Complex bile duct injuries after laparoscopic cholecystectomy: a comparative outcomes analysis of patients treated in tertiary private and public health facilities in Cape Town, South Africa

BACKGROUND

The South African healthcare system has an under-financed public sector serving most of the population and a better resourced private sector serving a small fraction of the population. This study evaluated management and outcome in patients with complex bile duct injuries (BDIs) after laparoscopic cholecystectomy referred from either private or public hospitals.

METHOD

The data of patients who underwent hepaticojejunostomy repair were retrieved from a prospectively maintained central departmental BDI database. Patients were treated either in the Surgical Gastroenterology Unit at Groote Schuur Hospital, University of Cape Town (UCT) or the Digestive Diseases Centre, UCT Private Academic Hospital by the same hepatobiliary surgical team. Relevant preoperative clinical data and postoperative complications and outcomes were compared between patients originating either in the public or private sector.

RESULTS

One hundred and twenty-five patients were included, 58 from the public and 67 from the private sector. The type of BDI, time to diagnosis, referral and repair were similar. Patients referred from the private sector underwent more percutaneous cholangiograms prior to referral (11.9% vs 1.7%, p = 0.037). Patients referred from the public sector underwent more CT examinations (p = 0.044) and endoscopic retrograde cholangiography (p = 0.038) after admission to our centre. There were no statistically significant differences in 30-day postoperative complications. Primary patency rates were similar for public and private referrals (90% vs 88%, respectively). There were two BDI-related mortalities at 90 days.

CONCLUSION

Despite differences in public and private healthcare system resources, patients were referred early and appropriately from both sectors and had similar postoperative outcomes when treated in a specialised unit.

Resection of benign liver tumours: an analysis of 62 consecutive cases treated in an academic referral centre

BACKGROUND

Benign tumours of the liver are increasingly diagnosed and constitute a substantial proportion of all hepatic tumours evaluated and resected at tertiary referral centres. This study assessed the safety and outcome after resection of benign liver tumours at a major referral centre.

METHOD

All patients with symptomatic benign liver tumours who underwent resection were identified from a prospective departmental database of a total of 474 liver resections (LRs). Demographic data, operative management and morbidity and mortality using the Accordion classification were analysed.

RESULTS

Sixty-two patients (56 women, 6 men, median age 45 years, range 17-82) underwent resection of symptomatic haemangiomata n=23 (37.1%), focal nodular hyperplasia n=19 (30.6%), biliary cystadenoma n=16 (25.8%) and hepatic adenomas n=4 (6.5%). A major resection was required in 25 patients, 14 patients had 4 segments resected, 11 had 3 segments and 37 patients had 2 or fewer segments resected. Median operating time was 169 minutes (range 80-410). Median blood loss was 300 ml (range 50-4500 ml) and an intra-operative blood transfusion was required in 6 patients. Median length of post-operative hospital stay was 7 days (range 4-32). Complications occurred in 11 patients (Accordion grades 1 n=1, 2 n=4, 3 n=1, 4 n=4, 6 n=1). Four patients required re-operation (bleeding n=2, bile leak n=1, small bowel obstruction n=1). An elderly patient died in hospital on day 16 following a postoperative cerebrovascular accident.

CONCLUSION

Clinically relevant symptomatic benign liver tumours comprise a substantial proportion of LRs. Our data suggest that resections can be performed safely with minimal blood loss and transfusion requirements. We advocate selective resection according to established indications. Despite the low postoperative mortality rate, the risk of postoperative complications emphasizes the need for careful selection of patients for resection.

Incidence and management of postoperative bile leaks: A prospective cohort analysis of 467 liver resections

BACKGROUND

Bile leaks from the parenchymal transection margin are a major cause of morbidity following major liver resections. The aim of this study was to benchmark the incidence and identify the risk factors for postoperative bile leakage after hepatic resection.

PATIENTS AND METHODS

A prospective database of 467 consecutive liver resections performed by the University of Cape Town HPB surgical unit between January 1990 and January 2016 was analysed. The relationship of demographic, clinical and perioperative factors to the development of bile leakage was determined. Bile leak and postoperative complications severity were graded using the International Study Group of Liver Surgery and Accordion classifications.

RESULTS

Overall morbidity was 24% (n = 112), with bile leaks occurring in 25 (5.4%) patients. Significantly more bile leaks occurred in patients who had major resections (≥ 3 segments) and longer total operative times (p < 0.05). There were 5 Grade A bile leaks which stopped spontaneously. Seventeen Grade B leaks required a combination of percutaneous drainage (n = 15), endoscopic biliary stenting (n = 8) and percutaneous transhepatic biliary drainage (n = 3). All 3 Grade C leaks required laparotomy for definitive drainage. Median hospital stay in the 442 patients without a bile leak was 8 days (IQR 1-98) compared with 12 days (IQR 6-30) for the 25 with bile leaks (p < 0.05) with no mortality. Major resections (≥ 3 segments) and total operative time (> 180mins) were significantly associated with bile leaks.

CONCLUSION

The incidence of bile leakage was 5.4% and occurred after major liver resections with longer operative times and resulted in significantly extended hospitalisation. Most were effectively treated nonoperatively by percutaneous drainage of the collection and/or endoscopic or percutaneous biliary drainage without mortality.

Endoscopic and operative treatment of delayed complications after pancreatic trauma: An analysis of 27 civilians treated in an academic Level 1 Trauma Centre

BACKGROUND

This study evaluated the efficacy of endoscopic treatment of delayed local complications including pseudocysts and persistent pancreatic fistulae in a cohort of civilian patients who had previously sustained a pancreatic injury.

METHOD

A large institutional database was interrogated to identify patients who developed a delayed pancreatic complication among those with pancreatic injuries treated between January 1990 and December 2013. The degree of the pancreatic duct injury was graded using a new duct injury grading system and endoscopic therapeutic outcome assessed according to the grade of injury.

RESULTS

During the period under review, 432 consecutive patients were treated for pancreatic injuries of whom 27 (20 men, 7 women, median age 31, range 15-68 years) presented with delayed complications related to the initial pancreatic injury. Sixteen patients had non-resolving symptomatic pancreatic pseudocysts, 10 had persistent pancreatic fistulae and 1 had a symptomatic duct stricture. Fourteen patients with grade 2a, 3a, 3b or 4c main pancreatic duct injuries were successfully treated endoscopically with either pancreatic duct stenting or pseudocyst drainage while 13 patients with grade 4a or 4b duct injuries who had complete duct division with a disconnected duct syndrome failed endoscopic management and required surgical intervention. The 27 patients underwent a total of 49 endoscopic procedures (47 elective, 2 emergency) of whom 4 developed complications related to the endoscopic treatment. All 4 resolved, 2 after urgent endoscopic re-intervention.

CONCLUSION

In this preliminary analysis the Cape Town pancreatic ductal injury grading classification showed a close correlation with outcome after endoscopic and operative intervention.

Prognostic factors, morbidity and mortality in pancreatic trauma: a critical appraisal of 432 consecutive patients treated at a Level 1 Trauma Centre

BACKGROUND

This large retrospective observational cohort study evaluated prognostic factors, 30-day morbidity and mortality and complications related to the pancreas in patients who had sustained pancreatic injuries.

METHODS

The records of 432 consecutive patients treated for pancreatic injuries at an urban Level 1 Trauma Centre in Cape Town between January 1982 and December 2012 were reviewed. Primary endpoints were postoperative morbidity and death. Bivariate and multivariate logistic regression analyses were used to assess significant predictors of morbidity and mortality.

RESULTS

Overall mortality in 432 patients [394 men, median age 26, median RTS 7.8] was 15.7% and morbidity 66%. Bivariate logistic regression analysis showed that nine factors, age, RTS, presence of shock, need for a transfusion, volume of blood transfused, damage control surgery, AAST grade of pancreatic injury, an associated vascular injury and a repeat laparotomy were significant predictors of morbidity. In the final multivariate logistic regression analysis model however only two variables, AAST grade of pancreatic injury and a repeat laparotomy were significant predictors of morbidity. When factors associated with mortality were considered, logistic regression analysis found that 11 variables, age, RTS, the presence of shock, patients who required a major blood transfusion, the median number of units transfused, the need for a damage control laparotomy, AAST grade 3, 4, 5 pancreatic injuries, associated vascular injuries, the number of associated injuries, postoperative complications and days in ICU were significant. However in the final stepwise multivariate logistic regression analysis model only five variables, age, shock, median number of units transfused and the presence of associated complications were significant factors associated with mortality.

CONCLUSIONS

Morbidity was 64% and AAST grade of pancreatic injury and a repeat laparotomy were significant predictors of morbidity. Overall mortality was 15.7%. Most deaths were due to associated injuries and were unrelated to the pancreatic injury. Five variables, age, shock, median number of units transfused and the presence of associated complications were significant factors associated with mortality. These data indicate that the magnitude of blood loss and haemorrhagic shock are primary determinants for survival and that urgent reversal of shock and control of bleeding are essential to reduce mortality in this cohort of patients.

An analysis of predictors of morbidity after stab wounds of the pancreas in 78 consecutive injuries

INTRODUCTION

Penetrating injuries of the pancreas may result in serious complications. This study assessed the factors influencing morbidity after stab wounds of the pancreas.

METHODS

A retrospective univariate cohort analysis was carried out of 78 patients (74 men) with a median age of 26 years (range: 16-62 years) with stab wounds of the pancreas between 1982 and 2011.

RESULTS

The median revised trauma score (RTS) was 7.8 (range: 2.0-7.8). Injuries involved the body (n=36), tail (n=24), head/uncinate process (n=16) and neck (n=2) of the pancreas. All 78 patients underwent a laparotomy. Sixty-five patients had AAST (American Association for the Surgery of Trauma) grade I or II pancreatic injuries and thirteen had grade III, IV or V injuries. Eight patients (10.3%) had an initial damage control operation. Sixty-nine patients (84.6%) had drainage of the pancreas only, six had a distal pancreatectomy and one had a pancreaticoduodenectomy. Most pancreas related complications occurred in patients with AAST grade III injuries; eight patients (10.2%) developed a pancreatic fistula. Four patients (5.1%) died. Grade of pancreatic injury (AAST grade I-II vs grade III-V injuries, p<0.001), RTS (odds ratio [OR]: 5.01, 95% confidence interval [CI]: 1.46-17.19, p<0.007), presence of shock on admission (OR: 3.31, 95% CI: 1.16-9.42, p=0.022), need for a blood transfusion (OR: 6.46, 95% CI: 2.40-17.40, p<0.001) and repeat laparotomy (p<0.001) had a significant influence on the development of general complications.

CONCLUSIONS

Although mortality was low after a pancreatic stab wound, morbidity was high. Increasing AAST grade of injury, high RTS, shock on admission to hospital, need for blood transfusion and repeat laparotomy were significant factors related to morbidity.

Morbidity and mortality after distal pancreatectomy for trauma: a critical appraisal of 107 consecutive patients undergoing resection at a Level 1 Trauma Centre

BACKGROUND

This study evaluated 30-day morbidity and mortality and assessed pancreas-specific complications in patients with major pancreatic injuries who underwent a distal pancreatectomy.

STUDY DESIGN

Records of 107 consecutive patients who underwent a distal pancreatectomy at a Level 1 Trauma Centre in Cape Town between January 1982 and December 2011 were reviewed. Primary endpoints were postoperative morbidity and death. Complications were graded according to the Clavien-Dindo severity classification and the International Study Group of Pancreatic Surgery (ISGPS) definitions.

RESULTS

A total of 107 patients [94 men, median age 26, median RTS 7.8, 69 penetrating injuries (63 gunshot wounds, 6 stabs wounds), 38 blunt injuries] underwent distal pancreatectomy. Overall mortality was 12%, 16% for gunshot injuries, 8% for blunt trauma and 0% in patients who had stab wounds. Eighty patients had a post-operative complication. A pancreatic leak (n=26) was the most common pancreatic related complication. Median postoperative stay in 28 patients with no or grade I complications was 9 days; in 11 patients with grade II complications was 18 days; in 14 grade IIIa, 31 days; in 19 grade IIIb, 38 days; in 8 grade IVa, 33 days in 14 grade IVb, and in 13 grade V the duration of postoperative stay was 14±39.4 days.

CONCLUSIONS

Overall mortality for distal pancreatectomy was 12%. Pancreatic leak was a common cause of morbidity. Length of hospitalisation increased with increasing Clavien-Dindo severity grading. There was a significant difference in the duration of hospitalisation in patients with no or grade I complications compared to those with grade II-IV injuries (p<0.05).

Surgical management and outcome of civilian gunshot injuries to the pancreas

BACKGROUND

Pancreatic injuries are uncommon but result in substantial morbidity and mortality. This study evaluated the factors associated with morbidity and mortality in civilian patients with pancreatic gunshot wounds.

METHODS

This was a single-institution, retrospective review of patients with gunshot wounds of the pancreas treated from 1976 to 2009 in Cape Town, South Africa. Univariable and multivariable analyses were performed.

RESULTS

A total of 219 patients (205 male, median age 27 years) had pancreatic American Association for the Surgery of Trauma grade I-II (111 patients) and grade III-V (108) gunshot injuries to the pancreatic head (72), neck (8), body (75) and tail (64). The patients underwent 239 laparotomies, including drainage of the pancreas (169), distal pancreatectomy (59) and pancreaticoduodenectomy (11). Some 218 patients had 642 associated intra-abdominal and 91 vascular injuries. Forty-three (19.6 per cent) required an initial damage control procedure. A total of 150 patients (68.5 per cent) had 407 postoperative complications (median 4, range 1-7). The 46 patients (21.0 per cent) who died had a median of 3 (range 1-7) complications. Median (range) intensive care unit and total hospital stay were 5 (1-153) and 11 (1-255) days respectively. Multivariable analyses identified age, high-grade pancreatic injury, associated vascular injuries and need for repeat laparotomy as predictors of morbidity. Age, shock on admission, need for damage control surgery, high-grade pancreatic injuries and associated vascular injuries were significant factors associated with mortality.

CONCLUSION

Morbidity and mortality rates were high after gunshot injuries to the pancreas. Initial shock and severe injury combined with need for damage control surgery were associated with the highest risk of death.

Pancreatic injuries after blunt abdominal trauma: an analysis of 110 patients treated at a level 1 trauma centre

BACKGROUND AND OBJECTIVE

Injuries to the pancreas are uncommon, but may result in considerable morbidity and mortality. This study evaluated the management of blunt pancreatic injuries using a previously defined protocol to determine which factors predicted morbidity and mortality.

METHODS

The study design was a retrospective chart review of all adult patients with blunt pancreatic injuries treated at a level 1 trauma centre between March 1981 and June 2009.

RESULTS

One hundred and ten patients (92 men, 18 women; mean age 30 years, range 13-68 years) were treated during the study period. Forty-six patients had American Association for the Surgery of Trauma (AAST) grade 1 or 2 pancreatic injuries and 64 had AAST grade 3, 4 or 5 pancreatic injuries. Injuries involved the head (N=21), neck (N=15), body (N=48) and tail (N=26) of the pancreas. The mean number of organs injured was 2.7 per patient (range 1-4). One hundred and one patients underwent a total of 123 operations, including drainage of the pancreatic injury (N=73), distal pancreatectomy (N=39) and Whipple resection (N=5). The overall complication rate was 74.5% and the mortality rate 16.4%. Only 2 of the 18 deaths were attributable to the pancreatic injury. Shock on presentation was highly predictive of death; 17 of 39 patients with shock died, compared with 1 of 71 patients who were not shocked (p < 0.0001). Fourteen of 46 patients with grade 1 and 2 pancreatic injuries died compared with 4 of 64 patients with grades 3, 4 and 5 injuries (p < 0.001). Mortality increased exponentially as the number of associated injuries increased. Two of 57 patients with injury to the pancreas only or one associated injury died, compared with 16 of 53 with two or more associated injuries (p < 0.0013).

CONCLUSIONS

This study demonstrated a significant correlation between the AAST grade of injury and pancreas-specific morbidity and between shock on admission, the number of associated injuries and death, in patients with blunt pancreatic injuries. Although morbidity and mortality rates after blunt pancreatic trauma are high, death was usually the result of major associated injuries and not related to the pancreatic injury.

Early rebleeding and death at 6 weeks in alcoholic cirrhotic patients with acute variceal bleeding treated with emergency endoscopic injection sclerotherapy

BACKGROUND

This study evaluated the incidence of rebleeding and death at 6 weeks after a first episode of acute variceal haemorrhage (AVH) treated by emergency endoscopic sclerotherapy in a large cohort of alcoholic cirrhotic patients.

METHODS

From January 1984 to December 2006, 310 alcoholic cirrhotic patients (242 men, 68 women; mean age 51.7 years) with AVH underwent 786 endoscopic variceal injection treatments (342 emergency, 444 elective) during 919 endoscopy sessions in the first 6 weeks after the first variceal bleed. Endoscopic control of initial bleeding, variceal rebleeding and survival at 6 weeks were recorded.

RESULTS

Endoscopic intervention controlled AVH in 304 of 310 patients (98.1%). Seventy-five patients (24.2%) rebled, 38 (12.3%) within 5 days and 37 (11.9%) within 6 weeks. No patient scored as Child-Pugh A died. Seventy-seven (24.8%) Child-Pugh B and C patients died, 29 (9.3%) within 5 days and 48 (15.4%) between 6 and 42 days. Mortality increased exponentially as the Child-Pugh score increased, reaching 80% when the score exceeded 13.

CONCLUSION

Despite initial control of variceal haemorrhage, 1 in 4 patients (24.2%) rebled within 6 weeks. Survival at 6 weeks was 75.2% and was influenced by the severity of liver failure, with most deaths occurring in Child-Pugh grade C patients.