Summary of findings 2. Duodenum‐preserving pancreatic resection versus pancreaticoduodenectomy for chronic pancreatitis (secondary outcomes).
| Patient or population: people requiring surgery for chronic pancreatitis Setting: surgical unit Intervention: duodenum‐preserving pancreatic resection Comparison: pancreaticoduodenectomy | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect (95% CI) | Number of participants (studies) | Quality of the evidence (GRADE) | Comments | |
| Risk with pancreaticoduodenectomy | Risk with duodenum‐preserving pancreatic resection | |||||
| Adverse events (proportion) | 363 per 1000 | 200 per 1000 (80 to 490) | RR 0.55 (0.22 to 1.35) | 226 (4 RCTs) | ⊕⊝⊝⊝ Very low1 2 3 | ‐ |
| Adverse events (number of adverse events) | 571 per 1000 | 543 per 1000 (246 to 1211) | Rate ratio 0.95 (0.43 to 2.12) | 43 (1 RCT) | ⊕⊝⊝⊝ Very low1 2 | ‐ |
| Length of hospital stay | The mean length of hospital stay was 14 days | The mean or median length of hospital stay in the intervention group was 1 to 5 days fewer | ‐ | 208 (4 RCTs) | ⊕⊕⊝⊝ Low1 | 2 trials that reported mean length of hospital stay reported statistically significant fewer hospital stay in intervention group (about 5 days fewer) (Farkas 2006; Klempa 1995), while 2 trials that reported the median length of hospital stay reported 1 day fewer hospital stay in intervention group, which was not statistically significant in 1 trial (Keck 2012), while the other trial did not report statistical significance (Buchler 1995) |
| Employed | 451 per 1000 | 694 per 1000 (451 to 1000) | RR 1.54 (1.00 to 2.37) | 189 (4 RCTs) | ⊕⊕⊝⊝ Low1 | ‐ |
| New‐onset diabetes | 239 per 1000 | 186 per 1000 (119 to 291) | RR 0.78 (0.50 to 1.22) | 269 (5 RCTs) | ⊕⊝⊝⊝ Very low1 2 | ‐ |
| Pancreatic exocrine insufficiency | 747 per 1000 | 620 per 1000 (508 to 762) | RR 0.83 (0.68 to 1.02) | 189 (4 RCTs) | ⊕⊝⊝⊝ Very low1,2,4 | ‐ |
| None of the trials reported the following outcomes: quality of life (4 weeks to 3 months), clinically significant pancreatic fistulas, serious adverse events, time to return to normal activity, time to return to work, and pain scores using a visual analogue scale. | ||||||
| *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group (mean control group proportion for all outcomes except short‐term mortality where an assumed risk of 1% was used as there was no short‐term mortality in the control group in the trials included in this review) and the relative effect of the intervention (and its 95% CI). CI: confidence interval; RCT: randomised controlled trial; RR: risk ratio | ||||||
| GRADE Working Group grades of evidence High quality: We are very confident that the true effect lies close to that of the estimate of the effect Moderate quality: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low quality: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect Very low quality: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect | ||||||
1 The trial(s) was/were of unclear or high risk of bias.
2 The confidence intervals were wide and the sample size was small.
3 The I2 value was high.
4 The I2 value was high and there was lack of overlap of confidence intervals.