Vennix 2015 — Ladies Trial (LOLA group)
Citation: Vennix S, Musters GD, Mulder IM, et al; on behalf of the Ladies trial collaborators. Laparoscopic peritoneal lavage or sigmoidectomy for perforated diverticulitis with purulent peritonitis: a multicentre, parallel-group, randomised, open-label trial. Lancet. 2015;386(10000):1269-1277. doi:10.1016/S0140-6736(15)61168-0.
Study design
- LOLA group of the Ladies trial (a two-part study; DIVA group — Hartmann’s vs. primary anastomosis — reported separately and now independently ingested, see Lambrichts 2019 - LADIES DIVA Trial). Multicenter (34 teaching + 8 academic hospitals; Belgium, Italy, Netherlands), open-label, randomized 2:1:1 (lavage : Hartmann’s : primary anastomosis, giving a 1:1 lavage-vs-sigmoidectomy comparison).
- Population: Hinchey III (purulent) perforated diverticulitis confirmed at diagnostic laparoscopy; excluded Hinchey I/II, IV, age >85, high-dose steroids, hemodynamic instability.
- Primary endpoint: composite of major morbidity + mortality within 12 months.
- Trial stopped early by the Data Safety Monitoring Board after 90 patients enrolled (of a planned 264), due to a high morbidity/reintervention rate in the lavage arm at interim analysis. 46 lavage / 42 resection patients in the final modified-ITT analysis.
Key findings
| Outcome | Lavage (n=46) | Resection (n=42) | p |
|---|---|---|---|
| Primary composite (12mo major morbidity/mortality) | 67% (30) | 60% (25) | .58 |
| Short-term (30-day) major morbidity | 39% (18) | 19% (8) | .043 |
| Abscess needing drainage (short-term) | 20% (9) | 0% | .003 |
| In-hospital reinterventions (interim-analysis trigger) | 18 pts | 2 pts | .001 |
| Mortality (12mo) | 9% (4) | 14% (6) | .62 |
| Recurrent diverticulitis | 20% (9) | 2% (1) | .03 |
| Stoma-free & alive at 12mo | 78% | 71% | .42 |
- Operating time shorter with lavage (60 vs 120 min); hospital stay and QoL (SF-36, GIQLI, EQ-5D) did not differ.
- Sepsis control failure in 24% of lavage patients — attributed mainly to misdiagnosed fecal peritonitis (Hinchey III vs IV is not reliably distinguishable preoperatively/intraoperatively by the study’s exploration protocol).
- 8% (7/88) of the overall cohort had an underlying sigmoid carcinoma, often diagnosed at follow-up colonoscopy/pathology — comparable to rates in other perforated-diverticulitis trials.
- No excess mortality in patients who “failed” lavage and needed reintervention — suggests failures are salvageable with timely reoperation.
Limitations
- Stopped early (33% of planned sample size) — underpowered for the primary endpoint; a significant DSMB-driven safety signal at interim analysis doesn’t necessarily generalize as a precise effect size.
- Open-label; not blinded.
- Low accrual rate (34% of eligible patients enrolled), though the authors did a chart-review check for selection bias and did not find one.
Relevance — cross-trial comparison with SCANDIV
This is the trial SCANDIV Trial’s discussion cited but this wiki hadn’t ingested yet. Reading them together:
- Recurrence signal replicates cleanly: Ladies trial 20% (lavage) vs 2% (resection); SCANDIV 5-year data 21% vs 4%. Two independent European RCTs, consistent direction and similar magnitude — this is the most robust finding across both trials.
- Severity of the safety signal differs: Ladies trial was stopped early specifically because of a stark interim reintervention signal (18 vs 2 patients, p=.001) and showed significantly higher short-term major morbidity (39% vs 19%, p=.043). SCANDIV’s 5-year report, by contrast, found no significant difference in severe complications (29% vs 25%, p=.58) — a materially more reassuring picture for lavage on that specific outcome.
- Both trials agree: no mortality or QoL difference; lower stoma burden with lavage; same underlying mechanism for lavage failure (missed fecal peritonitis / misdiagnosed cancer).
- Open question this raises rather than resolves: is the discrepancy in short-term morbidity signal (stark in Ladies, absent in SCANDIV by 5-year report) due to different endpoint definitions/timepoints, different patient selection, or does it reflect real heterogeneity in how “purulent peritonitis” was diagnosed/confirmed across trials? Worth flagging if a future source (e.g. a meta-analysis pooling DILALA/Ladies/SCANDIV) addresses this directly.