
SRKT DK Seitz/Speicher−1.07
(−3.31, 1.13)		−0.99
(−3.57, 1.51)		−0.98
(−3.56, 1.47)		−1.57
(−4.14, 0.94)		−0.77
(−2.97, 1.44)		−0.72
(−2.79, 1.25)		−1.23
(−3.44, 0.99)		−0.06
(−2.27, 2.15)		−0.03
(−0.27, 0.21)		1.04
(0.64, 1.44)		1.31
(0.75, 1.87)		1.48
(0.91, 2.05)		1.62
(1.03, 2.21)		0.52
(0.16, 0.88)		0.20
(−0.05, 0.45)		3.08
(2.52, 3.64)		1.38
(0.97, 1.79)
SRKT DK Seitz/Speicher/Savini−1.10
(−3.32, 1.12)		−1.02
(−3.59, 1.46)		−1.02
(−3.56, 1.49)		−1.59
(−4.16, 0.92)		−0.80
(−3.03, 1.41)		−0.75
(−2.77, 1.23)		−1.26
(−3.44, 0.96)		−0.09
(−2.29, 2.10)		−0.03
(−2.24, 2.19)		1.07
(0.67, 1.47)		1.34
(0.78, 1.90)		1.51
(0.94, 2.08)		1.65
(1.06, 2.24)		0.55
(0.19, 0.91)		0.23
(−0.02, 0.48)		3.11
(2.55, 3.67)		1.41
(1.00, 1.82)
SRKT DK Shammas−0.03
(−2.26, 2.22)		0.05
(−2.57, 2.55)		0.05
(−2.51, 2.56)		−0.52
(−3.16, 2.01)		0.26
(−1.94, 2.45)		0.32
(−1.74, 2.33)		−0.19
(−2.44, 2.08)		0.97
(−1.25, 3.22)		1.04
(−1.18, 3.25)		1.07
(−1.14, 3.29)		0.27
(−0.38, 0.92)		0.44
(−0.22, 1.10)		0.58
(−0.09, 1.25)		−0.52
(−1.00, −0.04)		−0.84
(−1.25, −0.43)		2.04
(1.39, 2.69)		0.34
(−0.18, 0.86)
SRKT Feiz-Mannis0.17
(−1.88, 2.25)		0.24
(−1.79, 2.24)		0.24
(−1.76, 2.21)		−0.34
(−2.37, 1.66)		0.46
(−1.57, 2.55)		0.51
(−1.02, 2.10)		0.00
(−2.01, 2.11)		1.17
(−0.81, 3.24)		1.24
(−0.77, 3.30)		1.26
(−0.74, 3.32)		0.20
(−1.81, 2.24)		0.17
(−0.59, 0.93)		−0.05
(−0.38, 0.27)		−0.79
(−1.41, −0.17)		−1.11
(−1.67, −0.55)		1.77
(1.01, 2.53)		0.07
(−0.58, 0.72)
SRKT Feiz-Mannis nomogram0.41
(−1.85, 2.68)		0.49
(−2.16, 3.01)		0.49
(−2.12, 3.02)		−0.09
(−2.72, 2.46)		0.71
(−1.55, 2.93)		0.76
(−1.32, 2.78)		0.25
(−1.98, 2.51)		1.42
(−0.78, 3.64)		1.48
(−0.73, 3.72)		1.51
(−0.72, 3.77)		0.44
(−1.83, 2.73)		0.25
(−1.84, 2.31)		0.14
(−0.65, 0.93)		−0.96
(−1.59, −0.33)		−1.28
(−1.86, −0.70)		1.60
(0.83, 2.37)		−0.10
(−0.76, 0.56)
SRKT Ladas-Stark0.20
(−1.77, 2.31)		0.27
(−1.70, 2.30)		0.28
(−1.63, 2.28)		−0.29
(−2.28, 1.73)		0.50
(−1.47, 2.59)		0.55
(−0.98, 2.12)		0.05
(−1.91, 2.15)		1.21
(−0.73, 3.30)		1.27
(−0.68, 3.41)		1.30
(−0.66, 3.42)		0.23
(−1.75, 2.35)		0.04
(−1.49, 1.65)		−0.20
(−2.23, 1.91)		−1.10
(−1.75, −0.45)		−1.42
(−2.01, −0.83)		1.46
(0.68, 2.24)		−0.24
(−0.92, 0.44)
SRKT Latkany−0.55
(−2.77, 1.63)		−0.47
(−3.02, 2.01)		−0.47
(−3.02, 2.03)		−1.05
(−3.60, 1.46)		−0.25
(−2.47, 1.94)		−0.20
(−2.26, 1.80)		−0.70
(−2.92, 1.54)		0.46
(−1.71, 2.65)		0.53
(−1.67, 2.71)		0.56
(−1.62, 2.76)		−0.51
(−2.71, 1.69)		−0.71
(−2.74, 1.30)		−0.96
(−3.18, 1.25)		−0.75
(−2.82, 1.21)		−0.32
(−0.69, 0.05)		2.56
(1.94, 3.18)		0.86
(0.37, 1.35)
SRKT Masket−0.87
(−3.06, 1.34)		−0.79
(−3.34, 1.71)		−0.78
(−3.36, 1.71)		−1.36
(−3.93, 1.13)		−0.57
(−2.78, 1.61)		−0.52
(−2.56, 1.44)		−1.03
(−3.22, 1.20)		0.14
(−2.03, 2.29)		0.20
(−1.97, 2.40)		0.23
(−1.95, 2.43)		−0.84
(−3.03, 1.33)		−1.03
(−3.08, 0.94)		−1.28
(−3.52, 0.97)		−1.07
(−3.15, 0.87)		−0.32
(−2.50, 1.83)		2.88
(2.31, 3.45)		1.18
(0.76, 1.60)
SRKT SK Ferrara2.01
(−0.24, 4.26)		2.08
(−0.53, 4.62)		2.09
(−0.49, 4.66)		1.52
(−1.09, 4.09)		2.31
(0.09, 4.55)		2.36
(0.29, 4.42)		1.86
(−0.36, 4.13)		3.02
(0.85, 5.24)		3.08
(0.88, 5.31)		3.11
(0.92, 5.36)		2.04
(−0.20, 4.28)		1.85
(−0.26, 3.93)		1.61
(−0.66, 3.86)		1.80
(−0.33, 3.83)		2.55
(0.35, 4.80)		2.87
(0.71, 5.11)		−1.70
(−2.35, −1.05)
SRKT SK Rosa0.31
(−1.94, 2.54)		0.39
(−2.21, 2.92)		0.39
(−2.20, 2.89)		−0.18
(−2.73, 2.34)		0.61
(−1.56, 2.83)		0.66
(−1.41, 2.69)		0.15
(−2.02, 2.38)		1.32
(−0.86, 3.51)		1.39
(−0.81, 3.56)		1.41
(−0.79, 3.63)		0.35
(−1.84, 2.57)		0.16
(−1.92, 2.17)		−0.10
(−2.35, 2.14)		0.12
(−2.00, 2.07)		0.86
(−1.34, 3.06)		1.18
(−1.01, 3.40)		−1.69
(−3.96, 0.53)
Figure 83. Myopic CRS Historical data methods: mean absolute error – relative effect of all options versus common comparator.

Figure 83Myopic CRS Historical data methods: mean absolute error – relative effect of all options versus common comparator

Table 140Myopic CRS Historical data methods: mean absolute error – rankings for each comparator

Probability bestMedian rank (95% CI)
Clinical history0.01311 (2, 17)
HofferQ DK0.02710 (1, 17)
Holladay2 DK0.02710 (1, 17)
SRKT Clinical history0.00815 (3, 18)
SRKT Diehl0.0278 (1, 17)
SRKT DK0.0117 (2, 15)
SRKT DK Awwad0.00712 (3, 17)
SRKT DK Savini0.1913 (1, 14)
SRKT DK Seitz/Speicher0.2313 (1, 13)
SRKT DK Seitz/Speicher/Savini0.2573 (1, 13)
SRKT DK Shammas0.01310 (2, 17)
SRKT Feiz-Mannis0.00212 (4, 17)
SRKT Feiz-Mannis nomogram0.00514 (4, 18)
SRKT Ladas-Stark0.00112 (5, 17)
SRKT Latkany0.0506 (1, 16)
SRKT Masket0.1254 (1, 14)
SRKT SK Ferrara0.00018 (12, 18)
SRKT SK Rosa0.00513 (3, 17)
Figure 84. Myopic CRS Historical data methods: mean absolute error – rank probability histograms.

Figure 84Myopic CRS Historical data methods: mean absolute error – rank probability histograms

Table 141Myopic CRS Historical data methods: mean absolute error – model fit statistics

Residual devianceDbarDhatpDDICtau
21.08
(compared to 21 datapoints)		−20.673−41.58220.9090.2370.816 (95% CI: 0.246, 1.896)

Table 142Myopic CRS Historical data methods: mean absolute error – notes

  • Continuous (normal; identity link); random effects
  • Prior distribution for between-study heterogeneity: uniform (Min=0; Max=2)
  • 50000 burn-ins; 10000 recorded iterations (thinned from 500000)
Figure 85. Myopic CRS Historical data methods: within 0.5D – evidence network.

Figure 85Myopic CRS Historical data methods: within 0.5D – evidence network

Table 143Myopic CRS Historical data methods: within 0.5D – input data

SRKT Clinical historyDouble-KFeiz-MannisHofferQ DKHolladay2 DKMasketModified MasketSRKT DKSRKT Feiz-MannisSRKT Ladas-Stark
Saiki et al. (2013)4/121/124/125/125/28
Fam & (2008)5/3713/3717/3719/3715/3717/37

Table 144Myopic CRS Historical data methods: within 0.5D – relative effectiveness of all pairwise combinations (RR and 95% credible interval)

SRKT Clinical historyDouble-KFeiz-MannisHofferQ DKHolladay2 DKMasketModified MasketSRKT DKSRKT Feiz-MannisSRKT Ladas-Stark
SRKT Clinical history--3.47
(1.09, 11.05)		5.44
(1.73, 17.06)		--6.76
(2.16, 21.16)		4.36
(1.38, 13.76)		5.44
(1.73, 17.06)
Double-K17.01
(2.37, 129.10)		0.18
(0.02, 1.95)		--1.00
(0.18, 5.46)		1.43
(0.27, 7.52)		0.43
(0.09, 2.03)		--
Feiz-Mannis2.25
(0.06, 27.09)		0.14
(0.00, 1.31)		--5.50
(0.51, 59.01)		7.86
(0.75, 82.13)		2.39
(0.25, 23.01)		--
HofferQ DK3.64
(1.18, 13.11)		0.22
(0.03, 1.40)		1.63
(0.15, 54.84)		1.57
(0.62, 4.00)		--1.95
(0.77, 4.96)		1.26
(0.49, 3.23)		1.57
(0.62, 4.00)
Holladay2 DK5.79
(1.90, 20.76)		0.35
(0.05, 2.21)		2.59
(0.24, 86.40)		1.59
(0.62, 4.16)		--1.24
(0.50, 3.09)		0.80
(0.32, 2.02)		1.00
(0.40, 2.50)
Masket17.16
(2.30, 133.70)		1.01
(0.17, 5.95)		7.55
(0.74, 251.90)		4.69
(0.70, 31.09)		2.93
(0.43, 19.25)		1.43
(0.27, 7.52)		0.43
(0.09, 2.03)		--
Modified Masket24.98
(3.69, 185.20)		1.47
(0.27, 8.50)		11.05
(1.18, 359.00)		6.82
(1.10, 44.26)		4.31
(0.70, 27.60)		1.47
(0.26, 8.61)		0.30
(0.07, 1.37)		--
SRKT DK7.18
(2.39, 25.50)		0.43
(0.09, 2.14)		3.19
(0.38, 96.87)		1.97
(0.78, 5.25)		1.24
(0.50, 3.19)		0.43
(0.09, 2.21)		0.29
(0.06, 1.39)		0.65
(0.26, 1.62)		0.81
(0.32, 2.01)
SRKT Feiz-Mannis4.61
(1.50, 16.22)		0.28
(0.04, 1.75)		2.04
(0.19, 70.47)		1.26
(0.49, 3.31)		0.80
(0.31, 2.03)		0.27
(0.04, 1.80)		0.18
(0.03, 1.16)		0.64
(0.25, 1.62)		1.25
(0.50, 3.13)
SRKT Ladas-Stark5.77
(1.90, 20.17)		0.35
(0.05, 2.18)		2.59
(0.24, 86.28)		1.59
(0.62, 4.11)		1.00
(0.40, 2.51)		0.34
(0.05, 2.27)		0.23
(0.04, 1.45)		0.80
(0.32, 2.02)		1.25
(0.50, 3.17)
Figure 86. Myopic CRS Historical data methods: within 0.5D – relative effect of all options versus common comparator.

Figure 86Myopic CRS Historical data methods: within 0.5D – relative effect of all options versus common comparator

Table 145Myopic CRS Historical data methods: within 0.5D – rankings for each comparator

Probability bestMedian rank (95% CI)
SRKT Clinical history0.00010 (9, 10)
Double-K0.2302 (1, 8)
Feiz-Mannis0.0039 (3, 10)
HofferQ DK0.0018 (4, 9)
Holladay2 DK0.0116 (2, 9)
Masket0.2342 (1, 8)
Modified Masket0.5031 (1, 6)
SRKT DK0.0054 (2, 8)
SRKT Feiz-Mannis0.0037 (3, 9)
SRKT Ladas-Stark0.0106 (2, 9)
Figure 87. Myopic CRS Historical data methods: within 0.5D – rank probability histograms.

Figure 87Myopic CRS Historical data methods: within 0.5D – rank probability histograms

Table 146Myopic CRS Historical data methods: within 0.5D – model fit statistics

Residual devianceDbarDhatpDDIC
11.44
(compared to 11 datapoints)		48.937.76811.13260.032

Table 147Myopic CRS Historical data methods: within 0.5D – notes

  • Dichotomous synchronic (binomial; logit link); fixed effects
  • 50000 burn-ins; 10000 recorded iterations (thinned from 500000)
Figure 88. Myopic CRS Historical data methods: within 1.0D – evidence network.

Figure 88Myopic CRS Historical data methods: within 1.0D – evidence network

Table 148Myopic CRS Historical data methods: within 1.0D – input data

SRKT Clinical historyDouble-KFeiz-MannisHofferQ DKHolladay2 DKMasketModified MasketSRKT DKSRKT Feiz-MannisSRKT Ladas-Stark
Saiki et al. (2013)8/126/1210/129/1214/28
Fam & (2008)11/3728/3730/3725/3723/3723/37

Table 149Myopic CRS Historical data methods: within 1.0D – relative effectiveness of all pairwise combinations (RR and 95% credible interval)

SRKT Clinical historyDouble-KFeiz-MannisHofferQ DKHolladay2 DKMasketModified MasketSRKT DKSRKT Feiz-MannisSRKT Ladas-Stark
SRKT Clinical history--7.35
(2.63, 20.60)		10.13
(3.43, 29.93)		--4.92
(1.84, 13.19)		3.88
(1.47, 10.23)		3.88
(1.47, 10.23)
Double-K10.80
(1.92, 69.43)		0.50
(0.10, 2.60)		--2.50
(0.36, 17.32)		1.50
(0.25, 8.84)		0.50
(0.12, 2.05)		--
Feiz-Mannis5.20
(0.93, 28.97)		0.48
(0.08, 2.57)		--5.00
(0.75, 33.21)		3.00
(0.53, 16.90)		1.00
(0.26, 3.87)		--
HofferQ DK7.76
(2.82, 23.22)		0.72
(0.11, 4.27)		1.51
(0.27, 8.84)		1.38
(0.45, 4.20)		--0.67
(0.24, 1.85)		0.53
(0.19, 1.44)		0.53
(0.19, 1.44)
Holladay2 DK10.94
(3.78, 34.76)		1.02
(0.15, 6.16)		2.14
(0.36, 12.56)		1.41
(0.45, 4.37)		--0.49
(0.17, 1.42)		0.38
(0.13, 1.10)		0.38
(0.13, 1.10)
Masket30.55
(4.54, 309.80)		2.80
(0.40, 27.85)		5.91
(0.92, 57.06)		3.89
(0.56, 40.10)		2.77
(0.38, 29.29)		0.60
(0.08, 4.45)		0.20
(0.04, 1.08)		--
Modified Masket16.88
(2.84, 124.70)		1.56
(0.25, 11.03)		3.27
(0.58, 22.60)		2.19
(0.34, 15.67)		1.55
(0.24, 11.70)		0.55
(0.06, 4.51)		0.33
(0.07, 1.50)		--
SRKT DK5.12
(1.93, 14.38)		0.48
(0.10, 1.98)		1.00
(0.25, 4.05)		0.66
(0.23, 1.83)		0.47
(0.15, 1.37)		0.17
(0.02, 0.86)		0.31
(0.05, 1.31)		0.79
(0.30, 2.05)		0.79
(0.30, 2.05)
SRKT Feiz-Mannis4.02
(1.53, 11.17)		0.37
(0.06, 2.07)		0.78
(0.14, 4.30)		0.52
(0.18, 1.41)		0.37
(0.12, 1.05)		0.13
(0.01, 0.89)		0.24
(0.03, 1.38)		0.79
(0.30, 2.05)		1.00
(0.39, 2.56)
SRKT Ladas-Stark4.02
(1.54, 11.05)		0.38
(0.06, 2.06)		0.78
(0.14, 4.32)		0.52
(0.18, 1.42)		0.37
(0.12, 1.04)		0.13
(0.01, 0.89)		0.24
(0.03, 1.39)		0.78
(0.29, 2.08)		1.00
(0.39, 2.58)
Figure 89. Myopic CRS Historical data methods: within 1.0D – relative effect of all options versus common comparator.

Figure 89Myopic CRS Historical data methods: within 1.0D – relative effect of all options versus common comparator

Table 150Myopic CRS Historical data methods: within 1.0D – rankings for each comparator

Probability bestMedian rank (95%CI)
SRKT Clinical history0.00010 (9, 10)
Double-K0.0764 (1, 9)
Feiz-Mannis0.0077 (2, 10)
HofferQ DK0.0215 (2, 8)
Holladay2 DK0.0784 (1, 7)
Masket0.5921 (1, 5)
Modified Masket0.2252 (1, 8)
SRKT DK0.0007 (4, 9)
SRKT Feiz-Mannis0.0008 (4, 9)
SRKT Ladas-Stark0.0008 (4, 9)
Figure 90. Myopic CRS Historical data methods: within 1.0D – rank probability histograms.

Figure 90Myopic CRS Historical data methods: within 1.0D – rank probability histograms

Table 151Myopic CRS Historical data methods: within 1.0D – model fit statistics

Residual devianceDbarDhatpDDIC
11.31
(compared to 11 datapoints)		49.35738.20411.15260.509

Table 152Myopic CRS Historical data methods: within 1.0D – notes

  • Dichotomous synchronic (binomial; logit link); fixed effects
  • 50000 burn-ins; 10000 recorded iterations (thinned from 500000)
Figure 91. Myopic CRS Historical data methods: within 2.0D – evidence network.

Figure 91Myopic CRS Historical data methods: within 2.0D – evidence network

Table 153Myopic CRS Historical data methods: within 2.0D – input data

SRKT Clinical historyHofferQ DKHolladay2 DKSRKT DKSRKT Feiz-MannisSRKT Ladas-Stark
Fam & (2008)33/3735/3734/3735/3732/3735/37

Table 154Myopic CRS Historical data methods: within 2.0D – relative effectiveness of all pairwise combinations (RR and 95% credible interval)

SRKT Clinical historyHofferQ DKHolladay2 DKSRKT DKSRKT Feiz-MannisSRKT Ladas-Stark
SRKT Clinical history2.12
(0.36, 12.36)		1.37
(0.29, 6.61)		2.12
(0.36, 12.36)		0.78
(0.19, 3.15)		2.12
(0.36, 12.36)
HofferQ DK2.32
(0.39, 19.84)		0.65
(0.10, 4.12)		1.00
(0.13, 7.50)		0.37
(0.07, 2.02)		1.00
(0.13, 7.50)
Holladay2 DK1.42
(0.27, 8.24)		0.61
(0.07, 4.32)		1.54
(0.24, 9.82)		0.56
(0.12, 2.56)		1.54
(0.24, 9.82)
SRKT DK2.33
(0.39, 19.87)		1.00
(0.10, 9.99)		1.63
(0.23, 15.74)		0.37
(0.07, 2.02)		1.00
(0.13, 7.50)
SRKT Feiz-Mannis0.76
(0.17, 3.29)		0.33
(0.04, 1.74)		0.53
(0.10, 2.50)		0.33
(0.04, 1.81)		2.73
(0.50, 15.09)
SRKT Ladas-Stark2.30
(0.39, 19.99)		0.99
(0.10, 9.99)		1.62
(0.23, 14.79)		0.99
(0.10, 10.03)		3.02
(0.56, 26.29)
Figure 92. Myopic CRS Historical data methods: within 2.0D – relative effect of all options versus common comparator.

Figure 92Myopic CRS Historical data methods: within 2.0D – relative effect of all options versus common comparator

Table 155Myopic CRS Historical data methods: within 2.0D – rankings for each comparator

Probability bestMedian rank (95%CI)
SRKT Clinical history0.0295 (1, 6)
HofferQ DK0.2922 (1, 6)
Holladay2 DK0.0934 (1, 6)
SRKT DK0.2942 (1, 6)
SRKT Feiz-Mannis0.0086 (2, 6)
SRKT Ladas-Stark0.2842 (1, 6)
Figure 93. Myopic CRS Historical data methods: within 2.0D – rank probability histograms.

Figure 93Myopic CRS Historical data methods: within 2.0D – rank probability histograms

Table 156Myopic CRS Historical data methods: within 2.0D – model fit statistics

Residual devianceDbarDhatpDDIC
6.394
(compared to 6 datapoints)		23.46517.5675.89829.364

Table 157Myopic CRS Historical data methods: within 2.0D – notes

  • Dichotomous synchronic (binomial; logit link); fixed effects
  • 50000 burn-ins; 10000 recorded iterations (thinned from 500000)

H.3.4. Intraocular lens constant optimisation – Network meta-analysis results

H.3.4.1. Model fit statistics for all outcomes

Table 158Model fit statistics used to select fixed or random effect models for all comparisons and outcomes

StudiesOutcomeModelTotal model DICTotal model DIC (FE – RE)Total residual devianceNo. of data-pointsBetween-study SD (95% CrI)Preferred model
4 (Charalampidou, Day, Lee, Sharma)Mean absolute errorFE−21.742.11617.8714-FE
RE−23.85613.560.216 (0.028, 1.093)
3 (Aristodemou, Day, Eom)Within 0.25DFE537.2426.3446.416-RE
RE110.915.31.05 (0.59, 1.87)
6 (Aristodemou, Charalampidou, Day, Eom, Fam, Lee)Within 0.5DFE254.756112.652137.420-RE
RE142.10419.720.900 (0.473, 1.776)
7 (Aristodemou, Charalampidou, Day, Eom, Fam, Lee, Sharma)Within 1.0DFE204.0654.23980.7324-RE
RE149.82122.770.611 (0.284, 1.489)
1 (Lee – pairwise comparison) Within 1.5DFE-----FE

H.3.4.2. Full dataset

Figure 94. Mean absolute error (RE) – evidence network.

Figure 94Mean absolute error (RE) – evidence network

Table 159Mean absolute error (RE) – input data

StandardIOLCOptimisedLenstarPezeroULIBULIBpersonalised
Sharma et al. (2014)0.56 (0.40)0.49 (0.50)
Day et al. (2012)0.47 (0.39)0.46 (0.39)
Day et al. (2012)0.84 (0.53)0.50 (0.37)
Day et al. (2012)0.89 (0.80)0.74 (0.58)
Day et al. (2012)0.88 (0.53)0.83 (0.61)
Charalampidou et al. (2010)0.38 (0.31)0.36 (0.30)
Lee et al. (2015)0.67 (0.52)0.55 (0.49)

Table 160Mean absolute error (RE) – relative effectiveness of all pairwise combinations (MD and 95% credible interval)

StandardIOLCOptimisedLenstarPezeroULIBULIBpersonalised
StandardIOLC−0.12
(−0.26, 0.02)		−0.13
(−0.33, 0.06)		−0.07
(−0.25, 0.11)		-
OptimisedLenstar−0.12
(−0.26, 0.02)		---
Pezero−0.08
(−0.17, 0.02)		0.05
(−0.12, 0.21)		--
ULIB−0.07
(−0.25, 0.10)		0.05
(−0.17, 0.27)		0.01
(−0.19, 0.20)		−0.02
(−0.08, 0.04)
ULIBpersonalised−0.09
(−0.27, 0.09)		0.03
(−0.20, 0.26)		−0.01
(−0.22, 0.19)		−0.02
(−0.08, 0.04)
Figure 95. Mean absolute error (RE) – relative effect of all options versus common comparator.

Figure 95Mean absolute error (RE) – relative effect of all options versus common comparator

Table 161Mean absolute error (RE) – rankings for each comparator

Probability bestMedian rank (95%CI)
StandardIOLC0.0015 (3, 5)
OptimisedLenstar0.4942 (1, 5)
Pezero0.1673 (1, 5)
ULIB0.0633 (1, 5)
ULIBpersonalised0.2752 (1, 5)
Figure 96. Mean absolute error (RE) – rank probability histograms.

Figure 96Mean absolute error (RE) – rank probability histograms

Table 162Mean absolute error (RE) – model fit statistics

Residual devianceDbarDhatpDDIC
17.87
(compared to 14 datapoints)		−32.725−43.70910.948−21.740

Table 163Mean absolute error (FE) – notes

  • Continuous (normal; identity link); fixed effects
  • 50000 burn-ins; 10000 recorded iterations
Figure 97. Within 0.25 dioptres (RE) – evidence network.

Figure 97Within 0.25 dioptres (RE) – evidence network

Table 164Within 0.25 dioptres (RE) – input data

StandardIOLCOptimisedKPezero
Eom et al. (2013)34/12338/123
Eom et al. (2013)49/11462/114
Day et al. (2012)4/3210/32
Day et al. (2012)33/10039/100
Day et al. (2012)2/193/19
Day et al. (2012)2/124/12
Aristodemou et al. (2011)1170/61592525/6159
Aristodemou et al. (2011)585/19491735/1949

Table 165Within 0.25 dioptres (RE) – relative effectiveness of all pairwise combinations (RR and 95% credible interval)

StandardIOLCOptimisedKPezero
StandardIOLC1.37
(0.94, 2.00)		 3.36
(1.11, 10.14)
OptimisedK1.37
(0.25, 7.44)		-
Pezero 3.66
(1.21, 10.00) 		2.66
(0.35, 19.12)
Figure 98. Within 0.25 dioptres (RE) – relative effect of all options versus common comparator.

Figure 98Within 0.25 dioptres (RE) – relative effect of all options versus common comparator

Table 166Within 0.25 dioptres (RE) – rankings for each comparator

Probability bestMedian rank (95%CI)
StandardIOLC0.0053 (2, 3)
OptimisedK0.1472 (1, 3)
Pezero0.8481 (1, 2)
Figure 99. Within 0.25 dioptres (RE) – rank probability histograms.

Figure 99Within 0.25 dioptres (RE) – rank probability histograms

Table 167Within 0.25 dioptres (RE) – model fit statistics

Residual devianceDbarDhatpDDICtau
15.33
(compared to 16 datapoints)		96.11381.32414.789110.9031.049 (95%CI: 0.585, 1.870)

Table 168Within 0.25 dioptres (RE) – notes

  • Dichotomous synchronic (binomial; logit link); random effects
  • Prior distribution for between-study heterogeneity: uniform (Min=0; Max=2)
  • 50000 burn-ins; 10000 recorded iterations (thinned from 500000)
Figure 100. Within 0.5 dioptres (RE) – evidence network.

Figure 100Within 0.5 dioptres (RE) – evidence network

Table 169Within 0.5 dioptres (RE) – input data

StandardIOLCOptimisedKOptimisedKALOptimisedLenstarPezero
Eom et al. (2013)68/12378/123
Eom et al. (2013)84/11490/114
Day et al. (2012)10/3218/32
Day et al. (2012)62/10060/100
Day et al. (2012)8/199/19
Day et al. (2012)4/124/12
Aristodemou et al. (2011)2587/61594373/6159
Aristodemou et al. (2011)1111/19491793/1949
Fam & (2009)43/9057/90
Lee et al. (2015)46/10062/100

Table 170Within 0.5 dioptres (RE) – relative effectiveness of all pairwise combinations (RR and 95% credible interval)

StandardIOLCOptimisedKOptimisedKALOptimisedLenstarPezero
StandardIOLC1.38
(0.93, 2.04)		1.89
(1.04, 3.43)		1.92
(1.09, 3.37)		2.51
(1.31, 4.81)
OptimisedK1.37
(0.31, 6.35)		---
OptimisedKAL1.91
(0.21, 16.86)		1.38
(0.10, 18.75)		--
OptimisedLenstar1.92
(0.22, 16.46)		1.40
(0.10, 19.34)		1.01
(0.05, 21.58)		-
Pezero2.49
(0.92, 5.88)		1.81
(0.28, 9.91)		1.30
(0.12, 13.16)		1.30
(0.12, 12.99)
Figure 101. Within 0.5 dioptres (RE) – relative effect of all options versus common comparator.

Figure 101Within 0.5 dioptres (RE) – relative effect of all options versus common comparator

Table 171SENSITIVITY ANALYSIS: Within 0.5 dioptres (RE) – rankings for each comparator

Probability bestMedian rank (95%CI)
StandardIOLC0.0014 (2, 5)
OptimisedK0.1033 (1, 5)
OptimisedKAL0.2823 (1, 5)
OptimisedLenstar0.2823 (1, 5)
Pezero0.3322 (1, 4)
Figure 102. SENSITIVITY ANALYSIS: Within 0.5 dioptres (RE) – rank probability histograms.

Figure 102SENSITIVITY ANALYSIS: Within 0.5 dioptres (RE) – rank probability histograms

Table 172SENSITIVITY ANALYSIS: Within 0.5 dioptres (RE) – model fit statistics

Residual devianceDbarDhatpDDICtau
19.72
(compared to 20 datapoints)		123.021103.93819.083142.1040.900 (95%CI: 0.473, 1.776)

Table 173SENSITIVITY ANALYSIS: Within 0.5 dioptres (RE) – notes

  • Dichotomous synchronic (binomial; logit link); random effects
  • Prior distribution for between-study heterogeneity: uniform (Min=0; Max=2)
  • 50000 burn-ins; 10000 recorded iterations
Figure 103. Within 1.0 dioptre (RE) – evidence network.

Figure 103Within 1.0 dioptre (RE) – evidence network

Table 174Within 1.0 dioptre (RE) – input data

StandardIOLCOptimisedKOptimisedKALOptimisedLenstarPezeroULIBULIBpersonalised
Sharma et al. (2014)44/5144/51
Eom et al. (2013)106/123111/123
Eom et al. (2013)110/114111/114
Day et al. (2012)23/3228/32
Day et al. (2012)91/10092/100
Day et al. (2012)12/1914/19
Day et al. (2012)6/126/12
Aristodemou et al. (2011)4989/61595851/6159
Aristodemou et al. (2011)1735/19491813/1949
Charalampidou et al. (2010)205/214205/214
Fam & (2009)69/9081/90
Lee et al. (2015)76/10082/100

Table 175Within 1.0 dioptre (RE) – relative effectiveness of all pairwise combinations (RR and 95% credible interval)

StandardIOLCOptimisedKOptimisedKALOptimisedLenstarPezeroULIBULIBpersonalised
StandardIOLC1.45
(0.72, 2.92)		2.74
(1.18, 6.37)		1.44
(0.72, 2.86)		2.01
(1.03, 3.91)		1.00
(0.32, 3.09)		-
OptimisedK1.45
(0.38, 5.49)		-----
OptimisedKAL2.86
(0.50, 16.69)		1.96
(0.22, 18.33)		----
OptimisedLenstar1.45
(0.28, 7.79)		1.01
(0.12, 8.56)		0.51
(0.04, 5.71)		---
Pezero2.12
(0.97, 4.17)		1.46
(0.30, 6.42)		0.74
(0.11, 4.55)		1.47
(0.22, 8.37)		--
ULIB1.02
(0.15, 7.02)		0.69
(0.07, 7.62)		0.36
(0.03, 4.93)		0.70
(0.06, 9.19)		0.48
(0.06, 4.03)		1.00
(0.39, 2.57)
ULIBpersonalised1.03
(0.07, 15.11)		0.71
(0.04, 15.08)		0.37
(0.01, 8.83)		0.72
(0.03, 17.42)		0.49
(0.03, 8.03)		1.02
(0.17, 6.18)
Figure 104. Within 1.0 dioptre (RE) – relative effect of all options versus common comparator.

Figure 104Within 1.0 dioptre (RE) – relative effect of all options versus common comparator

Table 176Within 1.0 dioptre (RE) – rankings for each comparator

Probability bestMedian rank (95%CI)
StandardIOLC0.0005 (3, 7)
OptimisedK0.0894 (1, 7)
OptimisedKAL0.4522 (1, 7)
OptimisedLenstar0.1214 (1, 7)
Pezero0.1553 (1, 6)
ULIB0.0465 (1, 7)
ULIBpersonalised0.1375 (1, 7)
Figure 105. Within 1.0 dioptre (RE) – rank probability histograms.

Figure 105Within 1.0 dioptre (RE) – rank probability histograms

Table 177Within 1.0 dioptre (RE) – model fit statistics

Residual devianceDbarDhatpDDICtau
22.77
(compared to 24 datapoints)		128.075106.3321.746149.8210.611 (95%CI: 0.284, 1.489)

Table 178Within 1.0 dioptre (RE) – notes

  • Dichotomous synchronic (binomial; logit link); random effects
  • Prior distribution for between-study heterogeneity: uniform (Min=0; Max=2)
  • 50000 burn-ins; 10000 recorded iterations
PROPORTION WITHIN 1.5D – pairwise comparison.

PROPORTION WITHIN 1.5D – pairwise comparison

H.3.4.3. Wang 2011 (people with axial lengths >25mm): unconnected to networks

Prediction error.

Prediction error

Proportion of eyes with a hyperopic refractive outcome.

Proportion of eyes with a hyperopic refractive outcome

H.3.5. Other considerations in biometry: Forest plots of outcomes

H.3.5.1. Second eye prediction refinement

50% adjusted of first eye prediction error vs unadjusted prediction
Mean absolute prediction errors.

Mean absolute prediction errors

Cumulative proportion of eyes within various ranges of absolute prediction errors.

Cumulative proportion of eyes within various ranges of absolute prediction errors

Mean prediction errors.

Mean prediction errors

H.3.5.2. Comparison of results from using 100% adjusted and 50% adjusted first eye prediction error

Mean absolute prediction errors.

Mean absolute prediction errors

Cumulative proportion of eyes within various ranges of absolute prediction errors.

Cumulative proportion of eyes within various ranges of absolute prediction errors

Mean prediction errors.

Mean prediction errors

H.4. Intraocular lens selection

  • Are different lens design (aspheric vs. spheric, plate vs. loop) effective in improving postoperative vision (refractive outcomes, optical aberrations) in cataract surgery?
  • Are different lens design (square-edged vs. round-edge, plate vs. loop) and material (hydrophilic acrylic, hydrophobic acrylic, collagen, hydroxyethyl methacrylate-based vs. silicone-based) effective in preventing posterior capsule opacification in cataract surgery?
  • Are tinted lenses effective in preventing the progression of age-related macular degeneration compared with colourless lenses in cataract surgery?
  • What is the optimal strategy to facilitate simultaneous distance and near vision following cataract surgery?
  • What is the optimal strategy to address pre-existing astigmatism in people undergoing cataract surgery?

H.4.1. Lens design

H.4.1.1. PMMA versus silicone

PCO score.

PCO score

Nd:YAG capsulotomy rate.

Nd:YAG capsulotomy rate

H.4.1.2. PMMA versus hydrophobic acrylic

Nd:YAG capsulotomy rate.

Nd:YAG capsulotomy rate

H.4.1.3. Hydrophobic acrylic versus silicone

BCDVA (decimal acuity).

BCDVA (decimal acuity)

PCO score.

PCO score

Nd:YAG capsulotomy rate.

Nd:YAG capsulotomy rate

Lens decentration (mm).

Lens decentration (mm)

Lens tilt (degrees).

Lens tilt (degrees)

H.4.1.4. Hydrophobic acrylic versus hydrophilic acrylic

BCDVA (decimal acuity).

BCDVA (decimal acuity)

Nd:YAG capsulotomy rate.

Nd:YAG capsulotomy rate

H.4.1.5. Network meta-analyses (lens material)

PCO score (hydrophobic acrylic as reference category).

PCO score (hydrophobic acrylic as reference category)

Pairwise mean differences from NMA

SiliconeHydrophobic acrylicPMMAHydrophilic acrylic
SiliconeN/A
Hydrophobic acrylic0.20 (−0.28, 0.68)N/A
PMMA7.99 (3.87, 12.11)7.79 (3.64, 11.94)N/A
Hydrophilic acrylic75.50 (68.90, 82.09)75.30 (68.71, 81.89)67.51 (59.94, 75.08)N/A

Quantifying heterogeneity/inconsistency:

tau^2 = 0.1953; I^2 = 94.8%

Network graph

Image apphf137

Comparison of direct and indirect evidence

Random effects model:  
                              comparison k prop      nma   direct    indir.     Diff     z  p-value 
 Hydrophilic acrylic:Hydrophobic acrylic 1 0.66 −75.2968 −94.2000  −39.0404 −55.1596 −7.79 < 0.0001 
                Hydrophilic acrylic:PMMA 1 0.25 −67.5094 −17.0000  −84.4988  67.4988  7.59 < 0.0001 
            Hydrophilic acrylic:Silicone 1 0.31 −75.4992 −45.2000  −89.2012  44.0012  6.06 < 0.0001 
                Hydrophobic acrylic:PMMA 0 0.00   7.7875        .    7.7875        .     .        . 
            Hydrophobic acrylic:Silicone 7 1.00  −0.2024  −0.2678   54.1836 −54.4513 −7.74 < 0.0001 
                           PMMA:Silicone 3 0.99  −7.9899  −6.6522 −209.3312 202.6790  7.80 < 0.0001 
 Legend: 
 comparison - Treatment comparison 
 prop       - Direct evidence proportion 
 nma        - Estimated treatment effect (MD) in network meta-analysis 
 direct     - Estimated treatment effect (MD) derived from direct evidence 
 indir.     - Estimated treatment effect (MD) derived from indirect evidence 
 Diff       - Difference between direct and indirect treatment estimates 
 z          - z-value of test for disagreement (direct versus indirect) 
 p-value    - p-value of test for disagreement (direct versus indirect)
PCO score (without hydrophilic acrylic - hydrophobic acrylic as reference category).

PCO score (without hydrophilic acrylic - hydrophobic acrylic as reference category)

Pairwise mean differences from NMA

SiliconeHydrophobic acrylicPMMA
SiliconeN/A
Hydrophobic acrylic0.23 (−0.16, 0.62)N/A
PMMA6.65 (2.54, 10.77)6.43 (2.29, 10.56)N/A

Quantifying heterogeneity/inconsistency:

tau^2 = 0.1242; I^2 = 92.9%

Network graph

Image apphf139

Comparison of direct and indirect evidence

                   comparison k prop     nma  direct indir. Diff z p-value 
     Hydrophobic acrylic:PMMA 0    0  6.4250       . 6.4250    . .       . 
 Hydrophobic acrylic:Silicone 7    1 −0.2293 −0.2293      .    . .       . 
                PMMA:Silicone 3    1 −6.6544 −6.6544      .    . .       .  
Legend: 
 comparison - Treatment comparison 
 prop       - Direct evidence proportion 
 nma        - Estimated treatment effect (MD) in network meta-analysis 
 direct     - Estimated treatment effect (MD) derived from direct evidence 
 indir.     - Estimated treatment effect (MD) derived from indirect evidence 
 Diff       - Difference between direct and indirect treatment estimates 
 z          - z-value of test for disagreement (direct versus indirect) 
 p-value    - p-value of test for disagreement (direct versus indirect)
Nd:YAG capsulotomy rate (hydrophobic acrylic as reference category).

Nd:YAG capsulotomy rate (hydrophobic acrylic as reference category)

Pairwise relative risks from NMA

SiliconeHydrophobic acrylicPMMAHydrophilic acrylic
SiliconeN/A
Hydrophobic acrylic1.02 (0.47, 2.23)N/A
PMMA2.98 (1.41, 6.32)2.90 (1.27, 6.63)N/A
Hydrophilic acrylic3.33 (1.17, 9.50)3.24 (1.38, 7.61)1.12 (0.41, 3.00)N/A

Quantifying heterogeneity/inconsistency:

tau^2 = 0.5280; I^2 = 54.9%

Network graph

Image apphf141

Comparison of direct and indirect evidence

Random effects model:  
                              comparison k prop    nma direct indir.    RoR     z  p-value 
 Hydrophilic acrylic:Hydrophobic acrylic 6 0.75 0.3090 0.2394 0.6723 0.3560 −1.02   0.3071 
                Hydrophilic acrylic:PMMA 1 0.47 0.8965 1.5472 0.5507 2.8094  1.02   0.3070 
            Hydrophilic acrylic:Silicone 0 0.00 0.3006      . 0.3006      .     .        . 
                Hydrophobic acrylic:PMMA 2 0.39 2.9013 6.0780 1.8222 3.3356  1.39   0.1645 
            Hydrophobic acrylic:Silicone 8 0.72 0.9727 0.6233 3.1085 0.2005 −1.82   0.0685 
                           PMMA:Silicone 6 0.74 0.3353 0.5477 0.0842 6.5060  2.15   0.0315 
 Legend: 
 comparison - Treatment comparison 
 prop       - Direct evidence proportion 
 nma        - Estimated treatment effect (RR) in network meta-analysis 
 direct     - Estimated treatment effect (RR) derived from direct evidence 
 indir.     - Estimated treatment effect (RR) derived from indirect evidence 
 RoR        - Ratio of Ratios (direct versus indirect) 
 z          - z-value of test for disagreement (direct versus indirect) 
 p-value    - p-value of test for disagreement (direct versus indirect)

H.4.1.6. Square-edge versus round-edge

BCDVA (decimal acuity).

BCDVA (decimal acuity)

PCO score.

PCO score

Nd:YAG capsulotomy rate.

Nd:YAG capsulotomy rate

H.4.1.7. Loop versus 3-piece

UCDVA (logMAR).

UCDVA (logMAR)

BCDVA (logMAR).

BCDVA (logMAR)

BCDVA (decimal acuity).

BCDVA (decimal acuity)

PCO score.

PCO score

Nd:YAG capsulotomy rate.

Nd:YAG capsulotomy rate

Lens decentration (mm).

Lens decentration (mm)

Lens tilt (degrees).

Lens tilt (degrees)

H.4.1.8. Aspheric versus spheric

UCDVA (logMAR).

UCDVA (logMAR)

BCDVA (logMAR).

BCDVA (logMAR)

BCDVA (decimal acuity).

BCDVA (decimal acuity)

Contrast sensitivity (Pelli-Robson test).

Contrast sensitivity (Pelli-Robson test)

Spherical aberrations.

Spherical aberrations

Higher-order aberrations.

Higher-order aberrations

Comatic aberrations.

Comatic aberrations

PCO score.

PCO score

H.4.2. Tinted vs colourless lenses

H.4.2.1. Post-operative best corrected visual acuity (logMAR)

Image apphf160

H.4.2.2. Post-operative best corrected visual acuity (logMAR) – OECD only

Image apphf161

H.4.2.3. Overall post-operative colour vision

Image apphf162

H.4.2.4. Post-operative colour vision in the blue light spectrum (photopic)

Image apphf163

H.4.2.5. Post-operative colour vision in the blue light spectrum (photopic) – OECD only

Image apphf164

H.4.2.6. Post-operative colour vision in the blue light spectrum (mesopic)

Image apphf165

H.4.2.7. Post-operative colour vision in the blue light spectrum (mesopic) – OECD only

Image apphf166

H.4.3. Multifocal vs monofocal intraocular lenses

H.4.3.1. Multifocal versus monofocal

Uncorrected distance visual acuity worse than 6/6.

Uncorrected distance visual acuity worse than 6/6

Corrected distance visual acuity worse than 6/6.

Corrected distance visual acuity worse than 6/6

Uncorrected near visual acuity worse than J3/J4 or equivalent.

Uncorrected near visual acuity worse than J3/J4 or equivalent

Mean uncorrected distance visual acuity (logMAR).

Mean uncorrected distance visual acuity (logMAR)

Mean corrected distance visual acuity.

Mean corrected distance visual acuity

Mean uncorrected intermediate visual acuity.

Mean uncorrected intermediate visual acuity

Mean corrected intermediate visual acuity.

Mean corrected intermediate visual acuity

Mean uncorrected near visual acuity.

Mean uncorrected near visual acuity

Mean corrected near visual acuity.

Mean corrected near visual acuity

Spectacle dependence (any).

Spectacle dependence (any)

Spectacle dependence (distance or near).

Spectacle dependence (distance or near)

Contrast sensitivity.

Contrast sensitivity

Visual function.

Visual function

Patient satisfaction.

Patient satisfaction

Glare.

Glare

Halos.

Halos

H.4.3.2. Multifocal versus monovision

Visual acuity.

Visual acuity

Spectacle dependence.

Spectacle dependence

Contrast sensitivity.

Contrast sensitivity

H.4.3.3. Refractive vs diffractive multifocal lenses

Uncorrected distance visual acuity.

Uncorrected distance visual acuity

Spectacle dependence.

Spectacle dependence

Glare.

Glare

Halo.

Halo

H.4.3.4. Bifocal versus trifocal intraocular lenses

Uncorrected visual acuity.

Uncorrected visual acuity

Corrected visual acuity.

Corrected visual acuity

H.4.4. Multifocal vs monofocal intraocular lenses: network meta-analyses (monofocal lenses used as reference category)

H.4.4.1. Uncorrected distance visual acuity

Class-level analysis.

Class-level analysis

Pairwise mean differences from NMA (higher number favour column)

MonovisionMonofocalMultifocal
MonovisionN/A
Monofocal0.01 (−0.09, 0.11)N/A
Multifocal0.02 (−0.07, 0.11)0.01 (−0.03, 0.05)N/A

Quantifying heterogeneity/inconsistency:

tau^2 = 0.0017; I^2 = 74.3%

Network graph

Image apphf193

Comparison of direct and indirect evidence

Random effects model:  
            comparison prop     nma direct  indir. Diff z p-value 
  Monofocal:Monovision    0 −0.0126      . −0.0126    . .       . 
  Monofocal:Multifocal    1  0.0074 0.0074       .    . .       . 
 Monovision:Multifocal    1  0.0200 0.0200       .    . .       . 
Legend: 
 comparison - Treatment comparison 
 prop       - Direct evidence proportion 
 nma        - Estimated treatment effect (MD) in network meta-analysis 
 direct     - Estimated treatment effect (MD) derived from direct evidence 
 indir.     - Estimated treatment effect (MD) derived from indirect evidence 
 Diff       - Difference between direct and indirect treatment estimates 
 z          - z-value of test for disagreement (direct versus indirect) 
 p-value    - p-value of test for disagreement (direct versus indirect)
Subdivided analysis.

Subdivided analysis

Pairwise mean differences from NMA (higher numbers favour column)

RefractiveMonofocalMonovisionDiffractive
RefractiveN/A
Monofocal0.01 (−0.03, 0.04)N/A
Monovision0.02 (−0.06, 0.10)0.01 (−0.07, 0.09)N/A
Diffractive0.04 (0.01, 0.07)0.03 (−0.00, 0.07)0.02 (−0.05, 0.09)N/A

Quantifying heterogeneity/inconsistency:

tau^2 = 0.0010; I^2 = 64.3%

Network graph

Image apphf195

Comparison of direct and indirect evidence

Random effects model:  
             comparison prop     nma  direct  indir.    Diff     z  p-value 
  Diffractive:Monofocal 0.68 −0.0311 −0.0190 −0.0567  0.0377  0.98   0.3255 
 Diffractive:Monovision 1.00 −0.0200 −0.0200       .       .     .        . 
 Diffractive:Refractive 0.90 −0.0393 −0.0460  0.0241 −0.0702 −1.50   0.1335 
   Monofocal:Monovision 0.00  0.0111       .  0.0111       .     .        . 
   Monofocal:Refractive 0.78 −0.0082  0.0107 −0.0755  0.0862  2.06   0.0394 
  Monovision:Refractive 0.00 −0.0193       . −0.0193       .     .        . 
Legend: 
 comparison - Treatment comparison 
 prop       - Direct evidence proportion 
 nma        - Estimated treatment effect (MD) in network meta-analysis 
 direct     - Estimated treatment effect (MD) derived from direct evidence 
 indir.     - Estimated treatment effect (MD) derived from indirect evidence 
 Diff       - Difference between direct and indirect treatment estimates 
 z          - z-value of test for disagreement (direct versus indirect) 
 p-value    - p-value of test for disagreement (direct versus indirect)

H.4.4.2. Uncorrected near visual acuity

Class-level analysis.

Class-level analysis

Pairwise mean differences from NMA (higher number favour column)

MultifocalMonovisionMonofocal
MultifocalN/A
Monovision0.04 (−0.36, 0.44)N/A
Monofocal0.20 (0.04, 0.37)0.16 (−0.27, 0.60)N/A

Quantifying heterogeneity/inconsistency:

tau^2 = 0.0416; I^2 = 98.3%

Network graph

Image apphf197

Comparison of direct and indirect evidence

Random effects model:  
            comparison k prop     nma  direct  indir. Diff z p-value 
  Monofocal:Monovision 0    0 −0.1631       . −0.1631    . .       . 
  Monofocal:Multifocal 6    1 −0.2031 −0.2031       .    . .       . 
 Monovision:Multifocal 1    1 −0.0400 −0.0400       .    . .       . 
Legend: 
 comparison - Treatment comparison 
 prop       - Direct evidence proportion 
 nma        - Estimated treatment effect (MD) in network meta-analysis 
 direct     - Estimated treatment effect (MD) derived from direct evidence 
 indir.     - Estimated treatment effect (MD) derived from indirect evidence 
 Diff       - Difference between direct and indirect treatment estimates 
 z          - z-value of test for disagreement (direct versus indirect) 
 p-value    - p-value of test for disagreement (direct versus indirect)
Subdivided analysis.

Subdivided analysis

Pairwise mean differences from NMA (higher numbers favour column)

DiffractiveMonovisionRefractiveMonofocal
DiffractiveN/A
Monovision0.04 (−0.33, 0.41)N/A
Refractive0.18 (−0.07, 0.43)0.14 (−0.31, 0.59)N/A
Monofocal0.30 (0.09, 0.50)0.26 (−0.17, 0.68)0.12 (−0.07, 0.30)N/A

Quantifying heterogeneity/inconsistency:

tau^2 = 0.0359; I^2 = 98.1%

Network graph

Image apphf199

Comparison of direct and indirect evidence

Random effects model:  
             comparison k prop     nma  direct  indir.    Diff     z  p-value 
  Diffractive:Monofocal 3 0.93  0.2967  0.2958  0.3074 −0.0115 −0.03   0.9776 
 Diffractive:Monovision 1 1.00  0.0400  0.0400       .       .     .        . 
 Diffractive:Refractive 1 0.46  0.1781  0.1376  0.2118 −0.0742 −0.29   0.7741 
   Monofocal:Monovision 0 0.00 −0.2567       . −0.2567       .     .        . 
   Monofocal:Refractive 4 0.95 −0.1186 −0.1075 −0.3348  0.2273  0.52   0.6034 
  Monovision:Refractive 0 0.00  0.1381       .  0.1381       .     .        . 
Legend: 
 comparison - Treatment comparison 
 prop       - Direct evidence proportion 
 nma        - Estimated treatment effect (MD) in network meta-analysis 
 direct     - Estimated treatment effect (MD) derived from direct evidence 
 indir.     - Estimated treatment effect (MD) derived from indirect evidence 
 Diff       - Difference between direct and indirect treatment estimates 
 z          - z-value of test for disagreement (direct versus indirect) 
 p-value    - p-value of test for disagreement (direct versus indirect)

H.4.4.3. Spectacle dependence

Class-level analysis.

Class-level analysis

Pairwise relative risks from NMA (higher number favour column)

MultifocalMonofocalMonovision
MultifocalN/A
Monofocal1.52 (1.31, 1.76)N/A
Monovision2.47 (1.62, 3.78)1.63 (1.04, 2.53)N/A

Quantifying heterogeneity/inconsistency:

tau^2 = 0.0436; I^2 = 81.0%

Network graph

Image apphf201

Comparison of direct and indirect evidence

Random effects model:   
            comparison  k prop    nma direct indir. RoR z p-value  
  Monofocal:Monovision  0    0 1.6253      . 1.6253   . .       .  
  Monofocal:Multifocal 11    1 0.6576 0.6576      .   . .       .  
 Monovision:Multifocal  2    1 0.4046 0.4046      .   . .       .  
Legend:  
 comparison - Treatment comparison  
 prop       - Direct evidence proportion  
 nma        - Estimated treatment effect (RR) in network meta-analysis  
 direct     - Estimated treatment effect (RR) derived from direct evidence  
 indir.     - Estimated treatment effect (RR) derived from indirect evidence  
 RoR        - Ratio of Ratios (direct versus indirect)  
 z          - z-value of test for disagreement (direct versus indirect)  
 p-value    - p-value of test for disagreement (direct versus indirect)
Subdivided analysis.

Subdivided analysis

Pairwise mean differences from NMA (higher numbers favour column)

DiffractiveRefractiveMonofocalMonovision
DiffractiveN/A
Refractive1.81 (1.33, 2.47)N/A
Monofocal2.18 (1.65, 2.89)1.21 (0.94, 1.54)N/A
Monovision3.16 (1.83, 5.46)1.75 (1.00, 3.06)1.45 (0.82, 2.57)N/A

Quantifying heterogeneity/inconsistency:

tau^2 = 0.0970; I^2 = 83.9%

Network graph

Image apphf203

Comparison of direct and indirect evidence

Random effects model:  
           comparison k prop    nma direct indir.    RoR     z  p-value 
  Diffractive:Monofocal 5 0.73 2.1829 1.8063 3.6192 0.4991 −2.16   0.0306 
 Diffractive:Monovision 1 0.61 3.1627 2.5830 4.3502 0.5938 −0.91   0.3623 
 Diffractive:Refractive 5 0.42 1.8101 3.2402 1.1804 2.7450  3.16   0.0016 
   Monofocal:Monovision 0 0.00 1.4488      . 1.4488      .     .        . 
   Monofocal:Refractive 6 0.81 0.8292 0.7148 1.5731 0.4544 −2.46   0.0139 
  Monovision:Refractive 1 0.47 0.5723 0.4345 0.7318 0.5937 −0.91   0.3623 
Legend: 
 comparison - Treatment comparison 
 prop       - Direct evidence proportion 
 nma        - Estimated treatment effect (RR) in network meta-analysis 
 direct     - Estimated treatment effect (RR) derived from direct evidence 
 indir.     - Estimated treatment effect (RR) derived from indirect evidence 
 RoR        - Ratio of Ratios (direct versus indirect) 
 z          - z-value of test for disagreement (direct versus indirect) 
 p-value    - p-value of test for disagreement (direct versus indirect)

H.4.4.4. Contrast sensitivity

Class-level analysis.

Class-level analysis

Pairwise relative risks from NMA (higher number favour row)

MultifocalMonovisionMonofocal
MultifocalN/A
Monovision0.04 (0.00, 0.07)N/A
Monofocal0.09 (−0.08, 0.26)0.05 (−0.12, 0.22)N/A

Quantifying heterogeneity/inconsistency:

tau^2 = 0.0010; I^2 = 23.8%

Network graph

Image apphf205

Comparison of direct and indirect evidence

Fixed effect model:  
            comparison prop     nma  direct  indir. Diff z p-value 
  Monofocal:Monovision    0 −0.0520       . −0.0520    . .       . 
  Monofocal:Multifocal    1 −0.0894 −0.0894       .    . .       . 
 Monovision:Multifocal    1 −0.0373 −0.0373       .    . .       . 
Legend: 
 comparison - Treatment comparison 
 prop       - Direct evidence proportion 
 nma        - Estimated treatment effect (MD) in network meta-analysis 
 direct     - Estimated treatment effect (MD) derived from direct evidence 
 indir.     - Estimated treatment effect (MD) derived from indirect evidence 
 Diff       - Difference between direct and indirect treatment estimates 
 z          - z-value of test for disagreement (direct versus indirect) 
 p-value    - p-value of test for disagreement (direct versus indirect)
Subdivided analysis.

Subdivided analysis

Pairwise mean differences from NMA (higher numbers favour row)

DiffractiveMonovisionRefractiveMonofocal
DiffractiveN/A
Monovision0.06 (0.02, 0.11)N/A
Refractive0.07 (−0.01, 0.15)0.01 (−0.06, 0.08)N/A
Monofocal0.15 (0.04, 0.27)0.09 (−0.02, 0.20)0.08 (−0.00, 0.17)N/A

Quantifying heterogeneity/inconsistency:

tau^2 = 0.0016; I^2 = 2.9%

Network graph

Image apphf207

Comparison of direct and indirect evidence

Fixed effect model:   
             comparison prop     nma  direct  indir. Diff z p-value  
  Diffractive:Monofocal    0  0.1524       .  0.1524    . .       .  
 Diffractive:Monovision    1  0.0600  0.0600       .    . .       .  
 Diffractive:Refractive    0  0.0700       .  0.0700    . .       .  
   Monofocal:Monovision    0 −0.0924       . −0.0924    . .       .  
   Monofocal:Refractive    1 −0.0824 −0.0824       .    . .       .  
  Monovision:Refractive    1  0.0100  0.0100       .    . .       .  
Legend:  
 comparison - Treatment comparison  
 prop       - Direct evidence proportion  
 nma        - Estimated treatment effect (MD) in network meta-analysis 
 direct     - Estimated treatment effect (MD) derived from direct evidence 
 indir.     - Estimated treatment effect (MD) derived from indirect evidence 
 Diff       - Difference between direct and indirect treatment estimates 
 z          - z-value of test for disagreement (direct versus indirect) 
 p-value    - p-value of test for disagreement (direct versus indirect)

H.4.4.5. Glare

Class-level analysis.

Class-level analysis

Pairwise relative risks from NMA (higher number favour column)

MonofocalMonovisionMultifocal
MonofocalN/A
Monovision0.86 (0.66, 1.12)N/A
Multifocal1.21 (1.03, 1.43)1.41 (1.14, 1.73)N/A

Quantifying heterogeneity/inconsistency:

tau^2 = 0; I^2 = 0%

Network graph

Image apphf209

Comparison of direct and indirect evidence

Fixed effect model:  
            comparison k prop    nma direct indir. RoR z p-value 
  Monofocal:Monovision 0    0 0.8596      . 0.8596   . .       . 
  Monofocal:Multifocal 8    1 1.2103 1.2103      .   . .       . 
 Monovision:Multifocal 1    1 1.4079 1.4079      .   . .       . 
Legend: 
 comparison - Treatment comparison 
 prop       - Direct evidence proportion 
 nma        - Estimated treatment effect (RR) in network meta-analysis 
 direct     - Estimated treatment effect (RR) derived from direct evidence 
 indir.     - Estimated treatment effect (RR) derived from indirect evidence 
 RoR        - Ratio of Ratios (direct versus indirect) 
 z          - z-value of test for disagreement (direct versus indirect) 
 p-value    - p-value of test for disagreement (direct versus indirect)
Subdivided analysis.

Subdivided analysis

Pairwise mean differences from NMA (higher numbers favour row)

MonovisionMonofocalDiffractiveRefractive
MonovisionN/A
Monofocal1.22 (0.93, 1.60)N/A
Diffractive1.41 (1.14, 1.73)1.15 (0.97, 1.37)N/A
Refractive1.82 (1.35, 2.43)1.49 (1.18, 1.88)1.29 (1.05, 1.59)N/A

Quantifying heterogeneity/inconsistency:

tau^2 = 0; I^2 = 0%

Network graph

Image apphf211

Comparison of direct and indirect evidence

Fixed effect model:  
             comparison k prop    nma direct indir.    RoR     z  p-value 
  Diffractive:Monofocal 3 0.84 0.8670 0.8690 0.8565 1.0146  0.06   0.9520 
 Diffractive:Monovision 1 1.00 0.7103 0.7103      .      .     .        . 
 Diffractive:Refractive 4 0.74 1.2892 1.2878 1.2931 0.9959 −0.02   0.9863 
   Monofocal:Monovision 0 0.00 0.8193      . 0.8193      .     .        . 
   Monofocal:Refractive 5 0.43 1.4870 1.4984 1.4785 1.0134  0.06   0.9556 
  Monovision:Refractive 0 0.00 1.8151      . 1.8151      .     .        . 
Legend: 
 comparison - Treatment comparison 
 prop       - Direct evidence proportion 
 nma        - Estimated treatment effect (RR) in network meta-analysis 
 direct     - Estimated treatment effect (RR) derived from direct evidence 
 indir.     - Estimated treatment effect (RR) derived from indirect evidence 
 RoR        - Ratio of Ratios (direct versus indirect) 
 z          - z-value of test for disagreement (direct versus indirect) 
 p-value    - p-value of test for disagreement (direct versus indirect)

H.4.4.6. Halo

Subdivided analysis.

Subdivided analysis

Pairwise relative risks from NMA (higher number favour column)

MonofocalDiffractiveRefractive
MonofocalN/A
Diffractive1.83 (1.50, 2.24)N/A
Refractive2.71 (2.05, 3.59)1.48 (1.21, 1.81)N/A

Quantifying heterogeneity/inconsistency:

tau^2 = 0; I^2 = 0%

Network graph

Image apphf213

Comparison of direct and indirect evidence

Fixed effect model:  
             comparison k prop    nma direct indir.    RoR     z  p-value 
  Diffractive:Monofocal 4 0.97 0.5459 0.5539 0.3372 1.6427  0.82   0.4150 
 Diffractive:Refractive 4 0.97 1.4808 1.4535 2.5310 0.5743 −0.97   0.3338 
   Monofocal:Refractive 4 0.07 2.7125 4.6366 2.6083 1.7777  1.01   0.3108 
Legend: 
 comparison - Treatment comparison 
 prop       - Direct evidence proportion 
 nma        - Estimated treatment effect (RR) in network meta-analysis 
 direct     - Estimated treatment effect (RR) derived from direct evidence 
 indir.     - Estimated treatment effect (RR) derived from indirect evidence 
 RoR        - Ratio of Ratios (direct versus indirect) 
 z          - z-value of test for disagreement (direct versus indirect) 
 p-value    - p-value of test for disagreement (direct versus indirect)

H.4.5. Optimal strategy to address pre-existing astigmatism

Mean Visual Acuity – uncorrected distance (logMAR).

Mean Visual Acuity – uncorrected distance (logMAR)

Proportion with visual acuity worse than 20/25 – uncorrected distance.

Proportion with visual acuity worse than 20/25 – uncorrected distance

Mean Visual Acuity – corrected distance (logMAR).

Mean Visual Acuity – corrected distance (logMAR)

Residual astigmatism (Refractive cylinder dioptres).

Residual astigmatism (Refractive cylinder dioptres)

Spectacle dependence.

Spectacle dependence

Note: Non OECD country studies removed for meta-analysis

H.5. Wrong lens implant errors

  • What are the procedural causes of wrong lens implant errors?
  • What strategies should be adopted to reduce the risk of wrong lens implant errors?

There were no meta-analyses conducted for these questions.

H.6. Surgical timing and technique

  • What is the effectiveness of laser-assisted phacoemulsification cataract surgery compared with standard ultrasound phacoemulsification cataract surgery?
  • What is the effectiveness of bilateral simultaneous (rapid sequential) cataract surgery compared with unilateral eye surgery?
  • What is the appropriate timing of second eye surgery, taking into account issues such as refractive power after first eye surgery?

H.6.1. Laser-assisted cataract surgery

H.6.1.1. Intra-operative complications

Image apphf219

H.6.1.2. Post-operative complications

Image apphf220

H.6.1.3. Visual acuity (logMAR)

Image apphf221

H.6.1.4. Duration of procedure (minutes)

Image apphf222

H.6.2. Bilateral surgery

H.6.2.1. Bilateral simultaneous versus unilateral cataract surgery

Any intraoperative complication.

Any intraoperative complication

Any postoperative complication.

Any postoperative complication

Any intra- or postoperative complication.

Any intra- or postoperative complication

Serious postoperative complications.

Serious postoperative complications

Visual function.

Visual function

H.6.2.2. Second-eye surgery versus no second-eye surgery

Visual acuity (logMAR).

Visual acuity (logMAR)

Contrast sensitivity.

Contrast sensitivity

Visual function.

Visual function

H.7. Anaesthesia

  • What is the optimal type and administration of anaesthesia for cataract surgery?
  • What is the effectiveness of sedation as an adjunct to local anaesthesia during cataract surgery?
  • What is the effectiveness of hyaluronidase as an adjunct to local anaesthesia during cataract surgery?
  • In what circumstances should general anaesthesia be considered in phacoemulsification cataract surgery?

H.7.1. Type and administration of anaesthesia

H.7.1.1. Pain on application

Warmed (37 degrees) vs room temperature anaesthetic.

Warmed (37 degrees) vs room temperature anaesthetic

Topical vs peribulbar anaesthesia.

Topical vs peribulbar anaesthesia

Topical vs sub-Tenon’s.

Topical vs sub-Tenon’s

H.7.1.2. Pain during surgery

Topical versus peribulbar.

Topical versus peribulbar

Topical versus retrobulbar.

Topical versus retrobulbar

Topical versus topical with intracameral.

Topical versus topical with intracameral

Topical versus sub-Tenon’s.

Topical versus sub-Tenon’s

Peribulbar versus retrobulbar.

Peribulbar versus retrobulbar

H.7.1.3. Surgical complications

Topical versus topical with intracameral (adverse surgical event).

Topical versus topical with intracameral (adverse surgical event)

Peribulbar vs retrobulbar (conjunctival chemosis).

Peribulbar vs retrobulbar (conjunctival chemosis)

H.7.2. Network meta-analyses

H.7.2.1. Anaesthetic drug (lidocaine used as reference category)

Pain on application.

Pain on application

Pairwise mean differences from NMA

BenoxinateBupivacaineLidocaineLevobupivacaine
BenoxinateN/A
Bupivacaine5.00 (3.61, 6.39)N/A
Lidocaine19.40 (16.08, 29.72)14.40 (11.98, 16.82)N/A
Levobupivacaine22.90 (17.03, 21.77)17.90 (11.06, 24.74)3.50 (−2.89, 9.89)N/A

Quantifying heterogeneity/inconsistency:

tau^2 = 0; I^2 = 0%

Network graph

Image apphf242

Comparison of direct and indirect evidence

Fixed effect model:  
                  comparison prop     nma  direct  indir. Diff z p-value 
      Benoxinate:Bupivacaine    1  5.0000  5.0000       .    . .       . 
  Benoxinate:Levobupivacaine    0 22.9000       . 22.9000    . .       . 
        Benoxinate:Lidocaine    1 19.4000 19.4000       .    . .       . 
 Bupivacaine:Levobupivacaine    0 17.9000       . 17.9000    . .       . 
       Bupivacaine:Lidocaine    1 14.4000 14.4000       .    . .       . 
   Levobupivacaine:Lidocaine    1 −3.5000 −3.5000       .    . .       . 
Legend: 
 comparison - Treatment comparison 
 prop       - Direct evidence proportion 
 nma        - Estimated treatment effect (MD) in network meta-analysis 
 direct     - Estimated treatment effect (MD) derived from direct evidence 
 indir.     - Estimated treatment effect (MD) derived from indirect evidence 
 Diff       - Difference between direct and indirect treatment estimates 
 z          - z-value of test for disagreement (direct versus indirect) 
 p-value    - p-value of test for disagreement (direct versus indirect)
Pain during surgery.

Pain during surgery

Pairwise mean differences from NMA

BenoxinateBupivacaineLidocaineLevobupivacaine
BenoxinateN/A
Bupivacaine−30.00 (−39.53, −20.47)N/A
Lidocaine−55.00 (−63.66, −46.34)−25.00 (−35.40, −14.60)N/A
Levobupivacaine−59.00 (−68.71, −49.29)−29.00 (−40.29, −17.71)−4.00 (−8.39, 0.39)N/A

Quantifying heterogeneity/inconsistency:

tau^2 = 0; I^2 = 0%

Network graph

Image apphf244

Comparison of direct and indirect evidence

Fixed effect model:  
                  comparison prop      nma   direct   indir. Diff z p-value 
      Benoxinate:Bupivacaine    1 −30.0000 −30.0000        .    . .       . 
  Benoxinate:Levobupivacaine    0 −59.0000        . −59.0000    . .       . 
        Benoxinate:Lidocaine    1 −55.0000 −55.0000        .    . .       . 
 Bupivacaine:Levobupivacaine    0 −29.0000        . −29.0000    . .       . 
       Bupivacaine:Lidocaine    1 −25.0000 −25.0000        .    . .       . 
   Levobupivacaine:Lidocaine    1   4.0000   4.0000        .    . .       . 
Legend: 
 comparison - Treatment comparison 
 prop       - Direct evidence proportion 
 nma        - Estimated treatment effect (MD) in network meta-analysis 
 direct     - Estimated treatment effect (MD) derived from direct evidence 
 indir.     - Estimated treatment effect (MD) derived from indirect evidence 
 Diff       - Difference between direct and indirect treatment estimates 
 z          - z-value of test for disagreement (direct versus indirect) 
 p-value    - p-value of test for disagreement (direct versus indirect)

H.7.2.2. Method of anaesthesia (topical used as reference category)

Pain on application.

Pain on application

Pairwise mean differences from NMA

Topicalsub-Tenon’sPeribulbarRetrobulbar
TopicalN/A
sub-Tenon’s6.15 (−4.81, 17.11)N/A
Peribulbar8.86 (−4.57, 17.11)2.71 (−14.63, 20.04)N/A
Retrobulbar49.10 (29.94, 68.26)42.95 (20.88, 65.02)40.24 (16.85, 63.64)N/A

Quantifying heterogeneity/inconsistency:

tau^2 = 89.5629; I^2 = 96.4%

Network graph

Image apphf246

Comparison of direct and indirect evidence

Random effects model:  
              comparison prop      nma   direct   indir. Diff z p-value 
  Peribulbar:Retrobulbar    0  40.2431        .  40.2431    . .       . 
  Peribulbar:sub-Tenon’s    0  −2.7073        .  −2.7073    . .       . 
      Peribulbar:Topical    1  −8.8569  −8.8569        .    . .       . 
 Retrobulbar:sub-Tenon’s    0 −42.9504        . −42.9504    . .       . 
     Retrobulbar:Topical    1 −49.1000 −49.1000        .    . .       . 
     sub-Tenon’s:Topical    1  −6.1496  −6.1496        .    . .       . 
Legend: 
 comparison - Treatment comparison 
 prop       - Direct evidence proportion 
 nma        - Estimated treatment effect (MD) in network meta-analysis 
 direct     - Estimated treatment effect (MD) derived from direct evidence 
 indir.     - Estimated treatment effect (MD) derived from indirect evidence 
 Diff       - Difference between direct and indirect treatment estimates 
 z          - z-value of test for disagreement (direct versus indirect) 
 p-value    - p-value of test for disagreement (direct versus indirect)
Pain during surgery.

Pain during surgery

Pairwise mean differences from NMA

TopicalTopical + intracameralPeribulbarRetrobulbarsub-Tenon’s
TopicalN/A
Topical + intracameral−2.73 (−7.84, 2.39)N/A
Peribulbar−6.49 (−10.96, −2.02)−3.76 (−10.55, 3.03)N/A
Retrobulbar−7.50 (−12.28, −2.72)−4.77 (−11.77, 2.23)−1.01 (−6.62, 4.61)N/A
sub-Tenon’s−12.97 (−21.62, −4.31)−10.24 (−20.29, −0.19)−6.48 (−16.21, 3.26)−5.47 (−15.35, 4.41)N/A

Quantifying heterogeneity/inconsistency:

tau^2 = 26.0174; I^2 = 82.2%

Network graph

Image apphf248

Comparison of direct and indirect evidence

Random effects model:  
                         comparison prop     nma  direct  indir.    Diff     z  p-value 
             Peribulbar:Retrobulbar 0.42 −1.0076 −0.3423 −1.4877  1.1454  0.20   0.8435 
             Peribulbar:sub-Tenon’s 0.00 −6.4774       . −6.4774       .     .        . 
                 Peribulbar:Topical 0.81  6.4896  6.2715  7.4172 −1.1457 −0.20   0.8435 
  Peribulbar:Topical + intracameral 0.00  3.7619       .  3.7619       .     .        . 
            Retrobulbar:sub-Tenon’s 0.00 −5.4699       . −5.4699       .     .        . 
                Retrobulbar:Topical 0.77  7.4972  7.7593  6.6139  1.1454  0.20   0.8435 
 Retrobulbar:Topical + intracameral 0.00  4.7695       .  4.7695       .     .        . 
                sub-Tenon’s:Topical 1.00 12.9671 12.9671       .       .     .        . 
 sub-Tenon’s:Topical + intracameral 0.00 10.2393       . 10.2393       .     .        . 
     Topical:Topical + intracameral 1.00 −2.7277 −2.7277       .       .     .        . 
Legend: 
 comparison - Treatment comparison 
 prop       - Direct evidence proportion 
 nma        - Estimated treatment effect (MD) in network meta-analysis 
 direct     - Estimated treatment effect (MD) derived from direct evidence 
 indir.     - Estimated treatment effect (MD) derived from indirect evidence 
 Diff       - Difference between direct and indirect treatment estimates 
 z          - z-value of test for disagreement (direct versus indirect) 
 p-value    - p-value of test for disagreement (direct versus indirect)

H.8. Preventing and managing complications

  • What is the effectiveness of interventions (for example, prophylactic laser surgery) to prevent retinal detachment in people with myopia undergoing cataract surgery?
  • What is the effectiveness of capsular tension rings applied during phacoemulsification cataract surgery?
  • What is the effectiveness of interventions to increase pupil size to improve visual outcomes and reduce complications during phacoemulsification cataract surgery?
  • What is the effectiveness of postoperative eye shields to prevent complications after cataract extraction?
  • What is the effectiveness of prophylactic antiseptics (for example, topical iodine) and antibiotics to prevent endophthalmitis after cataract surgery?
  • What is the effectiveness of prophylactic topical corticosteroids and/or NSAIDs to prevent inflammation and cystoid macular oedema after phacoemulsification cataract surgery?
  • What is the effectiveness of interventions to reduce the impact of perioperative posterior capsule rupture?
  • What is the effectiveness of interventions used to manage cystoid macular oedema following cataract surgery?

H.8.1. Capsular tension rings

H.8.1.1. Full population

CDVA (3 months postoperatively).

CDVA (3 months postoperatively)

UDVA (3 months postoperatively).

UDVA (3 months postoperatively)

H.8.1.2. People with pseudoexfoliation

Zonular dehiscence.

Zonular dehiscence

IOL in the bag successfully.

IOL in the bag successfully

H.8.2. Intervention to prevent cystoid macular oedema

H.8.2.1. Pairwise meta-analyses

NSAIDs plus steroids vs steroids
Poor vision due to CMO.

Poor vision due to CMO

CMO.

CMO

Inflammation (events).

Inflammation (events)

BCVA [logMAR].

BCVA [logMAR]

NSAIDs vs steroids
CMO.

CMO

Inflammation (flare) [photons/ms].

Inflammation (flare) [photons/ms]

BCVA [logMAR].

BCVA [logMAR]

NSAIDs plus steroids vs steroids (population with diabetic retinopathy)
CMO.

CMO

BCVA [logMAR].

BCVA [logMAR]

Proportion losing 5 or more letters of BCVA.

Proportion losing 5 or more letters of BCVA

H.8.2.2. Network meta-analyses (steroids used as reference category)

BCVA [logMAR].

BCVA [logMAR]

Number of studies: k=9

Number of treatments: n=3

Number of pairwise comparisons: m=11

Differences between treatments – mean and 95% confidence interval

NSAIDsNSAIDs + steroidsSteroids
NSAIDs0.0038
[−0.0298, 0.0373]		−0.0044
[−0.0319, 0.0232]
NSAIDs + Steroids−0.0038
[−0.0373, 0.0298]		−0.0081
[−0.0273, 0.0110]
Steroids0.0044
[−0.0232, 0.0319]		0.0081
[−0.0110, 0.0273]

Quantifying heterogeneity/inconsistency:

tau^2 = 0.0003; I^2 = 63.2%

Test of heterogeneity/inconsistency:

Q d.f. p-value

21.72 8 0.005

Network graph

Image apphf264

Comparison of direct and indirect evidence

Random effects model:  
                  comparison prop     nma  direct  indir.    Diff     z  p-value 
   NSAIDS:NSAIDS + steroids     0 −0.0038 −0.1050 −0.0033 −0.1017 −0.38   0.7029 
             NSAIDS:steroids    1  0.0044  0.0046 −0.1175  0.1220  0.32   0.7456 
 NSAIDS + steroids :steroids    1  0.0081  0.0081       .       .     .        . 
Legend: 
 comparison - Treatment comparison 
 prop       - Direct evidence proportion 
 nma        - Estimated treatment effect (MD) in network meta-analysis 
 direct     - Estimated treatment effect (MD) derived from direct evidence 
 indir.     - Estimated treatment effect (MD) derived from indirect evidence 
 Diff       - Difference between direct and indirect treatment estimates 
 z          - z-value of test for disagreement (direct versus indirect) 
 p-value    - p-value of test for disagreement (direct versus indirect)
CMO.

CMO

Number of studies: k=12

Number of treatments: n=3

Number of pairwise comparisons: m=14

Differences between treatments – mean and 95% confidence interval

NSAIDsNSAIDs + steroidsSteroids
NSAIDs1.0869
[0.5102, 2.3156]		0.2639
[0.1694, 0.4109]
NSAIDs + steroids0.9200
[0.4319, 1.9600]		0.2428
[0.1310, 0.4500]
Steroids3.7897
[2.4336, 5.9016]		4.1191
[2.2225, 7.6343]

Quantifying heterogeneity/inconsistency:

tau^2 = 0; I^2 = 0%

Test of heterogeneity/inconsistency:

Q d.f. p-value

4.68 11 0.9455

Network graph

Image apphf266

Comparison of direct and indirect evidence

Random effects model:  
                  comparison prop    nma direct  indir.    RoR     z  p-value 
   NSAIDS:NSAIDS + steroids  0.04 0.9200 1.0417  0.9155 1.1378  0.06   0.9490 
             NSAIDS:steroids 1.00 3.7897 3.7776 18.8152 0.2008 −0.32   0.7509 
 NSAIDS + steroids :steroids 1.00 4.1191 4.1019 11.6442 0.3523 −0.21   0.8343 
Legend: 
 comparison - Treatment comparison 
 prop       - Direct evidence proportion 
 nma        - Estimated treatment effect (RR) in network meta-analysis 
 direct     - Estimated treatment effect (RR) derived from direct evidence 
 indir.     - Estimated treatment effect (RR) derived from indirect evidence 
 RoR        - Ratio of Ratios (direct versus indirect) 
 z          - z-value of test for disagreement (direct versus indirect) 
 p-value    - p-value of test for disagreement (direct versus indirect)
Inflammation (flare) [photons/ms].

Inflammation (flare) [photons/ms]

Number of studies: k=5

Number of treatments: n=3

Number of pairwise comparisons: m=7

Differences between treatments – Mean and 95% confidence interval

NSAIDsNSAIDs + steroidsSteroids
NSAIDs2.1419 [−1.1857, 5.4694]−1.6413 [−3.4897, 0.2070]
NSAIDs + steroids−2.1419 [−5.4694, 1.1857]−3.7832 [−7.2631, −0.3033]
Steroids1.6413 [−0.2070, 3.4897]3.7832 [0.3033, 7.2631]

Quantifying heterogeneity/inconsistency:

tau^2 = 2.8678; I^2 = 68.5%

Test of heterogeneity/inconsistency:

Q d.f. p-value

12.69 4 0.0129

Network graph

Image apphf268

Comparison of direct and indirect evidence

Random effects model:  
                  comparison prop     nma  direct  indir.    Diff     z  p-value 
   NSAIDS:NSAIDS + steroids  0.85 −2.1419 −2.4000 −0.6558 −1.7442 −0.36   0.7152 
             NSAIDS:steroids 1.00  1.6413  1.6413       .       .     .        . 
 NSAIDS + steroids :steroids 0.64  3.7832  3.3000  4.6516 −1.3516 −0.36   0.7152 
Legend: 
 comparison - Treatment comparison 
 prop       - Direct evidence proportion 
 nma        - Estimated treatment effect (MD) in network meta-analysis 
 direct     - Estimated treatment effect (MD) derived from direct evidence 
 indir.     - Estimated treatment effect (MD) derived from indirect evidence 
 Diff       - Difference between direct and indirect treatment estimates 
 z          - z-value of test for disagreement (direct versus indirect) 
 p-value    - p-value of test for disagreement (direct versus indirect)

H.9. Postoperative assessment

  • What are the early and late complications of cataract surgery?
  • What should the postoperative assessment include?
  • Who and in what setting should carry out the postoperative assessment?
  • What issues should be considered when organising postoperative care?
  • What is the appropriate time to assess outcomes in the postoperative period?
  • If the postoperative assessment and care are undertaken outside of the hospital, how should outcomes between surgical units and these providers be effectively communicated?

H.9.1. Details of postoperative assessment

H.9.1.1. Complications

Image apphf269

H.9.1.2. CDVA

Image apphf270

H.9.1.3. Unscheduled visits

Image apphf271
Copyright © NICE 2017.
Bookshelf ID: NBK536568
Contents

Views

In this Page

Other titles in this collection

Recent Activity

ClearTurn OffTurn On

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...