Feasibility of Comparing the Results of Pancreatic Resections between Surgeons: A Systematic Review and Meta-Analysis of Pancreatic Resections

Detecting Hospital Outliers in Post-Pancreatectomy Care Using Funnel Plots from 2009-2018 Based on Nationwide Medico-Administrative Data

Objectives

Our objective was to identify hospitals with unusual mortality rates for major pancreatectomies over a period of ten years using 30-day mortality data from the French national database.

Methods

Data for all patients who underwent pancreatectomy were extracted from the national medico-economic database (Programme de Médicalisation des Systèmes d'Information). To identify quality outliers for each hospital, the observed-to-expected 30-day mortality rates were used as a quality indicator.

Results

A total of 19 494 patients underwent a major pancreatectomy in France between January 2009 and December 2018. The overall 30-day mortality rate was 4.8% (n = 944). For the 2009–2014 period, the funnel plot showed that 10 of the 176 hospitals lie outside the central 95% region and 7 lie outside the central 99.8% region. For the 2015–2018 period, out of 176 hospitals, 6 lie outside the central 95% region and 2 lie outside the central 99.8% region. The change in standardized mortality ratios between 2009–2014 and 2015–2018 testing for differences from the overall change, they were there 4 hospitals lie outside the central 95% region and 0 lie outside the central 99.8% region.

Conclusion

Over time, the improvement in hospital quality was weak. This study suggests that there is a pressing need to reorganize the supply of care for pancreatic surgery in France.

A Strategy for Risk-adjusted Ranking of Surgeons and Practices Based on Patient-reported Outcomes After Elective Lumbar Surgery

STUDY DESIGN

This study retrospectively analyzes prospectively collected data.

OBJECTIVE

The primary aim of this study is to present a scheme for patient-reported outcome (PRO)-based, risk-adjusted rankings of spine surgeons and sites that perform elective lumbar surgery, using the Quality and Outcomes Database (QOD).

SUMMARY OF BACKGROUND DATA

There is currently no means of determining which spine surgeons or centers are positive or negative outliers with respect to PROs for elective lumbar surgery. This is a critical gap as we move toward a value-based model of health care in which providers assume more accountability for the effectiveness of their treatments.

METHODS

Random effects regression models were fit for the following outcomes, with QOD site as a fixed effect but surgeon ID as a random effect: Oswestry Disability Index, EQ-5D, back pain and leg pain, and satisfaction. Hierarchical Bayesian models were also fit for each outcome, with QOD site as a random effect and surgeon as a nested random effect.

RESULTS

Our study cohort consists of 8834 patients who underwent surgery by 124 QOD surgeons, for the degenerative lumbar diseases. Nonoverlapping Bayesian credible intervals demonstrate that the variance attributed to QOD site was greater than the nested variance attributed to surgeon ID for the included PROs.

CONCLUSION

This study presents a novel strategy for the risk-adjusted, PRO-based ranking of spine surgeons and practices. This can help identify positive and negative outliers, thereby forming the basis for large-scale quality improvement. Assuming adequate coverage of baseline risk adjustment, the choice of surgeon matters when considering PROs after lumbar surgery; however, the choice of site appears to matter more.

LEVEL OF EVIDENCE

3.

One hundred and forty five total laparoscopic pancreatoduodenectomies: A single centre experience

INTRODUCTION

Laparoscopic pancreatoduodenectomy (LPD) remains one of the most challenging minimal invasive operations today.

PATIENTS AND METHODS

Between January 2007 and December 2016, 197 patients were scheduled for LPD but 162 patients (from January 2007 to July 2016) were analysed in this cohort series.

RESULTS

Total LPD concerned for 162 patients (five patients did not undergo PD and 12 underwent conversion): standard LPD in 104 patients (66%), and laparoscopic pylorus-preserving PD in 41 patients (26%). Median operative time was 415 (240-765) min. Median blood loss was 200 (50-2100) ml. Twelve patients required blood transfusion. Clinically relevant pancreatic fistula (ISGPF grades B and C) occurred in 21 (13%) patients: 16 (10.0%) grade B, and 5 (3%) grade C. Grades B and C delayed gastric emptying occurred in five patients each. Grades B and C post-pancreatectomy bleeding occurred in 9 (5.7%) and 3 (1.9%) patients, respectively. LPD was performed for 18 (11.4%) benign and 139 (88.5%) malignant lesions. Superior mesenteric and/or portal vein involvement required major venous resection in eight patients. The 90-day mortality 5.0%. The median overall survival for pancreatic ductal adenocarcinoma was 22.5 months.

CONCLUSION

Morbidity and mortality for LPD are comparable to open procedures rates in the literature. Laparoscopic major venous resection is feasible and safe.