Key points are not available for this paper at this time.
BACKGROUND: The optimal initial management of left-sided malignant colonic obstruction (LMCO) remains debated, as previous studies have reported inconsistent results. To address this uncertainty, we conducted a systematic review and Bayesian network meta-analysis comparing various initial management strategies for potentially curable LMCO. The objective was to evaluate and compare short-term and long-term outcomes among bridge-to-surgery (BTS) approaches and immediate surgery. METHODS: PubMed, Embase and the Cochrane Library were systematically searched until the end of September 2025. Eligible randomized controlled trials (RCTs) or propensity score (PS)-based comparative studies reported outcomes of immediate surgery or BTS using self-expandable metallic stents (SEMS), decompression stoma or transanal decompression tube (TDT). A Bayesian network meta-analysis was performed to assess survival outcomes, postoperative morbidity and stoma-related outcomes. RESULTS: Nine RCTs and 14 PS-based studies comprising 5164 patients were included. Procedure-related perforation occurred in 6.8% of SEMS (100/1467) and 7.5% of TDT (13/173) cases. No significant differences were observed among groups regarding survival outcomes or postoperative morbidity. The SEMS group was associated with a significantly lower overall stoma formation rate than immediate surgery (RR: 0.56, 95% CrI: 0.44-0.73) and TDT (RR: 0.44, 95% CrI: 0.23-0.82). Although decompression stoma required temporary stoma formation in all cases, it was associated with the lowest risk of permanent stoma formation. CONCLUSIONS: All strategies showed comparable survival and morbidity outcomes. The use of SEMS as a BTS approach was associated with a lower overall stoma formation rate. Decompression stoma as a BTS approach was associated with a lower risk of permanent stoma formation and may represent a reasonable alternative when SEMS is unsuitable and temporary stoma is acceptable.
Kitaguchi et al. (Mon,) studied this question.