HHV-6 is becoming increasingly recognized as emerging CNS pathogen, nevertheless, it has never been investigated in Sudan and very little is known in sub-Saharan Africa and the meningitis belt. In this study, HHV-6 DNA was identified in CSF of 2.6% of pediatric patients with suspected meningitis. Close findings of 1.8% (27/1,482) and 1.5% (1/65) were reported from New York, USA [21] and Southern Iran [24], respectively.
Another study in New York [11] revealed significantly higher prevalence of 40% (14/35) in patients with CNS infections who tested negative for other CNS pathogens. Among well-defined groups in our study, the prevalence was also high accounting for 33% out of 40 cases with proven infectious meningitis (Table 3). Unlike Yao [11] approach, we did not exclude cases with other detected CNS pathogens because of possible co-infections, as frequently reported [21, 25–27]. In fact, we were able to identify mixed microbial infections in three cases.
Primary HHV-6 infection almost invariably occurs in the first 2 years of life [5, 28, 29], but rare cases of CNS infections, presumed due to HHV-6 reactivation, have been reported in immunocompetent older children and adults [22]. Among 24 HHV-6 positive patients in Tavakoli study [21], 42% were infants ≤ 3 years and 12.5% were teenagers ≤ 17 years. Out of our 13 HHV-6 positive cases; 92% were infants ≤ 2.3 years and 8% was a 15 years old. The ratio of males to females was 1:1 which agrees with Tavakoli’s findings of 1.3:1.1.
Clinical signs and symptoms in HHV-6 meningitis are not specific [30]. Tavakoli [21] reported fever in 71%, altered mental status in 67%, headache in 29% and seizures in 33%. Other reported symptoms were muscle weakness, muscle pain and stiff neck; which are general symptoms for meningitis or encephalitis. We report fever and vomiting in 100% of HHV-6 positive cases, seizures in 86%, chills in 14.3% and stiff neck in 14.3%. None of our patients developed skin rash being the only specific -but rare- symptom in HHV-6 meningitis [30]; other studies [21, 24] concur.
Normal CSF glucose with normal or elevated proteins is the usual finding in viral CNS infections [17], as found in this study. Unfortunately, chemical analysis of the CSF is ineffective in case of viral infections; however, its significance is in distinguishing bacterial from aseptic aetiologies which is crucial preliminary step in deciding adequate treatment. CSF leukocytosis was seen in 33%. Increased numbers of CSF leukocytes (˃5cell/mm3) indicates inflammation, however, normal cellular counts do not rule out viral aetiologies. In fact, normal CSF cellular counts in patients with proven CNS infections were frequently reported [17–20]. Normal CSF profile was reported in 25% of HHV-6 positive cases by Tavakoli [21], accordingly, HHV-6 testing should not be limited to patients with abnormal CSF profiles.
Thermal cycle (Ct) is defined as the number of cycles required for the fluorescent signal to cross the threshold (exceed background level). Median Ct for our positive cases was 38 (range: 31.9–40.8). Tavakoli reported Ct values range 25.03 to 39.92. In quantitative real-time PCR, Ct values inversely correlate with viral loads, therefore, low Ct value indicates high viral load and vice versa. Our Ct values and viral loads were found to be significantly (p = 0.029) inversely correlated (-0.6, 95% CI:-0.9 to -0.1) indicating significant variation of viral loads among our patients. Substantial variation among viral loads in patients was also reported [21].
The phenomenon of HHV-6 chromosomal integration is in debate; while some [11–13] consider it an easily identifiable condition based on the presence or absence of nucleus containing blood cells in different body compartments, Ward [16] believes the few leukocytes that are usually present in normal CSF can reveal positive viral DNA in case of HHV-6 integration. In here, CSF was clear with no observed cells in most cases (86%), while single case (14%) showed increased CSF leukocytes. Ward [16] elaborated that in order to identify a condition as chromosomal integration; viral load should be high while low virus copies would indicate infection. The average concentration of CSF HHV-6 DNA in 9 children with primary infection (2.4 log10 copies/ml) was significantly lower than that of 21 patients with viral chromosomal integration (4.0 log10 copies/ml) in Ward study. In ours, the median CSF virus concentration was 1 × 105 copies/ml with minimum and maximum virus concentrations of 12 × 103 and 18 × 106 copies/ml. While Ward [16] recommended identifying low virus copies (≤ 103) as an acute HHV-6 infection and high virus copies (≥ 104) as chromosomal integration, Collot [31] identified viral integration in approximate concentrations of 103 to 106 copies/ml. We identified viral concentrations as low as 104 and as high as 107. Others [11, 21] reported high CSF viral loads in patients with HHV-6 CNS infections. Some authors [12, 13] insist that chromosomal integration is a rare condition. Accordingly, we assume the detected HHV-6 DNA in our mostly cell-free CSF specimens is more likely to be from free replicating virus than from chromosomally integrated virus.
Febrile seizures appear to be caused by primary HHV-6 infection in infants with incidence of 13% in the United States [4]. Knowing that febrile seizures and vomiting were dominant symptoms among our population and the most frequent age group was children ≤ 2.3 years, therefore further supporting our assumption. Despite this, and for the sake of scientific relevance, we are not ruling out the possibility of integration among our identified cases. For this reason, studies to identify the prevalence of HHV-6 integration among healthy Sudanese population are warranted.