The epidemiological and serological characteristics of measles in Dongguan, China, 2005–2014

ABSTRACT

This study examined the epidemiological and serological characteristics of measles in Dongguan, China. From 2005 to 2014, a total of 8,224 measles cases were reported in Dongguan, 33.5% of which were aged <1 y and 30.6% >14 y. From 2005 to 2014, the proportion of the <1 y measles cases increased year by year from 24.3% to 47.9%. Of the cases aged ≥8 months (n = 6,768 cases), only 11.6% had been immunized with at least one dose of measles vaccine. Of the 2,213 cases who had never been immunized with measles vaccine, immigrants accounted for 82.4%. 52.4% of the measles cases were diagnosed with pneumonia, and 12 cases died from respiratory failure. Seroprevalence rate in women and their newborns was 86.0% and 82.5%, respectively. Measurement of serum measles antibody levels for infants aged less than 8 months indicated that seroprevalence rate dramatically declined from 97.3% at birth to 9.3% and 13.2% at 6- and 7- month old. The existence of a sufficient pool of unvaccinated people (especially immigrants) and decreased level of passively transferred measles antibodies in infants from vaccinated mothers contributed to the sustained transmission observed in Dongguan. In addition to high routine vaccination coverage, new strategies and innovations for measles vaccination are needed to eliminate measles.

KEYWORDS:

• antibody
• epidemiology
• infant
• measles
• pregnant woman

Introduction

Measles is a highly contagious vaccine-preventable respiratory disease caused by measles virus. It is still a common and fatal disease, and a leading cause of death of children in many developing countries.

Measles is a significant public health threat, not only for developing countries, but also for developed ones. Although widespread use of measles vaccine has decreased significantly measles incidence and measles related mortality,^1,2 unfortunately, measles remains common in underdeveloped countries. Even in developed countries, maintenance of elimination proves difficult, despite apparently high vaccine coverage. It is estimated that measles have caused more than 100,000 deaths each year worldwide.¹ In 2014, 14,176 cases were reported in European countries.³

The elimination of endemic measles transmission had been achieved in American regions in 2002, but the US experienced a large multi-state measles outbreak that started in California in 2014. So far no source case of the outbreak has been identified.⁴ Brazil and Canada also experienced large outbreaks in the same year.^4-6

China introduced measles vaccine in 1967. Measles vaccine (MV) is included in the Chinese National Expanded Program on Immunization (CNEPI) and is offered to all eligible children for free. Infants receive 2 doses of measles vaccine-the first at 8 months of age and the second between 18–24 months of old.⁷ Similarly, the incidence of measles in China decreased markedly thanks to the widespread use of measles vaccine. By the end of the last century, the incidence of measles in China dropped to 5.7/100 000.⁸ As a member of the World Health Organization (WHO) Western Pacific Region (WPR), China aimed to eliminate measles by 2012.⁹ China's central government has made great efforts to eliminate measles by sustaining high 2-dose measles vaccination coverage (>95%), implementing supplementary immunization activities (SIAs), and maintaining an effective surveillance system. However, resurgence of measles in recent years has disrupted the momentum to eliminate the disease in China.¹⁰ Having failed to achieve the target, China along with other countries in the region nonetheless made impressive progression measles control, and the WPR is on track to eliminate measles by 2020.¹¹

An increased incidence has been reported in Dongguan, a southern city of Guangdong province, China, since 2005 (26.01, 19.66, 31.20, 24.08 per 100 000 during 2005–2008). And the incidence was as high as 6.02 and 5.49 per 100 000 in 2013 and 2014. Significant changes in epidemiological characteristics such as age distribution may be the cause. To explore the causes of measles persistence in Dongguan, observational study was conducted to examine the epidemiological and clinical characteristics of measles from 2005 to 2014. Serological survey was also conducted to analyze the prevalence of seropositivity for measles in <1 y old infants, women and their newborns. The study will explore the causes of measles persistence in Dongguan, and then provide useful information to identify the subpopulation at risk and address possible additional immunization strategies.

Results

Measles cases reported in Dongguan, 2005–2014

There were 8,224 measles cases reported to Dongguan Center for Disease Control and Prevention (CDC) from 1 January 2005 to 31 December 2014 with 8,137 serologically confirmed. As shown in Fig. 1, there was local measles prevalence in 2005–2008 with a peak incidence of 32/100 000 in 2007. Overall, the incidence went down from 2009 to 2012, but went up again in recent 2 y. The age distribution of those reported cases is shown in Table 1. 33.5% of the measles cases reported in Dongguan from 2005 to 2014 were infants aged <1 y and 30.6% were individuals aged >14 y (χ2 = 213.74, P<0.001). Only 25.2% of the measles cases were children aged 1–7 y. From 2005 to 2014, the proportion of <1 year measles cases grew year by year from 24.3% to 47.9%. The youngest infant diagnosed with measles was just 4 d old. The age distribution for the measles cases aged <1 y is shown in Fig. 2. As can be seen, from 2005 to 2014, the number of measles cases in Dongguan region increased from birth up to age 8 months and declined thereafter. 54.9% were male. Furthermore, 79.1% of the population was defined as immigrant population in Dongguan.

Figure 1. Number of measles cases and the incidence of measles in Dongguan, 2005–2014.

Figure 2. The age distribution of measles cases aged<1 y.

Table 1. The age distribution of reported measles cases in Dongguan, 2005–2014.

	age
Year	<1 yr	%	1–14 yr	%	>14 yr	%	Total
2005	423	24.3	663	38.1	655	37.6	1741
2006	338	28.5	494	41.7	352	29.7	1184
2007	760	35.0	786	36.2	625	28.8	2171
2008	599	35.3	617	36.4	481	28.3	1697
2009	103	36.5	92	32.6	87	30.9	282
2010	34	38.2	24	27.0	31	34.8	89
2011	26	40.0	10	15.4	29	44.6	65
2012	17	45.9	10	27.0	10	27.0	37
2013	234	46.7	129	25.7	138	27.5	501
2014	219	47.9	130	28.4	108	23.6	457
Total	2753	33.5	2955	35.9	2516	30.6	8224

Clinical manifestations

The main clinical manifestations of cases are shown in Table 2. Among the 8,224 cases, 100% cases had fever and rashes, 93.8% coughed, 73.0% had catarrh symptom, 74.7% conjunctivitis, 9.4% lymphadenectasis and 3.0% arthralgia. The most common complication was pneumonia (52.4%). Of these, 13 cases died of respiratory failure resulted from acute respiratory distress syndrome (ARDS) despite maximal supportive care in hospital. 707 cases (8.6%) were combined with laryngitis, 592 cases (7.2%) with hepatitis and 296 cases (3.6%) with encephalitis.

Table 2. The main clinical manifestations of measles Cases (n = 8824) in Dongguan, 2005–2014.

Clinical Manifestations	n	Proportion (%)
Fever	8824	100
Rashes	8824	100
Cough	7715	93.8
Catarrh symptom	6003	73.0
Conjunctivitis	6146	74.7
Koplik spot	3879	47.2
Lymphadenectasis	770	9.4
Arthralgia	247	3.0
Pnemonia	4309	52.4
Laryngitis	707	8.6
Hepatitis	592	7.2
Encephalitis	296	3.6

Immunization history of the measles cases

Of the measles patients aged ≥8 months (n = 6,768 cases), 11.6% (711 cases) had been immunized with at least one dose of measles vaccine, 32.7% (2213 cases) had never been immunized with the vaccine at all, and 55.7% had an unclear history of measles immunization. The majority of those immunized received one measles vaccination, while there were 74 patients received 2 doses of immunization as documented on their immunization card.

We analyzed the relation between age, gender, immigration status and measles vaccination history. The findings of the analysis were showed in Table 3. It indicated that there was no significant association between gender and measles vaccination history (χ2 = 0.57, P = 0.751). Compared with immigrants, local residents were more likely to receive measles vaccination (χ2 = 34.19, P<0.001). The vaccination rates differ greatly from age to age (χ2 = 798.22, P < 0.001). Even for children of the 8 m-1 yr old, only 19.8% of them were vaccinated, which indicated the delay of first dose of vaccination. The proportion of cases reported unknown vaccination status rose with age, but the proportion of vaccinated cases declined with age. Of the 2,213 cases who had never been immunized with measles vaccine, immigrant population accounted for 82.4% (1823/2213) (χ2 = 22.61, P < 0.001).

Table 3. The measles vaccination status of reported measles cases in different age groups in Dongguan, 2005–2014.

		Vaccinated		Unvaccinated		Unknown
Classification	Total	n	%	n	%	n	%	2 value	P value
Age								798.22	<0.001
8 m-1 yr	1242	246	19.8	644	51.9	352	28.3
1–3 yr	1715	272	15.9	637	37.1	806	47.0
4–7 yr	955	136	14.2	220	23.0	599	62.7
7–14 yr	285	42	14.7	57	20.0	186	65.3
>14 yr	2571	89	3.5	655	25.5	1827	71.1
Sex								0.57	0.751
Male	3607	427	11.8	1225	34.0	1955	54.2
Female	3161	358	11.3	1093	34.6	1710	54.1
Residential status								34.19	<0.001
Local residents	1397	195	14.0	390	27.9	812	58.1
Immigrants	5371	590	11.0	1928	35.9	2853	53.1
Total	6768	785	11.6	2318	34.2	3665	54.2

Prevalence of antibody against measles in pregnant women and their offspring

To determine measles immunity in pregnant women and their offspring, sera from 315 mother–infant pairs were examined for anti-measles antibody titer. The mean age of these pregnant women was 26.55 ± 4.97 y (range: 17–45 y), and most aged 20–29 y (n = 211, 67%). All offspring were healthy full-term newborn infants. The mothers of all these infants had been immunized with measles vaccine and none had a documented history of measles. The individual and mean serum measles-specific immunoglobulin G (IgG) levels in the pregnant women and neonatal group were shown in Fig. 3. There are 271 (86.0%) pregnant women and 260 (82.5%) newborn infants detected to meet protective measles IgG antibody levels in their serum (χ2 = 51.68, P < 0.001). A significant correlation was observed between anti-measles antibody titer in newborn infants and their paired mothers (r = 0.44, P < 0.001).

Figure 3. The individual and the mean serum measles-specific IgG levels in pregnant women and normal newborns.

Measles antibody concentrations of infants aged <1 y

To determine measles antibody of infants aged <1 year, 708 infants were studied. The results were shown in Table 4. The median anti-measles antibody titer was 336.40 mIU/mL (interquartile range: 50.0–1387.9 mIU/mL). By using 200 mIU/ml as a cut-off, 55.9% (95% confidence interval [CI]: 52.3–59.6% ) infants aged <1 y had protective measles IgG antibody levels. Seropositivity differed among the month groups (χ2 = 290.41, P < 0.001) Seroprevalence rates significantly decreased in children aged 0–7 months. Neonates at birth had protective measles IgG antibody levels of 97.3% (95% CI: 93.7–100%), compared with 9.3% (95% CI: 0.6–18.0%) and 13.2% (95% CI: 4.1–22.3%) of 6 and 7-month. These data suggest that, over 90% of infants whose mothers were immunized with measles vaccine and without a natural measles history maybe susceptible to measles by their age of 6 months. Protective measles IgG antibody levels raised in infants aged 8–11 months thanks to acquisition of immunity by vaccination. Seroprevalence rate significantly increased to 24.1% (95% CI: 13.1–35.2%) at age 8 months, and the age groups of 9-,10-,11- mo had seropositivity rates ≥80%. There was no significant association between gender, residential status and IgG positivity (P > 0.05).

Table 4. Measles seroprevalence in <1 y population in Donguan, by month group, sex and residential status (%).

Classification	Total	n	%	χ2 value	P value
Month				290.41	<0.001
0	75	73	97.3
1	46	36	78.3
2	59	40	67.8
3	66	41	62.1
4	57	13	22.8
5	74	17	23.0
6	43	4	9.3
7	53	7	13.2
8	58	14	24.1
9	59	47	79.7
10	57	46	80.7
11	61	58	95.1
Sex				0.43	0.51
Male	385	211	54.8
Female	323	185	57.3
Residential status				2.26	0.13
Local residents	426	248	58.2
Immigrants	282	148	52.5
Total	708	396	55.9

Discussion

Measles elimination is one of the major global public health priorities. Following wide application of measles vaccination, measles has been controlled in China and in other countries worldwide. In Dongguan, measles incidence decreased dramatically since the introduction of measles vaccine in 1965.

With influx of migrant workers to respond to the economic development, measles incidence soared from 2005 to 2008, reaching a national height of 31.20 per 100,000 people in 2007. In order to curb the endemicity of the measles virus, a variety of measures were carried out in 2007, including supplemental immunization activities targeting migrant workers and their children. Another catch-up SIA targeted all children aged 9 mo-14 y was conducted in Dongguan in 2009.¹⁶ Measles incidence fell from 2009 to 2012 before it rose in 2013–2014. Our study demonstrated that during the last decade measles outbreaks continued to occur in the Dongguan, mainly affecting infants aged <1 y and those aged >14 y. Similar results have been reported in other localities, including European countries, with a significant number of infants aged <1 y and older individuals being infected. ^17-19

Most of the clinical parameters, such as fever, rashes, eruption, typical catarr, cough symptoms, and Koplik spot were found among measles patients. Measles virus would weaken immune systems, and thus made people infected with the virus much more susceptible to opportunistic infections.²⁰ Pneumonia was found to be the most common lethal complication of measles.^21,22 Almost half of the measles patients in our study had pneumonia with 12 cases died of ARDS. Other complications such as laryngitis, liver dysfunction, and encephalitis were also common. The founding indicated that we should pay more attention to diagnosis/therapy of measles complications and to patients with other underlying diseases.

Despite measles vaccine is offered to all eligible children free of charge and a relatively broad reported vaccination coverage in Dongguan, a sufficient pool of children who did not receive scheduled measles immunization. In our study, almost 90% of the measles cases aged ≥ 8 months had not received vaccine injection or their immunization histories were not clear. When a baby is born in China, the parent of the baby is required to register the birth at the local immunization center within the first month of life for scheduled immunizations. The baby's immunization status is reinforced by periodic province-specific SIAs.^9,23 However, there is an increasing number of families relocate due to socioeconomic reasons and it has brought big challenge to this system.²³ If an immigrant parent fail to re-register at the local immunization center of the new city, the baby of the parent could be dropped out of the system and would not receive appropriate scheduled immunization. In Dongguan, 77.4% of the population was identified as immigrant population. The people who are residing in Dongguan but whose Hukou or registered residency are not are considered as immigrant population. In our study 82.4% of our measles patients were immigrant population. In 2010, a survey of 1,530 children from the immigrant families in Guangdong showed that only 71.8% of the children aged 1–3 y had received measles vaccine. Compared with immigrant children, local children were more likely to receive immunization (OR = 1.32; 95%CI: 1.06, 1.65). ²⁴ Children were less likely to receive timely first dose of measles vaccinations if they were from immigrant community.²⁵ In our study, most of the non-local population were from undeveloped areas of China. Immunization delivery in these areas still needed to be improved. Furthermore, immigrant population tended to live in poor and remote areas of the city, making it harder for immigrant children to be reached by vaccine providers. The low measles vaccination coverage among migrant population was probably one of the major causes of persistent measles outbreaks in Dongguan.

The study showed that a large proportion of reported measles cases were infants aged <1 y and the proportion rose from 24.3% in 2005 to 47.9% in 2014. Most of the cases were aged <8 months. The number of measles cases increased from neonates to 8 months old when infants started to be immunized with measles vaccine. Before measles vaccination was introduced, infants were usually protected against the disease via passive maternal antibodies transfer. The waning of maternally derived protection under current vaccination system may explain an increasing proportion of infants being susceptible to measles before the age of measles vaccination.^26,27 Accordingly, our serological data of <1 y normal infants clearly suggested that 55.9% were found to be seropositive. The rate was lower than that of in 2011 when our previous survey was carried out.¹³ Given that there were only 27 infants of <1 year involved in the previous survey but 708 infants in the current one, the result of the current one should be more representative. The difference of positive cut-off value may be another reason. Seroprevalence rates dropped significantly in children aged from 0 to 7 months old from 97.3% to 13.2% and then increased from age of 8 to 11 months.

Our study has found that the concentration of maternally derived measles antibodies in infants was significantly associated with concentration of maternal antibody, but newborn infants had higher levels than their corresponding mothers. Other studies had made similar observation.^28-30 This indicates a more efficient placental transfer of maternal measles antibody in mother-infant pairs, possibly from active placental transfer.³¹ We also found both mothers (most born after 1981) and infants had comparatively low rates of anti-measles antibody positivity and the results found in other parts of China were similar. ^32-34

In the pre-vaccination era, the average annual incidence of measles was 639.10 per 100,000 persons in Guangdong province and almost everyone had acquired measles in childhood.³⁵ People were more likely to acquire immunity from natural infection at that time. But in the vaccination era, people were more likely to receive routine measles vaccination. The paucity of enhanced immunity from natural infection, combined with waning of immunity from vaccination may have contributed to the increase in measles susceptibility of young fertile women and their infants. This could give rise to the disease burden of measles in this subpopulation. Chen et al. also found that the waning vaccine-induced immunity may have some impact on measles control.³⁶

An increasing number of mothers acquire immunity against measles by vaccination rather than natural infection. Our cross-sectional study showed that passively acquired maternal measles antibodies rapidly decreased with age among infants aged <8 months. The measles antibody titer in 77.2% of infants of 4 months old was lower than 1:200, compared with 90.7% and 87.8% of 6 and 7 months old infants lower than 1:200. The seroprevalence of measles antibodies in infants aged 8 months or younger in Dongguan was lower than the results of other researches in China (53.2–98.0% ).³⁷ The high proportion of migrant population maybe the main reason.

The injection of measles vaccine at early age may not be effective because maternal antibodies neutralize vaccine antigens before the development of a specific immune response³⁸, but any delay in vaccination may increase the risk of disease complications in infants. Some researchers have suggested that the timing of the first vaccine dose should be advanced.^39,40

Aaby et al. also found that child mortality may be reduced if measles vaccine is given earlier than currently recommended by international organizations.⁴¹ With the presence of maternal antibody, child mortality in low-income countries may be reduced via a 2-dose vaccination schedule with the first dose at 4–6 months (earlier than currently recommended) and a booster dose at 9–12 months of age.¹⁰ However, other studies have reported that vaccination in younger infants (6–9 mo) usually results in lower seroconversion rates and neutralizing titers compared with immunization in older infants (12–15 mo).⁴² It is important to identify an optimal age for children to receive measles vaccine in order to balance the risk of an early loss of maternal antibodies in the majority of infants with the risk of primary vaccine failure due to the presence of maternal antibodies.³⁰

There are some limitations in the current study. First, once someone's written immunization history was missing, the person would be regarded as unvaccinated. Therefore, the measles immunization coverage rates could be underestimated and in our study 55.7% of the subjects' immunization histories were unclear. Such underestimation is inevitable given that immunization records are not shared by clinics. Therefore, it is necessary to set up an electronic immunization system. Second, the serological investigation of infants aged less than 1 y old was carried out by a cross-sectional survey but not a cohort study. The results based on the data from 66 densely populated villages in Dongguan. The study's representation was limited. But all the infants were randomly selected from the same population and their birthday were close. So the results would provide important insights for measles control especially in other regions of Guangdong. Third, none of the 315 pregnant women reported had natural measles infection and all of them reported measles vaccine history. We cannot compare the difference in average maternal antibody titers induced by vaccination or natural infection. One possible explanation is that mothers could not remember whether they had natural measles infection during childhood, thus subject to recall bias.

In summary, measles remains a significant problem in Dongguan. A large proportion of measles cases are infants aged <1 y and individuals aged >14 y. Pneumonia remains the leading cause of measles-related hospitalization and mortality. The existence of a sufficient pool of measles-susceptible children who have missed scheduled measles immunization and the decreased level of passively transferred measles antibodies in infants from vaccinated mothers contribute to the sustained transmission observed in Dongguan. Similar problems existed in measles control of other regions of China.

Therefore, we suggest that it is important to maintain vaccination coverage rate of at least 95% with routine 2-dose measles vaccination. Large-scale follow-up supplementary immunization activities are also necessary every 3–4 y until routine immunization can achieve sufficiently high levels of population immunity to prevent the accumulation of measles susceptible individuals. Furthermore, new strategies and innovations for vaccination are needed in order to meet the target of complete elimination of measles in Dongguan. The relatively rapid decay of protective antibody in infants suggests that an earlier administration of the first dose of measles vaccination should be considered. Mass campaign activities to vaccinate women of childbearing age with measles vaccine will be another optional way to protect susceptible young adults as well as infants who are below the measles vaccination age.

Materials and methods

Surveillance and data collection

As a class B reportable disease, measles case data was extracted from the China Information System for Disease Control and Prevention (CISDCP), a web-based communicable disease surveillance system that allows for reporting of communicable disease information from local CDCs, hospitals and other health agencies. Healthcare providers in China are required to report cases of 37 infectious diseases, including measles, to a public health authority under provision of the Law of the People's Republic of China on Prevention and Treatment of Infectious Diseases.¹² The Dongguan Center for Disease Control and Prevention (DGCDC) is responsible for data collection in the region.

Dongguan had a total population of 669.5 million in 2005 and 891.5 million by 2014. We calculated the population data in other years based on the growth rate reported by the government. These population figures of each year were used as the denominators for the calculation of disease incidence. The CISDCP surveillance report for each measles case consists of basic demographic information (age, gender, occupation, district and residency status), measles disease summary information, and vaccination history.

In 2011, we carried out a seroprevalence study of measles in 0–60 y general population in Dongguan, a total of 1,969 individuals were enrolled in the study and the seropositivity rate was 93.4%. In 27 children aged <1 y, 23 (85.2%) were found to be seropositive. ¹³ During July to October, 2011, another study was also designed to investigate the seroprevalence of measles in women and their offspring in Humen hospital of Dongguan, 315 women who gave birth in hospital were interviewed. Maternal and cord blood samples were collected.¹³

Base on our previous studies and the changing of measles epidemiology, to explore the measles protective level of <1 year infants, we carried out a further serological survey from September to October, 2013. A multistage cluster sampling was designed. Three villages in each town and 5 out of the 33 towns in Dongguan were sampled randomly. 150 households were randomly chosen within each village, a total of 708 infants aged <1 y participated in the investigation. Personal information such as gender, date of birth, residential status, vaccination status, and date of sampling was obtained from their parents and blood samples of the infants were donated.

Measles diagnosis and serology testing

The measles case definition recommended by the WHO and adopted by China includes clinical and laboratory criteria for diagnosis, and both were used to define confirmed cases.¹⁴ Confirmatory serological testing is performed on patients with suspected measles by measuring specific measles IgM antibodies using a WHO-recommended measles specific enzyme-linked immunosorbent assay (ELISA; Virion-Serion, Germany).

A commercially available ELISA kit (Virion Serion) was used for measurement of anti-measles antibody concentrations. The laboratory results were interpreted according to the manufacturer's instructions. The antibody level was considered positive when the geometric mean titer (GMT) reached 1:200 mIU/ml.¹⁵

Statistical analyses

Statistical analysis was performed using SPSS for Windows version 17.0. The χ2 test was used for categorical data. Continuous variables were expressed as mean ± standard deviation (M ± SD). Correlation between anti measles antibody titer in pregnant women and their newborn infants was analyzed with Spearman rank correlation coefficients. A P value < 0.05 was taken as the level of significance.

Ethical statement

Dongguan CDC Ethics Committee approved the protocol and consent forms were signed before starting.

Abbreviations

ARDS	=	acute respiratory distress syndrome
CI	=	confidence interval
CISDCP	=	China Information System for Disease Control and Prevention
CNEPI	=	Chinese National Expanded Program on Immunization
CDC	=	centers for disease control and prevention
GMT	=	geometric mean titer
IgG	=	immunoglobulin G
MV	=	measles vaccine
SIAs	=	supplementary immunization activities
WHO	=	World Health Organization
WPR	=	Western Pacific Region

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.

Author contributions

The authors contributed equally to this work.

Acknowledgments

We would like to thank all infants and their mothers for participating in the study and all staff involved from the County Centers for Disease Control and Prevention in Dongguan for their support.