|
|
||||||||
1,2Antibiotic Resistance Unit, Centre of Bacteriology1 and National Observatory for Health2, National Institute of Health Dr Ricardo Jorge, Lisboa, Portugal 3Paediatric Infectious Diseases Unit, Hospital de Dona Estefânia, Lisboa, Portugal
Correspondence: Manuela Caniça, manuela.canica{at}insa.min-saude.pt
Received November 28, 2003
Accepted April 10, 2004
The first investigation of Neisseria meningitidis isolated from a large area covering an appreciable population in Portugal, before the voluntary vaccination period with the serogroup C conjugate vaccine, is reported. The serogroups and antimicrobial susceptibility of 116 isolates were studied. Serogroups C (50.0 %), B (47.4 %) and W135 (2.6 %) were found. Serogroup C was most common in the 115-years-old group and B in the less than 1-year-old and over 16-years-old groups (P = 0.042). Clinical diagnosis of meningococcal disease was primarily meningitis for patients with serogroup C and meningitis associated with sepsis for those with serogroup B. Penicillin resistance was significantly associated with serogroup C (P < 0.001). This work reinforces the importance for public health of monitoring the serogroup and antimicrobial susceptibility of isolates from patients with invasive meningococcal disease.
| INTRODUCTION |
|---|
|
|
|---|
The Antibiotic Resistance Unit (ARU) at the National Institute of Health Dr Ricardo Jorge (NIH) ran a voluntary laboratory-based surveillance programme of invasive N. meningitidis from 1995. The programme was independent of the notification provided by a Compulsory Notifiable Diseases (CND) system (DGS, 2003).
There is evidence that serogroup B was the most prevalent in the 1980s (Cunha & Carvalho, 1993), and that serogroup C emerged in Portugal in the 1990s (Ferreira et al., 1999, 2001). Nevertheless, we believed that it was important to investigate a broad population.
| METHODS |
|---|
|
|
|---|
Collection, transport and culture.
Biological samples collected in hospitals were immediately used to inoculate appropriate media and incubated at 35 °C. Samples were transported to the ARU in an incubator at 35 °C accompanied by an epidemiological questionnaire. A second questionnaire was completed and sent on the 15th day after collection.
The bacteriology laboratories of GEMVSA sent strains at 20 °C in tryptone soya broth (Oxoid) supplemented with 20 % (v/v) glycerol.
Strains, identification and serological typing.
GEMVSA laboratories provided 44 strains. The other 72 strains were isolated in the ARU from 611 suspected cases of meningitis or sepsis of microbiological origin. The non-N. meningitidis cases included viral, other bacterial, meningococcal diseases with antibiotic treatment prior to collection of CSF or blood and other diagnoses.
All 116 isolates were subcultured and identified by standard methods. Agglutination and PCR were used to determine serogroups (Taha, 2000; Ferreira et al., 2001).
Susceptibility testing.
MICs of six antibiotics were determined by an agar dilution method and the MIC of sulphadiazine was determined by E-test. One hundred and nine available isolates were tested against penicillin, and 108 were tested against ampicillin, cefotaxime, ceftriaxone, ciprofloxacin, rifampicin and sulphadiazine.
Data analysis.
We performed descriptive statistical analysis (Table 1). For the proportions of each serogroup in the sample, the respective 95 % confidence interval (CI) was calculated using the Normal approximation, except for serogroup W135, for which the Poisson approximation was used. Pearson's S2 test, with the Yates correction in the 2x2 tables was used to establish associations between serogroups B and C and the variables sex, age group, season and penicillin susceptibility. Serogroup W135 was not included in the analysis because the sample was too small. Similarly, the analysis of the association between quarter and serogroup did not include the trial study period because the numbers were too small. The S2 goodness-of-fit test was used to test if the distribution by age group in the sample was the same as that observed in the CND system to the year 2000 (DGS, 2003). For all tests we established the significance level at 5 % (0.05). SPSS software version 11.0 was used.
|
| RESULTS |
|---|
|
|
|---|
There was no association between the season and the serogroup (P = 0.665) (Table 1). There was no previous history of disease for 84.4 % of the 64 cases of IMD for which data were available (Table 1). For cases of serogroup C infection the clinical diagnosis was most often meningitis and for cases of B infection it was meningitis associated with sepsis (Table 1). Epidemiological questionnaires indicated 13 cases with complications associated with IMD, arthritis being the most frequent; the purpuric form occurred only in individuals with serogroup B meningococci (four of 13 complications, one being associated with arthritis). Among 55 questionnaires answered, only four reported sequelae (epilepsy, scabs, necrotic ulcer with haemorrhagic suffusion and one not specified), all in cases of IMD due to serogroup C. In 83.3 % of 62 cases from which meningococci were isolated, no antibiotic treatment was administered prior to sample collection. Ceftriaxone (87.1 %) was the antibiotic most widely used, followed by cefotaxime (9.7 %).
In the Alentejo-Region and in the Madeira-Autonomous-Region the number of isolates of serogroups B and C was the same (50 %). In the North-Region 48.6 % were serogroup B, 48.6 % C and 2.8 % W135, and in the Lisbon-Tagus-Valley-Region 47.3 % were serogroup B, 50 % C and 2.7 % W135. In the sample for the study only one strain from the Centre-Region was isolated and none were isolated in the Azores-Autonomous-Region. The case fatality rate was not possible to determine, because the questionnaires did not provide sufficient information.
Antimicrobial susceptibility
Sixty-eight (62.4 %) of 109 strains were susceptible to penicillin and 41 (37.6 %) had reduced susceptibility (Table 2). Overall, a statistically significant association between serogroups and penicillin susceptibility was found (P < 0.001) with a higher proportion of resistant strains in serogroup C (data not shown). Table 2 shows susceptibilities of the 108 isolates assayed against six other antibiotics.
|
| DISCUSSION |
|---|
|
|
|---|
Serogroup C meningococci were more frequently found than in previous Portuguese studies. The Paediatric Infectious Diseases Unit of the Dona Estefânia Hospital in Lisbon studied 139 meningococci between 1985 and 1988 of which 9 % were serogroup C, 90 % B and 1 % unclassifiable (L. Carvalho, personal observation). Between January 1999 and January 2002, 52 % of the meningococci in the same unit were serogroup B and 48 % C (Carvalho, 2002). Between January 1995 and December 1999, the surveillance programme conducted by ARU, with strains from nine hospitals, found 47.3 % serogroup C and 41.8 % B (Ferreira et al., 2001), consistent with a study in the Oporto District (45.5 % B and 50 % C in 22 cases) in the year 2000 (Blanco et al., 2000). These three studies reported values within the 95 % CI for this study. Thus, there is evidence of the emergence of serogroup C in the 1990s. This trend seems to continue, as shown between January 2001 and June 2003 by a hospital not included in our study: 84.2 % were serogroup C (16 of 19 strains), 10.5 % B (2 of 19 strains) and 5.3 % A (one of 19 strains) (Santos et al., 2003).
There is no evidence of substantial bias in the sample. The differences between the distributions by age group between cases reported under the CND system in 2000 (DGS, 2003) and our study were not significant (P = 0.2595). The percentages of serogroup B and C were not significantly different (P = 0.319) according to the source of N. meningitidis (72 from clinical samples and 44 from the GEMVSA laboratories). Other work suggests that there were no outbreaks during the period of the study (Gomes et al., 2001), as also indicated by the epidemiological questionnaires, and indeed the prevalence of serogroups was similar in each of the four quarters of the study. Our serogroup results were also confirmed by PCR.
The CND system reported 183 cases of meningococcal disease between January and July 2001 (DGS, 2003) including 63 serogrouped strains (39.7 % serogroup B and 55.6 % C) (Gomes et al., 2001). Our strain population over the same period includes 76 strains (35 of serogroup B, 39 of C and 2 of W135) (data not shown), with a higher proportion of serogroup B (46.1 vs 39.7 %) and a lower proportion of serogroup C (51.3 vs 55.6 %). The values for the CND system fall within our 95 % CI. This reinforces the value of laboratory collaboration for serogroup surveillance, as well as for resistance surveillance, of IMD strains in each country, as has been shown previously (Ferreira & Caniça, 2002).
The pattern of IMD in Portugal is similar to that elsewhere in Europe (Connolly et al., 1999; Noah & Henderson, 2001): most cases are children under 5 years old; a seasonal peak in winter; a higher proportion of serogroup C in the under 16-year-old group (groups attending nursery schools and schools). Of the 64 cases of IMD for which questionnaires were available, 10.9 % had previously suffered viral infection. Respiratory virus infections have been suggested as risk factors for subsequent development of meningococcal infection, possibly due to virus-induced immune suppression (Cartwright et al. 1991). Sequelae were only associated with serogroup C cases. Similarly, Healy et al. (2002) reported more multiple sequelae in patients with IMD due to N. meningitidis serogroup C than B.
Reduced susceptibility to penicillin in our strains was presumably due to the expression of PBP2 variants with low affinity for penicillin. Many of the serogroup C strains (82.9 %) had reduced susceptibility to penicillin, consistent with the 70.4 % in a previous study (Ferreira & Caniça, 2002) and high levels in other countries (Blondeau et al., 1995; Arreaza et al., 2000). More isolates had reduced susceptibility to ampicillin (60.2 %) than to penicillin, as previously shown (Arreaza et al., 2000; Ferreira & Caniça, 2002).
Resistance to sulphadiazine (56.5 %) was widespread and mostly associated with strains of serogroup C (57.4 %). This resistance has been considered a virulence marker (Holten et al., 1984). Rifampicin, the drug still used in Portugal for chemoprophylaxis, was active against 98 % of the isolates. However, two strains exhibiting intermediate resistance (MIC values of 2 mg l1) demonstrate that it is important to monitor susceptibility to this antibiotic. We know of no meningococcal isolate, including ours, with reduced susceptibility to cefotaxime or ceftriaxone.
The results of this study emphasize the importance of monitoring antimicrobial susceptibility and serogroup in IMD, because of changes in susceptibility of bacterial populations mainly to penicillin, the possibility of antigenic variation in N. meningitidis due to capsular switching and the eventual spread of new clones from other countries. Surveillance of serogroups plays a major role in evaluating vaccination strategies and then helping policy-making, but should be complemented with cost-benefit studies. Our work places the current trends in Portugal in a European context.
| ACKNOWLEDGEMENTS |
|---|
|
|
|---|
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
M. A. Hossain, D. Ahmed, T. Ahmed, N. Islam, and R. F. Breiman Increasing Isolations of Neisseria meningitides Serogroup A from Blood and Cerebrospinal Fluid in Dhaka, Bangladesh, 1999-2006 Am J Trop Med Hyg, April 1, 2009; 80(4): 615 - 618. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. du Plessis, A. von Gottberg, C. Cohen, L. de Gouveia, K. P. Klugman, and for the Group for Enteric Respiratory and Meningea Neisseria meningitidis Intermediately Resistant to Penicillin and Causing Invasive Disease in South Africa in 2001 to 2005 J. Clin. Microbiol., October 1, 2008; 46(10): 3208 - 3214. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. L. Kaplan, G. E. Schutze, J. A.D. Leake, W. J. Barson, N. B. Halasa, C. L. Byington, C. R. Woods, T. Q. Tan, J. A. Hoffman, E. R. Wald, et al. Multicenter Surveillance of Invasive Meningococcal Infections in Children Pediatrics, October 1, 2006; 118(4): e979 - e984. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Dias, D. Louro, the Antimicrobial Resistance Surveillance Program, and M. Canica Antimicrobial Susceptibility of Invasive Streptococcus pneumoniae Isolates in Portugal over an 11-Year Period. Antimicrob. Agents Chemother., June 1, 2006; 50(6): 2098 - 2105. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| INT J SYST EVOL MICROBIOL | J MED MICROBIOL | MICROBIOLOGY | J GEN VIROL | ALL SGM JOURNALS |