The Incidence and Prevalence of Paroxysmal Nocturnal Hemoglobinuria (PNH) and Survival of Patients in Yorkshire.

Abstract

Abstract Paroxysmal nocturnal hemoglobinuria (PNH) is an acquired clonal stem cell disorder characterized by the expansion of a population of blood cells deficient in glycosylphosphatidylinositol (GPI) linked proteins. This results in the classical clinical features of intravascular hemolysis and thrombosis. PNH is known to be a rare disorder, but its incidence and prevalence have so far been poorly defined with very few studies. In order to better define the incidence and prevalence of PNH, survival data was collected on all patients diagnosed with PNH in the strategic health authorities of North and East Yorkshire, Northern Lincolnshire and West Yorkshire between January 1991 and July 2006. All patients were diagnosed by flow cytometry for GPI-linked antigens on red cells and neutrophils at a single reference laboratory (HMDS). The population of the study region is 3,742,835 (based on the 2001 census of Britain). 76 PNH patients were diagnosed during this time period giving an incidence of 0.13/100,000/year. Based on incidence and survival rates, the estimated 15 year prevalence of PNH is 1.59 per 100,000 resulting in a predicted prevalence of 59 patients in the study region. We have previously demonstrated that a neutrophil clone size >50% is a predictor of increased thrombotic risk; the current study predicts that 25% of patients will have >50% PNH neutrophil clone size, 43% with >10%, and 82% with >1%. Platelet count >100 x 109/L has been used as a criteria to consider primary prophylactic anticoagulation in PNH patients with substantial hemolysis if the neutrophil clone size is >50%. In the current study, the platelet count is >100 x 109/L in 32% of patients and <30 x 109/L in 27%. The primary clinical manifestation of PNH is intravascular hemolysis and although levels of hemolysis vary considerably between patients even those with relatively small PNH clones will have some degree of hemolysis. Levels of LDH (a sensitive marker of hemolysis) were elevated above the upper limit of the normal range in 82.5% of patients. Of the 59 patients in the study region, 33% reported hemoglobinuria. Overall survival was 78% with a median follow-up of 6.25 years (range 0 to 15 years). Survival was compared between patients with a) hemoglobinuria (89%) vs. those without (76%); b) neutrophil clone size ≥50% (81%) vs. <50% (80%) and c) Platelet counts <30 x 109/L (61%) vs. ≥30 x 109/L (86%) by univariate analysis. Hazard ratios and 95% confidence intervals were estimated using Cox’s proportional hazards model adjusted for age and sex. Worse survival was only significantly predicted by a platelet count <30 x 109/L (log rank test, p=0.008). With a population of 57,105,375 (2001 census of Britain), Great Britain should have an estimated 75 new cases of PNH per year and a predicted prevalence of 908 patients. The U.S.A. will therefore have 4713 cases of PNH based on its July 1, 2005 census bureau population estimate of 296,410,404. This study is the first to accurately report the incidence and prevalence of PNH in a given population in a well-defined geographical area.