Defining benchmark values for outcomes of comprehensive resection of primary retroperitoneal liposarcoma: a retrospective multicenter study

Abstract

Background: Comprehensive resection represents the standard of care for patients affected by retroperitoneal well- or dedifferentiated liposarcoma (WDLPS/DDLPS). However, reference values to indicate the best achievable results are currently lacking. As such, the study aimed to define clinically relevant benchmark values for intra- and postoperative outcomes of patients undergoing comprehensive resection for primary retroperitoneal WDLPS/DDLPS.

Methods: The international, prospectively maintained Retroperitoneal Sarcoma Registry (RESAR; NCT03838718) was used to calculate benchmark values for 22 outcomes, including intraoperative factors, and rates of complications, recurrence and survival. Only low-risk patients undergoing comprehensive resection for WDLPS/DDLPS at high-volume centers between 1st January 2017 and 31st December 2021 were used to calculate the benchmark values. Specifically, "low risk" was defined as age <75 years, with minimal comorbidities, and undergoing a "standard" comprehensive resection including at least colon and kidney with or without other organs-excluding those associated with significant morbidity (e.g., pancreas). Benchmark values were defined based on the 25th or 75th percentiles of the center-level data. To validate the benchmark values, these were applied to two cohorts expected to have inferior outcomes, which were defined by changing one of the exclusion criteria; namely those treated in low-volume centers, and those with American Society of Anesthesiologists (ASA) score ≥3 ("ASA ≥ 3").

Findings: Of the 1510 patients undergoing surgery, 147 met the inclusion criteria and were included in the benchmarking analysis. This identified benchmark values including: median duration of surgery ≤278 min, intraoperative packed red cell transfusion rate ≤30%, R0/R1 resection rate ≥89%, median length of hospital stay ≤15 days, reoperation rate ≤13%, major postoperative complication rate ≤21%, and 90-day postoperative mortality/failure-to-rescue rates of 0%. The "low-volume centers" cohort failed to meet 10 of these benchmarks, including duration of surgery (median: 293 vs. ≤278 min), R0/R1 resection rate (82% vs. ≥ 89%), major postoperative complication rate (35% vs. ≤21%), and reoperation rate (35% vs. ≤13%), whilst the "ASA ≥ 3" cohort failed to meet seven benchmarks.

Interpretation: These novel benchmark values can act as reference values to which sarcoma centers or individual surgeons can compare, which may help to identify performance gaps and improve the quality of care.

Funding: "5 x mille" fund for healthcare research (Italian Ministry of Health).

Keywords: Benchmark; Liposarcoma; Quality of surgery; Retroperitoneal sarcoma; Surgical oncology.

Fig. 1

Study flowchart. Exclusions are applied cumulatively in the order stated. ^aIncludes N = 8 with missing data, who were excluded since it was unclear whether they had retroperitoneal tumors ^bPatients aged <18 years were already excluded from the RESAR database at source. ^cPatients who were not reported to meet any of the stated medical exclusion criteria, but who had missing data for at least one of the criteria. Abbreviations: ASA: American Society of Anesthesiologists score, BMI: body mass index, COPD: chronic obstructive pulmonary disease, DD(WD)LPS: dedifferentiated (well-differentiated) liposarcoma, ECOG: Eastern Cooperative Oncology Group performance status.

Fig. 2

Total volume and number of benchmark cases for individual centers.^aRepresents the number of benchmark cases for the high-volume centers; for low-volume centers, this represents the number of cases meeting all other benchmarking criteria (except for center volume, i.e., the “Low-Volume Centers” cohort).

Fig. 3

Center-level results and benchmark values for intra- and post-operative outcomes. Black points represent data for the N = 7 individual centers in the benchmarking cohort, namely medians for continuous variables and rates for binary outcomes. Survival and recurrence rates were estimated from Kaplan–Meier and cumulative incidence function curves, respectively; only the one-year outcomes are reported, for brevity. Overlapping points are stacked for clarity. Red diamonds represent the benchmark value for each outcome. Abbreviations: Intraop: intraoperative, Postop: postoperative, PRC: packed red cells, R0/R1: macroscopically complete.