Paroxysmal Nocturnal Hemoglobinuria Workup

Updated: Jun 06, 2023
  • Author: Emmanuel C Besa, MD; Chief Editor: Sara J Grethlein, MD, MBA, FACP  more...
  • Print

Approach Considerations

In addition to a complete blood cell count, the principal studies used to establish the diagnosis of paroxysmal nocturnal hemoglobinuria (PNH) are flow cytometry of peripheral blood and bone marrow analysis. Flow cytometry measures the percentage of cells that are deficient in the glycosyl phosphatidylinositol–anchored proteins (GPI-APs) and identifies discrete populations with different degrees of deficiency. Because of the missing GPI-APs, red blood cells (RBCs) and other cells in patients with PNH lack DAF (CD55) and MIRL (CD59), which regulate complement.

Hemosiderin is nearly always present in the urine sediment and can accumulate in the kidneys; this is visible on magnetic resonance images (MRI) or computed tomography (CT) scans. An elevated reticulocyte count and serum lactate dehydrogenase (LDH) level with a low serum haptoglobin level in the absence of hepatosplenomegaly are the hallmarks of intravascular hemolysis.

Bone marrow examination will differentiate classic PNH from PNH that develops in the setting of other bone marrow disorders. [28] In addition, bone marrow examination will identify an erythroid and hyperplastic bone marrow during the hemolytic phase or a hypoplastic bone marrow in the aplastic phase.

Imaging studies are indicated in patients with venous thrombosis.


Laboratory Studies

The tests involved in establishing the diagnosis of paroxysmal nocturnal hemoglobinuria (PNH) demonstrate the presence of red blood cells (RBCs) that are exceptionally sensitive to the hemolytic action of complement. These tests include the following:

  • Flow cytometry
  • Acidified serum lysis and Ham test
  • Complement lysis sensitivity test
  • Sugar water or sucrose lysis test

Most laboratories no longer perform the Ham test or the sugar water test.

Flow cytometry

The state-of-the-art laboratory test is flow cytometry of the patient's blood to detect CD59 (MIRL), a glycoprotein, and CD55 (DAF) in regulation of complement action. Absence or reduced expression of both CD59 and CD55 on RBCs is diagnostic of PNH.

The use of flow cytometry in PNH differs from many applications in that the diagnosis depends upon demonstrating the absence of relevant antigens. In this context, it is important that at least two glycosyl-phosphatidylinositol (GPI)–linked antigens are studied to exclude rare congenital deficiencies of single antigens (CD55 and CD59) and polymorphism with individual antigens (CD16), which render them undetectable by some monoclonal antibody clones.

Standard and high-sensitivity flow cytometric procedures for detecting PNH cells are now available. [29] For routine analysis and diagnosis of suspected PNH, the standard test is sufficient. This test can detect 1% or more PNH cells, bu; most laboratories report only 10% or more as a positive result. High-sensitivity analysis (in which as little as 0.01% PNH cells can be detected) may be helpful in aplastic anemia patients, who may eventually develop PNH, and possibly in those with hypoplastic myelodysplasia syndrome (MDS), to predict responses to immunosuppressive therapy.

Fluorescent aerolysin

A more accurate alternative reagent for PNH screening and PNH clone measurement is the bacterial toxin aerolysin, which binds to RBCs via GPI anchor and initiates hemolysis. A modified, nonhemolytic form of a fluorescently labeled molecule has been developed that can detect PNH cells to a level of 0.5% (fluorescently labeled inactive toxin aerolysin [FLAER] binding of peripheral blood granulocytes). The advantage of this assay is that it can detect the clone in all hematopoietic cell lineages in one assay.

This is the most specific test for PNH, as FLAER binds the GPI anchor specifically. Thus, the lack of FLAER binding to granulocytes (measured by flow cytometry) is sufficient for the diagnosis of PNH. The disadvantage of the test is in measuring binding in the absence of adequate granulocytes, such as in severe aplastic anemia, when the number of circulating granulocytes is extremely low.


Peripheral blood is the most suitable specimen for immunophenotyping for PNH, and it is important to screen both RBCs and granulocytes, because RBC transfusions are common among these patients and granulocytes may not be present in severe hypoplastic anemia patients.

Studies have shown that the size of the PNH clone correlates with the risk for venous thrombosis. Patients with less than 50% PNH granulocytes seldom develop thrombosis, whereas patients with larger clone sizes appear to be at great risk and will require anticoagulation.

Acidified serum lysis and Ham test

If performed properly, acidified serum lysis and the Ham test (from Thomas Hale Ham) are reliable ways to diagnose PNH (see image below). Dr. Ham demonstrated that the RBCs in PNH were lysed by complement when normal serum was acidified or activated by alloantibodies.

The Ham test (acidified serum lysis) establishes t The Ham test (acidified serum lysis) establishes the diagnosis of paroxysmal nocturnal hemoglobinuria (PNH), demonstrating a characteristic abnormality of PNH red blood cells by acidified fresh normal serum. Here is a PNH patient's (Pt) red blood cells lysed by normal serum at room temperature (RT) and at 37°C compared with normal red cells (no hemolysis) (control [C]). Heated serum at 56°C inactivates complement and prevents hemolysis in PNH cells. Permission to use this image has been granted by the American Society of Hematology Slide Bank, 3rd edition.

The serum pH is lowered to about 6.2 and the magnesium level is adjusted to 0.005 mol/L to achieve maximum sensitivity. The cells that are hemolyzed are the sensitive cells, and those that remain intact are normal cells, indicating 2-3 subpopulations of RBCs in the circulation.

A false-positive test result is seen in congenital dyserythropoietic anemia, type II (hereditary erythroblastic multinuclearity with positive acidified serum tests [HEMPAS]). These patients have a negative sucrose hemolysis ("sugar water test") result. Some normal serum can give a false-negative Ham test result; thus, the sucrose water test is more sensitive but less specific for paroxysmal nocturnal hemoglobinuria (PNH).

Complement lysis sensitivity test

The complement lysis sensitivity test of Rosse and Dacie is a more precise method for diagnosing PNH. RBCs are sensitized with a potent lytic anti-I antigen and hemolyzed with limiting amounts of normal serum as a source of complement. [30, 31, 32] This demonstrates threee groups of RBCs in patients with PNH, including the following:

  • PNH I cells are normal in sensitivity to complement

  • PNH II cells are moderately more sensitive to complement than normal cells

  • PNH III cells are markedly sensitive to complement, requiring one fifteenth to one twentieth of the amount of complement for an equal degree of lysis; this group of cells is increased in patients with more severe PNH, and it is associated with a mean life span of 10-15 days

Sugar water or sucrose lysis test

The sugar water or sucrose lysis test uses the ionic strength of serum that is reduced by adding an iso-osmotic solution of sucrose, which then activates the classic complement pathway, and complement-sensitive cells are lysed. This test is less specific but more sensitive for PNH than the Ham test, because some RBCs hemolyze from autoimmune hemolytic anemias, leukemia, and aplastic anemia to a minor degree. Although the tests are inexpensive and simple to perform, they are more labor intensive and less sensitive due to the short half-life of circulating PNH RBCs.

Other tests for intravascular hemolysis

Other tests to demonstrate intravascular hemolysis include the following:

  • Elevated serum lactate dehydrogenase (LDH)
  • Elevated reticulocyte count
  • Low-to-absent serum haptoglobin
  • Hemoglobinuria and hemosiderinuria; however, hemolysis may occur intermittently and hence can be missed easily, depending on when the tests are performed

Imaging Studies

Thromboses of major veins are best evaluated by radiographic means.

Investigate hepatic vein thrombosis with a routine technetium-99m (99m Tc) colloid scan of the liver and spleen. This study often reveals diminished function in all portions of the liver except the caudate lobe, which is spared because it is drained by the inferior vena cava rather than the hepatic vein. A magnetic resonance imaging (MRI) study or ultrasonogram can demonstrate the cessation of flow through the hepatic vein or by injection or use of a dye to demonstrate a thrombus in the vein.

MRI with contrast may demonstrate sagittal vein thrombosis.


Other Tests

PIG-A gene mutation analysis is still limited to research laboratories. In addition, although it is very specific, it is still not diagnostic for paroxysmal nocturnal hemoglobinuria (PNH).