Background: Surgery in neutropenic patients is challenging due to both atypical manifestations of common conditions and higher perioperative risk. We sought to describe the outcomes of neutropenic patients undergoing abdominal surgery and to identify factors contributing to morbidity and mortality.
Methods: A retrospective chart review was performed for all patients neutropenic in the 24-hours prior to an abdominal operation at our institution between 1998 and 2017. The primary and secondary outcomes were 30-day mortality and morbidity, respectively. The chi-square test and two-tailed Student's t test were used for univariable comparisons (non-parametric tests used when appropriate). To determine the optimal threshold of absolute neutrophil count (ANC) to discriminate 30-day mortality, we maximized the Youden index (J).
Results: Amongst 237 patients, mortality was 11.8% (28/237) and morbidity 54.5% (130/237). Absolute neutrophil count < 500 cells/μL (50% vs. 20.6%, P < 0.01) and perforated viscus (35.7% vs. 14.8%, P = 0.01) were associated with mortality. Perforated viscus (25.4% vs. 7.5%) was also associated with morbidity. Urgent operations were associated with higher morbidity (63.6% vs 34.7%, P < 0.001) and mortality (16.4% vs 1.4%, P = 0.002) when compared to elective operations. Transfer from an outside hospital (22.3% vs. 11.2%, P = 0.02) and longer median time from admission to operation (2 days (IQR 0-6) vs. 1 day (IQR 0-3), P < 0.01) were associated with morbidity. An ANC threshold of 350 provided the best discrimination for mortality.
Conclusions: Elective surgery in the appropriately chosen neutropenic patient is relatively safe. For patients with obvious surgical pathology, we advocate for earlier operation and a lower threshold for surgical consultation in an effort expedite the diagnosis and necessary treatment.
Keywords: General surgery; Neutropenia; Outcomes research.