Page 53 - 2019_11 Resto del Mondo-web

Guidelines for HHV-6 infections
occur in areas such as the entorhinal cortex or amygdala.38 Temporal lobe seizures are relatively frequent but focal neurological deficits are rare. Computed tomography of the brain is often normal. Electroencephalograms are usually diffusely abnormal sometimes involving the tem- poral region. Autopsy reveals hippocampal disease with HHV-6 protein in astrocytes and neurons suggesting local virus reactivation32 rather than an indirect effect of viral- ly-induced neuroinflammation. Notably, a retrospective study39 showed that only one-third of HHV-6 encephalitis patients had the typical features of PALE.
Different studies have used different definitions of HHV-6 encephalitis.40 Ideally the definition would require proof of HHV-6 infection in tissue samples from the affected part of the brain. However, given the impracti- cality of such an approach and the epidemiological evi- dence, the definition below can replace the need for brain biopsy.
• Diagnosis of HHV-6B encephalitis should be based on HHV-6 DNA in CSF coinciding with acute-onset altered mental status (encephalopathy), or short-term memory loss, or seizures.
• Other likely infectious or non-infectious causes must be excluded.
• CIHHV-6 in donor and recipient should be excluded.
• If CIHHV-6 is detected, evidence for CIHHV-6 reacti- vation in the CSF or brain tissue is necessary to implicate CIHHV-6.
Other central nervous system dysfunction
Apart from encephalitis post-HSCT, HHV-6 has been associated with CNS disease ranging from headache to delirium and neurocognitive decline;41-43 patients whose donors or recipients had CIHHV-6 were excluded in two of these studies.42,43 HHV-6 has also been associated with myelitis, pruritis and dysesthesia in Japanese patients.44 Notably, HHV-6 DNA can be found in CSF in patients without CNS symptoms.42
Risk factors for human herpesvirus 6B encephalitis
Human herpesvirus 6B reactivation in blood (i.e. viremia) is a major risk factor and occurs in approximate- ly half of allogeneic transplant recipients in the first few weeks post-HSCT.45,46 The highest rates are seen after umbilical cord blood transplantation (CBT); in a prospec- tive cohort of 125 cord blood recipients, HHV-6B reacti- vation was documented in 94%.47 In a multicenter prospective study, Ogata et al.48 showed that reactivation precedes or coincides with HHV-6 encephalitis and that ≥10,000 copies/mL in plasma correlated with onset of dis- ease with 100% sensitivity and 64.6% specificity. Similar values of 100% and 81% respectively were obtained in a much larger retrospective study.49
However, not all patients develop encephalitis when the plasma HHV-6 DNA level is high, and other factors are involved, usually related to poor T-cell function, such as T-cell depleted allografts, CBT, a mismatched or unre- lated donor, acute GvHD and treatment with glucocorti- coids.50 A retrospective cohort study of 1,344 patients showed CBT is a major risk factor [adjusted hazard ratio (aHR) 20.0; P<0.001], as well as acute GvHD grades II-IV (aHR 7.5; P<0.001) and use of mismatched unrelated donors (aHR 4.3; P<0.04).49 A subsequent systematic review and meta-analysis of all relevant HSCT studies also demonstrated the incidence of HHV-6 encephalitis
was significantly higher post-CBT than other stem cell sources (8.3% vs. 0.5%; P<0.001).40 Ogata et al.36 used the Japanese Adult Transplant Registry and identified 145 patients with HHV-6 encephalitis; the relative risk for CBT was 11.09 (P<0.001) and 9.48 (P<0.001) for HLA- mismatched unrelated donors. Haploidentical transplant recipients may also be at high risk of HHV-6B encephali- tis based on a combined report of two small studies51 where, in an attempt to improve engraftment and reduce GvHD, donor cells were depleted of naïve T cells and nat- ural killer (NK) cells, but memory T cells remained. Finally, pre-engraftment syndrome might be a risk factor for HHV-6 encephalitis.50
Prognosis of human herpesvirus 6B encephalitis
Zerr33 reviewed the outcome in the many previous detailed descriptions of individual patients; 11 of 44 (25%) died within 1-4 weeks of diagnosis, 6 (14%) showed improvement but died with various unrelated medical problems, 8 (18%) improved but with lingering neurological compromise, and 19 (43%) appeared to make a full recovery. In a single retrospective study, Hill et al.49 reported 19 patients with PALE; attributable mor- tality was higher after CBT (5 of 10) than in recipients of adult donor stem cells (0 of 9). In a much larger number of allogeneic HSCT recipients,36 neuropsychological sequelae were reported in 57% of encephalitic patients with an overall survival rate of 58.3% in those with encephalitis as opposed to 80.5% in those without.
Other retrospective surveys of small numbers of patients have reported variable outcomes in terms of mortality and neurological sequelae including temporal lobe epilepsy (TLE).50 Long-term consequences of HHV-6 encephalitis post-HSCT in children may include a new syndrome, involving generalized epilepsy (as opposed to TLE in adults) together with cognitive regression and delayed intellectual development.52,53
Human herpesvirus 6B myelosuppression and allograft failure
Evidence for a causal association is moderate (Table 4). HHV-6B infects hematologic progenitor cells in vitro there- by reducing colony formation.54 Virus reactivation post- HSCT has been frequently associated with myelosup- pression and delayed engraftment, particularly involving platelets46,55,56 and also allograft failure.57,58
• If there is failed engraftment, blood or bone marrow should be tested for HHV-6B DNA.
• Other likely infectious or non-infectious causes must be excluded.
• CIHHV-6 in donor and recipient should be excluded.
Other end-organ diseases
Evidence for a causal association of HHV-6 with other disease post-HSCT is weak (Table 4). Viral DNA in tissue is not diagnostic as it may reflect HHV-6 DNAemia or inflammation with consequent infiltrating HHV-6 infect- ed lymphocytes.
Pneumonitis remains a leading cause of morbidity and mortality post-HSCT, and HHV-6 has been implicated as a potential cause. 59 Studies using heterogeneous popula- tions and methods, including patients with hematologic malignancies with and without HSCT, have produced variable results.60-62 A recent study applied molecular test- ing for 28 pathogens in bronchoalveolar lavage samples
haematologica | 2019; 104(11)
2159