Human Heme Oxygenase-1 Deficiency Presenting with Hemolysis, Bleeding, Nephritis, Asplenia and Inflammation.

Heme oxygenase-1 (HO-1) is a stress induced enzyme that catalyses the oxidation of heme to biliverdin and bilirubin. HO-1 is not constitutively produced and has been shown to be an endogenous factor against inflammation. Deficiency of HO-1 in HO-1 knockout mice as well as in one human case has been found to have a pro-inflammatory phenotype. To the best of our knowledge, previously there is only one published case of human HO-1 deficiency in literature. We report a patient who presented with massive hemolysis, bleeding, nephritis and inflammation and was proved to have HO-1 deficiency.

A 15-year-old previously well girl presented with high-grade fever for 2 weeks and cervical and axillary lymphadenopathy. She had an insignificant family history and had no growth or developmental retardation. Initial evaluation was unremarkable except for thrombocytosis noted from first week of fever and asplenia that was noted on CT scan. There was no past history of any major infection or thrombocytosis. She developed intravascular hemolysis from the second week of illness followed by nephritis as evidenced by hematuria, proteinuria (2950 mg over 24 hr), hyaline cast with hypertension. C3 levels were normal. IgA levels were increased to 502 mg/dl. Her blood was agglutinating at room temperature and serum could be separated with great difficulty. Workup for hemolysis showed evidence of cold antibody mediated immune hemolysis. Her direct combs test was positive. LDH was 9262 IU/l. Urine was positive for hemoglobinuria. Her serum ferritin was 4219 ng/ml, ESR was 75mm/1^st hr, platelets were increased to 1200,000/ul and interleukin-6 levels were increased to 139 pg/ml. Blood cultures were negative repetitively. However etiology including infections, lymphoreticular malignancies was ruled out. She was started on treatment with methylprednisolone (20 mg/kg x 3 days) followed by oral prednisolone at 2mg/kg/day. A month later, she developed vasculitic skin rash. In renal biopsy, glomeruli showed normal cellularity with mild focal mesangeal prominence, normal basement membrane and open capillary loops compatible with IgA nephropathy or minimal change disease. Immunoflourescence was non-contributory due to absence of any glomerulus. Bone marrow was normal and skin biopsy was non-contributory. Hemolysis persisted despite giving steroids. She developed bleeding diathesis in the form of uncontrolled epistaxis, diffuse alveolar hemorrhage and skin bleeds. Evaluation for coagulopathy showed high platelet count and normal PT (14.9 sec), INR 1.24 and APTT (32 sec). Von Willebrand antigen was raised to 308 %; Fibrinogen was 2.4 g/l and FDP level was increased to 8.0 mg/l by screening method. Epistaxis was controlled with repeated transfusion of fresh plasma. In view of hemolysis and nephritis non-responsive to steroids, she received a pulse of cyclophosphamide (500mg/m²) followed by two doses of Rituximab (375 mg/m² each). Her condition deteriorated further over the next few weeks. She developed intracranial hemorrhage in right parietal lobe and required Factor VIIa for control of bleeding. She died with fungal sepsis (Trichosporon asahii) after prolonged hospitalization within 5 months from onset of the illness. Atypical features noticed in this case were near normal serum bilirubin despite severe intravascular hemolysis and presence of asplenia. In view of a similar case with hemolysis, asplenia and nephritis reported from Japan as the first case of human HO-1 deficiency, our patient's DNA was sent to the same lab for HO-1 gene mutation. Mutation analysis revealed homozygous mutations in exon 2 (R44X) on chromosome 22q12 which would result in absence of functional protein HO. Renal biopsy for HO-1 immunostaining showed little if any detectable HO-1 in renal tubules, despite the presence of massive hemolysis.

HO-1 has been found to have cytoprotective effect against stress mediated organ damage. Inhibition of HO-1 has been shown to increase the vulnerability of an individual to stressful stimuli like infections. Lack of this enzyme might explain the varied clinical features and exaggerated response to stress noted in our patient. However it is not clear whether the entire spectrum of manifestations noted in our patient can be explained solely by HO-1 deficiency. Further cases are required to characterize the clinical and pathological manifestations of HO-1 deficiency.

Off Label Use: Use of activated factor VII for control of intracranial hemorrhage.