Diagnostic value of post-operative platelet-to-white blood cell ratio after splenectomy in patients with advanced ovarian cancer

Introduction: Splenectomy-induced thrombocytosis and leukocytosis may obscure the early diagnosis of post-operative infection or sepsis. In trauma patients after splenectomy, a platelet-to-white blood cell ratio of <20 has been shown to reliably differentiate post-operative sepsis from transient physiological responses.

Objective: To determine whether the platelet-to-white blood cell ratio can be applied to differentiate between reactive post-operative changes and latent infection.

Methods: All consecutive patients with ovarian cancer who underwent splenectomy between January 2013 and October 2018 in two large European gynecological cancer centers were retrospectively evaluated. Main outcome measures were white blood cell count, platelet count, and platelet-to-white blood cell ratio on post-operative days 1, 5, and 7. These were correlated with surgical outcome and morbidity according to the Clavien-Dindo classification. A binomial logistic regression was applied to assess the predictive value of day 5 platelet-to-white blood cell ratio, white blood cell count, and platelet count for predicting grade III post-operative sepsis.

Results: Ninety-five patients with ovarian cancer (mean age 54 years, range 18-75) were identified. Seventeen patients (17.9%) developed a grade III post-operative sepsis. In all post-operative patients, mean white blood cell count on day 5 decreased (from 15.4×10³/μL to 11.4×10³/μL), while the mean platelet count rose (from 260.7×10³/μL to 385.3×10³/μL). A high platelet count (>313×10³/μL) failed to show any predictive value (OR=0.94; 95% CI 0.30 to 3.0; p=0.921). A low platelet-to-white blood cell ratio (<26) (OR=3.49; 95% CI 1.18 to 10.32; p=0.0241) and high white blood cell count (>14.5×10³/μL) on day 5 (OR=11.0; 95% CI 3.3 to 36.2; p<0.001) were significant for predicting sepsis. Despite a significant OR, the sensitivity and specificity were low; day 5 platelet-to-white blood cell ratio at a cut-off point of 26 achieved a sensitivity of 72% and specificity of 53% (area under the curve 0.637, 95% CI 0.480 to 0.796) in predicting grade III post-operative sepsis.

Conclusions: Platelet-to-white blood cell ratio after cytoreductive surgery for ovarian cancer with splenectomy does not appear to have a strong predictive value in differentiating between sepsis and reactive splenectomy-induced changes. Leukocytosis, in combination with clinical assessment, may remain the most useful tool for prediction of sepsis after cytoreductive surgery with splenectomy.