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.

Observational evidence on the effectiveness of endovascular aneurysm repair compared with open surgical repair of unruptured abdominal aortic aneurysms: Abdominal aortic aneurysm: diagnosis and management: Evidence review K2. London: National Institute for Health and Care Excellence (NICE); 2020 Mar. (NICE Guideline, No. 156.)

Cover of Observational evidence on the effectiveness of endovascular aneurysm repair compared with open surgical repair of unruptured abdominal aortic aneurysms

Observational evidence on the effectiveness of endovascular aneurysm repair compared with open surgical repair of unruptured abdominal aortic aneurysms: Abdominal aortic aneurysm: diagnosis and management: Evidence review K2.

Show details

Appendix FGRADE tables

F.1. Exclusively or predominantly infrarenal AAA

No. of studiesQuality assessmentSummary of findingsQuality
Risk of biasIndirectnessInconsistencyImprecisionNo. of patientsRelative effect estimate (95% CI)Absolute expectation
EVAROSROSREVAR (95% CI)
Perioperative mortality – relative effects less than 1 favour EVAR
21aV seriousbNot seriousV seriouscNo serious219,938166,528OR = 0.31 (0.27 to 0.36)29 per 1,000d20 fewer per 1,000 (18 to 21 fewer)VERY LOWe
Duration of procedure (minutes) – relative effects less than 0 favour EVAR
1fV seriousbNot seriousNANo serious157157MD = 86.6 shorter (71.9 to 101.3 shorter)215 minsg128 mins (114 to 143 mins)VERY LOWe
Perioperative complications (all) – relative effects less than 1 favour EVAR
6hV seriousbNot seriousNot seriousiNo serious124,92385,740OR = 0.35 (0.26 to 0.48)357 per 1,000j194 fewer per 1,000 (147 to 231 fewer)VERY LOWe
Perioperative complications (cardiovascular) – relative effects less than 1 favour EVAR
7kV seriousbNot seriousNot seriousiNo serious138,692124,481OR = 0.35 (0.27 to 0.45)88 per 1,000j55 fewer per 1,000 (46 to 63 fewer)VERY LOWe
Perioperative complications (respiratory) – relative effects less than 1 favour EVAR
5lV seriousbNot seriousNot seriousiNo serious134,633120,345OR = 0.20 (0.14 to 0.29)175 per 1,000j134 fewer per 1,000 (117 to 146 fewer)VERY LOWe
Perioperative complications (renal) – relative effects less than 1 favour EVAR
6mV seriousbNot seriousNot seriousiNo serious134,724120,436OR = 0.29 (0.22 to 0.38)81 per 1,000j55 fewer per 1,000 (47 to 62 fewer)VERY LOWe
Length of critical care stay (days) – relative effects less than 0 favour EVAR
2nV seriousbNot seriousNot seriousNo serious1,4466,131MD = 1.61 shorter (1.38 to 1.84 shorter)3.4 daysd1.8 days (1.5 to 2.0 days)VERY LOWe
Total length of hospital stay (days) – relative effects less than 0 favour EVAR
4oV seriousbNot seriousNot seriousiNo serious132,337121,319MD = 5.67 shorter (5.11 to 6.22 shorter)10.5 daysd4.8 days (4.3 to 5.4 days)VERY LOWe
Discharge to location other than home – relative effects less than 1 favour EVAR
3pV seriousbNot seriousV seriouscNo serious131,779120,761OR = 0.31 (0.25 to 0.38)175 per 1,000q113 fewer per 1,000 (100 to 125 fewer)VERY LOWe
Post-perioperative survival – relative effects less than 1 favour EVAR
10rV seriousbNot seriousV seriouscNo serious55,52354,104HR = 1.24 (1.13 to 1.35)441 surviving per 1,000 at 10 yearss79 fewer per 1,000 (45 to 110 fewer)VERY LOW
Reinterventions (all or unspecified) – relative effects less than 1 favour EVAR
8tV seriousbNot seriousV seriouscNo serious59,74662,408HR = 1.54 (1.21 to 1.96)206 per 1,000 at 8 yearsu93 more per 1,000 (38 to 158 more)VERY LOW
Reinterventions (vascular) – relative effects less than 1 favour EVAR
3vV seriousbNot seriousV seriouscNo serious43,19047,875HR = 2.90 (1.49 to 5.65)37 per 1,000 at 8 yearsu67 more per 1,000 (18 to 155 more)VERY LOWe
Reinterventions (non-vascular) – relative effects less than 1 favour EVAR
1wSeriousxNot seriousN/ANo serious39,96639,966HR = 0.38 (0.36 to 0.40)177 per 1,000 at 8 yearsu106 fewer per 1,000 (102 to 109 fewer)LOWy
a

Chadi et al. (2012), Choke et al. (2012), Hicks et al. (2015), Hua et al. (2005), Huang et al. (2015), Jetty et al. (2010), Johnson et al. (2006), Karthikesalingam et al. (2016) [UK], Karthikesalingam et al. (2016) [USA], Lee et al. (2004), Liang et al. (2018), Lo et al. (2013), Malas et al. (2014), Mark et al. (2013), Quintana et al. (2019), Salata et al. (2019), Schermerhorn et al. (2015) [2001–2004 cohort], Schermerhorn et al. (2015) [2005–2008 cohort], Sugimoto et al. (2017), Symonides et al. (2018), Tarbunou et al. (2019)

b

Majority of studies at high risk of bias

c

I2 > 66.7%

d

Source: UK National Vascular Registry (2017)

e

Eligible for uprating, owing to large effect size, but cancelled out by multiple other reasons for downgrading

f

Sugimoto et al. (2017)

g

Source: EVAR-1 (Brown et al., 2012)

h

Hua et al. (2005), Huang et al. (2015), Johnson et al. (2006), Locham et al. (2018b), Raval et al. (2012), Schwarze et al. (2009)

i

High I2, but all estimates agree there is a large effect; hence, no downgrading because statistical inconsistency is not important for decision uncertainty

j

Source: Schwarze et al. (2009)

k

Feringa et al. (2007), Huang et al. (2015), Liang et al. (2018), Nguyen et al. (2013), Salata et al. (2019), Schermerhorn et al. (2015), Schwarze et al. (2009)

l

Huang et al. (2015), Liang et al. (2018), Nguyen et al. (2013), Schermerhorn et al. (2015), Schwarze et al. (2009)

m

Huang et al. (2015), Liang et al. (2018), Nguyen et al. (2013), Schermerhorn et al. (2015), Schwarze et al. (2009), Zabrocki et al. (2018)

n

Huang et al. (2015), Jetty et al. (2010)

o

Huang et al. (2015), Jetty et al. (2010), Schermerhorn et al. (2015), Schwarze et al. (2009)

p

Jetty et al. (2010), Schermerhorn et al. (2015), Schwarze et al. (2009)

q

Derived from Karthikesalingam et al. (2016) [UK cohort]

r

Behrendt et al. (2017), de la Motte et al. (2013), Huang et al. (2015) [2000–2005 cohort], Huang et al. (2015) [2006–2011 cohort], Mark et al. (2013), Salata et al. (2019), Schermerhorn et al. (2015) [2001–2004 cohort], Schermerhorn et al. (2015) [2005–2008 cohort], Sugimoto et al. (2017), Symonides et al. (2018)

s

Derived from EVAR-1 – 10-year survival conditional on surviving 30 postoperative days

t

Chang et al. (2015), Huang et al. (2015), Jetty et al. (2010), Liang et al. (2018), Salata et al. (2019), Schermerhorn et al. (2015) [2001–2004 cohort], Schermerhorn et al. (2015) [2005–2008 cohort], Sugimoto et al. (2017)

u

Source: Schermerhorn et al. (2015)

v

Jetty et al. (2010), Schermerhorn et al. (2015), Symonides et al. (2018)

w

Schermerhorn et al. (2015)

x

Single study at moderate risk of bias

y

Uprated for large effect size

F.2. Complex AAA

No. of studiesQuality assessmentSummary of findingsQuality
Risk of biasIndirectnessInconsistencyImprecisionNo. of patientsRelative effect estimate (95% CI)Absolute expectation (95% CI)
EVAROSROSREVAR
Perioperative mortality – relative effects less than 1 favour EVAR0
6aV seriousbNot seriousSeriouscV seriousd9803,383OR = 0.90 (0.41 to 1.98)4 more per 1,000 (17 fewer to 46 more)35 per 1,000e,fVERY LOW
Duration of procedure (minutes) – relative effects less than 0 favour EVAR
2gV seriousbNot seriousNot serioushNot serious365365MD = 107.7 shorter (23.9 to 191.5 shorter)279 minsi171 mins (88 to 255 mins)VERY LOWj
Perioperative complications (all) – relative effects less than 1 favour EVAR
4kV seriousbNot seriousV seriouslV seriousd448614OR = 0.64 (0.23 to 1.76)96 more per 1,000 (96 fewer to 347 more)270 per 1,000f,mVERY LOW
Perioperative complications (cardiovascular) – relative effects less than 1 favour EVAR
4kV seriousbNot seriousNot seriousNot serious448614OR = 0.34 (0.21 to 0.55)138 more per 1,000 (64 to 233 more)92 per 1,000f,f,mVERY LOWj
Perioperative complications (respiratory) – relative effects less than 1 favour EVAR
4kV seriousbNot seriousSeriouscV seriousd448614OR = 0.50 (0.17 to 1.47)47 more per 1,000 (16 fewer to 192 more)52 per 1,000f,f,mVERY LOW
Perioperative complications (renal) – relative effects less than 1 favour EVAR
4kV seriousbNot seriousV seriouslV seriousd448614OR = 0.54 (0.21 to 1.40)32 more per 1,000 (11 fewer to 126 more)40 per 1,000f,f,mVERY LOW
Length of critical care stay (days) – relative effects less than 0 favour EVAR
2gV seriousbNot seriousV seriouscV seriousd365365MD = 1.32 shorter (3.9 shorter to 1.3 longer)3.1 days (0.5 to 5.7 days)1.8 dayse,fVERY LOW
Total length of hospital stay (days) – relative effects less than 0 favour EVAR
1nV seriousbNot seriousN/ANot serious263263MD = 5.65 shorter (5.2 to 6.1 shorter)14.5 days (14.0 to 14.9 days)8.8 dayse,fVERY LOWj
Discharge to location other than home – relative effects less than 1 favour EVAR
1nV seriousbNot seriousN/ANot serious263263OR = 0.23 (0.14 to 0.39)125 per 1,000o93 fewer per 1,000 (72 to 105 fewer)VERY LOWj
Post-perioperative survival – relative effects less than 1 favour EVAR
2pV seriousbNot seriousN/ANot serious143204HR = 1.94 (1.15 to 3.27)441 per 1,000 surviving at 10 yearsq237 fewer per 1,000 (51 to 372 fewer)VERY LOW
Reinterventions (all or unspecified) – relative effects less than 1 favour EVAR
1rV seriousbNot seriousN/ASeriouss102102HR = 3.02 (0.96 to 9.51)206 per 1,000 at 8 yearst296 more per 1,000 (7 fewer to 682 more)VERY LOW
a

Fiorucci et al. (2019), Michel et al. (2015), Orr et al. (2017), Raux et al. (2014), Tinelli et al. (2018), Tsilimparis et al. (2013)

b

Majority of studies at high risk of bias

c

33.3% < I2 < 66.7%

d

95% confidence interval encompasses meaningful benefit for either EVAR or OSR

e

Source: National Vascular Registry (2017)

f

EVAR preferred as baseline option as more reliable data available

g

Orr et al. (2017), Tinelli et al. (2018)

h

High I2, but all estimates agree there is a large effect; hence, statistical inconsistency is not important for decision uncertainty

i

Weighted average of OSR arms in included studies

j

Eligible for uprating, owing to large effect size, but cancelled out by reasons for downgrading

k

Fiorucci et al. (2019), Orr et al. (2017), Raux et al. (2014), Tinelli et al. (2018)

l

I2 > 66.7%

f,m

Source: Roy et al. (2017)

n

Orr et al. (2017)

o

Derived from Karthikesalingam et al. (2016) [UK cohort] (in absence of long-term data specific to complex AAA)

p

Fiorucci et al. (2019), Tinelli et al. (2018)

q

Source: EVAR-1 - 10-year survival conditional on surviving 30 postoperative days (in absence of long-term data specific to complex AAA)

r

Tinelli et al. (2018)

s

95% confidence interval encompasses meaningful harm for EVAR and no difference

t

Source: Schermerhorn et al. (2015) (in absence of long-term data specific to complex AAA)

Copyright © NICE 2020.
Bookshelf ID: NBK556888

Views

  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (2.0M)

Other titles in this collection

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...