| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect (95% CI) | No of Participants (studies) | Quality of the evidence (GRADE) | Comments | |
|---|---|---|---|---|---|---|
| Assumed risk | Correspon ding risk | |||||
| SEMS | Plastic | |||||
| Treatmentrelated mortality | 0 per 1000 | 0 per 1000 (0 to 0) | RR 2.88 (0.12 to 69.16) | 100 (1 study) | ⊕⊝⊝⊝ very low1,2 | |
| Overall Survival | Study population | HR 1 (0.75 to 1.31) | 247 (3 studies) | ⊕⊝⊝⊝ very low1,4,5,9,13,21,22 | ||
| See comment3 | See comment3 | |||||
| Moderate | ||||||
| 0 per 10003 | -214748364 8 per 1000 (-2147483648 to -2147483648)3 | |||||
| Time to stent dysfunction for unresectable PC - primary and/or secondary stent | Study population | HR 2.59 (1.67 to 4) | 229 (3 studies) | ⊕⊝⊝⊝ very low3,4,5,8,9,13,17,18 | ||
| See comment3 | See comment3 | |||||
| Moderate | ||||||
| 0 per 10003 | -214748364 8 per 1000 (-2147483648 to -2147483648)3 | |||||
| Time to stent dysfunction for unresectable PC - Covered or Partially Covered SEMS (Primary Stent only) | 257 per 1000 | 489 per 1000 (350 to 649) | HR 2.26 (1.45 to 3.53) | 224 (2 studies) | ⊕⊝⊝⊝ very low4,5,6,7,8 | |
| Time to stent dysfunction for unresectable PC - Uncovered SEMS (Primary Stent only) | 167 per 1000 | 421 per 1000 (232 to 677) | HR 3 (1.45 to 6.2) | 117 (1 study) | ⊕⊝⊝⊝ very low4,6,7,8 | |
| Time to stent dysfunction for unresectable PC - Partially Covered SEMS (Secondary Stent only) | 118 per 1000 | 567 per 1000 (160 to 982) | HR 6.69 (1.39 to 32.07) | 33 (1 study) | ⊕⊝⊝⊝ very low4,6,7,8 | |
| Time to stent dysfunction for unresectable PC - Uncovered SEMS (Secondary Stent only) | 67 per 1000 | 497 per 1000 (212 to 862) | HR 9.97 (3.46 to 28.74) | 31 (1 study) | ⊕⊝⊝⊝ very low4,6,7,8 | |
| Stent Dysfunction - Stent Occlusion | 191 per 1000 | 430 per 1000 (319 to 577) | RR 2.25 (1.67 to 3.02) | 471 (6 studies) | ⊕⊕⊝⊝ low1,4,5,9,10,11,12,13,14,15 | |
| Stent Dysfunction - Stent Migration | 91 per 1000 | 17 per 1000 (2 to 143) | RR 0.19 (0.02 to 1.57) | 113 (1 study) | ⊕⊝⊝⊝ very low2,4,5 | |
| Stent Dysfunction - Stent Occlusion or Migration | 167 per 1000 | 403 per 1000 (240 to 677) | RR 2.42 (1.44 to 4.06) | 171 (1 study) | ⊕⊝⊝⊝ very low4,6,7,8 | |
| Stent Occlusion - any type of SEMS | 176 per 1000 | 387 per 1000 (255 to 590) | RR 2.2 (1.45 to 3.35) | 258 (4 studies) | ⊕⊝⊝⊝ very low4,8,9,10,11,12,13,14,15 | |
| Stent Occlusion - Covered SEMS | 212 per 1000 | 487 per 1000 (319 to 738) | RR 2.3 (1.51 to 3.49) | 213 (2 studies) | ⊕⊝⊝⊝ very low1,4,5,8 | |
| Stent Occlusion - unresectable patients | 174 per 1000 | 410 per 1000 (295 to 570) | RR 2.36 (1.7 to 3.28) | 417 (5 studies) | ⊕⊕⊝⊝ low1,4,5,9,11,12,13,14 | |
| Stent Occlusion - resectable, borderline resectable or locally advanced | 303 per 1000 | 524 per 1000 (270 to 1000) | RR 1.73 (0.89 to 3.34) | 54 (1 study) | ⊕⊕⊝⊝ low4,10,15,16 | |
| Pancreatitis | 22 per 1000 | 18 per 1000 (7 to 46) | RR 0.81 (0.32 to 2.04) | 720 (7 studies) | ⊕⊝⊝⊝ very low1,2,4,5,6,9,10,11,13,14,15,17 | |
| Pancreatitis - any SEMS | 25 per 1000 | 26 per 1000 (9 to 73) | RR 1.02 (0.36 to 2.92) | 473 (4 studies) | ⊕⊝⊝⊝ very low2,4,6,7,10,11,14,15,17,18 | |
| Pancreatitis - covered SEMS | 19 per 1000 | 6 per 1000 (1 to 58) | RR 0.32 (0.03 to 3.01) | 213 (2 studies) | ⊕⊝⊝⊝ very low1,2,4,5 | |
| Pancreatitis - unresectable patients | 1 per 100 | 1 per 100 (0 to 4) | RR 1.52 (0.51 to 4.59) | 632 (5 studies) | ⊕⊝⊝⊝ very low1,2,4,5,6,7,9,11,14,17,18 | |
| Pancreatitis - resectable, borderline resectable or locally advanced patients | 182 per 1000 | 22 per 1000 (2 to 365) | RR 0.12 (0.01 to 2.01) | 54 (1 study) | ⊕⊝⊝⊝ very low2,4,10,15 | |
| Cholangitis - unresectable patients | 30 per 1000 | 93 per 1000 (38 to 224) | RR 3.1 (1.28 to 7.48) | 334 (4 studies) | ⊕⊕⊝⊝ low1,4,9,11,13,17,18 | |
| Cholangitis - any SEMS | 39 per 1000 | 67 per 1000 (19 to 229) | RR 1.71 (0.5 to 5.89) | 152 (2 studies) | ⊕⊝⊝⊝ very low2,4,9,11,13,14 | |
| Cholangitis - covered SEMS | 0 per 1000 | 0 per 1000 (0 to 0) | RR 4.81 (0.24 to 97.68) | 100 (1 study) | ⊕⊝⊝⊝ very low1,2 | |
| Cholangitis - partiallycovered SEMS | 49 per 1000 | 244 per 1000 (57 to 1000) | RR 5 (1.17 to 21.43) | 82 (1 study) | ⊕⊝⊝⊝ very low4,16,17,18 | |
| Cholecystitis -unresectable patients | 27 per 1000 | 13 per 1000 (4 to 41) | RR 0.47 (0.15 to 1.53) | 448 (4 studies) | ⊕⊝⊝⊝ very low2,4,5,6,7,9,13,17,18 | |
| Cholecystitis - any SEMS | 6 per 1000 | 16 per 1000 (2 to 123) | RR 2.56 (0.33 to 20.1) | 253 (2 studies) | ⊕⊝⊝⊝ very low2,4,6,7,9,13 | |
| Cholecystitis - partiallycovered SEMS | 49 per 1000 | 10 per 1000 (0 to 197) | RR 0.2 (0.01 to 4.04) | 82 (1 study) | ⊕⊝⊝⊝ very low2,4,17,18 | |
| Cholecystitis - Covered SEMS | 73 per 1000 | 8 per 1000 (1 to 139) | RR 0.11 (0.01 to 1.91) | 113 (1 study) | ⊕⊝⊝⊝ very low2,4,5 | |
| # patients with cholestatic symptoms to 2- FU Follow-up: 2 years | 250 per 1000 | 360 per 1000 (183 to 710) | RR 1.44 (0.73 to 2.84) | 79 (1 study) | ⊕⊝⊝⊝ very low2,4,17,18 | |
| Post-ES Haemorrhage | Study population | RR 3 (0.12 to 72.18) | 118 (1 study) | ⊕⊝⊝⊝ very low2,4,11,14 | ||
| 0 per 1000 | 0 per 1000 (0 to 0) | |||||
| Moderate | ||||||
| 0 per 1000 | 0 per 1000 (0 to 0) | |||||
| Hospitalisation Days | The mean hospitalisati on in the intervention groups was 0.49 standard deviations higher (0.21 to 0.77 higher) | 197 (2 studies) | ⊕⊝⊝⊝ very low4,11,14,16,17,18 | |||
| # >=30% decrease in serum bilirubin | 1000 per 1000 | 940 per 1000 (790 to 1000) | RR 0.94 (0.79 to 1.1) | 34 (1 study) | ⊕⊕⊝⊝ low9,16 | |
| % Reduction in total serum bilirubin levels | The mean % reduction in total serum bilirubin levels in the control groups was 74 percent age | The mean % reduction in total serum bilirubin levels in the intervention groups was 10.3 lower (32.51 lower to 11.91 higher) | 79 (1 study) | ⊕⊝⊝⊝ very low4,17,18,19,20 | ||
| Total Serum Bilirubin - rate of change | The mean total serum bilirubin - rate of change in the intervention groups was 0.23 standard deviations lower (0.62 lower to 0.17 higher) | 98 (1 study) | ⊕⊕⊝⊝ low1,16 | |||
The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; RR: Risk ratio; HR: Hazard ratio;
Soderlund et al. 2006 sample included 78% pancreatic cancer patients.
Crosses 2 default MIDs for dichotomous outcomes (0.8 and 1.25).
Not all included studies provided data regarding number of patients who were still alive or experienced stent dysfunction.
Majority of studies are high/unclear risk of bias due to insufficient reporting regarding blinding and incomplete reporting of outcomes.
Isayama et al. 2001 (all patients received endoscopic sphincterotomy).
Walter et al. 2015 (unclear whether blinding would affect outcome; selective reporting of outcomes).
Walter et al. 2015 included 75% pancreatic cancer patients.
Small sample size for dichotomous outcomes (<300 events).
Schmidt et al. 2015 (selective reporting of outcomes; study terminated early due to high rate of stent failure in plastic [winged] stent group).
Gardner et al. 2016 (unclear allocation concealment and blinding of outcome assessment; selective reporting of outcomes; participants were receiving 1 of 3 neoadjuvant chemoradiotherapy regimens).
Kaassis et al. 2003 (unclear randomisation method and allocation concealment; selective reporting of outcomes; significant difference in % weight loss at baseline; some patients also received sphincterotomy).
Travis et al. 1997 (unclear randomisation method, allocation concealment, blinding of personnel/participants/outcome assessment; imbalance in group numbers and selective reporting of outcomes).
Schmidt et al 2015 sample included 67% pancreatic cancer patients.
Kaassis et al. 2003 sample included 75% pancreatic cancer patients.
Gardner et al. 2016 includes both resectable (19%), borderline resectable (26%), and unresectable (55%) pancreatic cancer patients.
Crosses 1 default MID for dichotomous (0.8 or 1.25) or continuous outcomes (0.5 or −0.5).
Moses et al. 2013 (unclear randomisation method; selective reporting of outcomes).
Moses et al. 2013 sample included 68% pancreatic cancer patients.
MID for this outcome assumed to be 21.81/-21.81 (0.5 SD of control group at follow up; data from Moses et al. 2013).
Crosses 1 MID for this outcome.
The committee decided to consider all survival outcomes that were statistically significant, regardless of whether the 95% confidence interval crossed the default MIDs. Survival outcomes were therefore downgraded for imprecision by one level only if they were not statistically significant.
Not statistically significant.
From: 9, Interventions to relieve biliary and duodenal obstruction
NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.