An 11-year-old girl presented with chronic, bloody diarrhea. Endoscopic biopsies confirmed a diagnosis of chronic ulcerative colitis with features of Crohn’s disease. She was treated first with prednisolone and mesalazine, but azathioprine, 75 mg/day (2.5 mg/kg), was started after one year on account of continued active colitis. She responded well to this and remission was achieved. At the age of 16 years she was admitted to hospital on account of a high fever and poor general condition. Azathioprine was stopped on account of acute infection, as three-lineage cytopenia was detected on admission: her white blood cell count was 2.5 (4.5–11 × 10
9/L), platelets 38 (200–450 × 10
9/L) and hemoglobin 89 g/L (117–155 g/L). Splenomegaly, acute cardiac failure and acute hepatic failure, with 15% necrosis in a biopsy, were observed in the next few days. Symptoms of encephalitis were present, but a brain MRI was normal. Viral genome examinations by EBV polymerase chain reaction revealed > 350,000 copies/mL in plasma on admission and > 1 million copies/mL after 4 days of treatment (Fig.
1). A high viral load of > 8 million copies/mL was also detected in the liver biopsy specimen. Tests for IgM antibodies to VCA were positive on admission, whereas IgG antibodies to EBNA were negative, indicating acute EBV infection. She was treated once with rituximab and then with corticosteroids, acyclovir and intravenous immunoglobulin. She developed severe multiple organ failure, including cardiac, hepatic and respiratory failure, and was critically ill for 4 weeks in the intensive care unit. After this treatment in the intensive care unit, approximately 30 days after admission, the viral load started to increase again and she developed signs of severe hepatitis and pneumonitis, so that regular weekly rituximab infusions were given for a total of 8 weeks. Very low NK-cells were noted and the persistent viremia (detection level 300 copies/mL) stopped when the NK-cell count started to rise, approximately 90 days after the cessation of azathioprine (Fig.
1). During the course of illness, CD8+ T-cell counts were 1131 (220–1129 × 10
6/L) on day 32, 2535 on day 70, 2487 on day 93 and 2537 on day 106. Due to the prolongation of viremia and hepatitis despite the discontinuation of azathioprine, WES was performed. This did not reveal any known primary immunodeficiency, but the patient was found to be carrying two rare heterozygous variants of the
TLR3 and
OAS1 genes, in which variants have previously shown to be associated with susceptibility to viral infections. TPMT genotype leading to low enzyme activity was excluded in this case after the patient’s recovery. A high total IgG level (> 15 g/L) persisted for 6 months after clearance of the viremia, and the low CD19-cell count attributable to the rituximab treatment started to rise 6 months after cessation of the infusions. At the last follow-up visit, she was asymptomatic and healthy and had successfully returned to her studies 6 months after the onset of the EBV infection. She was still in remission with respect to her inflammatory bowel disease without any medication.