In Taiwan, as data from the Taiwan government's Department of Health (Year 2009) has shown, the elderly accounted for 10.63% of the total (23,119,772) Taiwanese population whereas the age group of 15-64 years accounted for 73.03% . These figures of age group distribution may mean that elderly persons accounted for 12.71% (10.63/83.66) of the group of persons with an age of ≥ 15 years. In the present study, we may find that the elderly group accounted for 34.8% (87/261) of the overall ABM patients with an age ≥ 17 years, and this figure of incidence is much higher than that of similar reports on bacterial meningitis in the US (1998-2007) which showed a 20% incidence of elderly patients among the overall ABM patients . This figure of incidence may also confirm the belief that the elderly adults are more vulnerable to infectious diseases including bacterial meningitis than non-elderly adults. The increasingly aged population in Taiwan  may also indicate that the burden of bacterial meningitis will increase gradually in the elderly population in Taiwan.
As to the implicated pathogens of the 87 elderly ABM patients, 88.5% (77/87) of them belonged to monomicrobial infection, while the other 11.5% (10/87), mixed infection. As to the implicated pathogens of the monomicrobial infection, G(-) pathogens accounted for 55.8% (43/77) of them, while the other 44.2%, G(+) pathogens. This distribution pattern of implicated pathogens was similar to that of non-elderly ABM cases, in which G(-) pathogen accounted for 56.9% (43/77) of them, while the other 43.1%, G(+) pathogens. Despite the fact that there were minor differences in the implicated G(-) and G(+) pathogens of the elderly and non-elderly AMB with monomicrobial infection, K. pneumoniae and staphylococcal species were the most common in the implicated G(-) and G(+) pathogens, respectively, of both elderly and non-elderly groups of ABM patients. These relative frequencies were also similar to the reported implicated pathogens of overall ABM in Taiwan , but were different to those reported in other studies of elderly ABM [2–4], in which Streptococcus pneumoniae, Neisseria meningitidis and Listeria monocytogenes were common pathogens. This difference in implicated pathogens may reflect the believing that several factors including geographic distribution may influence the epidemiologic trend of ABM [6, 7].
As shown in Table 2 several clinical and laboratory factors were different between the elderly and non-elderly ABM patients; but among them, only a lower incidence of peripheral blood leukocytosis was of statistical significance. This phenomenon of immune senescence reflected in the finding of blunt leukocytosis response in elderly patients [11–14]. Although several other factors did not reach a statistical significance, the relatively higher incidence of female patients in the elderly group can be attributed to the relatively longer life expectancy in females in Taiwan . NPC was the most common malignancy among the 24 non-elderly ABM cases with malignancy. The relatively high incidence of NPC in the patients aged 40 to 50 years and its association with G(-) bacterial meningitis in Taiwan has been previously studied [15–17]. DM is an important preceding factor of ABM in Taiwan, especially in those with K. pneumoniae infection . The relatively higher incidence of DM in the elderly ABM group can be attributed to the fact that the prevalence of DM increases with age .
It is known that, in acute bacterial meningitis, old-age is a grave prognostic factor [5–7], and as shown in this study, the elderly ABM patients had a relatively higher mortality rate (43%, 34/87) than non-elderly ABM patients (26%, 48/147), although this age difference did not show a significant influence on the mortality rate. Several factors are known to have an influence on the therapeutic result of ABM [5–7]. In present study, only the presence of seizure and shock were of significant influence. The important influence of the presence of seizure and shock on the prognosis of ABM is also noted in other studies of bacterial meningitis [2, 7, 20].
There are several limitations of this study that are worth noting. First, because of the defined criteria used to include studied cases, some of the patients with atypical clinical and laboratory presentations would be missed, e.g. patients without culture-proved bacterial meningitis were not included. Second, this is a retrospective study; several factors, such as nutritional status, were not recorded in charts; therefore, they were not included for clinical and prognostic analysis. Third, no data from nation-wide study are available and therefore the real estimated incidence of ABM among the elderly individuals cannot be estimated accurately. And fourth, the long study period may integrate different issues of the global epidemiology because of modification of surgical procedures, antibiotic prophylaxis and antimicrobial resistance.