Conference Coverage

Study finds no link between platelet count, surgery bleed risk in cirrhosis



Perioperative bleeding risk was not associated with platelet count in a fairly large cohort of patients with liver cirrhosis, according to a review of cases, including some involving severe thrombocytopenia.

The findings raise questions about current recommendations that call for transfusing platelet concentrates to reduce bleeding risk during surgery in cirrhosis patients with extremely low platelet counts, Gian Marco Podda, MD, PhD, said at the International Society on Thrombosis and Haemostasis virtual congress.

The overall rate of perioperative bleeding was 8.9% in 996 patients who underwent excision of hepatocellular carcinoma by resection (42%) or radiofrequency ablation (58%) without platelet transfusion between 1998 and 2018. The rates were slightly higher among 65 patients with platelet count of fewer than 50 × 109/L indicating severe thrombocytopenia, and in 292 patients with counts of 50-100 × 109/L, indicating moderate thrombocytopenia (10.8% and 10.2%, respectively), compared with those with a platelet count of higher than 100 × 109/L (8.1%), but the differences were not statistically significant, said Dr. Podda of the University of Milan (Italy).

The corresponding rates among those who underwent radiofrequency ablation were 8.6%, 5.9%, and 5%, and among those who underwent resection, they were 18.8%, 17.7%, and 15.9%.

On multivariate analysis, factors associated with an increased incidence of major bleeding were low hemoglobin level (odds ratio, 0.57), age over 65 years (OR, 1.19), aspartate aminotransferase level greater than twice the upper limit of normal (OR, 2.12), hepatitis B or C cirrhosis versus cryptogenic cirrhosis (OR, 0.08), and resection versus radiofrequency ablation (OR, 3.74), he noted. Logistic regression analysis showed no significant association between platelet count and major bleeding events.

Mortality, a secondary outcome measure, was significantly higher among those with moderate or severe thrombocytopenia (rate of 5.5% for each), compared with those with mild or no thrombocytopenia (2.4%), Dr. Podda said.

Factors associated with mortality on multivariate analysis were severe liver dysfunction as demonstrated by Model for End-Stage Liver Disease score of 10 or greater versus less than 10 (OR, 3.13) and Child-Pugh B and C score versus Child-Pugh A score (OR, 16.72), advanced tumor status as measured by Barcelona-Clínic Liver Cancer staging greater than A4 versus A1 (OR, 5.78), major bleeding (OR, 4.59), and resection versus radiofrequency ablation (OR, 3.31).

“Low platelet count was associated with an increased risk of mortality at 3 months. However, this association disappeared at the multivariate analysis, which took into account markers of severity of liver cirrhosis,” he said.

Dr. Podda and his colleagues conducted the study in light of a recommendation from a consensus conference of the Italian Association for the Study of Liver Disease and the Italian Society of Internal Medicine that called for increasing platelet count by platelet transfusions in patients with cirrhosis who undergo an invasive procedure and who have a platelet count lower than 50 × 109/L.

“This recommendation mostly stemmed from consideration of biological plausibility prospects rather than being based on hard experimental evidence,” he explained, noting that such severe thrombocytopenia affects about 10% of patients with liver cirrhosis.

Based on the findings of this study, the practice is not supported, he concluded.

Dr. Podda reported honoraria from Sanofi, Boehringer Ingelheim.

SOURCE: Ronca V et al. ISTH 2020, Abstract OC 13.4.

Next Article: