Susceptibility of pneumococci causing meningitis in Spain and prevalence among such isolates of serotypes contained in the 7-valent pneumococcal conjugate vaccine

Sir, Streptococcus pneumoniae remains the most common cause of bacterial meningitis in children in the USA,1 although since the introduction of the 7-valent pneumococcal conjugate vaccine (PCV-7) rates have decreased despite the increase in meningitis caused by non-PCV-7 serotypes.2 In Spain the decrease in the prevalence of PCV-7 serotypes among the global population of invasive pneumococci after the introduction of PCV-7 was not as marked as in the USA, probably due to irregular and lower coverage. After vaccine introduction in 2001, distribution was via the private market because of the selective introduction into childhood vaccination calendars (it was only introduced in the Madrid region from November 2006).3,4 A recently published ecological analysis of invasive isolates over time in Spain suggested that PCV-7 vaccination in children had produced a herd effect (with respect to prevalence of PCV-7 isolates and antibiotic susceptibility) in adults.3 It has also been reported that the incidence of pneumococcal meningitis among children <5 years old significantly decreased in Spain from 2001 to 2006, without evidence of changes in the incidence of meningitis caused by non-vaccine serotypes.5 Although a small proportion of invasive pneumococcal infections present as meningitis, it has a high case-fatality rate. Whether the empirical use of cefotaxime for meningitis needs to be continued in countries with increasing PCV-7 uptake needs to be monitored. We considered isolates from CSF received in the Spanish Reference Laboratory for Pneumococci (SRLP) in the current decade to analyse their susceptibility and prevalence of PCV-7 serotypes. All CSF isolates of S. pneumoniae sent voluntarily from all over the country to the SRLP (passive, laboratory-based surveillance system) from January 2000 to December 2008 were analysed. Isolates were serotyped by Quellung reaction and/or dot blot assay, and susceptibility was determined by agar dilution.3,4 Current CLSI meningitis susceptibility breakpoints for penicillin (MIC ≤ 0.06 mg/L) and cefotaxime (MIC ≤ 0.5 mg/L), and susceptibility breakpoints of MIC ≤ 1 mg/L for vancomycin and MIC ≤ 2 mg/L for levofloxacin were used.6 Trends over time were explored by linear regression analysis. P ≤ 0.05 was considered significant. Data are shown in Table 1. Of the 1397 CSF isolates received between January 2000 and December 2008, 923 (66.1%) were from adults and 474 (33.9%) from children ≤14 years of age. No significant trends in the percentage of CSF isolates among invasive isolates were found in the three populations: total population (R2 = 0.008, P = 0.823), adults (R2 = 0.240, P = 0.180) and children (R2 = 0.395, P = 0.070), although in children there was a continuous decrease from 15.5% in 2003 to 9.8% in 2008. Total number of S. pneumoniae invasive isolates, number and percentage of CSF isolates, percentage of PCV-7 serotypes among CSF isolates, and susceptibility to penicillin and cefotaxime of CSF isolates in the total population, in adults and in children PEN, penicillin; CTX, cefotaxime; St, serotypes. Percentage susceptibility refers to total CSF isolates: a≤0.06 mg/L; and b≤0.5 mg/L. Total number of S. pneumoniae invasive isolates, number and percentage of CSF isolates, percentage of PCV-7 serotypes among CSF isolates, and susceptibility to penicillin and cefotaxime of CSF isolates in the total population, in adults and in children PEN, penicillin; CTX, cefotaxime; St, serotypes. Percentage susceptibility refers to total CSF isolates: a≤0.06 mg/L; and b≤0.5 mg/L. The prevalence of PCV-7 serotypes among CSF isolates showed significant decreasing linear trends in the total population (R2 = 0.914, β = −0.956, P < 0.001), adults (R2 = 0.819, β = −0.905, P < 0.001) and children (R2 = 0.870, β = −0.933, P < 0.001), with a significantly higher decreasing slope in children than in adults (B coefficient = −6.307, 95% CI = −8.485 to −4.128 in children versus B coefficient = −3.495, 95% CI = −4.963 to −2.027 in adults). Penicillin susceptibility showed a significant increasing linear trend in the study period when analysing the total population (R2 = 0.741, β = 0.861, P = 0.003) and children (although with lower correlation: R2 = 0.584, β = 0.764, P = 0.017), but not in adults (R2 = 0.150, P = 0.303). In the period 2001–08 susceptibility rates were >80% to cefotaxime and 100% to vancomycin and levofloxacin. In the 2000s there was an increase in the number of invasive isolates sent voluntarily to the SRLP (Table 1), probably reflecting greater awareness of and interest in invasive pneumococcal disease by paediatricians, and on the effect of PCV-7 introduction on susceptibility and serotype ecology among the medical community.3,4 PCV-7 was introduced in Spain in 2001 with distribution (doses/1000 inhabitants ≤ 59 months age/year) increasing from 193.34 in 2002 through 392.64 in 2005 and 411.90 in 2007 up to ∼500 in 2008.3 Data in this study indicate that PCV-7 vaccination in children had a herd effect in adults since the prevalence of PCV-7 serotype isolates among CSF isolates significantly decreased not only in children but also in adults. However, this herd effect was not demonstrated when analysing penicillin susceptibility in CSF isolates, because the significant linear increase in penicillin susceptibility in the total population and in children could not be shown in adults, in contrast to previous studies analysing all invasive isolates where penicillin susceptibility increased in both populations.3 Nevertheless the non-susceptibility rates to penicillin (and to a lesser extent to cefotaxime), using meningitis breakpoints, indicate the need to maintain cefotaxime or ceftriaxone plus vancomycin as empirical treatment in suspected pneumococcal meningitis in Spain. This study was carried out as part of our routine work. O. R. received funding from the Spanish Network for Research in Infectious Diseases (REIPI RD06/0008). None to declare.