People with cancer are at high risk for developing venous thromboembolism (VTE).1 Unfortunately, they are also at high risk for bleeding, particularly in the context of thrombocytopenia. When a patient with severe thrombocytopenia (<50 × 109/L) develops VTE, walking the fence between recurrent thrombosis and bleeding can be daunting. The International Society of Thrombosis and Haemostasis (ISTH) Scientific Subcommittee recently published an article that offers clinicians guidance on how to manage this difficult situation.
First, it is important to realize that there are no randomized controlled studies evaluating the safety and efficacy of anticoagulant therapy in patients with cancer who have thrombocytopenia. Instead, the data are limited to case series and small observational cohort studies.2 For this reason, the ISTH article provides guidance statements instead of graded recommendations. Second, low-molecular-weight heparin (LMWH) is the anticoagulant most frequently used during thrombocytopenia, primarily because dose reductions are easy to achieve.
The ISTH statements can be broken down into three major components:
1) Age of VTE
The natural history of venous thrombosis tells us that the risk of extension and/or embolization is highest within the first 30 days.3 Consequently, the need for therapeutic anticoagulant therapy is critical during this time frame. After 30 days, the risk drops, which is likely why reduced or prophylactic doses of anticoagulants can be used during thrombocytopenia in this subgroup. It is also important to remember that malignancy is an ongoing risk factor for VTE, therefore anticoagulant therapy should be increased to therapeutic levels as soon as it is considered safe to do so.
2) Location and Extent of VTE
Proximal lower-limb deep vein thrombosis (DVT) is considered more serious than catheter-associated upper-limb DVT owing to the higher potential for clinically significant pulmonary embolism (PE) travelling from the larger caliber veins within the lower limbs.4 Similarly, PE located within the main, lobar, or segmental pulmonary arteries is more of a concern than subsegmental PE. Although anticoagulants do not lyse existing emboli, they reduce the likelihood of recurrent PE that could prove fatal in patients who are already compromised. It is therefore reasonable to be more aggressive (see point 3) about anticoagulating a patient with a proximal leg DVT or segmental (or larger) PE, and less aggressive in patients with isolated distal leg DVT, catheter-associated upper-limb DVT, subsegmental PE, or superficial thrombophlebitis.
3) Severity and Expected Duration of Thrombocytopenia
A platelet count of 40 to 50 × 109/L is generally considered adequate for therapeutic-dose anticoagulant therapy, whereas anticoagulants should be held if the platelet count falls below 25 to 30 × 109/L. For patients who have an acute proximal leg DVT or segmental (or larger) PE less than 30 days old, platelet transfusion is suggested to increase the platelet count to 40 to 50 × 109/L to support anticoagulation. Clearly, this strategy is not feasible for patients who are refractory to platelet transfusions or patients who have prolonged thrombocytopenia (i.e., weeks to months). For patients with chronic (older than 30 days) and/or less severe VTE (see point 2) with platelet counts of 25 to 50 × 109/L, either half-therapeutic or prophylactic-dose LMWH is suggested as an alternative.
Prior to implementing the anticoagulant strategies listed here, heparin-induced thrombocytopenia as an alternative diagnosis must be considered and additional risk factors for bleeding should be reviewed (e.g., coagulopathy, renal, or liver dysfunction).
In Brief
Why is a platelet count of 50 × 109/L so widely quoted as the safe threshold for therapeutic-dose anticoagulant therapy? Surprisingly, no studies have addressed this issue directly. Randomized trials evaluating different thresholds for platelet transfusion for patients with hematologic malignancies excluded those who required therapeutic anticoagulation.5 Retrospective cohort studies and case series of cancer patients with VTE and thrombocytopenia have generally assumed a threshold of 50 × 109/L.2
It could be argued that patients who require anticoagulant therapy during thrombocytopenia are at similar risk to patients who require surgery during thrombocytopenia. While anticoagulants will not provoke bleeding as would a surgical incision, the consequences of bleeding can be as severe.6
Finding direct evidence to support 50 × 109/L as a safe threshold for surgery is also surprisingly difficult. Most platelet transfusion guidelines reference expert opinion as the reason for selecting that particular threshold.7,8 The highest quality of evidence comes from a single retrospective review of 95 patients with acute leukemia who underwent 167 operations prior to 1986.9 All patients were given platelet transfusions to raise their counts to 50 × 109/L. The procedures ranged from major operations (e.g., laparotomy, craniotomy) to minor procedures (e.g., catheter insertion, tooth extraction). Reassuringly, only 7 percent of patients experienced more than 500 mL of blood loss, and there were no deaths due to surgery-related hemorrhage. Extrapolating from these results, it is reasonable to conclude that the risk of bleeding should be low in anticoagulated patients who have a platelet count of at least 50 × 109/L.10
In summary, within the limitations of this “no data zone”,11 the guidelines for management of cancer-associated thrombosis during thrombocytopenia outlined by ISTH provide clinicians with a fence to walk on between bleeding and recurrent thrombosis.
References
Competing Interests
Dr. Linkins indicated no relevant conflicts of interest.
