부산대학교 도서관

Aim: Percutaneous endoscopic colostomy (PEC) represents an important intervention in specific patients. Limited data currently exist. We present the largest recorded study of patients undergoing PEC. Methods: Retrospective analysis of consultant logbooks highlighted all patients from 1997 to 2020. Two independent reviewers assessed records. Parameters measured were age, sex, indication, number of sites, complications, mortality and survival. Three subgroups were identified: recurrent sigmoid volvulus (RSV), pseudo‐obstruction and neurogenic. ANOVA, chi‐squared and Fischer's exact tests were utilized; Kaplan–Meier curves estimated survival and the log‐rank test was applied. A p value of <0.05 was considered statistically significant. Results: Ninety‐six PEC insertions were done on 91 patients (five reinsertions). There were 66 men (69%) and the mean age was 73.1 years (interquartile range 23). The indications were RSV n = 72, pseudo‐obstruction n = 13, neurogenic n = 11. The 30‐day complication rate was overall n = 27 (28%), RSV n = 23, pseudo‐obstruction n = 4. Nine patients leaked (9.9%) (eight RSV, one pseudo‐obstruction), of whom five died. 90‐day mortality was 14.6% (14 patients), 18.5% (13/72) for RSV, 7.7% (1/13) for pseudo‐obstruction. Overall recurrence following PEC was 10.4%. The median follow‐up was 25 months (interquartile range 4.6–62.2 months). At 3, 5 and 10 years survival was 46%, 34% and 26% for RSV, 70%, 55% and 15% for pseudo‐obstruction and 91%, 91% and 81% for neurogenic respectively. Conclusion: Recurrent sigmoid volvulus and pseudo‐obstruction patients undergoing PEC compared to neurogenic patients have poorer outcomes with higher complication rates and shorter life expectancy. We advocate that high volume specialist units undertake PEC. The significant associated risks of PEC require careful consideration when determining patient suitability. Utilizing risk stratification scores may help guide shared decision making between patients, relatives and clinicians. [ABSTRACT FROM AUTHOR]