Risk Factors for Survival and Thrombosis in Patients with Paroxysmal Nocturnal Hemoglobinuria (PNH): A Single Institution Study of 68 Patients.

Background:

Paroxysmal Nocturnal Hemoglobinuria (PNH) is a clonal stem cell disorder characterized by intravascular hemolysis, thrombosis, bone marrow failure and reduced survival. In this current single institution study of patients with PNH, we investigate the risk factors predictive of survival and thrombosis.

Methods:

Data were abstracted from medical records of a consecutive cohort of patients with PNH seen at the Mayo Clinic. Diagnosis was confirmed by the presence of intravascular Hemolysis and either a positive Ham’s test, and/or by the presence of a GPI-deficient hematopoietic clone. Follow-up data were obtained by contacting the patients and by reviewing the social security death index.

Results:

i. Patients and treatment:

The study cohort included 68 patients with PNH (median age 32 years, range 10–77; 33 females) with a median follow up of 84.5 months (range, 1–538 months). Presenting features included infection (n=9), thrombosis (n=9), neutropenia (absolute neutrophil count (ANC) < 1000) (n=32) severe intravascular hemolysis (n=16) and splenomegaly (n=11). Prior to the diagnosis of PNH, 6 patients (8.8%) carried a diagnosis of aplastic anemia (AA) and 2 (2.9%) myelodysplastic syndrome (MDS). Complications documented after diagnosis included major thrombosis (n=11), major hemorrhage (n=4), and development of AA (n=7), MDS (n=4), and non-hematologic cancer (n=4) but none experienced leukemic transformation. Treatment included prednisone (n=30), other immunosuppressive agents (n=6), androgens (n=10), and bone marrow transplantation (n=4). Review of treatment with systemic anticoagulation disclosed only 2 patients that were receiving prophylactic warfarin therapy and both patients remain free of thrombosis after 1 and 2 years of therapy.

ii. Thrombotic events and risk factors:

Overall, 20 patients (29.4 %) experienced one or more episodes of major thrombosis. Among several clinical and laboratory parameters evaluated at presentation, only the detection of urine hemoglobin and infection were found to be significantly associated with the occurrence of thrombosis (p=0.05)

iii. Risk factors for survival

To date, 18 patients (27%) have died and median survival for the study cohort is projected at 21.1 years. Multivariate analysis identified thrombosis (p=0.003) and an ANC of > 1000 (p= 0.05) at presentation as independent risk factors for inferior survival. In addition, the occurrence of major hemorrhage was significantly associated with poor survival (p= 0.04).

Conclusion

The current study provides natural history information in a relatively large cohort of patients with PNH and identifies certain disease features that could be further explored for their potential as risk factors for both thrombosis and survival.