Previous surveys of meningococcal disease in the United States and elsewhere found serogroup W135 was rare [1]. An outbreak of serogroup W135 meningococcal infection in Saudi Arabia following the 2000 Hajj was reported, and then spread to several countries around the world [2–4]. Significantly higher carriage rate of N. meningitidis serogroup W135 in pilgrims returning from the Hajj [5, 6] and continuing diversification of the serogroup after its emergence in 2000 has been found [7, 8]. In Taiwan, only about 2% (50,000) of the population are Muslims. In the period from 2001 through 2003, about 20 to 40 pilgrims per year went from Taiwan to Saudi Arabia for the Hajj (data from Chinese Muslim Association, Taiwan). Little is known about the clinical features and outcome of serogroup W135 disease [9–11], but there are some reports of extrameningeal complications [11–13].

The incidence of meningococcal disease in Taiwan was below 0.001 from 1980 to 1987, and re-emerged in 2000 with a rate of 0.07/10⁵ population. In 2001 there was a further increase in incidence (0.19/10⁵) [14]. Serogroup B was the most common and W135 was secondmost predominant, which is different from other country [1, 14]. To determine the clinical characteristics and outcome of patients with serogroup W135 and non-W135 meningococcal disease in Taiwan, a nationwide study was conducted from January 1, 2001, through December 31, 2003. The relationship between N. meningitidis serogroups with respect to patient characteristics, clinical manifestations, and outcome was assessed. The factors associated with mortality in meningococcal disease were also investigated.

This nationwide study of meningococcal disease in Taiwan focused on the sporadic occurrence of serogroup W135 meningococcal disease. W135 serogroup accounted for 26% of all meningococcal disease in Taiwan, and its prevalence is second only to that of serogroup B. During the 1990s, N. meningitidis W135 represented less than 5% of all reported N. meningitidis in the UK, France, and the United States (1, 17–18). Taiwan is not a Muslim country and less than 40 persons per year are Hajj pilgrims from Taiwan. No case had contact with Hajj pilgrims in this series and analysis of the geographic distribution and timing of our serogroup W135 meningococcal disease cases did not suggest an epidemiologic link to Hajj pilgrims. The nasopharyngeal carrier rate of N. meningitidis was 2.3% in a nationwide surveillance from military recruits in Taiwan in 2001, and serogroup W135 constituted 15.2% [Taiwan CDC, unpublished data].

From pulsed-field gel electrophoresis and multilocus sequence typing genotyping in Taiwan CDC, serogroup B isolates in Taiwan were derived from several distinct lineages, but isolates of serogroup A, serogroups W135 and C, and serogroup Y, respectively, belonged to ST-7, ST-11, and ST-23 clones [19]. All our serogroup W135 meningococcal strains in Taiwan available for analysis in 2001–2002 belonged to ST-11/ electrophoretic type-37 complex, and they were apparently introduced before the Hajj pilgrimage outbreak [19]. And multilocus variable-number tandem repeat analysis with better genotyping discrimination power also showed that there were no a major epidemic N. meningitidis strain circulating in the country [Chiou CS, et el, unpublished data]. Our basic epidemiological data and the results of genotyping method by Chiou CS et al do not suggest that the 2000-Hajj epidemic directly lead to the sporadic cases of serogroup W135 meningococcal disease in Taiwan.

Few published papers discuss the clinical characteristics of serogroup W135 meningococcal disease, and all focus on cases caused by the epidemic strain [3, 4]. But these studies [3, 4] did not report finding more W135 than non-W135 patients who were older and in the 20–54 year-old age group. In case series from Europe and Saudi Arabia, patients 1–10 years-old accounted for nearly 50% of all cases of serogroup W135 meningococcal disease [4, 20]. The cause of shift in the age distribution for W135 cases in our series was not clear. Meningococcal disease due to uncommon serogroups such as X, Y, Z, W135, has been reported associated with complement deficiency [21]. In our meningococcal patients older than 16 years old with complement 50 level and C3a level available (31/62,50%), serogroup W135 did not have more patients with low complement 50 and C3a level than serogroup B (28.6%; 4/14 vs. 47.1%; 8/17)(Lin CY, unpublished data). In adult patients elder than 16 years old, only 4 patients (28.6%) of serogroup W135 meningococcal disease had an underlying immunocompromising disease (cancer, steroid use, and organ failure etc) and were not different from non-W135 group. Our study showed that 83.3% of serogroup W135 disease and 76.7% of non-serogroup B occurred in patients older than 10 years – an age group in which the disease is potentially preventable by use of the current quadrivalent vaccine. In Taiwan, the current policy is to recommend meningococcal vaccination for people travelling to an endemic country, or for use in controlling outbreaks in crowded settings such as a school or military establishment.

If continued bacteriologic surveillance showed the rapid clonal spread of non-serogroup B of meningococcus especially in the special population such as college students or military personnel, the policy of public meningococcal vaccination should be reevaluated.

In the Saudi Arabian study by Lingappa JR et al, comparing patients with the epidemic strain of serogroup W135 and those with serogroup A [3], the former were less likely to have meningitis than the latter [3]. Similarly, we found a statistically nonsignificant trend toward more pure bacteremia without meningitis among patients with sporadic W135 disease, compared to patients with non-W135 disease. Our finding that more pneumonia and less CSF inflammatory cells in patients with W135 meningitis than in patients with non-W135 meningitis was similar to findings in France that the predominant extrameningeal diseases (such as pneumonia, septic meningococcal arthritis and pericarditis) occurred predominantly in patients with W135 disease [10–13]. However, the use of a different definition, and failure to use PCR as diagnostic tool may have resulted in less association between arthritis and W135 in our series.

Similar to previous reports [22–24], we found that pure bacteremia without meningitis, altered mental status, peripheral white cell counts less than 10,000/mm³, thrombocytopenia < 100,000/mm³, initial systolic blood pressure < 85 mmHg, and purpura or petechiae on admission were factors associated with mortality in patients with meningococcal disease. There are several scoring systems developed for assessing the prognosis and mortality of meningococcal disease [22–25], Glasgow Meningococcal Septicemia Prognostic Score is a good example of a clinical prediction tool for assessing the mortality of meningococcal disease [22, 23]. Most of those scoring systems are based on easily available clinical and laboratory parameters such as differences in age distribution, period between onset of disease and admission, absence of meningitis, presence of widespread skin lesions, hypotension, metabolic acidosis, normal C-reactive protein level, absence of leukocytosis, presence of thrombocytopenia, and hypofibrinogenemia [22–25]. However, some of the above factors (e.g., C-reactive protein and fibrinogen levels) were not routinely checked in our cases.

Previous studies that included only a low percentage of patients with serogroup W135 meningococcal disease could not comment on the influence of W135 on outcome [23–25]. Our present series, with a higher percentage of patients with serogroup W135 disease, showed no difference in case-fatality rates between patients with W135 and patients with non-W135. Lingappa JR et al [3] also found no difference in the case fatality rate between patients with serogroup W135 and serogroup A in Saudi Arabia, but more patients with W135 needed intensive care. There was a statistically insignificant trend toward more shock and intubation in patients with W135 compared with those with non-W135 in our series.

In conclusion, sporadic W135 infections accounted for a substantial proportion of meningococcal disease in Taiwan from 2001 through 2003 and do not appear to be Hajj-related. More patients age 20–54 years, and not younger, developed the disease in Taiwan. The clinical features of W135 and non-W135 meningococcal disease were similar except that patients with W135 disease were more likely to have extrameningeal involvement such as pneumonia. Independent predictors of mortality included bacteremia without meningitis, altered mental status, and petechiae or purpura on admission.