Journal of Hospital Infection 90 (2015) 163e166 Available online at www.sciencedirect.com
Journal of Hospital Infection journal homepage: www.elsevierhealth.com/journals/jhin
Short report
Bordetella pertussis infection in paediatric healthcare workers K.S.A. Cunegundes*, M.I. de Moraes-Pinto, T.N. Takahashi, D.A.B. Kuramoto, L.Y. Weckx Department of Pediatrics, Federal University of Sa˜o Paulo, Sa˜o Paulo, SP, Brazil
A R T I C L E
I N F O
Article history: Received 14 April 2014 Accepted 14 February 2015 Available online 28 March 2015 Keywords: Healthcare workers Infection Pertussis Serology
S U M M A R Y
An increased incidence of pertussis has been observed recently in adults, and healthcare workers (HCWs) are considered a risk group for transmission to infants. Prevalence of recent pertussis infection was assessed in HCWs from a paediatric department of a tertiary care hospital in Brazil. Serum pertussis toxin IgG antibodies were measured by enzymelinked immunosorbent assay. Of 388 HCWs included in the analysis, 6.4% had serology suggestive of recent infection. Medical residents [odds ratio (OR): 4.15; 95% confidence interval (CI): 1.42e12.14; P ¼ 0.009] and those working >40 h a week (OR: 3.29; 95% CI: 1.17e9.26; P ¼ 0.024) had increased risk of pertussis infection. ª 2015 The Healthcare Infection Society. Published by Elsevier Ltd. All rights reserved.
Introduction An increasing incidence of pertussis has been reported in adolescents and adults in some countries. These groups are recognized as playing an important role in transmitting Bordetella pertussis infection to incompletely immunized young infants.1 Pertussis among healthcare workers (HCWs) is of special concern because of the potential nosocomial transmission to high-risk patients and other HCWs.2 In the State of Sa ˜o Paulo, Brazil, pertussis incidence was stable until 2010 when it increased in infants from 29 per 100,000 to 133 per 100,000 in 2011. However, this trend has not been observed in older age groups.3
* Corresponding author. Address: Rua Borges Lagoa, 770, Sa ˜o Paulo, CEP 04038-001, Brazil. Tel.: þ55 (11) 5576 4993. E-mail address: [email protected] (K.S.A. Cunegundes).
The aims of this study were to assess the prevalence of recent pertussis infection in HCWs working in the paediatric department of a tertiary care university hospital and to examine the factors associated with increased risk of infection in this group.
Methods This study was conducted in the paediatric department of Sa ˜o Paulo Hospital of the Federal University of Sa ˜o Paulo, Brazil, a 740-bed tertiary-care referral hospital, from October to December 2011. The department employs 689 HCWs: 343 physicians, 82 nurses, and 222 nursing assistants. A previous survey reported a prevalence of 50% of high-titre pertussis antibody in HCWs.4 Assuming a confidence level of 95% and statistical power of 80%, a minimum sample size of 385 participants was required to estimate the prevalence of pertussis infection in HCWs.
http://dx.doi.org/10.1016/j.jhin.2015.02.016 0195-6701/ª 2015 The Healthcare Infection Society. Published by Elsevier Ltd. All rights reserved.
164
K.S.A. Cunegundes et al. / Journal of Hospital Infection 90 (2015) 163e166
Physicians, nurses, and nursing assistants working in the paediatric department were contacted and asked to participate in the study. HCWs were included until the minimum sample size was achieved. The Ethics Research Committee of Federal University of Sao Paulo (0558/11) approved the study protocol and written informed consent was obtained from all participants. A self-completed questionnaire was used to assess: demographic characteristics, pertussis vaccination during childhood and adulthood, history of pertussis, cough symptoms (more than 14 days in the previous 12 months), contact with pertussis cases (suspected or confirmed) in the previous 12 months, and household contact with children. As high antibody titres of IgG anti-pertussis toxin could be a result of recent vaccination, participants were excluded from the analysis if they had been vaccinated with adult-formulation tetanusediphtheria toxoid and acellular pertussis vaccine (Tdap) during the previous 12 months.5 A blood sample was collected from each HCW. Quantitative IgG antibodies to pertussis toxin were determined by a commercial enzyme-linked immunosorbent assay (ELISA) (SERION ELISA classic Bordetella pertussis-IgG, VirionySerion, Wu ¨rzburg, Germany). A cut-off value of 62.5 IU/mL was employed as suggestive of pertussis infection within the previous 12 months. The reported sensitivity and specificity of this cut-off as an indication of recent pertussis infection are 80% and 95%, respectively.5 Statistical analysis was undertaken using Minitab 16 (Minitab, Inc., State College, Pennsylvania, USA). Categorical variables were analysed using the Pearson chi-square test or Fisher’s exact test. Continuous variables were analysed using Student’s t-test. Variables were incorporated into the multiple logistic regression analysis when they were significantly associated with recent pertussis infection in the univariate analysis. P < 0.05 was considered significant.
Results Among the 689 physicians, nurses, and nursing assistants working in the paediatric department (paediatric ward, neonatal unit, paediatric intensive care unit (PICU), emergency and outpatient clinics), 414 HCWs were contacted, and only 23 (5.5%) refused to participate in this study. Three were excluded because they reported having been vaccinated with Tdap in the previous 12 months, leaving 388 to be studied. The characteristics of participants are shown in Table I. Overall, 6.4% of HCWs had pertussis serology suggestive of infection in the past 12 months (Table II). Medical residents were analysed as a group because they work as trainees within the system and are not limited to any specific area cited above. They had the highest rate of recent pertussis infection at 19.5%. HCWs from the emergency clinic and PICU had an intermediate prevalence of recent infection (6.5% and 9.3%, respectively) when compared with other groups of HCWs. The data were analysed in three groups: low prevalence (10%) which were the medical residents.
Table I Characteristics of healthcare workers included in the study (N ¼ 388) Characteristic
Mean age (years) (range) Female Occupation Physician Nurse Nursing assistant Patient care area/group Outpatient clinics Paediatric wards Emergency Medical residents in paediatrics Paediatric intensive care unit Neonatal unit/postpartum unit Weekly working hours >40 Previous pertussis vaccine in childhood with at least three DTwP doses Knowledge on adult pertussis vaccine (Tdap) Previous vaccination with Tdap History of whooping cough Cough illness lasting 2 weeks in the previous 12 months Contact with whooping cough patients in the previous 12 months Household contact with children
Value
36.4 (21e67) 332 (85.4%) 201 (51.8%) 64 (16.5%) 123 (31.7%) 155 78 46 41 43 25 191/388 253/302a
(39.9%) (20.1%) (11.8%) (10.5%) (11.0%) (6.4%) (49.2%) (83.7%)
163/388 (42.4%) 28/388 (7.2%) 51/388 (13.1%) 67/385a (17.4%) 223/377a (59.1%) 152/387a (39.2%)
DTwP, diphtheria, tetanus and whole-cell pertussis vaccine. a Not all participants answered this question.
The variables associated with pertussis infection were patient care area, weekly working hours, and years since graduation (Table II). Comparing HCWs who reported, and those who did not report, cough in the previous 12 months, no difference in geometric mean antibodies was found (11.1 versus 11.6 IU/mL, t-test, P ¼ 0.338). On multivariate analysis, medical residents showed an increased risk of pertussis infection when compared with HCWs from other areas (OR: 4.15; 95% CI: 1.42e12.14; P ¼ 0.009). HCWs from emergency clinic and PICU (OR: 2.49; 95% CI: 0.90e6.89; P ¼ 0.080) and those who worked >40 h per week (OR: 3.29; 95% CI: 1.17e9.26; P ¼ 0.024) also had a higher risk of pertussis infection.
Discussion This study has found antibody levels suggestive of recent pertussis infection in 6.4% of paediatric HCWs. A wide range of rates of pertussis infection in HCWs have been reported by serological surveys, partly because different antigens and cut-off values were used, but also due to the different epidemiological conditions. Incidence rates vary from 1.3% among US medical residents to 6% among German paediatric personnel and 22% among emergency department staff in Australia.6e8
K.S.A. Cunegundes et al. / Journal of Hospital Infection 90 (2015) 163e166
165
Table II Incidence of Bordetella pertussis infection in healthcare workers and factors associated with this infection Characteristic
N
IgG anti-pertussis toxin 62.5 UI/mL
Age (years) (mean SD) 388 Female 332 Male 56 Occupation 388 Physician 201 Nurse 64 Nursing assistant 123 Patient care area/group 388 Emergency and PICU 89 Medical residents in paediatrics 41 Paediatric wards, neonatal unit and outpatient clinics 258 Weekly working hours 388 >40 h 191 14 days compared with the group without cough. This is similar to the findings from a study of Dutch adults and adolescents.10 The apparently low rate of symptoms reported by individuals with serological evidence of recent pertussis infection might be explained because the clinical presentation of pertussis is often atypical in adults, without classic symptoms and with a shorter duration of cough.1,2 In addition, some
166
K.S.A. Cunegundes et al. / Journal of Hospital Infection 90 (2015) 163e166
HCWs might not recall exactly what symptoms they had in the previous year. This study had some limitations. Since it was conducted in a single tertiary care centre, it may not represent the incidence rate of pertussis among paediatric staff from primary and secondary health units generally. The instrument used to assess information about participants was a questionnaire, which has a potential for recall bias. On the other hand, the results highlight that, among paediatric HCWs, the incidence of whooping cough is probably underestimated. Undiagnosed infected HCWs can act as a source of infection to susceptible contacts, including patients, other employees, and their household contacts. Outbreaks in healthcare facilities have occurred frequently and consume resources in terms of personnel, diagnostic tests, treatment, and prophylaxis.2 In an effort to control pertussis and prevent transmission to infants in healthcare facilities, some European countries have included pertussis vaccine in their vaccination programmes for HCWs. Similarly, the Advisory Committee on Immunization Practices recommended that HCWs who are unimmunized against pertussis should receive a single dose of Tdap as soon as feasible, regardless of when they last received tetanusediphtheria vaccination.2 In conclusion, a high prevalence of pertussis infection was found in paediatric HCWs, especially in medical residents. Paediatric HCWs should be considered a priority group to receive pertussis vaccine. Conflict of interest statement None declared. Funding sources FAP/UNIFESP.
References 1. Wiley KE, Zuo Y, Macartney KK, McIntyre PB. Sources of pertussis infection in young infants: a review of key evidence informing targeting of the cocoon strategy. Vaccine 2013;31:618e625. 2. Bechini A, Tiscione E, Boccalina S, Levi M, Bonanni P. Acellular pertussis vaccine use in risk groups (adolescents, pregnant women, newborns and health care workers): a review of evidences and recommendations. Vaccine 2012;30:5179e5190. 3. Centro de Vigila ˆncia Epidemiolo ´gica do Estado de Sa ˜o Paulo (CVE). Divisa ¸a ˜o de Imunizac ˜o. Informe Te ´cnico Coqueluche 2011 e Atualizac ¸˜ ao da Situac ¸˜ ao Epidemiolo ´gica. Sa ˜o Paulo: CVE; 2011. Available from: http://www.cve.saude.sp.gov.br/htm/resp/pdf/ IF11_COQUELUCHE.pdf. 4. Juanes JR, Gil A, Gonza ´lez A, Arrazola MP, San-Martin M, Esteban J. Seroprevalence of pertussis antibody among health care personnel in Spain. Eur J Epidemiol 2004;19:69e72. 5. Guiso N, Berbers G, Fry NK, He Q, Riffelmann M, Wirsing von Konig CH. What to do and what not to do in serological diagnosis of pertussis: recommendations from EU reference laboratories. Eur J Clin Microbiol Infect Dis 2011;30:307e312. 6. Wright SW, Decker MD, Edwards KM. Incidence of pertussis infection in healthcare workers. Infect Control Hosp Epidemiol 1999;20:120e123. 7. Riffelmann M, Koesters K, Saemann-Ischenko G, Schmitt HJ, Wirsing von Koenig CH. Antibodies to pertussis antigens in pediatric health care workers. Pediatric Infect Dis J 2002;21:381e383. 8. Faruque MO, Senanayake S, Meyer ADM, Dear KBG. Emergency department staff and susceptibility to pertussis: a seroprevalence study. Emerg Med Australas 2008;20:45e50. 9. Quinn HE, McIntyre PB, Backhouse JL, Gidding HF, Brotherton J, Gilbert GL. The utility of seroepidemiology for tracking trends in pertussis infection. Epidemiol Infect 2010;138:426e433. 10. de Greeff SC, Melker HE, van Gageldonk PGM, et al. Seroprevalence of pertussis in the Netherlands: evidence for increased circulation of Bordetella pertussis. PLoS One 2010;5:e14183.
Journal of Hospital Infection journal homepage: www.elsevierhealth.com/journals/jhin
Short report
Bordetella pertussis infection in paediatric healthcare workers K.S.A. Cunegundes*, M.I. de Moraes-Pinto, T.N. Takahashi, D.A.B. Kuramoto, L.Y. Weckx Department of Pediatrics, Federal University of Sa˜o Paulo, Sa˜o Paulo, SP, Brazil
A R T I C L E
I N F O
Article history: Received 14 April 2014 Accepted 14 February 2015 Available online 28 March 2015 Keywords: Healthcare workers Infection Pertussis Serology
S U M M A R Y
An increased incidence of pertussis has been observed recently in adults, and healthcare workers (HCWs) are considered a risk group for transmission to infants. Prevalence of recent pertussis infection was assessed in HCWs from a paediatric department of a tertiary care hospital in Brazil. Serum pertussis toxin IgG antibodies were measured by enzymelinked immunosorbent assay. Of 388 HCWs included in the analysis, 6.4% had serology suggestive of recent infection. Medical residents [odds ratio (OR): 4.15; 95% confidence interval (CI): 1.42e12.14; P ¼ 0.009] and those working >40 h a week (OR: 3.29; 95% CI: 1.17e9.26; P ¼ 0.024) had increased risk of pertussis infection. ª 2015 The Healthcare Infection Society. Published by Elsevier Ltd. All rights reserved.
Introduction An increasing incidence of pertussis has been reported in adolescents and adults in some countries. These groups are recognized as playing an important role in transmitting Bordetella pertussis infection to incompletely immunized young infants.1 Pertussis among healthcare workers (HCWs) is of special concern because of the potential nosocomial transmission to high-risk patients and other HCWs.2 In the State of Sa ˜o Paulo, Brazil, pertussis incidence was stable until 2010 when it increased in infants from 29 per 100,000 to 133 per 100,000 in 2011. However, this trend has not been observed in older age groups.3
* Corresponding author. Address: Rua Borges Lagoa, 770, Sa ˜o Paulo, CEP 04038-001, Brazil. Tel.: þ55 (11) 5576 4993. E-mail address: [email protected] (K.S.A. Cunegundes).
The aims of this study were to assess the prevalence of recent pertussis infection in HCWs working in the paediatric department of a tertiary care university hospital and to examine the factors associated with increased risk of infection in this group.
Methods This study was conducted in the paediatric department of Sa ˜o Paulo Hospital of the Federal University of Sa ˜o Paulo, Brazil, a 740-bed tertiary-care referral hospital, from October to December 2011. The department employs 689 HCWs: 343 physicians, 82 nurses, and 222 nursing assistants. A previous survey reported a prevalence of 50% of high-titre pertussis antibody in HCWs.4 Assuming a confidence level of 95% and statistical power of 80%, a minimum sample size of 385 participants was required to estimate the prevalence of pertussis infection in HCWs.
http://dx.doi.org/10.1016/j.jhin.2015.02.016 0195-6701/ª 2015 The Healthcare Infection Society. Published by Elsevier Ltd. All rights reserved.
164
K.S.A. Cunegundes et al. / Journal of Hospital Infection 90 (2015) 163e166
Physicians, nurses, and nursing assistants working in the paediatric department were contacted and asked to participate in the study. HCWs were included until the minimum sample size was achieved. The Ethics Research Committee of Federal University of Sao Paulo (0558/11) approved the study protocol and written informed consent was obtained from all participants. A self-completed questionnaire was used to assess: demographic characteristics, pertussis vaccination during childhood and adulthood, history of pertussis, cough symptoms (more than 14 days in the previous 12 months), contact with pertussis cases (suspected or confirmed) in the previous 12 months, and household contact with children. As high antibody titres of IgG anti-pertussis toxin could be a result of recent vaccination, participants were excluded from the analysis if they had been vaccinated with adult-formulation tetanusediphtheria toxoid and acellular pertussis vaccine (Tdap) during the previous 12 months.5 A blood sample was collected from each HCW. Quantitative IgG antibodies to pertussis toxin were determined by a commercial enzyme-linked immunosorbent assay (ELISA) (SERION ELISA classic Bordetella pertussis-IgG, VirionySerion, Wu ¨rzburg, Germany). A cut-off value of 62.5 IU/mL was employed as suggestive of pertussis infection within the previous 12 months. The reported sensitivity and specificity of this cut-off as an indication of recent pertussis infection are 80% and 95%, respectively.5 Statistical analysis was undertaken using Minitab 16 (Minitab, Inc., State College, Pennsylvania, USA). Categorical variables were analysed using the Pearson chi-square test or Fisher’s exact test. Continuous variables were analysed using Student’s t-test. Variables were incorporated into the multiple logistic regression analysis when they were significantly associated with recent pertussis infection in the univariate analysis. P < 0.05 was considered significant.
Results Among the 689 physicians, nurses, and nursing assistants working in the paediatric department (paediatric ward, neonatal unit, paediatric intensive care unit (PICU), emergency and outpatient clinics), 414 HCWs were contacted, and only 23 (5.5%) refused to participate in this study. Three were excluded because they reported having been vaccinated with Tdap in the previous 12 months, leaving 388 to be studied. The characteristics of participants are shown in Table I. Overall, 6.4% of HCWs had pertussis serology suggestive of infection in the past 12 months (Table II). Medical residents were analysed as a group because they work as trainees within the system and are not limited to any specific area cited above. They had the highest rate of recent pertussis infection at 19.5%. HCWs from the emergency clinic and PICU had an intermediate prevalence of recent infection (6.5% and 9.3%, respectively) when compared with other groups of HCWs. The data were analysed in three groups: low prevalence (10%) which were the medical residents.
Table I Characteristics of healthcare workers included in the study (N ¼ 388) Characteristic
Mean age (years) (range) Female Occupation Physician Nurse Nursing assistant Patient care area/group Outpatient clinics Paediatric wards Emergency Medical residents in paediatrics Paediatric intensive care unit Neonatal unit/postpartum unit Weekly working hours >40 Previous pertussis vaccine in childhood with at least three DTwP doses Knowledge on adult pertussis vaccine (Tdap) Previous vaccination with Tdap History of whooping cough Cough illness lasting 2 weeks in the previous 12 months Contact with whooping cough patients in the previous 12 months Household contact with children
Value
36.4 (21e67) 332 (85.4%) 201 (51.8%) 64 (16.5%) 123 (31.7%) 155 78 46 41 43 25 191/388 253/302a
(39.9%) (20.1%) (11.8%) (10.5%) (11.0%) (6.4%) (49.2%) (83.7%)
163/388 (42.4%) 28/388 (7.2%) 51/388 (13.1%) 67/385a (17.4%) 223/377a (59.1%) 152/387a (39.2%)
DTwP, diphtheria, tetanus and whole-cell pertussis vaccine. a Not all participants answered this question.
The variables associated with pertussis infection were patient care area, weekly working hours, and years since graduation (Table II). Comparing HCWs who reported, and those who did not report, cough in the previous 12 months, no difference in geometric mean antibodies was found (11.1 versus 11.6 IU/mL, t-test, P ¼ 0.338). On multivariate analysis, medical residents showed an increased risk of pertussis infection when compared with HCWs from other areas (OR: 4.15; 95% CI: 1.42e12.14; P ¼ 0.009). HCWs from emergency clinic and PICU (OR: 2.49; 95% CI: 0.90e6.89; P ¼ 0.080) and those who worked >40 h per week (OR: 3.29; 95% CI: 1.17e9.26; P ¼ 0.024) also had a higher risk of pertussis infection.
Discussion This study has found antibody levels suggestive of recent pertussis infection in 6.4% of paediatric HCWs. A wide range of rates of pertussis infection in HCWs have been reported by serological surveys, partly because different antigens and cut-off values were used, but also due to the different epidemiological conditions. Incidence rates vary from 1.3% among US medical residents to 6% among German paediatric personnel and 22% among emergency department staff in Australia.6e8
K.S.A. Cunegundes et al. / Journal of Hospital Infection 90 (2015) 163e166
165
Table II Incidence of Bordetella pertussis infection in healthcare workers and factors associated with this infection Characteristic
N
IgG anti-pertussis toxin 62.5 UI/mL
Age (years) (mean SD) 388 Female 332 Male 56 Occupation 388 Physician 201 Nurse 64 Nursing assistant 123 Patient care area/group 388 Emergency and PICU 89 Medical residents in paediatrics 41 Paediatric wards, neonatal unit and outpatient clinics 258 Weekly working hours 388 >40 h 191 14 days compared with the group without cough. This is similar to the findings from a study of Dutch adults and adolescents.10 The apparently low rate of symptoms reported by individuals with serological evidence of recent pertussis infection might be explained because the clinical presentation of pertussis is often atypical in adults, without classic symptoms and with a shorter duration of cough.1,2 In addition, some
166
K.S.A. Cunegundes et al. / Journal of Hospital Infection 90 (2015) 163e166
HCWs might not recall exactly what symptoms they had in the previous year. This study had some limitations. Since it was conducted in a single tertiary care centre, it may not represent the incidence rate of pertussis among paediatric staff from primary and secondary health units generally. The instrument used to assess information about participants was a questionnaire, which has a potential for recall bias. On the other hand, the results highlight that, among paediatric HCWs, the incidence of whooping cough is probably underestimated. Undiagnosed infected HCWs can act as a source of infection to susceptible contacts, including patients, other employees, and their household contacts. Outbreaks in healthcare facilities have occurred frequently and consume resources in terms of personnel, diagnostic tests, treatment, and prophylaxis.2 In an effort to control pertussis and prevent transmission to infants in healthcare facilities, some European countries have included pertussis vaccine in their vaccination programmes for HCWs. Similarly, the Advisory Committee on Immunization Practices recommended that HCWs who are unimmunized against pertussis should receive a single dose of Tdap as soon as feasible, regardless of when they last received tetanusediphtheria vaccination.2 In conclusion, a high prevalence of pertussis infection was found in paediatric HCWs, especially in medical residents. Paediatric HCWs should be considered a priority group to receive pertussis vaccine. Conflict of interest statement None declared. Funding sources FAP/UNIFESP.
References 1. Wiley KE, Zuo Y, Macartney KK, McIntyre PB. Sources of pertussis infection in young infants: a review of key evidence informing targeting of the cocoon strategy. Vaccine 2013;31:618e625. 2. Bechini A, Tiscione E, Boccalina S, Levi M, Bonanni P. Acellular pertussis vaccine use in risk groups (adolescents, pregnant women, newborns and health care workers): a review of evidences and recommendations. Vaccine 2012;30:5179e5190. 3. Centro de Vigila ˆncia Epidemiolo ´gica do Estado de Sa ˜o Paulo (CVE). Divisa ¸a ˜o de Imunizac ˜o. Informe Te ´cnico Coqueluche 2011 e Atualizac ¸˜ ao da Situac ¸˜ ao Epidemiolo ´gica. Sa ˜o Paulo: CVE; 2011. Available from: http://www.cve.saude.sp.gov.br/htm/resp/pdf/ IF11_COQUELUCHE.pdf. 4. Juanes JR, Gil A, Gonza ´lez A, Arrazola MP, San-Martin M, Esteban J. Seroprevalence of pertussis antibody among health care personnel in Spain. Eur J Epidemiol 2004;19:69e72. 5. Guiso N, Berbers G, Fry NK, He Q, Riffelmann M, Wirsing von Konig CH. What to do and what not to do in serological diagnosis of pertussis: recommendations from EU reference laboratories. Eur J Clin Microbiol Infect Dis 2011;30:307e312. 6. Wright SW, Decker MD, Edwards KM. Incidence of pertussis infection in healthcare workers. Infect Control Hosp Epidemiol 1999;20:120e123. 7. Riffelmann M, Koesters K, Saemann-Ischenko G, Schmitt HJ, Wirsing von Koenig CH. Antibodies to pertussis antigens in pediatric health care workers. Pediatric Infect Dis J 2002;21:381e383. 8. Faruque MO, Senanayake S, Meyer ADM, Dear KBG. Emergency department staff and susceptibility to pertussis: a seroprevalence study. Emerg Med Australas 2008;20:45e50. 9. Quinn HE, McIntyre PB, Backhouse JL, Gidding HF, Brotherton J, Gilbert GL. The utility of seroepidemiology for tracking trends in pertussis infection. Epidemiol Infect 2010;138:426e433. 10. de Greeff SC, Melker HE, van Gageldonk PGM, et al. Seroprevalence of pertussis in the Netherlands: evidence for increased circulation of Bordetella pertussis. PLoS One 2010;5:e14183.
