S1186 Mild Thrombocytopenia as a Predictor of Surgical Outcomes Following Cholecystectomy: A Tool for Perioperative Assessment in Patients With Non-Alcoholic Fatty Liver Disease?

Abstract

Introduction: As evidenced by its inclusion in estimations of liver fibrosis, thrombocytopenia is a well-defined abnormality in chronic hepatic disease. Studies show that platelet counts fewer than 160 x 109/L are an independent marker for severity of hepatic fibrosis. Non-alcoholic fatty liver disease (NAFLD) is now the most prevalent liver disorder in Western countries making characterization of morbidity increasingly important. There is a paucity of information available to characterize perioperative risk for patients with NAFLD. We used a threshold of 150 x 109 as a surrogate for NAFLD in patients undergoing laparoscopic cholecystectomy to study its effect on perioperative complications and mortality. Methods: We queried the American College of Surgeons National Surgical Quality Improvement Program (NSQIP) database for cholecystectomies occurring from 2005 through 2018. Demographic differences between patients with and without thrombocytopenia were evaluated using the t-test or the chi-square test, whereas differences in outcome risk was evaluated using log-binomial regression models. Results: We identified 437,630 patients who underwent cholecystectomy, of whom 6.9% had thrombocytopenia. Patients with thrombocytopenia were more often male, older, with chronic disease. As shown in Table and Figure, patients with thrombocytopenia had higher 30-day mortality rates risk ratio (RR) 5.3 (95% CI: 4.8-5.9) and higher complication rates RR 2.4 (95% CI: 2.3-2.5). The most frequent complications included respiratory, need for transfusion, and renal. Conclusion: Peri-operatively, patients with mild thrombocytopenia undergoing cholecystectomy had higher mortality rates and complications compared to patients with normal platelet counts, and this effect continued as thrombocytopenia became more pronounced. Other etiologies of mild thrombocytopenia could not be excluded; however, aside from cancer (a small proportion) these are unlikely to affect clinical outcomes. Currently, most society guidelines including the AASLD practice guidance do not recommend NAFLD screening in the general population even among high-risk patients with diabetes or obesity. Thrombocytopenia might be a promising, cost-effective tool for NAFLD screening especially if used in high-risk populations (male gender, hypertension or diabetes, with or without high BMI). Furthermore, our study suggests that platelet counts may have utility in predicting peri-operative outcomes in this population.Figure 1.: Estimated rate of 30-day Mortality (left) and having any complication (right; infection, cardiac, respiratory, thrombotic, transfusion, renal, and/or unplanned return to the OR) across observed pre-operative platelet counts. Shaded areas represent 95% confidence intervals. The solid vertical line defines thrombocytopenia threshold at 150 x 109/L. The dashed vertical lines identify the 1st and 99th percentile of pre-operative platelet counts (92 and 480, respectively) indicating that <1% of the pre-operative platelet counts were below 92 or above 480. Table 1. - Unadjusted outcomes Thrombocytopenia p No Yes Ratio (95% CI) Sample Size 407,380 30,250 - - 30-day Mortality 0.3 1.7 5.3 (4.8-5.9) <.001 Complication 5.2 12.4 2.4 (2.3-2.5) <.001 Cardiac 0.2 0.9 3.7 (3.2-4.3) <.001 Respiratory 0.5 2.2 4.1 (3.8-4.5) <.001 Thrombotic 0.0 0.0 - - Transfusion 0.2 1.2 5.7 (5.0-6.4) <.001 Renal 0.2 1.0 4.2 (3.7-4.8) <.001 Return to OR 1.2 2.2 1.9 (1.8-2.1) <.001 Prolonged Hospitalization 0.1 0.3 2.7 (2.1-3.5) <.001 Unplanned Readmission 55.0 72.9 1.3 (1.2-1.5) <.001 Length of Stay Same-day Discharge Yes 43.4 21.5 0.5 (0.5-0.5) <.001 No (days) 3.8 5.5 1.5 (1.4-1.5) <.001 Note. Data presented as estimated percent or days. Thrombocytopenia defined as pre-operative platelet count less than 150 x 109/L. The ratios for 30-day Mortality through Unplanned Readmission are risk ratios. For all ratios, a ratio greater than 1 indicates greater unadjusted outcome for patients with thrombocytopenia.