Diagnosing nocturnal paroxysmal hemoglobinuria: a single‐center 4‐year experience. (13th April 2017)

• Record Type: Journal Article
• Authors: Mercier, T.; Devos, T.; Mukovnikova, M.; Boeckx, N.
• Is Part Of: International journal of laboratory hematology. Volume 39:Number 3(2017:Jun.)
• Page Start: 329
• Page End: 336
• Publication Date: 2017-04-13
• Subjects: PNH -- nocturnal paroxysmal hemoglobinuria -- diagnostic testing -- screening -- indications; Hematology -- Periodicals; Blood -- Diseases -- Periodicals; Hematology -- Periodicals; 616.15005
• DOI: 10.1111/ijlh.12631
• Languages: English
• ISSNs: 1751-5521

Abstract

Summary: Introduction: Paroxysmal nocturnal hemoglobinuria (PNH) is a rare disease and can present as a wide range of signs and symptoms. As such, the indication for diagnostic testing for PNH is not always straightforward. Therefore, we analyzed all first‐time samples tested over a 56‐month period to determine the clinical settings with a high probability of detecting a PNH clone. Methods: We retrospectively analyzed 323 first‐time PNH flow cytometry tests, including LDH, cytopenias, direct antiglobulin test (DAT), and clinical indication for testing as available at the time of testing. Results: The probability of finding a PNH clone was 47% in patients tested because of aplastic/hypoplastic bone marrow disorders, 10% in DAT‐negative hemolytic anemia (HA), 5% in myelodysplastic syndromes (MDS), 3% in cytopenias other than HA, and 2% in thrombosis. When testing for another reason than the indications described before, there were no positive samples. Conclusion: Our findings reinforce guidelines from the International PNH Interest Group which suggest testing for PNH in the setting of unusual thrombosis, HA, aplastic/hypoplastic bone marrow disorders, or MDS, as these have a higher pretest probability. This probability drops to zero in our study in nonrecommended indications. This reflects the need for better education of clinicians about the disease PNH and the indications for diagnostic testing.