Shigella species are a common cause of acute diarrheal disease worldwide, with an estimated 167 million cases per year and resulting in approximately 1.1 million deaths; 97.6% of the cases occur in developing countries. According to the Chinese National Infectious Disease Internet Reporting System, the annual incidence of shigellosis in China made it rank in the top three of the most notable infectious diseases for four consecutive years (2005 to 2008), with close to 500000 cases of shigellosis per year (http://www.moh.gov.cn); this number is now widely believed to be underestimated.
Shigellosis is caused by four species, S. dysenteriae, S. flexneri, S. boydii, and S. sonnei. Shigella species can be identified by serotyping with group-specific antigens; serotyping is based on structural differences within the O-antigen repeating unit of lipopolysaccharide. A total of 47 serotypes of Shigella have been recognized, including 15 for S. flexneri, 13 for S. dysenteriae, 18 for S. boydii, and a single one for S. sonnei. The distribution of species and serotypes of Shigella is heterogeneous over time and place.
The World Health Organization has made the development of a safe and effective vaccine against S. flexneri[1,6-8], but the vaccine effectiveness depends on the distribution patterns of local species and serotypes, because only type-specific immunity has been demonstrated in humans and moreover cross-serotype protection is controversial.
According to a previous multicenter study of Shigella diarrhea in six Asian countries, S. flexneri is the most common species in Bangladesh, China, Pakistan, Indonesia, and Vietnam; whereas S. sonnei is predominant in industrialized countries. Two recent reports have indicated that S. flexneri 2a is the most frequently isolated Shigella organism in China. However, these reports may not be generalizable for the whole China; the time period of these studies is short, and surveillance is performed only in less-developed areas of China.
Little is known about the distribution of Shigella serotypes in Beijing, the political, educational, cultural, and economic center of China with a population of over 30 million. The present study describes the trends in Shigella species and their serotypes isolated from patients with diarrhea in a national infectious disease hospital in Beijing, China, from 1994 to 2010.
Study sites and settings
The location was a clinical diagnostic center at the 302nd Hospital of the People’s Liberation Army in Beijing, China. The 302nd Hospital is the largest infectious disease teaching hospital in Beijing, China, with 1300 beds and receiving more than 36400 patients annually. From January 1994 to December 2010, fresh stool specimens were collected from patients with diarrhea and clinically suspected dysentery. The specimens were submitted to the microbiology laboratory of the 302nd Hospital. All experimental research have been performed with the approval of ethics committee of 302nd Hospital of the People’s Liberation Army, with reference number 2004013D.
Serologic identification was performed by slide agglutination with polyvalent somatic (O) antigen grouping sera, followed by testing with available monovalent antisera for specific identification of serotypes according to the manufacturer’s instructions (Denka Seiken, Japan). Only one Shigella isolate per patient per diarrheal episode was included in the analysis.
Statistical comparisons were performed using the CHISS software (version 2001, Yuan YiTang Sci-Tech Co., Ltd., Beijing, China). Categorical data were expressed as percentages and calculated using a chi-square test, and p ≤ 0.05 was considered statistically significant.
A statistically significant decreasing trend in S. flexneri and an increasing trend in S. sonnei were observed by chi-square analysis (p < 0.01) (Figure 1). The trends in Shigella spp. isolated from Beijing between 1994 and 2010 are shown in Figure 2. The recording of annual Shigella isolation began in 1994, and the maximum number of isolates was reported in 1996 (n = 1194). The annual total number of isolated Shigella organisms had been decreasing since then, reaching a low point in 2008 (n = 22). This trend may be related to the strict hygiene inspection and adequate sanitation during the 2008 Olympic season. Four peaks were observed during the 17-year collection period. Peak 1 appeared in 1996, with subsequent peaks in 1998 (peak 2, n = 602), 2002 (peak 3, n = 398), and 2004 (peak 4, n = 251). A sudden decrease in Shigella isolation was observed more in 2003 than in 2002 and 2004; one possible explanation is that resources were redirected to identify severe acute respiratory syndrome cases in China in 2003, thereby limiting bacterial diarrheal isolation. It should be noted that as the numbers of observed cases of shigellosis were decreasing, China’s per capita gross domestic product (GDP) was increasing (Figure 3).
The distribution of typeable Shigella during the study period was S. flexneri, 71.7% (n = 4295); S. sonnei, 27.3% (n = 1639); S. dysenteriae, 0.55% (n = 33); and S. boydii, 0.33% (n = 20). The distribution of Shigella species changed over the 17-year observation period (Figure 4). Between 1994 and 2005, Shigella isolation rates were largely driven by S. flexneri, reaching peak numbers in 1996 when 90% of all isolated Shigella were S. flexneri. In 2006, S. sonnei became the dominant subgroup. In 2009, the lowest percentage of isolated S. flexneri (6%) was recorded. The apparent isolation rates of S. boydii and S. dysenteriae increased during this period, e.g., S. boydii isolation rates increased from 0% to 4.2% (n = 5), 3.8% (n = 3), 4.5% (n = 1), 8.3% (n = 3), and 7.6% (n = 2) in 2006, 2007, 2008, 2009, and 2010, respectively. However, the absolute numbers of S. boydii and S. dysenteriae did not change during this period, remaining between 0 and 5 per year. This result suggests that although uncommon, sources of S. dysenteriae and S. boydii remain in the Beijing area.
Serotypes of S. flexneri and S. sonnei
As revealed by 17 years’ worth of data, 27/47 (57.4%) Shigella serotypes were identified; these serotypes included S. boydii serotypes 1, 2, 5, 15, and 17; S. dysenteriae serotypes 1, 2, 3, 4, 5, 7, and 8; S. flexneri serotypes 1a, 1b, 2a, 2b, 3a, 3b, 4a, 4b, 4c, 5, 6, x, and y; and S. sonnei. Given the small numbers of S. boydii and S. dysenteriae (n = 53 combined), only isolates of S. flexneri and S. sonnei were further analyzed. Surveillance data indicated large differences in the serotypes of S. flexneri isolated between 1994 and 2010, in addition to the change in dominant species in 2006 (Table 1). S. flexneri 2a and S. sonnei were observed in every year of the study, and S. sonnei surpassed S. flexneri 2a as the predominant single Shigella type in 2000. The number and proportion of S. flexneri 2a isolates decreased from 42.6% (n = 262) in 1994 to 13.6% (n = 3) in 2010, whereas the proportion of S. sonnei increased from 36.6% (n = 225) to 82.3% (n = 20) during the entire period of data collection. S. flexneri 2a ranked the first among the S. flexneri subtypes except in 2003, 2004, and 2005. S. flexneri 4c, 4a, and S. flexneri x were seldom isolated, accounting for only about 3% (n = 180) of the 5934 S. flexneri and S. sonnei isolates. However, S. flexneri 4c was the most frequently isolated S. flexneri serotype in 2004 (26.4%, n = 65) and 2005 (20.3%, n = 30). S. flexneri 4c and S. sonnei were isolated in equal numbers in 2004.
Shigella isolates were recovered routinely throughout the study but were frequently recovered in the summer months (June to September; t = 7.83, p < 0.001; Table 2). Isolation of Shigella almost always peaked in July and August; 2003 and 2008 were exceptional years during which a September peak was observed. As indicated in the Beijing weather information, the temperature in September 2003 reached 33.7°C, the highest on record for the past 42 years (Provided by China Meteorological Data Sharing Service System, http://cdc.cma.gov.cn/home.do,). In September 2008, the amount of rainfall was 98.1 mm, which was twice the amount of rainfall during the same month in 2007 (Provided by China Meteorological Data Sharing Service System, http://cdc.cma.gov.cn/home.do). Both factors may have contributed to the late seasonal peaks.
Epidemiological information was available for all 5999 patients. Patient age ranged from 3 months to 90 years. The age distribution for the 5934 cases of S. flexneri and S. sonnei is shown in Table 3. Adults aged between 21 and 25 years were the most commonly affected group (n = 978; 16.3%), followed closely by children aged less than 6 years (n = 821; 13.6%). S. flexneri 2a and S. sonnei were recovered from patients in each age group, although most infections caused by S. sonnei were found in children (n = 1639; 48%); children aged 0 to 5, 6 to 10, and 11 to 15 years accounted for 17.6% (n = 289), 15.2% (n = 250), and 14.8% (n = 242) of the cases, respectively. S. flexneri 2a occurred frequently in adults, especially those in the 21 to 25 (n = 594; 18.8%) and 26 to 30 (n = 380, 12%) age groups, although a high percentage (n = 400; 12.7%) was found to affect children aged less than 6 years.
Information about patient gender was known for all of the patients infected with S. flexneri and S. sonnei; distribution was slightly biased toward male patients (n = 3616; 60%; Table 4). Unexpectedly, the ratio for some of the predominant serotypes of S. flexneri and S. sonnei differed in distribution between male and female patients. The most prevalent S. flexneri serotype (2a) was found more frequently in males (63.5%, p < 0.0001) than in females (36.5%). This result suggests that in Beijing, males either have greater exposure or are more susceptible to this subserotype than females. Similarly, S. flexneri 5 affected more males than females (ratio of infected males to females, 28 [46%]:5 [44%]; p < 0.001), although this serotype is not common in Beijing. By contrast, S. flexneri 2b and 6 as well as S. sonnei were more often associated with women than with men (p < 0.003, p < 0.04, and p < 0.0001, respectively).
Written informed consent was obtained from the patient for the publication of this report and any accompanying images.
The authors declared that they have no competing interests.
YM, EC, CB, ZL, SC, JZ, HW, CZ performed the experiments and provided clinical samples and patient data, JZ analyzed the data, JDK and BZ wrote the first draft of the manuscript, FQ and ZW designed the study, supervised the experiments and contributed to the manuscript. All authors read and approved the final manuscript.