- Email: [email protected]
High prevalence of mumps in Lao People’s Democratic Republic
ORIGINAL ARTICLE
VIROLOGY
High prevalence of mumps in Lao People’s Democratic Republic J. M. H€ ubschen1,2, K. Vilivong1, C. Souvannaso1, A. P. Black1,2, N. L€ utteke1,2, B. Samountry3, V. Phongsavath4, B. Khamphaphongphane5, J. Denny6, C. Sayyavong7, G. K. S. Woo8, K. Sengsaya5, A. Sausy1,2, P. Vongphrachanh5, P. Jutavijittum9, D. Phonekeo5,10 and C. P. Muller1,2 1) LaoLuxLab, Institut Pasteur du Lao PDR, Vientiane, Lao PDR, 2) Institute of Immunology, Centre de Recherche Public de la Sante/Laboratoire National de Sante, Luxembourg City, Luxembourg, 3) Department of Pathology, Faculty of Medicine, University of Health Sciences, Vientiane, Lao PDR, 4) Section of Laboratory, Luangprabang Provincial Hospital, Luangprabang, 5) National Center for Laboratory and Epidemiology, 6) WHO Lao Country Office, Vientiane, 7) Vientiane Capital Health Office, Vientiane, Lao PDR, 8) Public Health Laboratory Services Branch, Department of Health, Centre for Health Protection, Hong Kong SAR, China, 9) Department of Pathology, Faculty of Medicine, Chiang Mai University, Chiang Mai, Thailand and 10) Institut Pasteur du Lao PDR, Vientiane, Lao PDR
Abstract In the Lao People’s Democratic Republic (PDR), mumps is not a notifiable disease and mumps vaccine is currently not included in the routine childhood immunization programme. In order to assess the burden of disease, we investigated the seroprevalence of mumps-specific IgG antibodies across four provinces. In addition, we genetically characterized mumps viruses from the past 3 years from several outbreaks and single cases. Blood and/or throat swabs from suspected cases were investigated for specific IgM antibodies or viral RNA. Mumps cases occurred between March and November in 2011–2013 and 5- to 15-year-olds were most affected. Four sequences from an outbreak in the north of Lao PDR in 2011 were identical and belonged to genotype G. Eight sequences from two outbreaks and two individual cases from 2012 and 2013 belonged to genotype J. In addition, sera collected from 2379 healthy infants and school pupils aged between 9 months and 19 years and from pregnant women aged between 16 and 46 years were investigated for mumps-specific IgG. Overall, 58.2% were positive, 39.5% were negative and the remaining 2.3% were equivocal. The seropositivity increased with age, with the lowest percentage found in <1-year-old infants (9.1%) and the highest in the cohort of pregnant women (69.2%). More female subjects than male subjects were seropositive (60.4 vs. 54.9%). There were some differences between the locations. Mumps should be a notifiable disease in Lao PDR in order to get more accurate case numbers and cost estimates for public health-care, and vaccination of children and high-risk groups should be considered. Keywords: Genotype, Lao PDR, mumps, outbreak, seroprevalence Original Submission: 28 October 2013; Revised Submission: 24 January 2014; Accepted: 31 January 2014
Editor: E. Bottieau Article published online: 11 February 2014 Clin Microbiol Infect 2014; 20: O664–O671 10.1111/1469-0691.12586
Corresponding author: C. P. Muller, Institute of Immunology, Centre de Recherche Public de la Sante / Laboratoire National de Sante, 20A, rue Auguste Lumiere, L-1950 Luxembourg City, Luxembourg E-mail: [email protected]
Introduction Mumps is a vaccine preventable disease caused by mumps virus (family, Paramyxoviridae; genus, Rubulavirus) with a
subclinical course in up to one-third of all cases. Common manifestations include parotitis and respiratory symptoms, and orchitis in postpubertal male subjects [1]. Laboratory diagnosis relies on detection of specific IgM antibodies or mumps virus RNA. Based on 316 nucleotides covering the small hydrophobic (SH) gene region, 12 different genotypes of mumps virus have been proposed with genotypes C, F, G and H predominating in Asia [1]. In 2012, 120 out of 194 WHO member states (62%) were using mumps vaccine in their national immunization schedule [2]. In the Lao People’s Democratic Republic (PDR), mumps is not a notifiable disease
ª2014 The Authors Clinical Microbiology and Infection ª2014 European Society of Clinical Microbiology and Infectious Diseases
CMI
H€ ubschen et al.
and mumps vaccine is currently not included in the routine childhood immunization programme. Except for 2011, when only 54 cases were recorded by WHO [3], no data exist. In order to assess the real burden of disease, we investigated the seroprevalence of mumps-specific IgG antibodies across four provinces. In addition, we genetically characterized mumps viruses from the past 3 years from several outbreaks and single cases brought to our attention by the National Center for Laboratory and Epidemiology (NCLE) and medical doctors. Thus we provide here the first comprehensive information about mumps in Lao PDR as a guide for future public health decisions.
Materials and Methods Data concerning the 2011 and 2013 outbreaks and the 2012 outbreaks in Saravane and Champasack province were collected by NCLE, in collaboration with the WHO Lao Country Office for 2011 data. The data for the 2012 Vientiane outbreak were obtained from the Vientiane Capital Health Office. Data for the sporadic cases were collected by the LaoLuxLab of the Institut Pasteur du Lao PDR. Blood and/or throat swabs were taken during the outbreaks from some or all clinical cases and from the sporadic cases and were investigated for mumps-specific IgM antibodies (Euroimmun, L€ubeck, Germany) or viral RNA, respectively. For RNA positive samples, the 316 nucleotide SH gene region was amplified and sequenced and the data were used for phylogenetic analyses according to WHO guidelines [1] with MEGA 4 software [4]. The sequences were submitted to GenBank under accession numbers KF438181-KF438184 and HG423644-HG423651. Sera collected from healthy infants and school pupils aged between 9 months and 19 years and from pregnant women aged between 16 and 46 years were investigated for mumps-specific IgG antibodies using a commercially available ELISA kit (Euroimmun) according to the manufacturer’s instructions. Equivocal samples were retested and classified according to the second test result as positive, negative or equivocal. Before blood collection, informed consent was obtained from all participants or their parents. Data such as age, sex and location were recorded and used for the seroprevalence analyses. For the blood collection, four different locations were selected: Savannakhet city in the south (participants between 5 and 19 years), Luangprabang city in the north (participants aged between 9 months and 19 years and pregnant women), Vientiane city (participants aged <1 year and pregnant women) and Borikhamxay province (participants aged between 5 and 19 years), both in the
High prevalence of mumps in Laos
O665
centre of the country. Fig. 1 was constructed using ArcView 9.3.1 software (Esri, California, USA). Free spatial data were downloaded from DIVA-GIS (http://www.diva-gis.org/Data). Ethical approval was given by the ethics committee of Lao PDR (NECHR 001/2011).
Results Outbreak 2011 Oudomxay province
Between 10 March and 23 April 2011, a total of 61 suspected cases (based on sore throat or cervical/mandibular lymphadenopathy or parotid swelling), including three deaths, were investigated in Mokkha village, located in the Hoon district in Oudomxay province in Northern Lao PDR (Fig. 1). The majority of the patients were female (n = 32, 52.5%) and <16 years old (n = 56, 91.8%). The most affected age group was the 5–9 year olds (n = 26, 42.6%), followed by the 10– 15 year olds (n = 17, 27.9%) and the 1–4 year olds (n = 13, 21.3%). The highest incidence of disease was observed in week 14 with 28 new mumps cases (Fig. 2a). Two 8-year-old girls and an 11-year-old boy died during week 12. Besides lymphadenopathy, headache, malaise or anorexia, also dizziness, lethargy, muscle pain or paralysis of the extremities were observed. For two of the three deceased patients, encephalitis was diagnosed as the cause of death. Mumps-specific IgM antibodies were detected in five out of six sera and mumps virus RNA was detected in four out of ten throat swabs. Thus, nine of the 16 patients from whom clinical specimens were available (56.3%), were laboratory-confirmed mumps cases. Interestingly, Coxsackievirus A16 was detected in one of the mumps PCR-positive samples and in one of the six mumps virus-negative throat swabs (both cases with disease onset in week 16). Rhinovirus was found in another mumps PCR-negative throat swab (case with disease onset also in week 16). Outbreak 2012 Vientiane capital
Between 20 July and 2 August 2012, 12 mumps cases were recorded in Donexingxou village in the Sikhottabong district of Vientiane capital (Fig. 1). Most patients were male (n = 10, 83.3%) and the ages ranged from 4 to 19 years (median 10.5 years). The most affected age group was the 10–15 year olds (n = 7, 58.3%), followed by the 5–9 year olds (n = 3, 25.0%). Most cases occurred in week 31 (n = 7, 58.3%), followed by week 30 (n = 3, 25.0%) and week 29 (n = 2, 16.7%) (Fig. 2a). All patients were reported to have fever, parotitis and headache and five patients were admitted to the hospital. Throat swabs were collected from all hospitalized patients and four of the samples (80.0%) were positive for the
ª2014 The Authors Clinical Microbiology and Infection ª2014 European Society of Clinical Microbiology and Infectious Diseases, CMI, 20, O664–O671
O666
CMI
Clinical Microbiology and Infection, Volume 20 Number 10, October 2014
FIG. 1. Map of Lao PDR showing the different sample collection sites. PV stands for province and PF for prefecture.
mumps virus-specific PCR. All four patients were also IgM positive. Outbreak 2012 Saravane province
A total of 32 mumps cases were recorded between 20 October and 29 November 2012 in Saneumnai village in the Lao-ngam district in Saravane province (Fig. 1). All patients were reported to have fever, a sore throat and parotitis and some also suffered from dizziness and fatigue. The patients were between 1 and 45 years old (mean 12.1 years) and slightly more male (n = 17, 53.1%) than female subjects were affected. None of the four collected throat swabs was PCR positive, but three of the four sera from the same patients were mumps IgM antibody positive. Outbreak 2012 Champasack province
Nine mumps cases were observed between 21 and 29 November 2012 in Kadun village in the Mounlapamok
district (n = 1) and in Donhee village in the Champasack district (n = 8) in Champasack province (Fig. 1). All patients had parotitis and some reported fever and headaches. Eight of the patients were pupils aged between 8 and 12 years (n = 7 female, 87.5%) and one was a 45-year-old male teacher. None of the nine throat swabs were PCR positive, but seven out of nine sera were mumps IgM antibody positive. Outbreak 2013 Attapeu province
Between 30 April and 4 May 2013 four mumps cases were noticed among pupils from Nong-Hin village in the Sanamxai district in Attapeu province (Figs 1 and 2a). The children were between 7 and 15 years old and two were male and two female. All patients were hospitalized on 6 May with fever, headache, sore throat and parotitis. Throat swabs were collected from all children and two of them were positive by mumps virus-specific PCR.
ª2014 The Authors Clinical Microbiology and Infection ª2014 European Society of Clinical Microbiology and Infectious Diseases, CMI, 20, O664–O671
CMI
H€ ubschen et al.
High prevalence of mumps in Laos
O667
(a) 30 2011 2012
25
Number of cases
2013 20
15
10
5
0
10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48
Epidemiological week (b) 30 F
Number of cases
25
FIG. 2. Number between
2011
of and
mumps 2013
cases (a)
by
epidemiological week and (b) according
M
20 15 10 5 0
1–4
5–9
10–15
> 15
Age in years
to age and gender. Individual case 2012 Vientiane capital
Phylogenetic analysis of the mumps virus sequences
On 11 September 2012 a 24-year-old male Lao citizen from Vientiane capital was clinically diagnosed as having mumps. A throat swab was collected, which was positive by PCR.
From all mumps virus-positive throat swabs, the 316 nucleotide region of the SH gene was amplified and sequenced. Phylogenetic analysis of the sequences identified four distinct virus variants (Fig. 3). The sequences from the 2011 outbreak were all identical and belonged to genotype G. The closest relatives identified by BLAST and phylogenetic analyses were sequences found in Japan between 1999 and 2005 (e.g. MuV-Yokohama.JPN/05-1), which differed by five nucleotides in the 316 nucleotide region of the SH gene. The sequences from the 2012 and 2013 outbreaks were also all identical, but belonged to genotype J. Strains found in Thailand in 2008 (e.g. MuVs/Phang-nga.THA/4.08-14) were the closest relatives and differed by a single nucleotide in the SH gene region (Fig. 3). The individual case from 2012 also belonged to genotype J, but was very distinct from the 2012 and 2013 outbreak strains (12 mutations, genetic distance 3.95%), while the single case from 2013 differed by only one nucleotide from the 2012 outbreak
Individual case 2013 Vientiane capital
On 28 June 2013 a 15-year-old girl from Phakhao village, Vientiane capital, fell ill with mumps-like symptoms. A throat swab collected on 1 July was positive by PCR. Overall summary of laboratory investigation and confirmation
A total of 120 suspected mumps cases observed between 2011 and 2013 were brought to our attention. Sera and/or throat swabs for laboratory investigation were available from one-third of them (n = 40). Laboratory confirmation was achieved for 27 patients, which is 67.5% of all patients investigated and 22.5% of all suspected cases.
ª2014 The Authors Clinical Microbiology and Infection ª2014 European Society of Clinical Microbiology and Infectious Diseases, CMI, 20, O664–O671
O668
CMI
Clinical Microbiology and Infection, Volume 20 Number 10, October 2014
MuVs/Ban Mokka.LAO/15.11 MuVs/Ban Mokka.LAO/16.11/2 MuVs/Ban Mokka.LAO/16.11/3 95 Genotype G MuVs/Ban Mokka.LAO/16.11/1 95 MuV-Yokohama.JPN/05-1 AB699705 MuVi/Gloucester.GBR/32.96-G MuVi/Sheffield.GBR/1.05-G MuVi/London.GBR/3.02(UK02-19) MuVi/Bedford.GBR/0.89-H MuVi/Ulaanbaatar.MNG/22.09-H 87 89 MuVi/RW154.USA/0.70s-K MuVi/Stockholm.SWE/26.83-K MuVi/Tokyo.JPN/0.93(MP93-N) MuVi/Ge9.DEU/0.77-D MuVi/Nottingham.GBR/19.04-D 96 99 MuVi/Zagreb.HRV/39.98-C MuVi/Stockholm.SWE/46.84-C MuVi/Fukuoka.JPN/41.00-L 100 MuVi/Tokyo.JPN/6.01-L 100 MuVi/Shandong.CHN/4.05-F MuVi/Zhejiang.CHN/11.06/1-F MuVi/Taylor.GBR/0.50s(Tay/UK50s) MuVi/Urabe.JPN/0.67-B 99 MuVi/Himeji.JPN/24.00-B MuVi/Vector.RUS/0.53VAC-N 99 MuVi/L-Zagreb.HRV/0.71VAC-N MuVi/Akita.JPN/42.93-I MuVi/Dg1062.KOR/46.98-I 100 100 MuVi/Boston.USA/0.45-A MuVi/Pennsylvannia.USA/13.63VAC-A MuVi/Sapporo.JPN/12.00-J MuVi/Leeds.GBR/9.04-J 100 MuVs/Vientiane.LAO/37.12 91 MuVs/Phang-nga.THA/4.08-14 EU497657 MuVs/Vientiane.LAO/26.13 Genotype J MuVs/Vientiane.LAO/31.12/2 MuVs/Sanamxai.LAO/18.13/2 MuVs/Vientiane.LAO/31.12/1 MuVs/Sanamxai.LAO/18.13/1 MuVs/Vientiane.LAO/31.12/3 MuVs/Vientiane.LAO/31.12/4
98
FIG. 3. Phylogenetic tree based on 316 nucleotides of the mumps virus SH gene region and the Kimura 2-parameter model and the neighbour-joining algorithm. Light grey dots mark the sequences obtained from the mumps outbreak in 2011, grey triangles highlight sequences from the 2012 outbreak and black diamonds show sequences from the 2013 outbreak. The grey square and the black triangle mark the individual cases from 2012 and 2013, respectively. White dots indicate the closest relatives of the sequences from
0.01
Lao PDR identified by BLAST.
strains. The closest relatives identified by BLAST were also the mumps strains found in Thailand in 2008, which differed by 13 nucleotides from the 2012 strain and by two nucleotides from the 2013 strain. Seroprevalence data
A total of 2379 sera collected from four different locations (Luangprabang, n = 896; Borikhamxay, n = 599; Savannakhet, n = 587; Vientiane, n = 297) were included in the seroprevalence study (Fig. 1). Overall, 58.2% of the participants were positive for mumps-specific IgG antibodies, 39.5% were negative and the remaining 2.3% were equivocal (Table 1). The highest seropositivity of 61.2% was found in Luangprabang, the lowest in Vientiane (52.5%). Of the two locations sampled in Borikhamxay province, Paksane had a considerably higher overall seropositivity than the more rural
Pakkading (68.3% of 303 samples compared with 43.2% of 296 samples). As expected, the seropositivity increased with age, with the lowest percentage found in the <1-year-old infants (9.1%) and the highest in the cohort of pregnant women (69.2%) (Table 1). The <1-year-old infants and the pregnant women from Vientiane had higher seropositivity rates than the corresponding cohorts from Luangprabang (14.3 vs. 4.0% and 71.4 vs. 67.0%). Among the 5–9 and the 10–14-year-old children, the highest seropositivity was observed in Luangprabang (60.5 and 70.0%) and the lowest in Borikhamxay (42.7 and 61.5%). The highest percentage of seropositives among the 15–19-year-olds was found in Luangprabang and the lowest in Savannakhet (75.3 vs. 62.3%, Table 1). Overall, more female than male subjects were seropositive (60.4 vs. 54.9%, p <0.01), but without the pregnant women, who included many
TABLE 1. Seroprevalence of mumps-specific IgG antibodies according to geographical location and age group/cohort Total
Luangprabang Borikhamxay Savannakhet Vientiane Total
N
Positive (%)
Negative (%)
Equivocal (%)
Infants Positive (%)
5–9 year olds Positive (%)
10–14 year olds Positive (%)
15–19 year olds Positive (%)
Pregnant women Positive (%)
896 599 587 297 2379
548 335 346 156 1385
329 244 230 136 939
19 20 11 5 55
4/100 (4.0) / / 14/98 (14.3) 18/198 (9.1)
112/185 85/199 78/166 / 275/550
149/213 123/200 141/217 / 413/630
149/198 127/200 127/204 / 403/602
134/200 (67.0) / / 142/199 (71.4) 276/399 (69.2)
(61.2) (55.9) (58.9) (52.5) (58.2)
(36.7) (40.7) (39.2) (45.8) (39.5)
(2.1) (3.3) (1.9) (1.7) (2.3)
(60.5) (42.7) (47.0) (50.0)
(70.0) (61.5) (65.0) (65.6)
ª2014 The Authors Clinical Microbiology and Infection ª2014 European Society of Clinical Microbiology and Infectious Diseases, CMI, 20, O664–O671
(75.3) (63.5) (62.3) (66.9)
CMI
H€ ubschen et al.
TABLE 2. Seroprevalence of mumps-specific IgG antibodies according to geographical location, cohort and gender
Location
Cohort
Luangprabang
Infants/pupils Pregnant women Borikhamxay Pupils Savannakhet Pupils Vientiane Infants Pregnant women All All All without pregnant women
Female
Male
N
Positive (%)
N
Positive (%)
364 200 290 335 51 199 1439 1040
219 134 169 195 10 142 869 593
332 / 309 252 47 / 940 940
195 / 166 151 4 / 516 516
(60.2) (67.0) (58.3) (58.2) (19.6) (71.4) (60.4) (57.0)
(58.7) (53.7) (59.9) (8.5) (54.9) (54.9)
participants older than 19 years, this difference was no longer statistically significant (57.0% vs. 54.9%, p >0.05, Table 2). This was also the case for all locations except Savannakhet, where a slightly higher percentage of positives was found among male subjects (59.9 vs. 58.2%). The biggest gender difference was observed in infants from Vientiane (8.5% seropositivity in male infants and 19.6% in female infants, Table 2), but overall only 98 children were recruited.
Discussion In the present manuscript, we investigate mumps seroprevalence amongst almost 2400 infants, school children and pregnant women from five different locations in four provinces and describe all mumps cases and outbreaks between 2011 and 2013 brought to our attention. While northern, central and southern regions of Lao PDR are represented both in the seroprevalence study and the case investigations, samples were mostly from urban settings. The high seroprevalence detected during this study indicates that Lao PDR sustains many more cases than were reported here. Nevertheless, our study is one of the very few comprehensive studies on mumps in a specific country and exceptional in the region because it combines seroprevalence and case investigation data. Two different genotypes of mumps virus were found: genotype G in the north (2011) and genotype J in the central and the southern parts (2012, 2013). As only 12 sequences were obtained in total, it is unclear to what extent our findings are representative for the country. However, genotype J was found at four different time-points with some sequence variation and except for the very distinct strain from 2012 all strains were very similar to viruses detected in 2008 in Thailand. This strongly suggests that this genotype is endemic in the region. A recent report from Nan Province of Thailand close to the border with Laos identified genotype J during a mumps outbreak in 2010 [5], but the sequences are not available for comparison. The individual patient from 2012
High prevalence of mumps in Laos
O669
declared having been in places frequented by tourists during the incubation period. Thus this distinct genotype J virus could either be endemic or could have been introduced directly from a location outside of Lao PDR. Genotype J has been detected in several countries, including Ireland [6], the UK [7,8], Japan [9] and Australia [10], but to date the sequences from Thailand are the most similar to the distinct strain identified in Lao PDR in 2012. Whether genotype G continuously circulates in northern Lao PDR requires further investigation. As the genotype G strains differed by five nucleotides from the most similar sequences identified by BLAST, it remains open whether this variant is endemic in Lao PDR or whether it was introduced from outside. While no genotype information is available from Myanmar, genotype F was reported from across the border with China [11]. During a recent outbreak in Cambodia, genotype K was detected [12] and the same genotype was presumably introduced from Vietnam to Canada (GenBank accession number KF212192), but so far no genotype G was reported from the neighbouring countries. Throat swabs for genotype determination were also collected during mumps outbreaks in the very south of the country, namely in Saravane and Champasack provinces, but the swabs were collected very late after onset of disease (7 days in one case, between 11 and 19 days in all others). This late collection is probably the main reason why all samples were PCR negative because viral RNA is known to be present during disease onset and then to decline over 8– 10 days [1]. Mumps may be found throughout the year in hot climates [13]. In Lao PDR mumps cases occurred between the beginning of March and end of November. In contrast to the other outbreaks, the one from 2011 in the north showed a large number of cases, the laboratory-confirmed co-circulation of two different enteroviruses and three deaths. The co-circulation of different viruses, the high level of malnutrition in the area and poor access to healthcare may potentially explain disease severity and fatal outcome. Two of the deceased patients reportedly died from encephalitis. Encephalitis is a very rare complication of mumps [1] and may also occur after enterovirus infections [14,15], but no samples for laboratory confirmation were available from the deceased patients. While in Europe mumps has become a disease of adolescents and young adults [16–20], most mumps cases in Lao PDR were found among 5–15-year-old children (Fig. 2b), which corresponds more with the ‘classical’ age range for an unvaccinated population [13]. In fact, our seroprevalence data show that there is a considerable increase in seropositivity between the age groups 5–9 and 10–14 years (Savannakhet
ª2014 The Authors Clinical Microbiology and Infection ª2014 European Society of Clinical Microbiology and Infectious Diseases, CMI, 20, O664–O671
O670
Clinical Microbiology and Infection, Volume 20 Number 10, October 2014
47.0 and 65.0%, Luangprabang 60.5 and 70.0%, Borikhamxay 42.7 and 61.5%), indicating that many become infected as pupils. In other age-stratified seroprevalence studies in unvaccinated populations, seropositivity was often even higher than in Lao PDR [21–24]. As about 30% of pregnant women are still susceptible to infection, mumps may be contracted at a wide age-range in Lao PDR. This is of concern because mumps may cause foetal loss during the first trimester of pregnancy and adults are more prone to severe disease complications such as meningitis or orchitis [13]. The seroprevalence study identified some differences between the locations. The overall seropositivity rate for Vientiane is largely biased due to the high proportion of infants (98/297, 33.0%), because both infants and pregnant women showed considerably higher rates in Vientiane than in Luangprabang, possibly because of the higher population density in the capital. This may also explain the differences between Paksane and the more rural Pakkading, where considerable differences in seroprevalence are observed even in neighbouring districts within the same province. For all age groups of pupils, the highest seropositivity rates were detected in Luangprabang, a town highly frequented by foreign tourists. The mean and median ages within the age groups or cohorts were not very different and cannot explain differences in seropositivity between the locations. As mumps vaccination is currently not included in the national immunization schedule in Lao PDR, the overall seropositivity of 58% is due essentially to wild-type infection. Besides the currently most affected children between 5 and 15 years, adults with professional contact with these children are particularly at risk of infection. In neighbouring Thailand many thousands of mumps cases are reported each year despite vaccination (9901 and 7422 in 2011 and 2012, respectively) [3]. The high seropositivity rate found during the present study suggests that a considerable annual number of mumps cases must occur also in Lao PDR. Although under-reporting would be likely from remote places, mumps should be a notifiable disease in Lao PDR to get more accurate case numbers and cost estimates for public healthcare. In this context, the importance of laboratory-based surveillance will increase considerably and it will become imperative that appropriate specimens are collected in a timely manner to not only confirm clinically suspected cases, but also to obtain genotype information for molecular epidemiology. Recently, rubella vaccination has been introduced into the national immunization programme in combination with measles vaccine. The cost-benefit of the incremental cost of using the combined measles, mumps and rubella vaccine should be carefully assessed.
CMI
Acknowledgements The authors wish to thank Dr Paul Brey for facilitating the present study and the Ministry of Cooperation and the Centre de Recherche Public de la Sante in Luxembourg for financial support. We also thank Dr Jean Marie Hospied from the French Clinic in Vientiane and Dr Chansamone Vanulasy from Hospital 103, Vientiane, for providing material for laboratory investigations. For the preparation of Fig. 1, we are very grateful to Dr Anna L. Reye. Parts of the data obtained during the present study have been shown during oral presentations at the Francophone research meeting in Vientiane, Lao PDR, on 20 November 2012 and at the 5th National Health Research Forum 2013 in Vientiane, Lao PDR, on 15 October 2013.
Transparency Declaration The authors declare no conflicts of interest.
References 1. WHO. Mumps virus nomenclature update: 2012. Wkly Epidemiol Rec 2012; 22: 217–224. 2. WHO. Countries using mumps vaccine in national immunization schedule, 2012. Available: http://www.who.int/immunization_monitoring/ diseases/Mumps_map_schedule.JPG. Accessed September 30 2013. 3. WHO. Mumps reported cases, July 14 2013. Available: http://apps.who.int/ immunization_monitoring/globalsummary/timeseries/tsincidencemumps. html. Accessed August 28 2013. 4. Tamura K, Dudley J, Nei M, Kumar S. MEGA4: Molecular Evolutionary Genetics Analysis (MEGA) software version 4.0. Mol Biol Evol 2007; 24: 1596–1599. 5. Phanawadee M, Ieowongjaroen I, Makaroon J et al. Investigation of Mumps Outbreak in a Rural District, Thailand, June–October 2010. Weekly Epidemiological Surveillance Report. 2012;43(Supplement: March 2012):15–22. 6. Reid F, Hassan J, Irwin F, Waters A, Hall W, Connell J. Epidemiologic and diagnostic evaluation of a recent mumps outbreak using oral fluid samples. J Clin Virol 2008; 41: 134–137. 7. Cui A, Brown DW, Xu W, Jin L. Genetic variation in the HN and SH genes of mumps viruses: a comparison of strains from mumps cases with and without neurological symptoms. PLoS One 2013; 8: e61791. 8. Cui A, Myers R, Xu W, Jin L. Analysis of the genetic variability of the mumps SH gene in viruses circulating in the UK between 1996 and 2005. Infect Genet Evol 2009; 9: 71–80. 9. Inou Y, Nakayama T, Yoshida N et al. Molecular epidemiology of mumps virus in Japan and proposal of two new genotypes. J Med Virol 2004; 73: 97–104. 10. Bangor-Jones RD, Dowse GK, Giele CM, van Buynder PG, Hodge MM, Whitty MM. A prolonged mumps outbreak among highly vaccinated Aboriginal people in the Kimberley region of Western Australia. Med J Aust 2009; 191: 398–401. 11. Ma SH, Shi HJ, He CY, Chen JY, Yang HJ, Sun QM. [Genetic characteristics on the small hydrophobic protein and hemaggluti-
ª2014 The Authors Clinical Microbiology and Infection ª2014 European Society of Clinical Microbiology and Infectious Diseases, CMI, 20, O664–O671
CMI
12.
13. 14.
15.
16.
17.
H€ ubschen et al.
nin-neuraminidase genes of mumps virus in Yunnan province, China from 2007 to 2009]. Zhonghua Liu Xing Bing Xue Za Zhi 2011; 32: 370– 375. Tek S, Rin R, Muth S, Bun S, Chea N, Concepcion Roces M. Mumps Outbreak in Siem Pang district, Stung Treng Province, Cambodia, 2012. The 4th International Conference on Public Health among Greater Mekong Sub-regional Countries; Kunming, China, 2012. WHO. Mumps virus vaccines. Wkly Epidemiol Rec 2007; 82: 51–60. Joshi R, Mishra PK, Joshi D et al. Clinical presentation, etiology, and survival in adult acute encephalitis syndrome in rural Central India. Clin Neurol Neurosurg 2013; 115: 1753–1761. Koskiniemi M, Rantalaiho T, Piiparinen H et al. Infections of the central nervous system of suspected viral origin: a collaborative study from Finland. J Neurovirol 2001; 7: 400–408. Hukic M, Ravlija J, Dedeic Ljubovic A et al. Ongoing large mumps outbreak in the Federation of Bosnia and Herzegovina, Bosnia and Herzegovina, December 2010 to July 2011. Euro Surveill 2011;16. pii: 19959. Karagiannis I, van Lier A, van Binnendijk R et al. Mumps in a community with low vaccination coverage in the Netherlands. Euro Surveill 2008; 13. pii: 18901.
High prevalence of mumps in Laos
O671
18. Kuzmanovska G, Polozhani A, Mikik V et al. Mumps outbreak in the former Yugoslav Republic of Macedonia, January 2008-June 2009: epidemiology and control measures. Euro Surveill 2010; 15. pii: 19586. 19. Rajcevic S, Seguljev Z, Petrovic V et al. Ongoing mumps outbreak in Novi Sad, the autonomous province of Vojvodina, Serbia, January to April 2012. Euro Surveill 2012; 17. pii: 20169. 20. Walker J, Huc S, Sinka K, Tissington A, Oates K. Ongoing outbreak of mumps infection in Oban, Scotland, November 2010 to January 2011. Euro Surveill 2011; 16. pii: 19803. 21. Aksit S, Egemen A, Ozacar T, Kurugol Z. Mumps seroprevalence in an unvaccinated population in Izmir, Turkey. Acta Paediatr 2000; 89: 370– 371. 22. Gurgoze MK, Yilmaz E, Godekmerdan A et al. Seroprevalence of mumps, varicella and rubella antibodies in children 1-16 years of age in eastern Turkey. Turk J Pediatr 2006; 48: 185–188. 23. Kanbur NO, Derman O, Kutluk T. Age-specific mumps seroprevalence of an unvaccinated population of adolescents in Ankara, Turkey. Jpn J Infect Dis 2003; 56: 213–215. 24. Tolfvenstam T, Enbom M, Ghebrekidan H et al. Seroprevalence of viral childhood infections in Eritrea. J Clin Virol 2000; 16: 49–54.
ª2014 The Authors Clinical Microbiology and Infection ª2014 European Society of Clinical Microbiology and Infectious Diseases, CMI, 20, O664–O671
VIROLOGY
High prevalence of mumps in Lao People’s Democratic Republic J. M. H€ ubschen1,2, K. Vilivong1, C. Souvannaso1, A. P. Black1,2, N. L€ utteke1,2, B. Samountry3, V. Phongsavath4, B. Khamphaphongphane5, J. Denny6, C. Sayyavong7, G. K. S. Woo8, K. Sengsaya5, A. Sausy1,2, P. Vongphrachanh5, P. Jutavijittum9, D. Phonekeo5,10 and C. P. Muller1,2 1) LaoLuxLab, Institut Pasteur du Lao PDR, Vientiane, Lao PDR, 2) Institute of Immunology, Centre de Recherche Public de la Sante/Laboratoire National de Sante, Luxembourg City, Luxembourg, 3) Department of Pathology, Faculty of Medicine, University of Health Sciences, Vientiane, Lao PDR, 4) Section of Laboratory, Luangprabang Provincial Hospital, Luangprabang, 5) National Center for Laboratory and Epidemiology, 6) WHO Lao Country Office, Vientiane, 7) Vientiane Capital Health Office, Vientiane, Lao PDR, 8) Public Health Laboratory Services Branch, Department of Health, Centre for Health Protection, Hong Kong SAR, China, 9) Department of Pathology, Faculty of Medicine, Chiang Mai University, Chiang Mai, Thailand and 10) Institut Pasteur du Lao PDR, Vientiane, Lao PDR
Abstract In the Lao People’s Democratic Republic (PDR), mumps is not a notifiable disease and mumps vaccine is currently not included in the routine childhood immunization programme. In order to assess the burden of disease, we investigated the seroprevalence of mumps-specific IgG antibodies across four provinces. In addition, we genetically characterized mumps viruses from the past 3 years from several outbreaks and single cases. Blood and/or throat swabs from suspected cases were investigated for specific IgM antibodies or viral RNA. Mumps cases occurred between March and November in 2011–2013 and 5- to 15-year-olds were most affected. Four sequences from an outbreak in the north of Lao PDR in 2011 were identical and belonged to genotype G. Eight sequences from two outbreaks and two individual cases from 2012 and 2013 belonged to genotype J. In addition, sera collected from 2379 healthy infants and school pupils aged between 9 months and 19 years and from pregnant women aged between 16 and 46 years were investigated for mumps-specific IgG. Overall, 58.2% were positive, 39.5% were negative and the remaining 2.3% were equivocal. The seropositivity increased with age, with the lowest percentage found in <1-year-old infants (9.1%) and the highest in the cohort of pregnant women (69.2%). More female subjects than male subjects were seropositive (60.4 vs. 54.9%). There were some differences between the locations. Mumps should be a notifiable disease in Lao PDR in order to get more accurate case numbers and cost estimates for public health-care, and vaccination of children and high-risk groups should be considered. Keywords: Genotype, Lao PDR, mumps, outbreak, seroprevalence Original Submission: 28 October 2013; Revised Submission: 24 January 2014; Accepted: 31 January 2014
Editor: E. Bottieau Article published online: 11 February 2014 Clin Microbiol Infect 2014; 20: O664–O671 10.1111/1469-0691.12586
Corresponding author: C. P. Muller, Institute of Immunology, Centre de Recherche Public de la Sante / Laboratoire National de Sante, 20A, rue Auguste Lumiere, L-1950 Luxembourg City, Luxembourg E-mail: [email protected]
Introduction Mumps is a vaccine preventable disease caused by mumps virus (family, Paramyxoviridae; genus, Rubulavirus) with a
subclinical course in up to one-third of all cases. Common manifestations include parotitis and respiratory symptoms, and orchitis in postpubertal male subjects [1]. Laboratory diagnosis relies on detection of specific IgM antibodies or mumps virus RNA. Based on 316 nucleotides covering the small hydrophobic (SH) gene region, 12 different genotypes of mumps virus have been proposed with genotypes C, F, G and H predominating in Asia [1]. In 2012, 120 out of 194 WHO member states (62%) were using mumps vaccine in their national immunization schedule [2]. In the Lao People’s Democratic Republic (PDR), mumps is not a notifiable disease
ª2014 The Authors Clinical Microbiology and Infection ª2014 European Society of Clinical Microbiology and Infectious Diseases
CMI
H€ ubschen et al.
and mumps vaccine is currently not included in the routine childhood immunization programme. Except for 2011, when only 54 cases were recorded by WHO [3], no data exist. In order to assess the real burden of disease, we investigated the seroprevalence of mumps-specific IgG antibodies across four provinces. In addition, we genetically characterized mumps viruses from the past 3 years from several outbreaks and single cases brought to our attention by the National Center for Laboratory and Epidemiology (NCLE) and medical doctors. Thus we provide here the first comprehensive information about mumps in Lao PDR as a guide for future public health decisions.
Materials and Methods Data concerning the 2011 and 2013 outbreaks and the 2012 outbreaks in Saravane and Champasack province were collected by NCLE, in collaboration with the WHO Lao Country Office for 2011 data. The data for the 2012 Vientiane outbreak were obtained from the Vientiane Capital Health Office. Data for the sporadic cases were collected by the LaoLuxLab of the Institut Pasteur du Lao PDR. Blood and/or throat swabs were taken during the outbreaks from some or all clinical cases and from the sporadic cases and were investigated for mumps-specific IgM antibodies (Euroimmun, L€ubeck, Germany) or viral RNA, respectively. For RNA positive samples, the 316 nucleotide SH gene region was amplified and sequenced and the data were used for phylogenetic analyses according to WHO guidelines [1] with MEGA 4 software [4]. The sequences were submitted to GenBank under accession numbers KF438181-KF438184 and HG423644-HG423651. Sera collected from healthy infants and school pupils aged between 9 months and 19 years and from pregnant women aged between 16 and 46 years were investigated for mumps-specific IgG antibodies using a commercially available ELISA kit (Euroimmun) according to the manufacturer’s instructions. Equivocal samples were retested and classified according to the second test result as positive, negative or equivocal. Before blood collection, informed consent was obtained from all participants or their parents. Data such as age, sex and location were recorded and used for the seroprevalence analyses. For the blood collection, four different locations were selected: Savannakhet city in the south (participants between 5 and 19 years), Luangprabang city in the north (participants aged between 9 months and 19 years and pregnant women), Vientiane city (participants aged <1 year and pregnant women) and Borikhamxay province (participants aged between 5 and 19 years), both in the
High prevalence of mumps in Laos
O665
centre of the country. Fig. 1 was constructed using ArcView 9.3.1 software (Esri, California, USA). Free spatial data were downloaded from DIVA-GIS (http://www.diva-gis.org/Data). Ethical approval was given by the ethics committee of Lao PDR (NECHR 001/2011).
Results Outbreak 2011 Oudomxay province
Between 10 March and 23 April 2011, a total of 61 suspected cases (based on sore throat or cervical/mandibular lymphadenopathy or parotid swelling), including three deaths, were investigated in Mokkha village, located in the Hoon district in Oudomxay province in Northern Lao PDR (Fig. 1). The majority of the patients were female (n = 32, 52.5%) and <16 years old (n = 56, 91.8%). The most affected age group was the 5–9 year olds (n = 26, 42.6%), followed by the 10– 15 year olds (n = 17, 27.9%) and the 1–4 year olds (n = 13, 21.3%). The highest incidence of disease was observed in week 14 with 28 new mumps cases (Fig. 2a). Two 8-year-old girls and an 11-year-old boy died during week 12. Besides lymphadenopathy, headache, malaise or anorexia, also dizziness, lethargy, muscle pain or paralysis of the extremities were observed. For two of the three deceased patients, encephalitis was diagnosed as the cause of death. Mumps-specific IgM antibodies were detected in five out of six sera and mumps virus RNA was detected in four out of ten throat swabs. Thus, nine of the 16 patients from whom clinical specimens were available (56.3%), were laboratory-confirmed mumps cases. Interestingly, Coxsackievirus A16 was detected in one of the mumps PCR-positive samples and in one of the six mumps virus-negative throat swabs (both cases with disease onset in week 16). Rhinovirus was found in another mumps PCR-negative throat swab (case with disease onset also in week 16). Outbreak 2012 Vientiane capital
Between 20 July and 2 August 2012, 12 mumps cases were recorded in Donexingxou village in the Sikhottabong district of Vientiane capital (Fig. 1). Most patients were male (n = 10, 83.3%) and the ages ranged from 4 to 19 years (median 10.5 years). The most affected age group was the 10–15 year olds (n = 7, 58.3%), followed by the 5–9 year olds (n = 3, 25.0%). Most cases occurred in week 31 (n = 7, 58.3%), followed by week 30 (n = 3, 25.0%) and week 29 (n = 2, 16.7%) (Fig. 2a). All patients were reported to have fever, parotitis and headache and five patients were admitted to the hospital. Throat swabs were collected from all hospitalized patients and four of the samples (80.0%) were positive for the
ª2014 The Authors Clinical Microbiology and Infection ª2014 European Society of Clinical Microbiology and Infectious Diseases, CMI, 20, O664–O671
O666
CMI
Clinical Microbiology and Infection, Volume 20 Number 10, October 2014
FIG. 1. Map of Lao PDR showing the different sample collection sites. PV stands for province and PF for prefecture.
mumps virus-specific PCR. All four patients were also IgM positive. Outbreak 2012 Saravane province
A total of 32 mumps cases were recorded between 20 October and 29 November 2012 in Saneumnai village in the Lao-ngam district in Saravane province (Fig. 1). All patients were reported to have fever, a sore throat and parotitis and some also suffered from dizziness and fatigue. The patients were between 1 and 45 years old (mean 12.1 years) and slightly more male (n = 17, 53.1%) than female subjects were affected. None of the four collected throat swabs was PCR positive, but three of the four sera from the same patients were mumps IgM antibody positive. Outbreak 2012 Champasack province
Nine mumps cases were observed between 21 and 29 November 2012 in Kadun village in the Mounlapamok
district (n = 1) and in Donhee village in the Champasack district (n = 8) in Champasack province (Fig. 1). All patients had parotitis and some reported fever and headaches. Eight of the patients were pupils aged between 8 and 12 years (n = 7 female, 87.5%) and one was a 45-year-old male teacher. None of the nine throat swabs were PCR positive, but seven out of nine sera were mumps IgM antibody positive. Outbreak 2013 Attapeu province
Between 30 April and 4 May 2013 four mumps cases were noticed among pupils from Nong-Hin village in the Sanamxai district in Attapeu province (Figs 1 and 2a). The children were between 7 and 15 years old and two were male and two female. All patients were hospitalized on 6 May with fever, headache, sore throat and parotitis. Throat swabs were collected from all children and two of them were positive by mumps virus-specific PCR.
ª2014 The Authors Clinical Microbiology and Infection ª2014 European Society of Clinical Microbiology and Infectious Diseases, CMI, 20, O664–O671
CMI
H€ ubschen et al.
High prevalence of mumps in Laos
O667
(a) 30 2011 2012
25
Number of cases
2013 20
15
10
5
0
10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48
Epidemiological week (b) 30 F
Number of cases
25
FIG. 2. Number between
2011
of and
mumps 2013
cases (a)
by
epidemiological week and (b) according
M
20 15 10 5 0
1–4
5–9
10–15
> 15
Age in years
to age and gender. Individual case 2012 Vientiane capital
Phylogenetic analysis of the mumps virus sequences
On 11 September 2012 a 24-year-old male Lao citizen from Vientiane capital was clinically diagnosed as having mumps. A throat swab was collected, which was positive by PCR.
From all mumps virus-positive throat swabs, the 316 nucleotide region of the SH gene was amplified and sequenced. Phylogenetic analysis of the sequences identified four distinct virus variants (Fig. 3). The sequences from the 2011 outbreak were all identical and belonged to genotype G. The closest relatives identified by BLAST and phylogenetic analyses were sequences found in Japan between 1999 and 2005 (e.g. MuV-Yokohama.JPN/05-1), which differed by five nucleotides in the 316 nucleotide region of the SH gene. The sequences from the 2012 and 2013 outbreaks were also all identical, but belonged to genotype J. Strains found in Thailand in 2008 (e.g. MuVs/Phang-nga.THA/4.08-14) were the closest relatives and differed by a single nucleotide in the SH gene region (Fig. 3). The individual case from 2012 also belonged to genotype J, but was very distinct from the 2012 and 2013 outbreak strains (12 mutations, genetic distance 3.95%), while the single case from 2013 differed by only one nucleotide from the 2012 outbreak
Individual case 2013 Vientiane capital
On 28 June 2013 a 15-year-old girl from Phakhao village, Vientiane capital, fell ill with mumps-like symptoms. A throat swab collected on 1 July was positive by PCR. Overall summary of laboratory investigation and confirmation
A total of 120 suspected mumps cases observed between 2011 and 2013 were brought to our attention. Sera and/or throat swabs for laboratory investigation were available from one-third of them (n = 40). Laboratory confirmation was achieved for 27 patients, which is 67.5% of all patients investigated and 22.5% of all suspected cases.
ª2014 The Authors Clinical Microbiology and Infection ª2014 European Society of Clinical Microbiology and Infectious Diseases, CMI, 20, O664–O671
O668
CMI
Clinical Microbiology and Infection, Volume 20 Number 10, October 2014
MuVs/Ban Mokka.LAO/15.11 MuVs/Ban Mokka.LAO/16.11/2 MuVs/Ban Mokka.LAO/16.11/3 95 Genotype G MuVs/Ban Mokka.LAO/16.11/1 95 MuV-Yokohama.JPN/05-1 AB699705 MuVi/Gloucester.GBR/32.96-G MuVi/Sheffield.GBR/1.05-G MuVi/London.GBR/3.02(UK02-19) MuVi/Bedford.GBR/0.89-H MuVi/Ulaanbaatar.MNG/22.09-H 87 89 MuVi/RW154.USA/0.70s-K MuVi/Stockholm.SWE/26.83-K MuVi/Tokyo.JPN/0.93(MP93-N) MuVi/Ge9.DEU/0.77-D MuVi/Nottingham.GBR/19.04-D 96 99 MuVi/Zagreb.HRV/39.98-C MuVi/Stockholm.SWE/46.84-C MuVi/Fukuoka.JPN/41.00-L 100 MuVi/Tokyo.JPN/6.01-L 100 MuVi/Shandong.CHN/4.05-F MuVi/Zhejiang.CHN/11.06/1-F MuVi/Taylor.GBR/0.50s(Tay/UK50s) MuVi/Urabe.JPN/0.67-B 99 MuVi/Himeji.JPN/24.00-B MuVi/Vector.RUS/0.53VAC-N 99 MuVi/L-Zagreb.HRV/0.71VAC-N MuVi/Akita.JPN/42.93-I MuVi/Dg1062.KOR/46.98-I 100 100 MuVi/Boston.USA/0.45-A MuVi/Pennsylvannia.USA/13.63VAC-A MuVi/Sapporo.JPN/12.00-J MuVi/Leeds.GBR/9.04-J 100 MuVs/Vientiane.LAO/37.12 91 MuVs/Phang-nga.THA/4.08-14 EU497657 MuVs/Vientiane.LAO/26.13 Genotype J MuVs/Vientiane.LAO/31.12/2 MuVs/Sanamxai.LAO/18.13/2 MuVs/Vientiane.LAO/31.12/1 MuVs/Sanamxai.LAO/18.13/1 MuVs/Vientiane.LAO/31.12/3 MuVs/Vientiane.LAO/31.12/4
98
FIG. 3. Phylogenetic tree based on 316 nucleotides of the mumps virus SH gene region and the Kimura 2-parameter model and the neighbour-joining algorithm. Light grey dots mark the sequences obtained from the mumps outbreak in 2011, grey triangles highlight sequences from the 2012 outbreak and black diamonds show sequences from the 2013 outbreak. The grey square and the black triangle mark the individual cases from 2012 and 2013, respectively. White dots indicate the closest relatives of the sequences from
0.01
Lao PDR identified by BLAST.
strains. The closest relatives identified by BLAST were also the mumps strains found in Thailand in 2008, which differed by 13 nucleotides from the 2012 strain and by two nucleotides from the 2013 strain. Seroprevalence data
A total of 2379 sera collected from four different locations (Luangprabang, n = 896; Borikhamxay, n = 599; Savannakhet, n = 587; Vientiane, n = 297) were included in the seroprevalence study (Fig. 1). Overall, 58.2% of the participants were positive for mumps-specific IgG antibodies, 39.5% were negative and the remaining 2.3% were equivocal (Table 1). The highest seropositivity of 61.2% was found in Luangprabang, the lowest in Vientiane (52.5%). Of the two locations sampled in Borikhamxay province, Paksane had a considerably higher overall seropositivity than the more rural
Pakkading (68.3% of 303 samples compared with 43.2% of 296 samples). As expected, the seropositivity increased with age, with the lowest percentage found in the <1-year-old infants (9.1%) and the highest in the cohort of pregnant women (69.2%) (Table 1). The <1-year-old infants and the pregnant women from Vientiane had higher seropositivity rates than the corresponding cohorts from Luangprabang (14.3 vs. 4.0% and 71.4 vs. 67.0%). Among the 5–9 and the 10–14-year-old children, the highest seropositivity was observed in Luangprabang (60.5 and 70.0%) and the lowest in Borikhamxay (42.7 and 61.5%). The highest percentage of seropositives among the 15–19-year-olds was found in Luangprabang and the lowest in Savannakhet (75.3 vs. 62.3%, Table 1). Overall, more female than male subjects were seropositive (60.4 vs. 54.9%, p <0.01), but without the pregnant women, who included many
TABLE 1. Seroprevalence of mumps-specific IgG antibodies according to geographical location and age group/cohort Total
Luangprabang Borikhamxay Savannakhet Vientiane Total
N
Positive (%)
Negative (%)
Equivocal (%)
Infants Positive (%)
5–9 year olds Positive (%)
10–14 year olds Positive (%)
15–19 year olds Positive (%)
Pregnant women Positive (%)
896 599 587 297 2379
548 335 346 156 1385
329 244 230 136 939
19 20 11 5 55
4/100 (4.0) / / 14/98 (14.3) 18/198 (9.1)
112/185 85/199 78/166 / 275/550
149/213 123/200 141/217 / 413/630
149/198 127/200 127/204 / 403/602
134/200 (67.0) / / 142/199 (71.4) 276/399 (69.2)
(61.2) (55.9) (58.9) (52.5) (58.2)
(36.7) (40.7) (39.2) (45.8) (39.5)
(2.1) (3.3) (1.9) (1.7) (2.3)
(60.5) (42.7) (47.0) (50.0)
(70.0) (61.5) (65.0) (65.6)
ª2014 The Authors Clinical Microbiology and Infection ª2014 European Society of Clinical Microbiology and Infectious Diseases, CMI, 20, O664–O671
(75.3) (63.5) (62.3) (66.9)
CMI
H€ ubschen et al.
TABLE 2. Seroprevalence of mumps-specific IgG antibodies according to geographical location, cohort and gender
Location
Cohort
Luangprabang
Infants/pupils Pregnant women Borikhamxay Pupils Savannakhet Pupils Vientiane Infants Pregnant women All All All without pregnant women
Female
Male
N
Positive (%)
N
Positive (%)
364 200 290 335 51 199 1439 1040
219 134 169 195 10 142 869 593
332 / 309 252 47 / 940 940
195 / 166 151 4 / 516 516
(60.2) (67.0) (58.3) (58.2) (19.6) (71.4) (60.4) (57.0)
(58.7) (53.7) (59.9) (8.5) (54.9) (54.9)
participants older than 19 years, this difference was no longer statistically significant (57.0% vs. 54.9%, p >0.05, Table 2). This was also the case for all locations except Savannakhet, where a slightly higher percentage of positives was found among male subjects (59.9 vs. 58.2%). The biggest gender difference was observed in infants from Vientiane (8.5% seropositivity in male infants and 19.6% in female infants, Table 2), but overall only 98 children were recruited.
Discussion In the present manuscript, we investigate mumps seroprevalence amongst almost 2400 infants, school children and pregnant women from five different locations in four provinces and describe all mumps cases and outbreaks between 2011 and 2013 brought to our attention. While northern, central and southern regions of Lao PDR are represented both in the seroprevalence study and the case investigations, samples were mostly from urban settings. The high seroprevalence detected during this study indicates that Lao PDR sustains many more cases than were reported here. Nevertheless, our study is one of the very few comprehensive studies on mumps in a specific country and exceptional in the region because it combines seroprevalence and case investigation data. Two different genotypes of mumps virus were found: genotype G in the north (2011) and genotype J in the central and the southern parts (2012, 2013). As only 12 sequences were obtained in total, it is unclear to what extent our findings are representative for the country. However, genotype J was found at four different time-points with some sequence variation and except for the very distinct strain from 2012 all strains were very similar to viruses detected in 2008 in Thailand. This strongly suggests that this genotype is endemic in the region. A recent report from Nan Province of Thailand close to the border with Laos identified genotype J during a mumps outbreak in 2010 [5], but the sequences are not available for comparison. The individual patient from 2012
High prevalence of mumps in Laos
O669
declared having been in places frequented by tourists during the incubation period. Thus this distinct genotype J virus could either be endemic or could have been introduced directly from a location outside of Lao PDR. Genotype J has been detected in several countries, including Ireland [6], the UK [7,8], Japan [9] and Australia [10], but to date the sequences from Thailand are the most similar to the distinct strain identified in Lao PDR in 2012. Whether genotype G continuously circulates in northern Lao PDR requires further investigation. As the genotype G strains differed by five nucleotides from the most similar sequences identified by BLAST, it remains open whether this variant is endemic in Lao PDR or whether it was introduced from outside. While no genotype information is available from Myanmar, genotype F was reported from across the border with China [11]. During a recent outbreak in Cambodia, genotype K was detected [12] and the same genotype was presumably introduced from Vietnam to Canada (GenBank accession number KF212192), but so far no genotype G was reported from the neighbouring countries. Throat swabs for genotype determination were also collected during mumps outbreaks in the very south of the country, namely in Saravane and Champasack provinces, but the swabs were collected very late after onset of disease (7 days in one case, between 11 and 19 days in all others). This late collection is probably the main reason why all samples were PCR negative because viral RNA is known to be present during disease onset and then to decline over 8– 10 days [1]. Mumps may be found throughout the year in hot climates [13]. In Lao PDR mumps cases occurred between the beginning of March and end of November. In contrast to the other outbreaks, the one from 2011 in the north showed a large number of cases, the laboratory-confirmed co-circulation of two different enteroviruses and three deaths. The co-circulation of different viruses, the high level of malnutrition in the area and poor access to healthcare may potentially explain disease severity and fatal outcome. Two of the deceased patients reportedly died from encephalitis. Encephalitis is a very rare complication of mumps [1] and may also occur after enterovirus infections [14,15], but no samples for laboratory confirmation were available from the deceased patients. While in Europe mumps has become a disease of adolescents and young adults [16–20], most mumps cases in Lao PDR were found among 5–15-year-old children (Fig. 2b), which corresponds more with the ‘classical’ age range for an unvaccinated population [13]. In fact, our seroprevalence data show that there is a considerable increase in seropositivity between the age groups 5–9 and 10–14 years (Savannakhet
ª2014 The Authors Clinical Microbiology and Infection ª2014 European Society of Clinical Microbiology and Infectious Diseases, CMI, 20, O664–O671
O670
Clinical Microbiology and Infection, Volume 20 Number 10, October 2014
47.0 and 65.0%, Luangprabang 60.5 and 70.0%, Borikhamxay 42.7 and 61.5%), indicating that many become infected as pupils. In other age-stratified seroprevalence studies in unvaccinated populations, seropositivity was often even higher than in Lao PDR [21–24]. As about 30% of pregnant women are still susceptible to infection, mumps may be contracted at a wide age-range in Lao PDR. This is of concern because mumps may cause foetal loss during the first trimester of pregnancy and adults are more prone to severe disease complications such as meningitis or orchitis [13]. The seroprevalence study identified some differences between the locations. The overall seropositivity rate for Vientiane is largely biased due to the high proportion of infants (98/297, 33.0%), because both infants and pregnant women showed considerably higher rates in Vientiane than in Luangprabang, possibly because of the higher population density in the capital. This may also explain the differences between Paksane and the more rural Pakkading, where considerable differences in seroprevalence are observed even in neighbouring districts within the same province. For all age groups of pupils, the highest seropositivity rates were detected in Luangprabang, a town highly frequented by foreign tourists. The mean and median ages within the age groups or cohorts were not very different and cannot explain differences in seropositivity between the locations. As mumps vaccination is currently not included in the national immunization schedule in Lao PDR, the overall seropositivity of 58% is due essentially to wild-type infection. Besides the currently most affected children between 5 and 15 years, adults with professional contact with these children are particularly at risk of infection. In neighbouring Thailand many thousands of mumps cases are reported each year despite vaccination (9901 and 7422 in 2011 and 2012, respectively) [3]. The high seropositivity rate found during the present study suggests that a considerable annual number of mumps cases must occur also in Lao PDR. Although under-reporting would be likely from remote places, mumps should be a notifiable disease in Lao PDR to get more accurate case numbers and cost estimates for public healthcare. In this context, the importance of laboratory-based surveillance will increase considerably and it will become imperative that appropriate specimens are collected in a timely manner to not only confirm clinically suspected cases, but also to obtain genotype information for molecular epidemiology. Recently, rubella vaccination has been introduced into the national immunization programme in combination with measles vaccine. The cost-benefit of the incremental cost of using the combined measles, mumps and rubella vaccine should be carefully assessed.
CMI
Acknowledgements The authors wish to thank Dr Paul Brey for facilitating the present study and the Ministry of Cooperation and the Centre de Recherche Public de la Sante in Luxembourg for financial support. We also thank Dr Jean Marie Hospied from the French Clinic in Vientiane and Dr Chansamone Vanulasy from Hospital 103, Vientiane, for providing material for laboratory investigations. For the preparation of Fig. 1, we are very grateful to Dr Anna L. Reye. Parts of the data obtained during the present study have been shown during oral presentations at the Francophone research meeting in Vientiane, Lao PDR, on 20 November 2012 and at the 5th National Health Research Forum 2013 in Vientiane, Lao PDR, on 15 October 2013.
Transparency Declaration The authors declare no conflicts of interest.
References 1. WHO. Mumps virus nomenclature update: 2012. Wkly Epidemiol Rec 2012; 22: 217–224. 2. WHO. Countries using mumps vaccine in national immunization schedule, 2012. Available: http://www.who.int/immunization_monitoring/ diseases/Mumps_map_schedule.JPG. Accessed September 30 2013. 3. WHO. Mumps reported cases, July 14 2013. Available: http://apps.who.int/ immunization_monitoring/globalsummary/timeseries/tsincidencemumps. html. Accessed August 28 2013. 4. Tamura K, Dudley J, Nei M, Kumar S. MEGA4: Molecular Evolutionary Genetics Analysis (MEGA) software version 4.0. Mol Biol Evol 2007; 24: 1596–1599. 5. Phanawadee M, Ieowongjaroen I, Makaroon J et al. Investigation of Mumps Outbreak in a Rural District, Thailand, June–October 2010. Weekly Epidemiological Surveillance Report. 2012;43(Supplement: March 2012):15–22. 6. Reid F, Hassan J, Irwin F, Waters A, Hall W, Connell J. Epidemiologic and diagnostic evaluation of a recent mumps outbreak using oral fluid samples. J Clin Virol 2008; 41: 134–137. 7. Cui A, Brown DW, Xu W, Jin L. Genetic variation in the HN and SH genes of mumps viruses: a comparison of strains from mumps cases with and without neurological symptoms. PLoS One 2013; 8: e61791. 8. Cui A, Myers R, Xu W, Jin L. Analysis of the genetic variability of the mumps SH gene in viruses circulating in the UK between 1996 and 2005. Infect Genet Evol 2009; 9: 71–80. 9. Inou Y, Nakayama T, Yoshida N et al. Molecular epidemiology of mumps virus in Japan and proposal of two new genotypes. J Med Virol 2004; 73: 97–104. 10. Bangor-Jones RD, Dowse GK, Giele CM, van Buynder PG, Hodge MM, Whitty MM. A prolonged mumps outbreak among highly vaccinated Aboriginal people in the Kimberley region of Western Australia. Med J Aust 2009; 191: 398–401. 11. Ma SH, Shi HJ, He CY, Chen JY, Yang HJ, Sun QM. [Genetic characteristics on the small hydrophobic protein and hemaggluti-
ª2014 The Authors Clinical Microbiology and Infection ª2014 European Society of Clinical Microbiology and Infectious Diseases, CMI, 20, O664–O671
CMI
12.
13. 14.
15.
16.
17.
H€ ubschen et al.
nin-neuraminidase genes of mumps virus in Yunnan province, China from 2007 to 2009]. Zhonghua Liu Xing Bing Xue Za Zhi 2011; 32: 370– 375. Tek S, Rin R, Muth S, Bun S, Chea N, Concepcion Roces M. Mumps Outbreak in Siem Pang district, Stung Treng Province, Cambodia, 2012. The 4th International Conference on Public Health among Greater Mekong Sub-regional Countries; Kunming, China, 2012. WHO. Mumps virus vaccines. Wkly Epidemiol Rec 2007; 82: 51–60. Joshi R, Mishra PK, Joshi D et al. Clinical presentation, etiology, and survival in adult acute encephalitis syndrome in rural Central India. Clin Neurol Neurosurg 2013; 115: 1753–1761. Koskiniemi M, Rantalaiho T, Piiparinen H et al. Infections of the central nervous system of suspected viral origin: a collaborative study from Finland. J Neurovirol 2001; 7: 400–408. Hukic M, Ravlija J, Dedeic Ljubovic A et al. Ongoing large mumps outbreak in the Federation of Bosnia and Herzegovina, Bosnia and Herzegovina, December 2010 to July 2011. Euro Surveill 2011;16. pii: 19959. Karagiannis I, van Lier A, van Binnendijk R et al. Mumps in a community with low vaccination coverage in the Netherlands. Euro Surveill 2008; 13. pii: 18901.
High prevalence of mumps in Laos
O671
18. Kuzmanovska G, Polozhani A, Mikik V et al. Mumps outbreak in the former Yugoslav Republic of Macedonia, January 2008-June 2009: epidemiology and control measures. Euro Surveill 2010; 15. pii: 19586. 19. Rajcevic S, Seguljev Z, Petrovic V et al. Ongoing mumps outbreak in Novi Sad, the autonomous province of Vojvodina, Serbia, January to April 2012. Euro Surveill 2012; 17. pii: 20169. 20. Walker J, Huc S, Sinka K, Tissington A, Oates K. Ongoing outbreak of mumps infection in Oban, Scotland, November 2010 to January 2011. Euro Surveill 2011; 16. pii: 19803. 21. Aksit S, Egemen A, Ozacar T, Kurugol Z. Mumps seroprevalence in an unvaccinated population in Izmir, Turkey. Acta Paediatr 2000; 89: 370– 371. 22. Gurgoze MK, Yilmaz E, Godekmerdan A et al. Seroprevalence of mumps, varicella and rubella antibodies in children 1-16 years of age in eastern Turkey. Turk J Pediatr 2006; 48: 185–188. 23. Kanbur NO, Derman O, Kutluk T. Age-specific mumps seroprevalence of an unvaccinated population of adolescents in Ankara, Turkey. Jpn J Infect Dis 2003; 56: 213–215. 24. Tolfvenstam T, Enbom M, Ghebrekidan H et al. Seroprevalence of viral childhood infections in Eritrea. J Clin Virol 2000; 16: 49–54.
ª2014 The Authors Clinical Microbiology and Infection ª2014 European Society of Clinical Microbiology and Infectious Diseases, CMI, 20, O664–O671