Anticoagulation with intravenous unfractionated heparin and major bleeding in cirrhosis.

Abstract

Potential conflict of interest: Nothing to report. TO THE EDITOR: We read with great interest the article by Drolz et al. regarding the prediction of new onset of major bleeding in 211 patients with cirrhosis admitted to the intensive care unit (ICU).1 The investigators used intravenous unfractionated heparin as therapeutic anticoagulation with an activated partial thromboplastin time (aPTT) target of 50‐60 seconds in 38 patients. Notably, the proportion of use of therapeutic anticoagulation was very similar between patients who developed and did not develop major bleeding during their hospitalizations (6 of 35 [17%] vs. 32 of 176 [18%]; P = 0.884, see Table 2), which suggested no association between anticoagulation and major bleeding. However, these results should be cautiously interpreted. As the authors described in the Methods section, therapeutic anticoagulation was prescribed in the absence of bleeding.1 In other words, anticoagulation was not considered in patients with bleeding on admissions. If so, the patients with bleeding on admission should be excluded from the statistical analysis regarding the association between the use of anticoagulation and new occurrence of major bleeding. Importantly, bleeding on admission was the most important factor for predicting the occurrence of major bleeding (18 of 35 [51%] vs. 10 of 176 [6%]; P < 0.001; see Table 2; odds ratio [OR] = 14.819; see Table 5).1 Thus, it might be necessary to reevaluate whether anticoagulation increased the risk of major bleeding. A revised statistical analysis did not find any statistically significant difference, but suggested a higher proportion of use of anticoagulation in patients who developed major bleeding than in those who did not (6 of 17 [35%] vs. 32 of 166 [19%]; P = 0.121). More important, an aPTT value >100 seconds (i.e., >1.6‐fold the upper limit of the therapeutic range on unfractionated heparin) was significantly associated with an increased risk of major bleeding in patients with cirrhosis (OR = 7.01; see Table 5; OR = 6.04, after excluding the patients with bleeding on admission). Accordingly, we could not exclude the possibility that unfractionated heparin prolonged aPTT, thereby increasing the risk of major bleeding. In our meta‐analysis of eight studies regarding therapeutic anticoagulation in patients with cirrhosis with portal vein thrombosis, the pooled rate of bleeding was 3.3% (95% confidence interval: 1.1‐6.7).2 Notably, subcutaneous low‐molecular‐weight heparin, rather than intravenous unfractionated heparin, was used in nearly all of included studies. By comparison, in the article by Drolz et al. with intravenous unfractionated heparin, the rate of major bleeding appeared to be higher (17% in all patients or 35% in patients without bleeding at admission).1 Accordingly, the use of therapeutic anticoagulation by intravenous unfractionated heparin, instead of subcutaneous low‐molecular‐weight heparin, may be suspected to influence the onset of major bleeding in ICU patients with liver cirrhosis.