Background: HHV6 reactivation has been increasingly recognized in allogeneic haematopoietic stem cell transplant recipients, particularly in umbilical cord blood transplantation. In adult recipients, HHV6 has been associated with limbic encephalitis (HHV6-LE), which causes anterograde amnesia and epilepsy. The CNS disorder confined to the hippocampal regions; the presence of abnormal MRI signals involving the hippocampus, and detection of HHV6 DNA inCSF. However, the clinical features of the syndrome have not been well characterized.

Methods: In this study we retrospectively reviewed the medical records of 139 umbilical cord blood transplantation recipients with reduced-intensity conditioning between January 2006 and December 2007 at Toranomon Hospital in Japan to determine the incidence of HHV6-LE among these patients. The diagnosis was based on the acute development of retrograde amnesia, confusion, coma and seizures accompanied by charascteristic MRI findings of abnormal high signal intensity on flair and T2 weighted images in the hippocampus and mesial temporal lobes, and the detection of HHV6 DNA in CSF. Eighty-six were received foscarnet before the onset of HHV6-LE as prophylaxis. All the patients who manifested suspicious symptoms of HHV6-LE were treated with foscarnet.

Results: HHV6-LE was diagnosed in 11/139 for a cumulative incidence of 7.2%. The median age was 49 years (range 36–82), 3 were female. Diseases treated were AML/MDS (8), ALL (1), diffuse large B-cell lymphoma (1), and aplastic anemia (1). All 11 were conditioned with fludarabine-based reduced-intensity regimen, and tacrolimus alone (5) or tacrolimus+mycophenolate mofetil (6) were used as GVHD prophylaxis. Five of them had started foscarnet before the onset of HHV6-LE as a prophylaxis. Median symptom onset was on day +20 (range, 12–50). Initial symptoms included confusion in 9, retrograde amnesia in 8, and seizures in 5. MRI demonstrated hyppocampal and mesial temporal lobe FLAIR/T2 high signal intensity in 8/11. Initial CSF analysis demonstrated a median WBC count was 7 (range, 1–21)/μl, median protein was 49 (range, 44–337) mg/dl, median glucose was 107 (range, 62–269) mg/dl. No positive results of bacterial or viral cultures were observed in any of the patients. Median HHV6 DNA levels in the CSF was 20000 (range, 200–80000) copies/ml. Median HHV6 DNA levels in peripheral blood was 200 (range, 0–300000) copies/ml. Remarkably, 7 patients were negative for HHV6 in peripheral blood even within 6 days prior to the onset of CNS symptoms. All were treated with foscarnet at 50–100mg/kg of body weight, and clinical symptoms improved in 9 evaluable patients (2 were not evaluable due to severe pulmonary complications). The virus was cleared in the CSF successfully in 7 patients evaluated. Seven died with the median survival of 32 days (range, 7–346) from diagnosis of limbic encephalitis. HHV6-LE was not a direct cause of death in any of them (2 were from infection, 2 from IP, 2 from relapse, and 1 from MOF). Four were alive as of July 31, 2008, and what is of extreme interest is that all were those who received foscarnet as a prophylaxis. None of the factors analysed failed to reach statistical significance regarding the impact on the onset of HHV6-LE.

Conclusions: High incidence of HHV6-LE as well as reactivation of the virus was observed in CB recipients. HHV6 DNA monitoring in peripheral blood may not be sufficient to predict the development of HHV6-LE. Although clinical manifestations were severe in some cases, foscarnet therapy was effective in all evaluable cases. Possible positive impact of prophylactic foscarnet therapy on better survival was suggested, which needs further evaluation by prospective study.

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author