Defining the role of surgery for complications after pancreatoduodenectomy

Deviations from a clinical pathway post pancreatoduodenectomy predict 90-day unplanned re-admission

Aim: To determine the frequency and relevance of deviations from a post pancreatoduodenectomy (PD) clinical care pathway. Materials & methods: A retrospective analysis using a prospectively maintained database of a post-PD clinical care pathway was carried out between May 2016 and March 2018. Patients were divided based on the number of factors deviating from the clinical care pathway (Group I: no deviation; Group II: deviation in 1-4 factors; Group III: deviation in 5-8 factors). The analysis included profiling of patients on different demographic and clinical as well as medical and surgical outcome parameters (discharge by postoperative day 8 and 90-day unplanned re-admission rate). Results: Post-PD clinical care pathways are feasible but deviations from the pathway are frequent (91%). An increase in frequency of deviations from the pathway was significantly associated with increased risk of POPF and delayed gastric emptying, delayed discharge, risk of mortality and 90-day unplanned re-admission rate. Conclusion: Deviations from a post-PD clinical care pathway are common. Poor nutrition and cardiac co-morbidities are associated with an increased likelihood of deviation. As the number of deviations increase, so does the risk of significant complications and interventions, delayed discharge and 90-day re-admission rate.

Risk factors for 30-day unplanned reoperation after pancreatoduodenectomy: A single-center experience

Objective

The purpose of this study was to investigate the rate and reasons and also the risk factors for unplanned reoperation after pancreatoduodenectomy (PD) in a single center.

Patients and Methods

This retrospective analysis included patients who underwent PD in the First Affiliated Hospital of Nanchang University between January 2010 and January 2018. The patients were divided into nonreoperation and reoperation groups according to whether they underwent unplanned reoperation following the primary PD. The incidence and reasons were examined. In addition, multivariate logistic regression analysis was performed to identify the risk factors for unplanned reoperation.

Results

Of the 330 patients who underwent PD operations, 22 (6.67%) underwent unplanned reoperation. The main reasons for reoperation were postpancreaticoduodenectomy hemorrhage (PPH) (12/22 [54.5%]) and pancreaticoenteric anastomotic (PEA) leak (5/22 [22.7%]). Multivariate logistic regression analyses identified that diabetes (odds ratio [OR], 3.70; 95% confidence interval [CI], 1.06-12.90; P = 0.04), intraoperative blood loss ≥400 mL (OR, 4.06; 95% CI, 1.29-12.84; P = 0.02), occurrence of postoperative complications in the form of PPH (OR, 30.67; 95% CI, 8.85-106.31; P < 0.001), and PEA leak (OR, 11.53; 95% CI, 3.03-43.98, P < 0.001) were independent risk factors for unplanned reoperation.

Conclusions

Our results suggest that diabetes, intraoperative blood loss ≥400 mL, PPH, and PEA leak were independent risk factors for unplanned reoperation after primary PD.

Risk Factors of Reoperation After Pancreatic Resection

BACKGROUND

Pancreatic resection is associated with a high incidence of postoperative complications, some of which require reoperation.

AIMS

To analyze the incidence of and risk factors for reoperation following pancreatectomy.

METHODS

Pre- and postoperative information and procedure characteristics of 15,549 patients having undergone pancreatectomy in 435 hospitals participating in the American College of Surgeons National Surgical Quality Improvement Program from 2011 to 2014 were analyzed.

RESULTS

A total of 773 (5.0%) patients required reoperation within 30 days of their index pancreatectomy. Patients requiring reoperation were more likely to be younger (mean ± standard deviation, 62.6 ± 13.2 vs. 64.1 ± 12.2 years, p < 0.001), male (60 vs. 49%, p < 0.001), to have respiratory comorbidities, lower preoperative serum albumin (3.7 ± 0.68 vs. 3.8 ± 0.62 mg/dl, p < 0.001), higher total bilirubin (1.7 ± 2.7 vs. 1.5 ± 2.4 mg/dl, p = 0.02), and higher American Society of Anesthesiologists (ASA) class than those who did not undergo reoperation. Other factors associated with increased incidence of reoperation included longer mean operative duration at the index procedure, postoperative transfusion requirement, wound complications, and cardiorespiratory, renal, thromboembolic, and infectious events. Multivariate regression analysis identified male sex, preoperative serum albumin <3.5 mg/dl, ASA class of 3 or 4, pancreaticoduodenectomy, and total pancreatectomy as the strongest predictors for reoperation after index pancreatic resection. Complication and readmission rates were significantly higher for those undergoing reoperation.

CONCLUSION

Patient characteristics and procedural factors contribute to reoperation after pancreatectomy in this largest and most diverse sample to date. Further investigation to identify perioperative strategies for mitigating this risk is required to improve the safety of pancreatic resection.

Randomized clinical trial: nasoenteric tube or jejunostomy as a route for nutrition after major upper gastrointestinal operations

BACKGROUND

Curative treatment of upper gastrointestinal tract neoplasms is complex and associated with high morbidity and mortality. In general, the patients are already malnourished, and early postoperative enteral nutrition is recommended. However, there is no consensus concerning the best enteral access route in these cases.

METHODS

A prospective randomized trial was performed from 2008 to 2012 with 59 patients who underwent esophagectomy, total gastrectomy, or pancreaticoduodenectomy. In all, 4 patients declined to participate, and 13 did not meet the inclusion criteria and were excluded. Of the 42 included patients, 21 had nasoenteric tubes, and 21 had a jejunostomy.

RESULTS

The two groups were similar in demographic and clinical aspects. The nasoenteric (NE) and jejunostomy groups underwent early enteral therapy in 71 and 62 % of cases (p > 0.05), respectively. The median length of enteral therapy use was less in the NE group (5.0 vs. 8.5 days), but the difference was not statistically significant. The NE group required introduction of parenteral therapy more frequently than the jejunostomy group (p < 0.05). Complications related to the enteral route occurred in 38.0 and 28.5 % of patients (p > 0.05) in the NE and jejunostomy groups, respectively. In the NE group, there were four losses and four tube obstructions. In the jejunostomy group, there were two losses, four obstructions, and two cases of leakage around the tube. In the latter group, patients who underwent therapy for a longer time had tubal complications (p < 0.05) and longer intensive care unit and hospital stays (p < 0.05).

CONCLUSION

The two enteral routes were associated with the same number of complications. However, the presence of a jejunostomy allowed enteral therapy for longer periods, especially in patients with complications, thus avoiding the need for parenteral nutrition.

A comparison of two invagination techniques for pancreatojejunostomy after pancreatoduodenectomy

Background. The aim of the study was to compare two invagination techniques for pancreatojejunostomy after pancreatoduodenectomy. Methods. For effective prevention of the development of pancreatic leakage, we modified invagination technique that we term the "serous touch." We analysed the diameter of the main pancreatic duct, the texture of the remnant pancreas, the method of the reconstruction, pancreatic external drainage, anastomotic procedure time, histopathological examination, and postoperative complications. Results. Fifty-two patients underwent pancreatoduodenectomy with pancreatojejunostomy using "serous touch" technique (ST group) and 52 classic pancreatojejunostomy (C group). In the ST group one patient (1.9%) was diagnosed as grade B pancreatic fistula, and no patient experienced fistula grade A or C. In the C group 6 patients (11.5%) were diagnosed as fistula grade A, 1 (1.9%) patient as fistula grade B, and 1 (1.9%) patient as fistula grade C. There was a significant statistical difference in incidents of pancreatic fistula (P < 0.05) and no statistical difference in other postoperative complications or mortality in comparison group. Anastomosis time was statistically shorter in the ST group. Conclusions. "Serous touch" technique appeared to be easy, safe, associated with fewer incidences of pancreatic fistulas, and less time consuming in comparison with classical pancreatojejunostomy.

Use and results of consensus definitions in pancreatic surgery: a systematic review

BACKGROUND

Because of the lack of standardized definitions of complications in gastrointestinal operations, consensus definitions have been developed in recent years. The aim of the current study was to systematically review the available consensus definitions and to report their use, acceptance, and results.

METHODS

A systematic search of the literature was conducted of the Medline, Cochrane, and ISI Web of Science databases. All articles published until August 2011 and that applied the identified consensus definitions were considered. Inclusion criteria for quantitative analysis were studies with correct usage of the definition and 100 or more patients who were treated after the year 2000.

RESULTS

Seven consensus definitions were identified: postoperative pancreatic fistula, postpancreatectomy hemorrhage, delayed gastric emptying, posthepatectomy liver failure, bile leakage after hepatobiliary and pancreatic surgery, posthepatectomy hemorrhage, and anastomotic leakage after anterior resection of the rectum. Of 1,637 articles retrieved from the literature search, 59 articles that correctly applied the definitions met the inclusion criteria. Subanalyses were feasible for definitions after pancreatic surgery. According to the consensus definitions, the median complication rates of retrospective studies were 21.9% (postoperative pancreatic fistula, n = 11,244 patients), 5.9% (postpancreatectomy hemorrhage, n = 3,311 patients), and 22.8% (delayed gastric emptying, n = 4,553 patients) after pancreatic resections. The incidences were not substantially different in prospective trials. Validation was performed for all three definitions, demonstrating that the severity grades significantly correlated with the clinical course of the patients.

CONCLUSION

The available consensus definitions were increasingly cited and facilitate scientific comparability and transparency if appropriately applied. The present data update the incidences of major pancreatic complications.

Evolution of pancreatoduodenectomy in a tertiary cancer center in India: improved results from service reconfiguration

BACKGROUND

Pancreatic cancer incidence in India is low. Over the years, refinements in technique of pancreatoduodenectomy (PD) may have improved outcomes. No data is available from India, South-Central, or South West Asia to assess the impact of these refinements.

PURPOSE

To assess the impact of service reconfiguration and standardized protocols on outcomes of PD in a tertiary cancer center in India.

METHODS

Three specific time periods marking major shifts in practice and performance of PD were identified, viz. periods A (1992-2001; pancreaticogastrostomy predominantly performed), B (2003-July 2009; standardization of pancreaticojejunal anastomosis), and C (August 2009-December 2011; introduction of neoadjuvant chemo-radiotherapy and increased surgical volume).

RESULTS

500 PDs were performed with a morbidity and mortality rate of 33% and 5.4%, respectively. Over the three periods, volume of cases/year significantly increased from 16 to 60 (p < 0.0001). Overall incidence of post-operative pancreatic anastomotic leak/fistula (POPF), hemorrhage, delayed gastric emptying (DGE), and bile leak was 11%, 6%, 3.4%, and 3.2%, respectively. The overall morbidity rates, as well as, the above individual complications significantly reduced from period A to B (p < 0.01) with no statistical difference between periods B and C.

CONCLUSION

Evolution of practice and perioperative management of PD for pancreatic cancer at our center improved perioperative outcomes and helped sustain the improvements despite increasing surgical volume. By adopting standardized practices and gradually improving experience, countries with low incidence of pancreatic cancer and resource constraints can achieve outcomes comparable to high-incidence, developed nations.

SYNOPSIS

The manuscript represents the largest series on perioperative outcomes for pancreatoduodenectomy from South West and South-Central Asia - a region with a low incidence of pancreatic cancer and a disproportionate distribution of resources highlighting the impact of high volumes, standardization and service reconfiguration.

Reoperation following Pancreaticoduodenectomy

Introduction. The literature on reoperation following pancreaticoduodenectomy is sparse and does not address all concerns. Aim. To analyze the incidence, causes, and outcome of patients undergoing reoperations following pancreaticoduodenectomy. Methods. Retrospective analysis of 520 consecutive patients undergoing pancreaticoduodenectomy from May 1989 to September 2010. Results. 96 patients (18.5%) were reoperated; 72 were early, 18 were late, and 6 underwent both early and late reoperations. Indications for early reoperation were post pancreatectomy hemorrhage in 53 (68%), pancreatico-enteric anastomotic leak in 10 (13%), hepaticojejunostomy leak in 3 (3.8%), duodenojejunostomy leak in 4 (5%), intestinal obstruction in 1 (1.2%) and miscellaneous causes in 7 (9%). Patients reoperated early did not fare poorly on long-term follow up. Indications for late reoperations were complications of index surgery (n = 12), recurrence of the primary disease (n = 8), complications of adjuvant radiotherapy (n = 3), and gastrointestinal bleed (n = 1). The median survival of 16 patients reoperated late without recurrent disease was 49 months. Conclusion. Early reoperations following pancreaticoduodenectomy, commonly for post pancreatectomy hemorrhage, carries a high mortality due to associated sepsis, but has no impact on long-term survival. Long-term complications related to pancreaticoduodenectomy and adjuvant radiotherapy can be managed successfully with good results.

Superior mesenteric artery first combined with uncinate process approach versus uncinate process first approach in pancreatoduodenectomy: a comparative study evaluating perioperative outcomes

PURPOSE

During pancreatoduodenectomy (PD), two techniques have been described to dissect the head of pancreas, viz. the superior mesenteric artery (SMA) approach by dissecting the uncinate process and the uncinate process first approach.

METHODS

Forty-four consecutive patients, who underwent PD between June 2009 and April 2010, were analyzed. Thirty patients underwent the SMA first approach along with uncinate dissection (group 1), while 14 patients underwent the uncinate process first approach (group 2).

RESULTS

There were 30 male and 14 female patients. The median age was 51 years (range 19-76 years). Median intraoperative blood loss in group 1 was 800 ml, while that in group 2 was 600 ml. A mean of 0.52 units of blood were transfused in group 1 (range 0-3) compared to 0.2 units in group 2 (range 0-1). The median operative time in group 1 was 457.5 min and the median operative time was 450 min in group 2. Complication rate was 40% and 14.3% in groups 1 and 2, respectively. Median duration of hospital stay was 14 days in group 1 and 12.5 days in group 2. Median nodes resected in group 1 were 8 (range 0-26), while in group 2 they were 9 (range 2-14). Resection margins were positive in two cases (one in each group). There were two mortalities in group 1 and no mortalities in group 2. None of the above differences were significant.

CONCLUSIONS

SMA first is a safe technique. It compares well with the uncinate first approach in terms of operative time, blood loss, number of lymph nodes retrieved, margin positivity and operative morbidity. Both techniques may be useful in situations such as a large uncinate process tumor or when superior mesenteric vein/portal vein/superior mesenteric artery involvement is suspected or present. Further studies, evaluating data related to specific predefined uncinate tumors, would be the next logical step in further defining the precise role of these techniques.

Borderline resectable pancreatic tumors: is there a need for further refinement of this stage?

BACKGROUND

The ideal treatment of patients with "borderline resectable pancreatic tumors (BRTs)" needs to be established. Current protocols advise neoadjuvant chemo(radio)therapy, although some patients may appear to have BRT on preoperative imaging and a complete resection may be achieved without the need for vascular resection. The aim of the present study was to identify specific findings on preoperative imaging that could help predict in which patients with BRT a complete resection, with or without vascular resection (VR), could be achieved.

METHODS

Twelve patients with BRTs were identified. Tumor location, maximum degree of circumferential contact (CC), length of contact of the tumor with major vessels (LC), and luminal narrowing of vessels at the point of contact with the tumor (venous deformity, VD) were graded on preoperatively acquired multidetector computed tomography (MDCT) images and then compared with the intraoperative findings and need for VR.

RESULTS

A complete resection (R0) was achieved in 10 patients with 2 having microscopic positive margins (R1) on histopathology at the uncinate margin. Four of the 10 patients required VR (40%). In 3 of the 4 patients whose tumors required VRs, CC was ≥grade III and VD was grade 2. LC did not influence the need for VR.

CONCLUSIONS

It is possible to achieve a complete resection at the first instance in patients found to have BRTs on preoperative imaging. Preoperative MDCT-based grading systems and our proposed criteria may help identify such patients, thus avoiding any delay in curative resections in such patients.

Vascular anomalies encountered during pancreatoduodenectomy: do they influence outcomes?

BACKGROUND

Because of the potential risk of hemorrhage or ischemia, the presence of vascular anomalies adds to the surgical challenge in pancreatoduodenectomy (PD).

OBJECTIVE

To analyze the literature concerning the influence of aberrant peripancreatic arterial anatomy on outcomes of PD.

MATERIALS AND METHODS

A systematic search using Medline and Embase for the years 1950-2008.

RESULTS

The most common aberration in hepatic arterial anatomy is the replaced right hepatic artery. Other vascular abnormalities such as replaced common hepatic artery with a hepatomesenteric trunk and celiomesenteric trunk and arcuate ligament syndrome leading to celiac artery stenosis are also associated with post-PD complications. Damage to the biliary branches of the hepatic arteries increases the risk of postoperative biliary anastomotic leak.

CONCLUSION

The most common abnormalities of the hepatic vasculature include a replaced RHA, replaced LHA, and accessory RHA or LHA. Celiac artery stenosis secondary to median arcuate ligament compression may also be encountered. Every attempt should be made to preserve the aberrant vessel unless their resection is oncologically indicated. Routine preoperative computerized tomography angiography helps to identify the hepatic vascular anatomy and thereby prepares the surgeon to better deal with the vascular anomalies intraoperatively. Increased awareness of the vascular anatomy would decrease the chances of intraoperative vascular injury and consequent postoperative complications such as biliary anastomotic leaks as well as the chances of postoperative hemorrhage.

Peri-operative outcomes for pancreatoduodenectomy in India: a multi-centric study

BACKGROUND

There have been an increasing number of reports world-wide relating improved outcomes after pancreatic resections to high volumes thereby supporting the idea of centralization of pancreatic resectional surgery. To date there has been no collective attempt from India at addressing this issue. This cohort study analysed peri-operative outcomes after pancreatoduodenectomy (PD) at seven major Indian centres.

MATERIALS AND METHODS

Between January 2005 and December 2007, retrospective data on PDs, including intra-operative and post-operative factors, were obtained from seven major centres for pancreatic surgery in India.

RESULTS

Between January 2005 and December 2007, a total of 718 PDs were performed in India at the seven centres. The median number of PDs performed per year was 34 (range 9-54). The median number of PDs per surgeon per year was 16 (range 7-38). Ninety-four per cent of surgeries were performed for suspected malignancy in the pancreatic head and periampullary region. The median mortality rate per centre was four (range 2-5%). Wound infections were the commonest complication with a median incidence per centre of 18% (range 9.3-32.2%), and the median post-operative duration of hospital stay was 16 days (range 4-100 days).

CONCLUSIONS

This is the first multi-centric report of peri-operative outcomes of PD from India. The results from these specialist centers are very acceptable, and appear to support the thrust towards centralization.