Root-Cause Analysis of Mortality Following Pancreatic Resection (CARE Study): A Multicenter Cohort Study

Preoperative body composition measured by bioelectrical impedance analysis can predict pancreatic fistula after pancreatic surgery

BACKGROUND

Postoperative pancreatic fistula (POPF) remains one of the most severe complications after pancreatic surgery. The methods for predicting pancreatic fistula are limited. We aimed to investigate the predictive value of body composition parameters measured by preoperative bioelectrical impedance analysis (BIA) on the development of POPF.

METHODS

A total of 168 consecutive patients undergoing pancreatic surgery from March 2022 to December 2022 at our institution were included in the study and randomly assigned at a 3:2 ratio to the training group and the validation group. All data, including previously reported risk factors for POPF and parameters measured by BIA, were collected. Risk factors were analyzed by univariable and multivariable logistic regression analysis. A prediction model was established to predict the development of POPF based on these parameters.

RESULTS

POPF occurred in 41 of 168 (24.4%) patients. In the training group of 101 enrolled patients, visceral fat area (VFA) (odds ratio [OR] = 1.077, P = 0.001) and fat mass index (FMI) (OR = 0.628, P = 0.027) were found to be independently associated with POPF according to multivariable analysis. A prediction model including VFA and FMI was established to predict the development of POPF with an area under the receiver operating characteristic curve (AUC) of 0.753. The efficacy of the prediction model was also confirmed in the internal validation group (AUC 0.785, 95% CI 0.659-0.911).

CONCLUSIONS

Preoperative assessment of body fat distribution by BIA can predict the risk of POPF after pancreatic surgery.

Root cause analysis of mortality after esophagectomy for cancer: a multicenter cohort study from the FREGAT database

BACKGROUND

Esophagectomy is associated with significant mortality. A better understanding of the causes leading to death may help to reduce mortality. A root cause analysis of mortality after esophagectomy was performed.

METHODS

Root cause analysis was retrospectively applied by an independent expert panel of 4 upper gastrointestinal surgeons and 1 anesthesiologist-intensivist to patients included in the French national multicenter prospective cohort FREGAT between August 2014 and September 2019 who underwent an esophagectomy for cancer and died within 90 days of surgery. A cause-and-effect diagram was used to determine the root causes related to death. Death was classified as potentially preventable or non-preventable.

RESULTS

Among the 1,040 patients included in the FREGAT cohort, 70 (6.7%) patients (male: 81%, median age 68 [62-72] years) from 17 centers were included. Death was potentially preventable in 37 patients (53%). Root causes independently associated with preventable death were inappropriate indication (odds ratio 35.16 [2.50-494.39]; P = .008), patient characteristics (odds ratio 5.15 [1.19-22.35]; P = .029), unexpected intraoperative findings (odds ratio 18.99 [1.07-335.55]; P = .045), and delay in diagnosis of a complication (odds ratio 98.10 [6.24-1,541.04]; P = .001). Delay in treatment of a complication was found only in preventable deaths (28 [76%] vs 0; P < .001). National guidelines were less frequently followed (16 [43%] vs 22 [67%]; P = .050) in preventable deaths. The only independent risk factor of preventable death was center volume <26 esophagectomies per year (odds ratio 4.71 [1.55-14.33]; P = .006).

CONCLUSIONS

More than one-half of deaths after esophagectomy were potentially preventable. Better patient selection, early diagnosis, and adequate management of complications through centralization could reduce mortality.

Contemporary Outcomes of Grade-C Postoperative Pancreatic Fistula in a Nationwide Database

INTRODUCTION

Grade-C postoperative pancreatic fistulas (POPFs) are dreaded complications following pancreaticoduodenectomy. The aim of this study was to quantify the incidence and risk factors associated with grade C POPF in a national database.

METHODS

The National Surgical Quality Improvement Program targeted user files were queried for patients who underwent elective pancreaticoduodenectomy (2014-2020). Outcomes were compared between clinically relevant (CR) grade B POPF and grade C POPF.

RESULTS

Twenty-six thousand five hundred fifty-two patients were included, of which 90.1% (n = 23,714) had No CR POPF, 8.7% (n = 2287) suffered grade B POPF, and 1.2% (n = 327) suffered grade C POPF. There was no change in the rate Grade-C fistula overtime (m = 0.06, P = 0.63), while the rate of Grade-B fistula significantly increased (m = +1.40, P < 0.01). Fistula Risk Scores were similar between grade B and C POPFs (high risk: 34.9% versus 31.2%, P = 0.21). Associated morbidity was increased with grade C POPF, including delayed gastric emptying, organ space infections, wound dehiscence, respiratory complications, renal complications, myocardial infarction, and bleeding. On multivariate logistic regression, diabetes mellitus (odds ratio: 1.41 95% confidence interval: 1.06-1.87, P = 0.02) was associated with grade C POPF.

CONCLUSIONS

This study represents the largest contemporary series evaluating grade C POPFs. Of those suffering CR POPF, the presence of diabetes mellitus was associated with grade C POPF. While modern management has led to grade C POPF in 1% of cases, they remain associated with alarmingly high morbidity and mortality, requiring further mitigation strategies to improve outcomes.

Management and outcomes of hemorrhage after distal pancreatectomy: a multicenter study at high volume centers

BACKGROUND

Data on clinically relevant post-pancreatectomy hemorrhage (CR-PPH) are derived from series mostly focused on pancreatoduodenectomy, and data after distal pancreatectomy (DP) are scarce.

METHODS

All non-extended DP performed from 2014 to 2018 were included. CR-PPH encompassed grade B and C PPH. Risk factors, management, and outcomes of CR-PPH were evaluated.

RESULTS

Overall, 1188 patients were included, of which 561 (47.2 %) were operated on minimally invasively. Spleen-preserving DP was performed in 574 patients (48.4 %). Ninety-day mortality, severe morbidity and CR-POPF rates were 1.1 % (n = 13), 17.4 % (n = 196) and 15.5 % (n = 115), respectively. After a median interval of 8 days (range, 0-37), 65 patients (5.5 %) developed CR-PPH, including 28 grade B and 37 grade C. Reintervention was required in 57 patients (87.7 %). CR-PPH was associated with a significant increase of 90-day mortality, morbidity and hospital stay (p < 0.001). Upon multivariable analysis, prolonged operative time and co-existing POPF were independently associated with CR-PPH (p < 0.005) while a chronic use of antithrombotic agent trended towards an increase of CR-PPH (p = 0.081). As compared to CR-POPF, the failure-to-rescue rate in patients who developed CR-PPH was significantly higher (13.8 % vs. 1.3 %, p < 0.001).

CONCLUSION

CR-PPH after DP remains rare but significantly associated with an increased risk of 90-day mortality and failure-to-rescue.

Imaging after Pancreatic Surgery: Expected Findings and Postoperative Complications

Pancreatic surgery is considered one of the most technically challenging surgical procedures, despite the evolution of modern techniques. Neoplasms remain the most common indication for pancreatic surgery, although inflammatory conditions may also prompt surgical evaluation. The choice of surgical procedure depends on the type and location of the pathologic finding because different parts of the pancreas have separate vascular supplies that may be shared by adjacent organs. The surgical approach could be conventional or minimally invasive (laparoscopic, endoscopic, or robotic assisted). Because of the anatomic complexity of the pancreatic bed, perioperative complications may be frequently encountered and commonly involve the pancreatic-biliary, vascular, lymphatic, or bowel systems, irrespective of the surgical technique used. Imaging plays an important role in the assessment of suspected postoperative complications, with CT considered the primary imaging modality, while MRI, digital subtraction angiography, and molecular imaging are considered ancillary diagnostic tools. Accurate diagnosis of postoperative complications requires a solid understanding of pancreatic anatomy, surgical indications, normal postoperative appearance, and expected postsurgical changes. The practicing radiologist should be familiar with the most common perioperative complications, such as anastomotic leak, abscess, and hemorrhage, and be able to differentiate these entities from normal anticipated postoperative changes such as seroma, edema and fat stranding at the surgical site, and perivascular soft-tissue thickening. In addition to evaluation of the primary operative fossa, imaging plays a fundamental role in assessment of the adjacent organ systems secondarily affected after pancreatic surgery, such as vascular, biliary, and enteric complications. Published under a CC BY 4.0 license. Test Your Knowledge questions are available in the supplemental material. See the invited commentary by Winslow in this issue.

Predictors of In-Hospital Mortality Following Pancreatectomy

BACKGROUND

In-hospital mortality rates following all types of pancreatic resections (PRs) have decreased over recent decades. Our aim was to identify predictors of in-hospital mortality following pancreatic resection.

METHODS

All patients undergoing pancreatic resection were sampled from the National Inpatient Sample (NIS) in the years 2007-2012. Predictors of in-hospital mortality were identified and incorporated into a binary logistic regression model.

RESULTS

A total of 111,568 patients underwent pancreatectomy. Annual mortality rates decreased from 4.3% in 2007 to 3.5% in 2012. Independent predictors of in-hospital mortality included age ≥75 years (vs. <65 years, OR = 2.04; 95% CI: 1.61-2.58), nonelective procedure status (OR = 1.46; 95% CI: 1.19-1.80), resection other than distal pancreatic resection (vs. Whipple, OR = 2.14; 95% CI: 1.71-2.69; other partial, OR = 2.48; 95% CI: 1.76-3.48), lower hospital volume (OR = 1.28; 95% CI: 1.09-1.49), indication for pancreatic resection other than benign diseases (vs. malignant, OR = 1.63; 95% CI: 1.25-2.15; other, OR = 2.48; 95% CI: 1.76-3.48), pulmonary complications (OR = 12.36; 95% CI: 10.11-15.17), infectious complications (OR = 2.17; 95% CI: 1.78-2.64), noninfectious wound complications and pancreatic leak (OR = 1.94; 95% CI: 1.53-2.46), and acute myocardial infarction (OR = 2.03; 95% CI: 1.32-3.06).

DISCUSSION

Our findings identify predictors of inpatient mortality following pancreatectomy, with pulmonary complications representing the single most significant factor for increased mortality. These findings complement and expand on previously published data and, if applied to perioperative care, may enhance survival following pancreatectomy.

Factors predicting 30-day mortality after pancreaticoduodenectomy-the impact of elevated aspartate aminotransferase

PURPOSE

High preoperative bilirubin levels and cholangitis are associated with poor peri-operative outcomes following pancreaticoduodenectomy (PD). However, the impact of deranged preoperative aspartate aminotransferase (AST) and alanine aminotransferase (ALT) levels on immediate postoperative outcomes is relatively unexplored. We hypothesized that deranged AST and ALT lead to worse postoperative outcomes after PD. The aim of this study was to assess the factors contributing to postoperative mortality (POM) following PD, and to study the impact of deranged aminotransferases.

METHODS

This is a retrospective analysis of 562 patients. Risk factors for POM were computed using a multivariate logistic regression model.

RESULTS

The rate of POM was 3.9%. On univariate analysis, the American Society of Anaesthesiologists grades, diabetes mellitus, cardiac comorbidity, preoperative biliary stenting, elevated serum bilirubin, AST, elevated serum creatinine, clinically relevant pancreatic fistula (CRPF), and grade B+C post-pancreatectomy hemorrhage (PPH) were associated with 30-day mortality. On multivariate analysis, preoperative elevated AST was independently predictive of 30-day POM (OR = 6.141, 95%CI 2.060-18.305, p = 0.001). Other factors independently predictive of POM were elevated serum creatinine, preoperative biliary stenting, CRPF and grade B and C PPH. The ratio of AST/ALT > 0.89 was associated with 8 times increased odds of POM.

CONCLUSION

Elevated preoperative AST emerged as a predictor of 30-day POM after PD, with an 8-times increased odds of death with an AST/ALT ratio > 0.89.

Preoperative liver arterial conditioning in patients scheduled for a Mayo Clinic class Ia distal pancreatectomy: embolization or ligation?

BACKGROUND

Liver ischemia may occur during intraoperative common hepatic artery ligation in Mayo Clinic class I distal pancreatectomy with en bloc celiac axis resection (DP-CAR). Preoperative liver arterial conditioning could be used to avoid this outcome. This retrospective study compared arterial embolization (AE) or laparoscopic ligation (LL) of the common hepatic artery before class Ia DP-CAR.

METHODS

From 2014 to 2022, 18 patients were scheduled for class Ia DP-CAR after neoadjuvant FOLFIRINOX treatment. Two were excluded due to hepatic artery variation, six underwent AE, ten underwent LL.

RESULTS

Two procedural complications occurred in the AE group: an incomplete dissection of the proper hepatic artery and a distal migration of coils in the right branch of the hepatic artery. Neither complication prevented surgery. The median delay between conditioning and DP-CAR was 19 days; decreased to five days in the last six patients. None required arterial reconstruction. Morbidity and 90-day mortality rates were 26.7% and 12.5%, respectively. No patient developed postoperative liver insufficiency after LL.

CONCLUSION

Preoperative AE and LL seem comparable in averting arterial reconstruction and postoperative liver insufficiency in patients scheduled for class Ia DP-CAR. However, serious complications that may arise during AE led us to prefer the LL technique.

Covered stent treatment for arterial complications after pancreatic surgery: risk assessment for recurrence and peri-stent implantation management

OBJECTIVES

To explore the risk factors for recurrence of arterial complications after pancreatectomy during the period of covered stent implantation and to provide some opinions on peri-stent implantation management.

METHODS

Data on patients implanted with covered stents due to arterial complications after pancreatectomy between January 2017 and December 2021 were analyzed retrospectively. Technical success, clinical success, recurrence, and survival were evaluated to elucidate the practicability of covered stents. Wilson score, Random Forest, logistic regression, and Pearson's chi-square test with bootstrap aggregation were performed for determining the perioperative risk factors for recurrence.

RESULTS

Among all fifty-five patients, success stent implantation (technical success) was achieved 100%. Patients who were hemodynamically stabilized without further treatment for artery complications in situ (clinical success) accounted for 89.1%. Based on statistical analysis, pre-stent implantation pancreatic fistula was identified as a robust recurrence-related risk factor for preoperative assessment (p = 0.02, OR = 4.5, 95% CI [1.2, 16.9]; p_bootstrap = 0.02). Post-stent implantation pancreatic fistula (p = 0.01, OR 4.5, 95% CI [1.4, 14.6]; p_bootstrap < 0.05) and SMA branches or GDA stumps (p = 0.02, OR 3.4, 95% CI [1.1, 10.3]) were relevant to recurrence. The survival rate during hospitalization was 87.3%. All survivors were free from recurrence during the subsequent follow-up. Vasospasm and stent occlusion were observed as short-term and long-term complications, respectively.

CONCLUSION

A covered stent implantation is a feasible and effective treatment option for post-pancreatectomy arterial complications. Rigorous management of pancreatic fistula, timely detection of problems, sensible strategies during stent implantation, and reasonable anticoagulation therapy are necessary for a better prognosis.

KEY POINTS

• A covered stent is feasible for various artery-related complications after pancreatectomy and has an ideal therapeutic effect. • Pancreatic fistula during the perioperative period of the covered stent is an independent risk factor for recurrent arterial complications and SMA branches or GDA stumps are prone to be recurrent offending arteries. • Rigorous management of pancreatic fistula, timely detection of problems, sensible strategies during stent implantation, and reasonable anticoagulation therapy are necessary for a better prognosis.

Serum lipase on postoperative day one is a strong predictor of clinically relevant pancreatic fistula after pancreaticoduodenectomy: A retrospective cohort

BACKGROUND

Increased postoperative serum amylase has been recently reported to be associated with increased postoperative morbidity, but studies on postoperative serum lipase are limited. The aim of this study was to evaluate the value of postoperative serum lipase in predicting clinically relevant postoperative pancreatic fistula (CR-POPF) after pancreaticoduodenectomy (PD).

METHOD

A retrospective analysis was performed on 212 patients who underwent PD from September 2018 and March 2021, focusing on the association between postoperative day (POD) 1 serum lipase and CR-POPF.

RESULTS

Overall, 108 (50.9%) patients had elevated serum lipase levels (>68 U/L) on POD 1. Patients with elevated serum lipase exhibited a significantly higher incidence of CR-POPF (37.0% vs. 6.7%, p < 0.001). Receiver operating characteristic (ROC) analyses showed improved diagnostic accuracy for POD 1 serum lipase compared with POD 1 serum amylase in predicting CR-POPF (AUC: 0.801 vs. 0.745, p = 0.029). Elevated serum lipase on POD 1 and elevated serum CRP on POD 3 were identified as independent predictors of CR-POPF. A simple early postoperative model, consisting of POD 1 serum lipase levels and POD 3 serum CRP levels, showed good discrimination (AUC 0.76, 95% CI 0.69-0.83) to identify the onset of CR-POPF.

CONCLUSION

Serum lipase on POD 1 outperformed serum amylase on POD 1 in predicting CR-POPF after PD. The combination of POD 1 serum lipase and POD 3 serum CRP provides a reliable predicting model for CR-POPF.