Diagnostic utility of the total nucleated cell count for differentiation of septic and sterile peritoneal effusions in dogs

Abstract

Background

Rapid and accurate diagnosis of septic peritonitis is critical for initiating appropriate medical and surgical management.

Objectives

The aim of this study was to determine the diagnostic utility of the total nucleated cell count (TNCC), absolute neutrophil count, neutrophil percentage, and total protein (TP) to distinguish septic versus non-septic peritoneal effusions in dogs.

Methods

Electronic medical records were retrospectively searched for peritoneal fluid samples from 2008 to 2018 and classified as septic or non-septic based on bacterial culture and/or cytology results. Receiver operator characteristic curves (ROCs) were used to describe the overall diagnostic utility of each test, with optimal cutpoints analyzed to dichotomize continuous variables. Positive and negative likelihood ratios were calculated at these cutpoints.

Results

A total of 166 unique samples, including 87 septic and 79 non-septic peritoneal effusions, were included. There were no significant differences in dog sex, age, or days hospitalized between groups. Septic effusions had significantly higher TP, TNCC, absolute neutrophil count, and neutrophil percentage compared with non-septic effusions. The area under the curve of the ROC curves was TNCC (0.80), absolute neutrophil count (0.80), neutrophil percentage (0.64), and TP (0.63). For TNCC and absolute neutrophil count, optimal cutoffs were 17.13 × 10³ cells/μL and 19.88 × 10³ cells/μL, resulting in positive and negative likelihood ratios of 2.39 and 0.28 and 2.85 and 0.28, respectively.

Conclusions

Total nucleated cell counts and absolute neutrophil counts aid in the differentiation of septic and non-septic peritoneal effusions with similar diagnostic utility but are not sufficiently sensitive or specific to use without concurrent microscopic evaluation.