Age-Associated Risk of 90-Day Postoperative Mortality After Cytoreductive Surgery for Advanced Ovarian Cancer

Abstract

This study explores the association of the risk of 90-day mortality with age for US women with advanced ovarian cancer undergoing cytoreductive surgery.

Surgical resection of all visible tumors is associated with improved long-term survival and is a critical component of primary therapy for advanced-stage ovarian cancer.¹ However, cytoreductive surgery carries a substantial risk of complications and death.² We describe associations between age and 90-day postoperative mortality after cytoreductive surgery and investigate if administrating neoadjuvant chemotherapy diminishes the age-associated risk of 90-day mortality.

Methods

Using the National Cancer Database, we analyzed age-associated trends in 90-day postoperative mortality after cytoreductive surgery among women with stage IIIC or IV epithelial ovarian cancer who were treated in Commission on Cancer–accredited hospitals in the United States between 2004 and 2013. We fit logistic joinpoint models to quantify the probability of 90-day mortality as a function of age for women undergoing primary and interval cytoreductive surgery. The number and location of joinpoints was determined using the Bayesian information criterion.³ Separate models were fit to estimate crude and adjusted age group–specific odds ratios for postoperative death after primary surgery compared with interval surgery. Analyses were performed in Stata, version 14 (StataCorp) and statistical significance was set at P < .05.

This study was deemed exempt from review by the Partners Healthcare ethics committee. The requirement for informed consent was waived because this study used an existing deidentified data set.

Results

We identified 47 117 women, of whom 37 024 (78.5%) underwent primary and 10 153 (21.5%) underwent interval cytoreductive surgery. Patients undergoing primary surgery were younger (mean [SD] age, 61.5 [12.3] vs 63.1 [11.4] years; P < .001) and had any comorbidities less often (17.2% vs 19.4%; P < .001) than those who underwent interval cytoreductive surgery but had longer postoperative hospitalizations (mean [SD] days, 8.0 [9.3] vs 6.1 [6.1] days; P < .001). Overall, 90-day mortality was higher after primary (2658 deaths [7.2%]) than interval (312 deaths [3.1%]) cytoreductive surgery. Age-associated trends in 90-day mortality differed between primary and interval surgery (P for interaction < .001; Figure, A). We observed no age-associated increase in the risk of 90-day mortality after interval (odds ratio, 0.99; 95% CI, 0.97-1.01; P = .36) or primary (odds ratio, 1.00; 95% CI, 0.97-1.02; P = .75) cytoreduction before age 47 years. Among women undergoing primary cytoreductive surgery, the odds of 90-day postoperative mortality began rising at age 47 years, increasing by 5.7% per year of age (95% CI, 5.0-6.5; P < .001) until age 71 years, after which they accelerated to 9.9% per year (95% CI, 8.8-10.9; P < .001). By contrast, the odds of 90-day mortality after interval cytoreductive surgery began to increase at age 62 years and increased steadily by 5.7% per year (95% CI, 3.9-7.5, P < .001) without evidence of acceleration in risk among the oldest patients. By age 75 years, the probability of 90-day postoperative mortality was 4.2% (95% CI, 3.6-4.9) after interval and 12.3% (95% CI, 11.4-12.7) after primary cytoreduction. These probabilities increased to 7.2% (95% CI, 5.5-9.2) and 26.0% (95 CI, 24.1-27.9), respectively, by age 85 years. Age group–specific odds ratios for 90-day mortality associated with primary surgery are shown in the Figure.

Figure. Probability of 90-Day Mortality.

A, Observed age-specific probabilities of 90-day mortality after primary cytoreductive surgery (PCS; blue dots) and interval cytoreductive surgery (ICS; orange dots) are plotted along with predicted probabilities (solid lines) and 95% confidence intervals (shaded areas) from piecewise joinpoint regression models. B, Number of operations, crude odds ratios, and adjusted odds ratios for 90-day mortality associated with PCS are tabulated by age group. Adjusted odds ratios were estimated from a model that included year of diagnosis, histologic type, grade, stage, comorbidity index, geographic region, insurance type, hospital volume, and cancer program type.

Discussion

Women undergoing primary cytoreductive surgery incurred an age-associated risk of 90-day postoperative mortality from a younger age and at a larger magnitude than those who received neoadjuvant chemotherapy and interval debulking surgery. The goal of cytoreductive surgery is to resect all grossly visible disease, which is associated with improved long-term survival.⁴ While death within 90 days of cytoreductive surgery cannot be exclusively attributed to surgery, undergoing a cytoreductive surgery within 3 months of death is an undesirable outcome because such operations fail to deliver oncologic benefit. This study is limited by an absence of information about the causes and circumstances of postoperative deaths, precluding the attribution of deaths to surgery or cancer. Furthermore, while we adjusted for many clinical, demographic, and hospital-level confounders, other potential confounders are unavailable in the National Cancer Database. Nonetheless, women who undergo primary surgery are younger and fitter than those selected for neoadjuvant chemotherapy,⁵ which is expected to bias the present results toward the null. In conclusion, among older women, neoadjuvant chemotherapy administered before cytoreductive surgery may substantially reduce the risk of postoperative mortality.