Pertussis is an increasingly recognized public health problem. Despite immunization during childhood, immunity wanes leading to susceptibility to infection in a significant proportion of young adults. Reliance on bacterial culture to confirm Bordetella pertussis infections has limited the ability to define the current impact of this pathogen; many cases confirmed by polymerase chain reaction have been shown to be culture negative.1 While many countries have reported increasing incidences of B. pertussis infections, in others the number of cases is stable or has decreased. Between 2000 and 2011, the annual number of confirmed B. pertussis cases in Mexico ranged from 53 to 568.2 However, studies performed in teenagers and young adults suggest that the number of B. pertussis infections may have been underestimated. Sandoval et al3 reported that 32.7% of 12- to 15-years-old students with cough of at least 14 days had pertussis. In San Luis Potosí (SLP), the number of suspected B. pertussis cases during a 10-year period was 166; only 4 were confirmed by culture.2 However, an increase in the number of cases has been noted. The introduction of molecular detection methods for B. pertussis in December 2011 in SLP allowed for a more accurate estimate of the incidence of this infection.
MATERIALS AND METHODS
Results from all probable B. pertussis cases reported to the State Epidemiological Surveillance System and analyzed at the Laboratorio Estatal de Salud Pública (SLP, México) between December 2011 and November 2012 were reviewed. Only individuals whose residence was the state of SLP were included. This analysis was approved by the Research and Ethics Committee, Servicios de Salud en el Estado de SLP.
Cases of B. pertussis infection were identified based on definitions established by the Mexican Health Ministry; patients with cough of 14 or more days with at least 2 of the following characteristics: paroxysmal, in accesses, spasmodic or with stridor and at least 1 of the following: cyanosing cough, associated hemorrhages (conjunctival patechiae, facial petechiae or epistaxis), leukocytosis with lymphocytosis or history of contact with similar cases during the 4 weeks before symptom onset were considered as probable pertussis cases.4 Nasopharyngeal swab samples were obtained for B. pertussis identification by culture and real-time PCR (qPCR) using primer sets as previously detailed.5 This assay allows for the identification of B. pertussis, B. parapertussis and B. holmesii. Quality control was carried out by the Instituto de Diagnóstico y Referencia Epidemiológicos in Mexico City; all positive samples were confirmed by repeated analyses at Instituto de Diagnóstico y Referencia Epidemiológicos.
Comparisons between cases with confirmed and negative B. pertussis results were performed using SPSS 14.0 for Windows. Categorical variables (gender and outcome) were compared using Fisher’s exact test or the χ2 test. Continuous variables (age, number of household contacts, leukocyte and lymphocyte counts) were compared using Student´s t test and Mann-Whitney U test.
During the 12-month study period, there were 147 probable B. pertussis cases reported in residents of SLP, of which 128 were <24 months, 4 were 2–12 years of age, 1 was 12–18 years of age and 14 were >18 years of age. Most patients were hospitalized, particularly those <12 months of age. Bordetella pertussis was detected by qPCR in 59 (40.1%) of 147 samples; 84 samples were negative and 4 were reported as indeterminate. There were 143 samples available for culture and B. pertussis was isolated in 12 (8.4%). All samples positive by culture were positive by qPCR. The average number of cases studied each month was 12.25 (range 5–16) and the monthly number of positive samples ranged from 1 to 7.
Based on the population of SLP in 2010 (2,585,518 inhabitants),6 the incidence of B. pertussis infection was 2.3/100,000 inhabitants; for infants <1 year of age, the incidence was 109.32/100,000 inhabitants, while for infants 1 to <2 years of age the incidence was 10.56/100,000 inhabitants.
Characteristics of confirmed and negative B. pertussis cases are presented in Table 1. Among suspected cases that were 2 months of age and older, only 36 (37.9%) had received the recommended number of immunizations according to age. There were 6 deaths among the cases included in this analysis; 5 of them corresponded to children with confirmed B. pertussis infection by qPCR. All infants who died were 1–4 months of age. Of the infants that died, 5 had not received any vaccine (4 with confirmed B. pertussis and 1 with negative result) and 1 child (with confirmed B. pertussis infection) had received 1 vaccine dose. The death rate for B. pertussis-confirmed cases was 0.19 per 100,000 inhabitants, while the estimated rate in children <1 year of age was 10.72 per 100,000 children.
Samples from household contacts of 141 suspected cases were also analyzed; in average, 5.8 contacts were analyzed for every index case (range 1–24). Overall, 816 contacts were studied and 66 were positive by qPCR and/or culture (B. pertussis, 64; B. parapertussis, 2); results were indeterminate in 23 contacts. Carriers of Bordetella sp. were found in the household of 39 suspected cases (31 cases in which B. pertussis had been identified and 8 index cases with negative or indeterminate results). At least 1 positive contact was found in 52.5% of households of B. pertussis-confirmed cases compared with 9.1% of households in which the index case was negative (P < 0.001). Thus, in addition to the 59 confirmed B. pertussis infections, there were 8 cases with epidemiologic evidence of Bordetella infection. The probability to detect a positive contact increased as the number of tested persons increased: at least 1 positive result was obtained in 11.1% households where 1–4 contacts were studied, 27.4% in households where 5–9 subjects were sampled and 87.5% in households where ≥10 individuals were sampled.
Pertussis continues to be a significant public health problem worldwide.7 Several countries have reported increasing numbers of B. pertussis infections while in others there continues to be a reduction in the number of cases. The reasons for the variability in epidemiology of this pathogen have not been fully determined. Factors that could be responsible for differences observed between countries include demographic changes, changes in immunization policies and vaccine composition, B. pertussis evolution, differences in patient management and changes in diagnostic methods. In Mexico, the reported incidence of pertussis over the last 2 decades has been low (<1 per 100,000 inhabitants). The current pertussis immunization schedule in Mexico includes administration of 4 doses of acellular pertussis vaccine and a booster at 4 years of age with DPT vaccine.2 However, recent analyses suggest that the number of cases may be increasing and that the incidence of this disease may be significantly higher than previously estimated.2,3 The use of qPCR for B. pertussis detection resulted in a significantly higher number of confirmed cases in our state compared with previous years. Between 2000 and 2010 there had been only 4 confirmed cases of B. pertussis infection (by culture) in the state of SLP.2 In contrast, during the 12-month period included in this analysis there were 59 confirmed B. pertussis index cases; the incidence of laboratory-confirmed pertussis cases was 2.3 per 100,000 inhabitants. This figure is significantly higher than previous estimates for our country.2
Most cases occurred in young children, an age distribution similar to that reported in other Latin American countries.8 Although pertussis infections in several developed countries are reported frequently in adults, most hospitalizations occur in infants, and nearly all mortality from pertussis occurs in the first few months of life, as confirmed by the current study.7
The overall mortality rate was low (0.19/100,000 population) compared with the estimated pertussis mortality rate for Latin American/Caribbean countries (1.14/100,000 population, based on data presented by Lopez et al9). However, the case-fatality rate for confirmed B. pertussis infections was 8.5%. This high case-fatality rate probably reflects the fact that many cases occurred in hospitalized young infants who had not been fully immunized due to their young age. In contrast, fatality rates in developed countries are lower (0.5–2%).10 Because most infants with pertussis (including all of those who died) were very young, additional prevention strategies should be considered.
A limitation in our study is that the case definition that we used may not identify older children and adults with B. pertussis infections who can present with less prominent clinical features. In addition, detection of B. pertussis in contacts of patients with negative results suggests that even with the use of qPCR some cases can be missed if a complete epidemiologic evaluation is not carried out.
In summary, pertussis incidence in our region is significantly higher than previously estimated. Most cases and all deaths were identified in infants ≤4 months of age who are too young to be protected by postnatal immunization. The implementation of new strategies to reduce the mortality associated with this disease is warranted.
1. Heininger U. Pertussis
: what the pediatric infectious disease specialist should know. Pediatr Infect Dis J. 2012;31:78–79
3. Sandoval PT, Arreola LP, Quechol GR, et al. [Bordetella pertussis
in adolescent students in Mexico
City]. Rev Saude Publica. 2008;42:679–683
4. Secretaría de Salud. . .. Dirección General de Epidemiología; Manuales simplificados enfermedades prevenibles por vacunación. 2005 México, DF:
5. Tatti KM, Sparks KN, Boney KO, et al. Novel multitarget real-time PCR assay for rapid detection of Bordetella species in clinical specimens. J Clin Microbiol. 2011;49:4059–4066
7. Tan T, Trindade E, Skowronski D. Epidemiology
. Pediatr Infect Dis J. 2005;24(5 suppl):S10–S18
8. Ulloa-Gutierrez R, Hozbor D, Avila-Aguero ML, et al. The global pertussis
initiative: meeting report from the Regional Latin America
Meeting, Costa Rica, 5-6 December, 2008. Hum Vaccin. 2010;6:876–880
9. Lopez AD, Mathers CD, Ezzati M, et al. Global Burden of Disease and Risk Factors. 2006 Washington, DC: The World Bank;
10. Tanaka M, Vitek CR, Pascual FB, et al. Trends in pertussis
among infants in the United States, 1980-1999. JAMA. 2003;290:2968–2975