Abstract
Introduction: Central venous catheters are frequently used in hematooncological patients for the administration of chemotherapy, antibiotics and parenteral feeding. Despite low platelet counts, observed in this category of patients, catheter related thrombosis (VTE), with the need for anticoagulant therapy, still occurs in 9.7% of the patients. Treatment of this complication in our institute consists of the removal of the catheter and anti-coagulant therapy, consisting of conventional heparin or low molecular weight heparin. Although solely based on expert opinion, the generally accepted platelet transfusion threshold of 10 x 109/l is raised to 20 – 40 x 109/l. In our institute a trigger of 30 x 109/l is used. We investigated whether this trigger policy is effective in the prevention of bleeding complications.
Methods: We performed a case-control study to study bleeding complications and threshold policy. From 1 February 2000 to 1 May 2005 we reviewed medical records of patients, admitted at the hematology department of the Leiden University Medical Center. A total of 22 patients were identified having thrombosis confirmed by ultrasound or phlebography. The database of a recently performed clinical trial in our institution was used to select a control group. Thirty-one controls were selected, matched for age, gender, diagnosis and treatment. For each patient, we calculated the days at risk of bleeding, i.e. days of thrombocytopenia with or without anti-coagulant therapy and recorded bleeding complications, platelet and red cell transfusions from the medical charts. Statistical analysis was performed using SPSS.
Results: There were no significant differences between patients with or without VTE in regard to age, gender, diagnosis and treatment. The number of days at risk of bleeding for patients without VTE was 511 days and for VTE group 239 days. The table shows the number of bleeding complications, the mean platelet count, the mean platelet transfusion threshold and the number of platelet and red cell transfusions. Both univariate as multivariate analysis (including age, gender, diagnosis, treatment and VTE-status) showed a significant effect of the existence of VTE on the occurrence of bleeding complications. The Odds’ ratio of bleeding in the VTE group compared to the non-VTE group was 2.9 (CI 95: 1.7 – 5.0; p = 0.0002). More then 95% of the bleeding complications consisted of WHO grades I and II. No lethal bleeding was observed.
Discussion: Patients experiencing (catheter-related) venous thrombosis treated with anticoagulant therapy during a period of thrombocytopenia showed an increased risk of bleeding complications, despite a higher platelet transfusion threshold. Whether steering between Scylla and Charibdis can be facilitated by raising the platelet transfusion threshold remains to established.
Study outcome No VTE VTE p value n = number Days at risk 511 239 n bleeding complications per day at risk 0.05 0.13 0.0002 mean platelet count (10E9/l) +/− SD 28 +/− 16 38 +/− 17 < 0.0001 mean platelet transfusion threshold (10E9/l) +/− SD 12 +/− 9 28 +/− 12 < 0.0001 n platelet transfusions per day at risk 0.35 0.49 0.0003 n red cell transfusions per day at risk 0.34 0.41 0.073
Perioperative treatment of hemophilia A patients: blood group O patients are at risk of bleeding complications