The differential diagnosis of the patient with dark urine includes other causes of intravascular hemolysis such as mechanical damage, antibody mediated hemolysis (such as in PCH, mycoplasma infection, and cold agglutinin disease), unstable hemoglobin variants (such as hemoglobin Zurich), and G6PD deficiency. Myoglobinuria, porphyria, and some medications can also darken the urine. PNH can mimic MDS, as well as Budd Chiari syndrome produced by a myeloproliferative neoplasm and other hypercoagulable states. The bouts of abdominal pain in PNH can be confused with porphyria, pancreatitis, and familial Mediterranean fever. The combination of thrombocytopenia and thrombosis can be seen in a few other disorders, such as DIC, the antiphospholipid antibody syndrome, aHUS, and TTP. Patients with PNH will often have a prior (mis)diagnosis of hematuria or immune thrombocytopenic purpura (ITP). “Laboratory PNH”/“Subclinical PNH” produces mild cytopenias and macrocytosis and can be confused with B12 or folate deficiency or alcoholism.
Fortunately, the diagnosis of PNH by flow cytometry is now straightforward if done following standard recommendations. For red cells, loss of CD59 is highly sensitive and specific, though the proportion of abnormal red cells is typically under estimated due to hemolysis and transfusions. For the analysis of granulocytes and monocytes, a combination of an anti-CD24 anti body conjugated to PE together with the FLAER reagent (Alexa-488 conjugated to a non-pore forming variant of proaerolysin that binds to cells in a GPI-dependent manner) is particularly robust. The flow cytometry analysis should be performed only on anticoagulated peripheral blood and never on marrow. Red cells should be identified by forward and side scatter on a log-log scale. Unlike most flow cytometry protocols, here the abnormality is identified as a population that lacks expression of a marker, so it is important to ensure that the entire cell population comes in contact with the staining reagents. An antibody specific for non-GPI-linked proteins can be used to identify specific lineages, and this renders the method extremely sensitive to small populations, provided that a large number of events are collected. This can identify patients who have “laboratory PNH” or “subclinical PNH” at a low level, and this can have implications for prognosis and treatment in patients who might otherwise have a diagnosis of MDS, for example.
Congenital loss of CD59 can present similarly to PNH, from which it can be distinguished if an appropriate panel of reagents is used in the flow cytometry protocol. Of historical interest, type II congenital dyserythropoietic anemia146 can produce a false-positive Ham test, a condition known as HEMPAS (hereditary erythroblastic multinuclearity with positive acidified serum lysis test), but will produce a negative flow cytometry test.
