Early Thrombotic Events Following Splenectomy for Myelofibrosis: Incidence, Risk Factors and Effect of Pre-Emptive Anticoagulant Therapy

Background:

Current drug therapy in myelofibrosis (MF) is mostly palliative. Other treatment modalities with palliative value in MF include red cell transfusion support, involved field radiotherapy, and splenectomy. Indications for splenectomy in MF include drug-refractory symptomatic splenomegaly, frequent transfusion need, refractory thrombocytopenia and complications from portal hypertension. The salutary effects of splenectomy include symptomatic relief and improvement in anemia and thrombocytopenia. Complications following splenectomy include perioperative bleeding, infections and thrombosis. The current study addresses the latter, in terms of prevalence, vascular distribution, risk factors and the effect of preventive anticoagulant therapy.

Methods:

Study patients were selected from institutional databases. Diagnosis of MF, including primary (PMF) and post-essential thrombocythemia (post-ET) and post-polycythemia vera (post-PV), was according to World Health Organization (WHO) (Blood 2016;127:2391) or international working group for MPN research and treatment (IWG-MRT) (Leukemia 2008;22:437) criteria. All surgical procedures were performed at our institution. Statistical analyses considered clinical and laboratory data collected at time of splenectomy. Early thrombotic events constituted those occurring in the first two months of splenectomy. Pre-emptive anticoagulant therapy was acknowledged if documented in the first two weeks of splenectomy, prior to an event.

Results:

Between 2001 and 2016, therapeutic splenectomies were performed on 123 consecutive patients with chronic phase MF; amongst these patients, 120 were evaluable for the occurrence of post-operative thrombotic events. Median time between initial diagnosis and date of splenectomy was 3.2 years.

• Clinical and laboratory features at time of splenectomy

Median age was 66 years (range 43-84) and 58% were males. MF variants were PMF 76%, post-PV MF 18% and post-ET MF 6%. Risk distribution, according to the dynamic international prognostic scoring system (DIPSS) (Blood . 2010;115:1703) was 21% high, 55% intermediate-2, 21% intermediate-1 and 3% low. 60% of the patients had constitutional symptoms, 61% were red cell transfusion-dependent and median platelet and leukocyte counts were 106 x 10(9)/L (range 1-699) and 13.5 x 10(9)/L (range 0.5-113), respectively. Driver mutational status was known in 89 patients and included JAK2 75%, CALR 15%, MPL 4% and triple-negative 6%. Karyotype was abnormal in 60%. History of thrombosis prior to splenectomy was documented in 22% (arterial 15% and venous 7%).

• Early post-splenectomy thrombosis and its risk factors

Median follow-up from the time of splenectomy was 1.3 years (range 0.02-8.5). 34 (28%) patients experienced post-splenectomy thrombosis in the first 2 months of surgery (median 7 days; range 1-60); these constituted venous events in 91% of the cases and involved portal and/or splenic veins in 20 and 13 episodes, respectively. During a comprehensive analysis of clinical and laboratory variables, post-splenectomy thrombosis was significantly associated with pre-splenectomy history of venous thrombosis (p=0.02) and higher platelet count at time of splenectomy (p=0.03); there were no associations with driver mutational status, leukocyte count, karyotype, age or gender.

• Pre-emptive anticoagulant therapy within two weeks of surgery and its effect on post-splenectomy thrombosis

Pre-emptive anticoagulant therapy within two weeks of surgery and a minimum treatment duration of 3 consecutive days was instituted in 77 (62%) patients and included heparin therapy in full therapeutic (n=17; 14%) or prophylactic (n=58; 48%) dose and continued warfarin therapy for a history of pre-splenectomy thrombosis (n=2; 2%). Post-splenectomy thrombosis occurred in none (0%) of the patients receiving full therapeutic dose heparin, 26% of those on prophylactic heparin therapy, both (100%) patients on warfarin therapy and 40% of those on no anticoagulant therapy (p=0.002).

Conclusions:

More than a quarter of MF patients undergoing therapeutic splenectomy might experience mostly venous post-splenectomy thrombosis, in the first two months of surgery; pre-emptive full dose heparin therapy might be protective in this regard and risk factors include history of venous thrombosis and higher platelet count at time of splenectomy.