Perioperative scores to predict mortality in surgical oncologic patients: a review of 1,362 cases

Introduction

Surgical outcomes depend on many factors, including baseline characteristics and differences in surgical and anesthetic practices. Few data are available in general surgical oncological patients. We aimed to evaluate the predictive value of Porthsmouth POSSUM and American Society of Anesthesiologists (ASA) scores in oncologic patients submitted to surgical procedures.

Methods

Data were collected from a database filled with demographic characteristics, neoplasm definitions, co-morbidities and data from intraoperative room and hospital mortality. The predicted mortality risk was calculated using Porthsmouth POSSUM and ASA physical status grading.

Results

From 1,362 surgical patients, the observed general hospital mortality was 2.34%. ASA grading underestimated the mortality for low-complexity patients, overestimated in patients with severe systemic disease (ASA III), and was a strong predictor of mortality in patients with life-threatening conditions (ASA IV) (Table 1). On the other hand, the Porthsmouth POSSUM scoring system overestimated mortality in all groups (Table 2).

Table 1 Observed and predicted mortality of oncologic patients evaluated through ASA physical status grading

Table 2 Observed and predicted mortality of oncologic patients evaluated through Portsmouth POSSUM scoring

Conclusions

In surgical oncologic patients, Porthsmouth POSSUM and ASA are not good predictors of mortality. These data suggest that a more sensible and specific score is needed in this population.