| Quality assessment | No of patients | Effect | Quality | Importance | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| No of studies | Design | Risk of bias | Inconsistency | Indirectness | Imprecision | Other considerations | Stenting + planned bowel resection or palliative care | Emergency bowel surgery | Relative (95% CI) | Absolute | ||
| Clinically successful bowel decompression, stent arm only - Palliative intent | ||||||||||||
| 2 | randomised trials | no serious risk of bias | serious inconsistency1 | serious2 | serious3 | none |
30/37 (81.1%) | - | Risk 0.84 (0.43 to 0.97) | 840 per 1000 (from 430 to 970) | VERY LOW | CRITICAL |
| Clinically successful bowel decompression, stent arm only - Curative intent | ||||||||||||
| 5 | randomised trials | very serious4,5 | serious inconsistency1 | no serious indirectness | serious3 | none |
123/177 (69.5%) | - |
Risk 0.69 (0.53 to 0.82) | 690 per 1000 (from 530 to 820) | VERY LOW | CRITICAL |
| 30-day mortality - Palliative intent | ||||||||||||
| 3 | randomised trials | serious4 | serious inconsistency1 | serious2 | serious3 | none |
4/48 (8.3%) |
4/47 (8.5%) | RD −0.00 (−0.12 to 0.12) | 0 more per 1000 (from 120 fewer to 120 more) | VERY LOW | CRITICAL |
| 30-day mortality - Curative intent | ||||||||||||
| 5 | randomised trials | very serious4,5 | no serious inconsistency | no serious indirectness | serious3 | none |
9/168 (5.4%) |
10/172 (5.8%) | Peto OR 0.92 (0.36 to 2.34) | 4 fewer per 1000 (from 34 fewer to 63 more) | VERY LOW | CRITICAL |
| Disease free survival, event is disease recurrence or death from any cause (follow-up 4 to 5 years) - Curative intent | ||||||||||||
| 2 | randomised trials | serious6 | no serious inconsistency | no serious indirectness | serious3 | none |
22/50 (44%) |
16/56 (29%) | HR 0.56 (0.29 to 1.06) | At 4 years ES 28.1%b, SBTS 57.2% (27.4% to 78.6%) | LOW | CRITICAL |
| 3-year progression free survival, event is disease recurrence or relapse or death from any cause – Curative intent | ||||||||||||
| 1 | randomised trials | no serious risk of bias | no serious inconsistency | no serious indirectness | serious3 | none |
17/56 (30%) |
12/59 (20%) | HR 0.95 (0.45 to 2.01) | At 3 years ES 20.3%c, SBTS 22% (4.2% to 48.8%) | MODERATE | CRITICAL |
| 1-year overall survival, event is death from any cause - Palliative intent | ||||||||||||
| 1 | randomised trials | no serious risk of bias | no serious inconsistency | serious2 | serious3 | none |
17/26 (65%) |
19/26 (73%) | HR 0.84 (0.44 to 1.6) | At 1 year ES 73.1%d, stenting 76.8% (60.5% to 87.1%) | LOW | IMPORTANT |
| 5-year overall survival, event is death from any cause - Curative intent | ||||||||||||
| 1 | randomised trials | no serious risk of bias | no serious inconsistency | no serious indirectness | serious3 | none | 12/24 | 16/24 | HR 0.51 (0.24 to 1.08) | At 5 years ES 67%a, SBTS 81.5% (64.9% to 90.8%) | MODERATE | IMPORTANT |
| 4-year overall survival, event is death from any cause - Curative intent | ||||||||||||
| 1 | randomised trials | serious6 | no serious inconsistency | no serious indirectness | serious3 | none | 10/26 | 10/32 | HR 0.72 (0.3 to 1.73) | At 4 years ES 31.3%b, SBTS 43.3% (13.4% to 70%) | LOW | IMPORTANT |
| 3-year overall survival, event is death from any cause - Curative intent | ||||||||||||
| 1 | randomised trials | no serious risk of bias | no serious inconsistency | no serious indirectness | serious3 | none | 18/56 | 16/59 | HR 1.00 (0.51 to 1.96) | At 3 years ES 27.1%c, SBTS 27.2% (7.7% to 51.4%) | MODERATE | IMPORTANT |
| Hospital stay - Palliative intent - Fiori 2004 | ||||||||||||
| 1 | randomised trials | serious4 | no serious inconsistency | no serious indirectness | serious3 | none |
N=15 Median= 2.6 |
N=13 Median= 8.1 | p<0.0001 | - | not assessable6 | IMPORTANT |
| Hospital stay - Palliative intent - Dutch Stent-In-1 trial (Van Hooft 2008) | ||||||||||||
| 1 | randomised trials | serious4 | no serious inconsistency | no serious indirectness | serious3 | none |
N=11 Median=12 Range=7-19 |
N=10 Median=11 Range=6.25-17.25 | p=0.46 | - | not assessable6 | IMPORTANT |
| Hospital stay - Palliative intent - Young 2015 | ||||||||||||
| 1 | randomised trials | no serious risk of bias | no serious inconsistency | serious2 | serious3 | none |
N=26 Median=7 Range=3-12 |
N=26 Median=11 Range=8-17 | p=0.03 | - | not assessable6 | IMPORTANT |
| Hospital stay - Curative intent - Alcantara 2011 | ||||||||||||
| 1 | randomised trials | very serious4,5 | no serious inconsistency | no serious indirectness | serious3 | none |
N=15 Median=13 |
N=13 Median=10 | p=0.105 | - | not assessable6 | IMPORTANT |
| Hospital stay - Curative intent - Cheung 2009 | ||||||||||||
| 1 | randomised trials | no serious risk of bias | no serious inconsistency | no serious indirectness | serious3 | none |
N=24 Median=13.5 Range=7-29 |
N=24 Median=14 Range=7-55 | p=0.7 | - | not assessable6 | IMPORTANT |
| Hospital stay - Curative intent - ESCO trial (Arezzo 2017) | ||||||||||||
| 1 | randomised trials | no serious risk of bias | no serious inconsistency | no serious indirectness | serious3 | none |
N=56 Median=10 Range=7-13 |
N=59 Median=11 Range=8-15 | - | - | not assessable6 | IMPORTANT |
| Hospital stay - Curative intent - Ghazal 2013 | ||||||||||||
| 1 | randomised trials | serious4 | no serious inconsistency | no serious indirectness | serious3 | none |
N=30 Median=13 |
N=30 Median=8 | p=0.102 | - | not assessable6 | IMPORTANT |
| Hospital stay - Curative intent - Ho 2012 | ||||||||||||
| 1 | randomised trials | no serious risk of bias | no serious inconsistency | no serious indirectness | serious3 | none |
N=20 Median=6 Range=4-28 |
N=19 Median=8 Range=6-39 | p=0.028 | - | not assessable6 | IMPORTANT |
| Hospital stay - Curative intent - Pirlet 2011 | ||||||||||||
| 1 | randomised trials | no serious risk of bias | no serious inconsistency | no serious indirectness | serious3 | none |
N=30 Median=23 Range=9-67 |
N=30 Median=17 (7-126) | p=0.13 | - | not assessable6 | IMPORTANT |
| Anastomotic leak - Palliative intent | ||||||||||||
| 1 | randomised trials | no serious risk of bias | no serious inconsistency | serious2 | serious3 | none |
0/26 (0%) |
0/26 (0%) | RD 0.00 (−0.07 to 0.07) | 0 more per 1000 (from 7 fewer to 7 more) | LOW | IMPORTANT |
| Anastomotic leak - Curative intent | ||||||||||||
| 7 | randomised trials | very serious4,5 | no serious inconsistency | no serious indirectness | serious3 | none |
11/221 (5%) |
12/226 (5.3%) | Peto OR 0.92 (0.40 to 2.13) | 4 fewer per 1000 (from 29 fewer to 49 more) | VERY LOW | IMPORTANT |
| Perforation rate, stent arm only - Curative intent | ||||||||||||
| 3 | randomised trials | no serious risk of bias | no serious inconsistency | no serious indirectness | serious3 | none |
13/133 (9.8%) | - | Risk 0.10 (0.06 to 0.17) | 100 per 1000 (from 60 to 170) | MODERATE | IMPORTANT |
| Surgical site infection - Palliative intent | ||||||||||||
| 1 | randomised trials | no serious risk of bias | no serious inconsistency | serious2 | serious3 | none |
0/26 (0%) |
1/26 (3.8%) | Peto OR 0.14 (0.00 to 6.82)7 | 33 fewer per 1000 (from 38 fewer to 176 more) | LOW | IMPORTANT |
| Surgical site infection - Curative intent | ||||||||||||
| 6 | randomised trials | very serious4,5 | no serious inconsistency | no serious indirectness | serious3 | none |
14/191 (7.3%) |
35/196 (17.9%) | RR 0.4 (0.22 to 0.71) | 107 fewer per 1000 (from 52 fewer to 139 fewer) | VERY LOW | IMPORTANT |
| Stoma rate - Palliative intent - Postprocedure | ||||||||||||
| 1 | randomised trials | no serious risk of bias | no serious inconsistency | serious2 | serious3 | none |
7/26 (26.9%) |
24/26 (92.3%) | RR 0.29 (0.15 to 0.55) | 655 fewer per 1000 (from 415 fewer to 785 fewer) | LOW | IMPORTANT |
| Stoma rate - Curative intent - Postprocedure | ||||||||||||
| 4 | randomised trials | no serious risk of bias | no serious inconsistency | no serious indirectness | serious3 | none |
50/153 (32.7%) |
84/159 (52.8%) | RR 0.62 (0.48 to 0.81) | 201 fewer per 1000 (from 100 fewer to 275 fewer) | MODERATE | IMPORTANT |
| Stoma rate - Curative intent - At last follow up | ||||||||||||
| 4 | randomised trials | no serious risk of bias | no serious inconsistency | no serious indirectness | serious3 | none |
37/147 (25.2%) |
57/153 (37.3%) | RR 0.70 (0.51 to 0.94) | 112 fewer per 1000 (from 22 fewer to 183 fewer) | MODERATE | IMPORTANT |
| Technically successful stent placement, stent arm only - Palliative intent | ||||||||||||
| 3 | randomised trials | serious4 | serious inconsistency1 | serious2 | serious3 | none |
44/52 (84.6%) | - | Risk 0.86 (0.61 to 0.96) | 860 per 1000 (from 610 to 960) | VERY LOW | IMPORTANT |
| Technically successful stent placement, stent arm only - Curative intent | ||||||||||||
| 5 | randomised trials | very serious4,5 | serious inconsistency1 | no serious indirectness | serious3 | none |
174/222 (78.4%) | - | Risk 0.69 (0.66 to 0.72) | 690 per 1000 (from 660 to 720) | VERY LOW | IMPORTANT |
| Stent failure, stent arm only - Curative intent | ||||||||||||
| 2 | randomised trials | no serious risk of bias | serious inconsistency1 | no serious indirectness | serious3 | none |
12/76 (15.8%) | - | Risk 0.18 (0.06 to 0.44) | 180 per 1000 (from 60 to 440) | LOW | IMPORTANT |
| Quality of life - Palliative intent - EQ-5D change score, change from baseline to 1 year (Better indicated by lower values) | ||||||||||||
| 1 | randomised trials | no serious risk of bias | no serious inconsistency | serious2 | serious3 | none | 26 | 26 | - | MD 0.26 higher (0.05 to 0.47 higher) | LOW | IMPORTANT |
| Quality of life - Curative intent - EORTC-C30 QL2 subscale, change from baseline to 6-months (Better indicated by lower values) | ||||||||||||
| 1 | randomised trials | serious9 | no serious inconsistency | no serious indirectness | serious3 | none | 36 | 39 | - | MD 10.1 higher (1.87 to 18.33 higher) | LOW | IMPORTANT |
CI: confidence interval; EQ-5D: EuroQol five dimensions questionnaire: ES: emergency surgery; HR: hazard ratio; MD: mean difference: N: number; OR: odds ratio; RD: risk difference; RR: relative risk; SBTS: stenting as a bridge to surgery
Quality of evidence downgraded by 1 due to moderate-high heterogeneity (I2 > 40%)
Quality of evidence downgraded by 1 due to indirectness of the study population - 6/30 (20%) patients had ovarian cancer (Xinopoulos 2004); 12/52 (23%) patients had non-colorectal cancer primaries (Young 2015)
Quality of evidence downgraded by 1 because of imprecision of the effect estimate (< 300 events for dichotomous outcomes or < 400 patients for continuous outcomes)
Quality of evidence downgraded by 1 due to failure to report random sequence generation procedure, unclear how attrition was managed, outcomes not pre-specified (Alcantara 2011; Fiori 2004; Ghazal 2013; Xinopoulos 2004)
Quality of evidence downgraded by 1 because interim safety analyses and termination procedure not determined a priori (Alcantara 2011)
Quality of evidence downgraded by 1 due to 69% attrition from original sample (Dutch Stent-In-2-Trial [Sloothaak 2014])
Peto OR used due to zero events in one arm
Not calculable because of 0 events in both arms
Quality of evidence downgraded by 1 because lack of blinding could have affected quality of life outcomes (Dutch Stent-In-2 trial [Van Hooft 2011])
The absolute risk at 5 years in the control group taken from Cheung 2009 (Tung 2013)
The absolute risk at 4 years in the control group taken from the Dutch Stent-In-2 trial (Sloothaak 2014)
The absolute risk at 3 years in the control group taken from the ESCO trial (Arezzo 2017)
The absolute risk at 1 year in the control group taken from the Young 2015
From: Effectiveness of stenting for acute large bowel obstruction
NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.