Varicella zoster and fever rash surveillance in Lao People’s Democratic Republic

All protocols were approved by the Lao National Ethics Committee for Health Research (NECHR). Written informed consent was obtained from adult participants and the parents or legal guardians of children and infants. Additional file 1: Figure S1 shows an overview of the methodology.

The National Centre for Laboratory and Epidemiology (NCLE) was notified of a suspected measles outbreak in a remote region of Pakseng district, Luang Prabang Province in January 2015. The investigation team included provincial health staff, epidemiologists and laboratory scientists from NCLE, WHO, and Institut Pasteur du Laos (IPL). Clinical diagnosis was performed by the local physicians and 29 blood samples were collected from children and 2 from adults who had rash within the previous 28 days. Measles and rubella IgM testing was performed by NCLE as described previously [12]. Samples were further tested at IPL for anti-varicella IgM antibodies (anti-VZV IgM ELISA, Euroimmun). In addition to the sera, throat swabs were collected from 42 cases and investigated by PCR and sequencing as described before [13].

Samples from 3139 participants aged 9 months to 46 years, living in different Lao Provinces, were collected between 2011 and 2015. Infants (n = 207; mean age 9.4 ± 0.7 months) were recruited when coming for routine measles vaccination to a central hospital in Vientiane and to Luang Prabang provincial hospital in 2011 and 2012 (NECHR ref. 001/2012). Pre-school children (n = 803; mean age 2.0 ± 0.9 years) were randomly selected from 26 villages in Kuan and Xamtai districts in Houaphan province in 2013 as described elsewhere [14] (NECHR ref. 732/2013). School children aged between 5 and 19 years (n = 1731; mean age 13.1 ± 3.7 years) were recruited in four different provinces (Vientiane Capital, Savannakhet, Bolikhamxay and Luang Prabang) during a measles and rubella (M/R) supplementary immunisation activity in 2011/2012 (NECHR ref. 001/2012). Pregnant women (n = 398; mean age 26.3 ± 5.4 years) were recruited from antenatal clinics in Vientiane and Luang Prabang in 2012 (NECHR ref. 001/2012). IgG antibodies were detected using anti-varicella IgG ELISA kits according to manufacturer’s instructions (anti-IgG VZV ELISA, Euroimmun).

Between January and December 2015, a total of 511 children presenting with F/R were recruited at the Out-Patients-Department of the Children Hospital, Vientiane, which served here as a sentinel site for active M/R surveillance in the framework of F/R surveillance. Following parental informed consent, throat swabs, oral fluid and blood samples were obtained. IgM antibodies against measles, rubella, parvovirus B19, adenovirus and VZV were measured using Euroimmun IgM ELISA kits.

DNA and RNA were extracted using Nucleospin kits extraction kits (Macherey Nagel). F/R multiplex kits (Fast Track Diagnostics) were used to detect parvovirus B19 and HHV6 and 7 DNA in serum, and measles and enterovirus RNA in throat swabs. Because HHV6 and HHV7 detection in throat swabs is likely to be due to chronic infection and detection in blood is more indicative of acute infection [15, 16], we chose serum for detection of these herpesviruses by PCR.

In participants who were IgM borderline or positive, rubella RNA and VZV DNA were detected from throat swabs and oral fluid, respectively, using published primers [13, 17]. Sequencing was attempted for all of the above samples that were positive by PCR for measles, rubella, parvovirus B19 and VZV. Sequences of open reading frames (ORF) 22 and ORF 62 of the VZV genome were analyzed as previously described [13], using the revised VZV nomenclature [18] . Rubella sequences of the 739 nucleotide region of the E1 gene were analyzed together with reference sequences [19] using MEGA 7 [20].

Between March 2012 and September 2013, 108 plasma samples were collected in Vientiane Capital from dengue suspected cases. All these adults and children presented with F/R disease as part of the Lao National dengue surveillance network had no laboratory evidence for an active dengue virus infection. These dengue negative samples were tested for IgM antibodies against measles, rubella, VZV and parvovirus B19 by ELISA (Euroimmun).

A total of 35 sera were collected during a F/R outbreak in Houitone village, Pakseng district, clinically suspected to be measles, were negative for both measles and rubella IgM but 30/34 (88.2%; excluding 1 equivocal) of them were varicella IgM positive. The median age of cases was 6 years (interquartile range, IQR (4.0–7.5)) with 50% females. Clinical signs and epidemiology were compatible with varicella diagnosis [12]. Three of 42 throat swabs (including those from the above 35 patients) were PCR positive and sequencing showed that all viruses in this group of patients belong to clade 2 (Table 1).

Anti-VZV IgG was determined in a total of 3139 participants (42.2% male) including 36.9% from urban areas (Vientiane Capital and Luang Prabang) and 63.1% from semi-urban areas (Bolikhamxay, Houaphan and Savannakhet provinces). Overall, 55.6% had IgG antibodies against VZV. The seroprevalence increased with age in both urban and semi-urban areas from 2.4% in infants to 91.5% in women of child bearing age and was significantly correlated with age (r² = 0.94, p = 0.017). The largest increase in seroprevalence was between the 1 to 5 age bracket (9.3%) and the 6 to 7 year olds (59.5%; Fig. 1). For all age groups, equivocal samples accounted for between 1.5 and 2.5% and were excluded from the positive results. There were no significant differences between gender, urban and semi-urban areas for any of the cohorts or age groups.

Following exclusion of 9 participants for whom there was not enough sample for testing and 4 for whom the age was unknown, 498 patients were included in the final analyses. The average age of the children was 2.5 years with a range of 0 to 15.4 years and 45% of the children were female. The majority of F/R cases (88.2%) occurred in children between 0 and 5 years of age. 28.6% (146/511) of participants had confirmed viral etiology (Table 2).

Adenovirus IgM was the most commonly detected (7.8%), with 17 (43.6%) co-infections or with persisting IgM from a past infection.

VZV IgM was detected in 20 subjects, two thirds of which were below the age of 5 (Table 2). In addition, 10 samples were “equivocal” for VZV IgM. DNA was extracted from oral fluid samples of the 30 participants with positive or equivocal IgM results. Sequencing was attempted for the 19 PCR positive samples. Sequence analyses of the ORFs 22 and 62 showed that 2 viruses belong to clade 5 and 15 to clade 2. Two sequences could not be assigned to specific clades, because of wobbles at important positions (Table 1). Measles IgM was detected in 12 (2.5%) patients, two of whom had positive throat swabs in the diagnostic PCR (Table 2), but not in the sequencing PCR. Rubella IgM was detected in 25 (5.0%) samples, two with PCR-positive throat swabs, one of which was assigned to the genotype 1a vaccine strain. No information about the vaccination status of this patient was available. Parvovirus B19 IgM was detected in 4 (0.8%) samples, with PCR detectable DNA in one of these sera. No sequence was obtained. HHV6 and HHV7 DNA were detected respectively in 38 (7.6%) and 11 (2.2%) sera, all but one in children between 0 and 5 years of age. Only one throat swab was positive for enterovirus by PCR (Table 2). Overall, 20 samples were IgM positive for more than one virus. The number of samples tested was too low to detect any seasonal patterns.

A total of 108 cases with F/R, that were suspected to have dengue, but were not confirmed, were included in this part of the study. The median age of the participants was 20 years with a range of 0 to 60 years. In this cohort IgM antibodies were detected against parvovirus B19 (4; 3.7%), measles (3; 2.8%), rubella (22; 20.4%) and VZV (6/100; 6%). Most IgM positive cases (31/35; 88.6%) were above the age of 10 years.