Complications of supraduodenal choledochotomy: A comparison of three methods of management
โ Scribed by M. R. B. Keighley; D. W. Burdon; R. M. Baddeley; N. J. Dorricott; G. D. Oates; G. T. Watts; J. Alexander-Williams
- Publisher
- John Wiley and Sons
- Year
- 1976
- Tongue
- English
- Weight
- 498 KB
- Volume
- 63
- Category
- Article
- ISSN
- 0007-1323
No coin nor oath required. For personal study only.
โฆ Synopsis
Abstract
The complications of elective choledochotomy in a prospective study of 116 patients with suspected intraduct calculi are reported. Management included T tube drainage (n = 59), primary closure of the bile duct (n = 29) and choledochoduodenostomy (n = 28).
Septicaemia occurred in 12 patients (10 per cent), with 1 death, and was unrelated to the type of operation. Thirty patients (26 per cent) developed wound infection; this complication was more common after T tube drainage than the other procedures. Intra-abdominal abscess occurred in 3 patients only. Thrombo-embolism was recorded in 10 patients (9 per cent), 7 of whom had an intraduct drain. Postoperative pancreatitis occurred in 5 patients (4 per cent), with 2 deaths; a third of the patients in whom sphincteroplasty had been combined with supraduodenal choledochotomy developed this complication. Reoperation for stones was required in 3 patients with T tube; 3 patients developed a temporary biliary fistula after choledochoduodenostomy. The hospital stay was 9ยท5 days after primary closure, 14ยท0 days after choledochoduodenostomy and 16ยท8 days after T tube drainage.
Wound sepsis (32 per cent) and thrombo-embolism (12 per cent) were more common in patients with bacterbilia than in patients where the bile was sterile at operation (13 and 3 per cent respectively). Furthermore, wound sepsis, septicaemia and thrombo-embolism were reduced in patients who were given effective antibiotic cover.
๐ SIMILAR VOLUMES
Three inferential morphometric methods, Euclidean distance matrix analysis (EDMA), Bookstein's edge-matching method (EMM), and the Procrustes method, were applied to facial landmark data. A Monte Carlo simulation was conducted with three sample sizes, ranging from n = 10 to 50, to assess type I erro