A report of a rare case of autoimmune haemolytic anaemia in a patient with Hodgkin’s disease in whom routine serology was negative

Introduction

Autoimmune haemolytic anaemia (AIHA) is an acquired clinical condition which is characterised by the production of auto-antibodies that bind to the surface of circulating erythrocytes, leading to haemolysis and decreased survival of the red blood cells. The estimated yearly incidence of AIHA is 1–3 cases per 100,000 persons in the general population¹. The direct antiglobulin test (DAT) is considered to be a cornerstone in establishing the diagnosis of AIHA, since there is uptake of autoantibodies and/or complement components onto the affected red blood cells. In suspected cases of AIHA, a positive DAT is predictive in 83% of patients, but not all cases of AIHA are DAT-positive². Between 5% to 10% of all cases of AIHA are DAT-negative³. Three causes for this situation have been identified: (i) red blood cell bound IgG molecules, below the threshold of detection of the DAT, (ii) low-affinity IgG autoantibodies that are washed off the red cells during the washing phase for the test, and (iii) red cell bound IgA and rare warm IgM autoantibodies that are not detectable by the routine anti-human globulin reagent³. Given these different possibilities, a negative DAT must be weighed in the light of clinical suspicion. If there is clinical evidence of haemolysis, a thorough investigation with more specialised testing is needed. We report here a case of DAT-negative AIHA in a patient with Hodgkin’s lymphoma which was detected in our hospital blood bank laboratory.

Case report

A 36-year old male was referred to our institute with the main complaints of low-grade intermittent fever along with left cervical lymphadenopathy and abdominal discomfort. A cervical lymph node was biopsied as a diagnostic procedure and was reported as showing mixed-cellularity Hodgkin’s disease (HD). Abdominal spiral computed tomography scanning from the diaphragm to pubic symphysis revealed enlarged para-aortic lymph nodes and multiple hypodense space-occupying lesions in the liver and spleen. A bone marrow biopsy did not reveal infiltration and so the patient was classified as having stage IIIB HD. Subsequently, he received eight cycles of chemotherapy. Post-treatment computed tomography scans showed complete resolution of the disease. There were no enlarged lymph nodes, no organomegaly and complete blood counts were within the normal range.

Four years later the patient came to the hospital again with complaints of malaise, intermittent fever, dizziness, yellow discoloration of the sclera and dark urine which had been present for 1 month. Abdominal and thoracic positron emission tomography showed a space-occupying lesion in the liver. He was admitted as a relapsed case of Hodgkin’s disease. Laboratory investigations showed a haemoglobin concentration of 4.3 g/dL, haematocrit 13%, mean corpuscular volume 121 fL, reticulocyte count 36%, white blood cell count 13.1×10⁹/L and platelet count 194×10⁹/L. His plasma haemoglobin was 28.4 mg/dL and his lactate dehydrogenase level was 1,200 IU/L. Liver function tests showed albumin 4.2 g/dL, total proteins 7.5 g/dL, total bilirubin 7.47 mg/dL, an elevated indirect bilirubin level of 6.5 mg/dL, alkaline phosphatase 111 IU/L, alanine transaminase 24 IU/L and aspartate transaminase 80 IU/L. His peripheral blood smear showed marked anisocytosis, macrocytosis and 5–8 normoblasts/100 white blood cells. His bone marrow aspirate and biopsy showed megaloblastic erythroid hyperplasia.

On immunohaematological workup, his blood group was found to be A Rh(D) positive and the DAT was negative when tested by both a column agglutination technique (DiaMed GmbH, Cressier, Switzerland), as well as by the conventional tube technique using polyspecific anti-human globulin. The indirect antiglobulin test (IAT) was also negative when commercially available antibody screening cells were used. In the mean time, because of his persistently low haemoglobin level, the patient was transfused with 13 units of packed red blood cells.

Due to ongoing severe haemolysis and markedly raised plasma haemoglobin, the patient’s sample was sent to the immunohaematology laboratory again for complete serological evaluation of AIHA. The DAT and IAT were repeatedly negative using the afore-mentioned techniques. Donath-Landsteiner antibody testing was negative, ruling out paroxysmal cold haemoglobinuria. A test for paroxysmal nocturnal haemoglobinuria on a gel card (DiaMed GmbH, Switzerland) was also negative. These immunohaematological results, when reviewed in conjunction with the clinical and laboratory findings of severe haemolysis, led to the suspicion of Coomb’s negative AIHA. A DAT, performed after washing the red blood cells in cold saline to rule out low affinity antibodies, was negative. Subsequently, DAT was performed using a monospecific antiglobulin gel card (DiaMed GmbH, Switzerland) and showed a grade 3 reaction with IgA antiglobulin. The reactions in other columns containing antiglobulin for IgG, IgM, C3c, C3d and control were negative, which established the diagnosis of secondary IgA-mediated AIHA.

The patient was treated with a protocol of gemcitabine, dexamethasone and cisplatin (GDC). He received six cycles of this GDP regime and full-dose steroid therapy was given for 3 months, before the dose was tapered down. One year later the steroid was discontinued and the patient is now in good clinical condition with remission, without any laboratory or clinical evidence of further haemolysis.

Discussion

In the absence of a positive DAT, the diagnosis of AIHA is not considered seriously and other causes of haemolysis are sought using pertinent investigations, thus overlooking the relatively uncommon entity called Coomb’s negative AIHA. The standard Coombs’ reagent consists of polyclonal rabbit antiserum that binds to human IgG and complement, usually C3. Therefore, in cases in which a warm IgM or IgA antibody is mediating the process, a standard DAT is unlikely to elucidate the diagnosis. This case demonstrates the importance of performing a monospecific antiglobulin test if there is a strong suspicion of AIHA in apparently DAT-negative cases.

Most reported cases of IgA-mediated AIHA present initially as severe haemolytic anaemia with a negative DAT. In a study by Petz and Garratty, 7.8% of all cases of AIHA and 11% of cases of warm-reactive AIHA had a negative DAT⁴. It has been estimated that from 3 to 11% of cases of AIHA are DAT-negative^4,5. In 1997 Sokol et al. published the results of a study in 5235 patients who were referred to their centre over a 14-year period⁶. They found that 124 patients had IgA auto-antibodies out of whom only six had IgA alone for an incidence of 0.14%. The rarity of patients in whom IgA alone coats the red blood cells is supported by all authorities, with the reported incidence in some series ranging from 0.2 to 2.7%^7,8. Sokol et al. found that all six patients with IgA-only AIHA had IgA and C3d coating their red blood cells and two of them had severe intravascular haemolysis⁶ in concordance with the evidence that IgA antibodies are capable of activating complement by an alternative or lectin pathway⁹. Our patient had a plasma haemoglobin of 28.4 mg/dL and it is, therefore, probable that complement-dependent mechanisms of red cell destruction were operating and had caused intravascular haemolysis.

AIHA may occur in lymphoproliferative diseases, especially chronic lymphocytic leukaemia and non-Hodgkin’s lymphoma, but is rarely seen in HD¹⁰. IgA-mediated AIHA in HD is an exceedingly rare entity. Sporadic case reports and reviews have shown that when AIHA occurs in HD, it happens mostly in stages III and IV of nodular sclerosis HD or mixed cellularity HD¹¹. This report presents an original case of severe IgA-mediated AIHA in a patient with relapsed stage IIIB HD, illustrating that AIHA in Hodgkin’s lymphoma does not always present in its classical form with a strongly positive DAT. A similar case was reported by Moncharmont et al. in 2007: a 21-year old male patient with stage IIB, nodular sclerosis HD, had features of haemolysis but routine immunohaematological tests were negative¹². In this case, too, the diagnosis of IgA-mediated AIHA was established using monospecific antiglobulin gel cards.

Corticosteroids are the mainstay of therapy for warm-reactive AIHA. Although the serological findings in this case are different from those in classical cases of AIHA, our patient had a rapid and sustained response to oral steroid therapy. His clinical course improved without further need for packed red blood cell transfusions, with an increase in haemoglobin concentration as well as decreased reticulocyte counts and serum lactate dehydrogenase levels.

In conclusion, additional testing may be warranted in patients with an unusual presentation of haemolytic anaemia and negative DAT specifically to identify the presence of IgA autoantibodies on the surface of circulating erythrocytes, which may lead to clinically significant haemolytic anaemia. The most instructive lesson from this case for clinicians and transfusion specialists is that early diagnosis of DAT-negative, IgA-mediated AIHA can curtail unnecessary transfusions to a patient.