Measures to Control an Outbreak of Pertussis in a Neonatal Intermediate Care Nursery After Exposure to a Healthcare Worker
Background. Hospitalized premature infants are particularly vulnerable to morbidity and mortality from pertussis. Effective prevention and investigative and control measures are not well described.
Objective. To identify the source of nosocomial pertussis in a 2‐month‐old premature infant in a neonatal intermediate care nursery (ICN) and to critically review the investigation and outbreak control measures.
Setting. An ICN and a neonatal intensive care unit.
Methods. We queried healthcare workers (HCWs) and family members about cough illness and contacted potentially exposed patients to determine whether they had symptoms of pertussis. Culture and polymerase chain reaction (PCR) testing for Bordetella pertussis were performed by the hospital laboratory with specimens collected from symptomatic patients and HCWs. Levels of pertussis toxin immunoglobulin G antibodies were measured in HCWs with cough of at least 14 days' duration at a public health laboratory. Extensive control measures were instituted.
Results. Four ICN HCWs met the clinical case definition for presence of pertussis. Serologic test results were positive for 3 of the HCWs. The primary case patient was a 36‐year‐old HCW with a cough illness of 3‐weeks' duration that was accompanied by paroxysms, whoop, posttussive emesis, and pneumothorax. Among the 4 affected HCWs, the duration of cough illness prior to identification of the infant index patient ranged from 11 to 25 days. Outbreak control measures included isolation of the infant case patient, furlough and treatment of symptomatic HCWs, administration of chemoprophylaxis to contacts, and surveillance for additional cases. Seventy‐two infant patients and 72 HCWs were exposed and were given antibiotic prophylaxis. One additional case of pertussis, confirmed by PCR and culture, occurred in a resident physician who declined prophylaxis; she had cared for the index patient but had no contact with symptomatic HCWs.
Conclusion. HCWs or patients may serve as the source of pertussis in nosocomial outbreaks, which can result in substantial morbidity and outlay of resources for control measures. Our review suggested that a diagnosis of pertussis should be an early consideration for HCWs with cough illness. Targeted pertussis immunization of HCWs, employee health policies that provide for testing and furlough of HCWs with prolonged cough, and monitoring of HCWs for compliance with infection control measures could reduce the morbidity and costs associated with pertussis outbreaks. These measures will require evaluation of their effectiveness.
Received July 28, 2005; accepted December 7, 2005; electronically published May 24, 2006.
As the number of pertussis cases reported annually in the United States has risen, nosocomial outbreaks have been recognized with increasing regularity. Sources of nosocomial pertussis have included pediatric patients, their parents or visitors, and occasionally healthcare workers (HCWs).1‐4 Because pertussis can cause fatal infection in young infants, hospitalized neonates constitute a particularly vulnerable population. Although outbreaks in newborn nurseries and neonatal intensive care units (NICUs) are uncommonly reported, they can be associated with significant morbidity, and are labor‐intensive and costly to control.5,6 We evaluated the measures used to control an outbreak of pertussis in a neonatal intermediate care nursery (ICN) in order to identify policies and interventions that might prevent future outbreaks or improve the effectiveness of control measures in such high‐risk settings.
On August 6, 2003, a 2‐month‐old infant (the index patient), born at 26 weeks' gestation, developed cough and apnea. The infant had been hospitalized in an ICN since birth for complications related to prematurity. On August 8, the infant was transferred from the ICN to a NICU. Mechanical ventilation was required from August 8 through 21. Pertussis was suspected on August 15 and confirmed by polymerase chain reaction (PCR) testing of a nasopharyngeal wash fluid specimen. Droplet precautions were instituted on August 18. The infant was treated with a 5‐day course of azithromycin. The infant recovered and was discharged from the hospital 2 months after the diagnosis of pertussis.
Methods
An investigation was conducted according to the recommendations of the Centers for Disease Control and Prevention Guidelines for Pertussis Case Investigation and Surveillance7: (1) identification, testing, and treatment of suspected case patients; (2) administration of chemoprophylaxis to close and high‐risk contacts; (3) vaccination of infants; and (4) active surveillance for 42 days. A clinical case of pertussis was defined as a cough illness of 2 weeks' duration in an individual who had contact with the index patient.7 Paroxysms, whoop, and posttussive emesis were not required to meet the case definition. A case was confirmed by a positive result of culture or PCR for Bordetella pertussis or by detection of IgG antibody to pertussis toxin at a level of greater than 20 μg/mL with a standardized assay developed by the Department of Public Health, State Laboratory Institute (Jamaica Plain, MA) for patients with cough for more than 2 weeks.8 An exposure was defined as direct contact with the respiratory, oral, or nasal secretions of a case patient, direct face‐to‐face contact with a case patient who was symptomatic, or close proximity (eg, adjacent chairs in a conference room) to a case patient for more than 1 hour.
All adult contacts of the index patient, including family members and HCWs, were queried about cough illness. Adults with cough illness of any duration were tested by culture and PCR. HCWs with cough illness that met the clinical case definition were furloughed with pay until they had completed a 5‐day course of azithromycin (500 mg on day 1 followed by 250 mg daily for 4 more days) or until after the 21st day of cough illness.7 HCWs with cough illness that did not meet the clinical case definition were furloughed until B. pertussis DNA was no longer detectable by PCR in specimens from the nasopharynx. Other ICN patients with apnea, respiratory distress, or cough had droplet precautions instituted and were tested for nasopharyngeal carriage of B. pertussis by PCR and culture.
Adult contacts were treated with azithromycin for 5 days. Infant case patients and contacts were also treated with azithromycin (10 mg/kg on day 1 followed by 5 mg/kg daily for 4 more days). Exposed infants 6 weeks of age or older were vaccinated with diphtheria‐tetanus‐acellular pertussis vaccine. Use of an accelerated vaccination schedule was discussed with the primary care physicians of eligible infants.
Parents of exposed infants were given oral and written information about the potential benefits of azithromycin therapy for the prevention of pertussis and about the hypothetical risk of infantile hypertrophic pyloric stenosis.9‐11 The primary care physician of every infant for whom prophylaxis was recommended was also provided with oral and written information.
Outbreak Evaluation and Control Measures
Identification, Testing, and Treatment of Suspected Case Patients
The index patient’s 33‐year‐old mother was asymptomatic. She was the sole family contact of the infant, and she had held, kissed, and provided care to the infant throughout the cough illness. The mother was treated with azithromycin prophylaxis and did not develop cough.
Four HCWs had cough illness that met the clinical case definition for pertussis. HCW A was a 36‐year‐old nurse in the ICN with onset of cough illness on July 12. Despite paroxysms, whoop, posttussive emesis, a spontaneous pneumothorax, and multiple visits for medical care, pertussis was not suspected until diagnosis of the case in the infant. PCR and culture performed 38 days after the onset of cough illness were negative for B. pertussis. IgG to pertussis toxin was detected at a level suggesting recent infection.
Between the onset of her cough illness and the onset of symptoms in the ICN patient, HCW A worked 7 shifts in the ICN. She was never assigned to care for the index patient but recalled feeding the infant once. She reportedly wore a mask during this encounter. HCW A had contact with 27 other ICN nurses during the shifts worked after the onset of cough; contact with other HCWs, such as physicians, respiratory therapists, and ancillary personnel, could not be determined by review of ICN records.
Two additional HCWs (HCW B and HCW C, both nurses in the ICN) reported the onset of cough illness on July 31. The first day worked after onset of cough was August 4 for HCW B, and August 5 for HCW C. A fourth ICN nurse (HCW D) developed cough on August 5 (Figure). PCR and culture performed on August 20 and 21 were negative for B. pertussis, but serologic test results confirmed acute B. pertussis infection in HCW C and HCW D. HCW C and HCW D had each worked with HCW A on at least 1 occasion after onset of her cough, but none of the three had been assigned to care for the infant case patient.
Figure. Time line of the pertussis outbreak in the neonatal intermediate care nursery (ICN). “Infant” designates the index patient.
Four additional HCWs, including 2 nurses and 2 respiratory therapists, reported cough illnesses that lasted less than 14 days. For these HCWs, PCR and culture were negative for B. pertussis. Acute‐phase and convalescent‐phase serologic testing was not performed.
Chemoprophylaxis for Close and High‐Risk Contacts
Azithromycin prophylaxis was recommended for all infants in the ICN from July 12 to August 8, infants in the NICU within a one‐bed radius of the index patient, and infants with exposure to a HCW who had cough illness that met the clinical case definition. Seventy‐two patients met criteria for prophylaxis. No case of pyloric stenosis developed in treated infants. Azithromycin prophylaxis was also recommended for 61 HCWs, including all HCWs in the ICN and any other HCW who had contact with the patient while in the ICN or NICU.
Vaccination
Two hospitalized infants were given diphtheria‐tetanus‐acellular pertussis vaccine. No HCW was reimmunized against pertussis (no pertussis vaccine was licensed for adult use at the time of the outbreak).
Active Surveillance
Eight exposed infants developed cough, apnea, or feeding difficulties and were tested for pertussis; PCR and culture results were negative. On August 23, a 29‐year‐old resident physician in her first trimester of pregnancy (HCW E) developed rhinorrhea and, subsequently, cough. The physician had examined the index patient on August 16, 17, and 18 without wearing a surgical or procedural mask. The infant was mechanically ventilated at the time of the examination; the physician denied suctioning the patient but did examine the oropharynx. Azithromycin prophylaxis was recommended for HCW E on August 20 but was not initiated. A specimen obtained from HCW E on August 27 was positive for B. pertussis by PCR on August 28, and B. pertussis was isolated from culture. The physician denied contact with HCWs A, B, C, and D and had no other known exposure to pertussis or individuals with prolonged cough illness. Azithromycin prophylaxis was offered to contacts of HCW E, including 7 patients, 11 HCWs, and 3 members of her family. No additional case of pertussis was identified.
Data about the financial impact of this outbreak were limited. Verifiable costs included employee testing and prophylaxis ($3,451) and employee furloughs ($6,312). Costs related to nonemployees ($731) were primarily for testing and prophylaxis for medical staff, residents, and students. The cost of prophylaxis for exposed infants, most of whom received therapy after discharge, was not available. Health departments in 3 counties provided prophylaxis for family members of the case patients. Costs related to the care of the infant case patient were not available.
Discussion
Control of pertussis outbreaks in hospitals requires prompt recognition and isolation of cases, furlough and treatment of infected HCWs, and administration of chemoprophylaxis to close contacts. This outbreak in an ICN highlights some of the difficulties of effectively instituting control measures.
Physicians may fail to consider pertussis as a cause of cough illness in adults.12‐15 In this outbreak, HCW A repeatedly sought medical care, but despite the presence of whoop, posttussive emesis, and pneumothorax, the diagnosis of pertussis was not considered until the diagnosis of the infant case. Pertussis can be a similarly elusive diagnosis in premature infants; the associated apnea can be difficult to distinguish from the apnea of prematurity, bacterial sepsis, nosocomial pneumonia, or viral infection. In the infant described in this report, pertussis was not initially considered as a cause of the illness, despite cough and apnea. Early consideration of pertussis and institution of droplet precautions could have reduced the number of exposed individuals.
The sensitivity of culture and PCR for detection of B. pertussis decreases with the duration of symptoms. In this outbreak, IgG antibodies against pertussis toxin were assayed by the Massachusetts Department of Health, which has population‐based norms for individuals 11 years of age or older (at the Massachusetts State Laboratory Institute, testing for anti–pertussis toxin IgG has a specificity of 99% and a sensitivity of 63%).8 The lack of such standardization by commercial assays precludes the use of serologic testing for routine diagnosis of pertussis except in Massachusetts.
Early treatment of close contacts of a pertussis case patient with a macrolide antibiotic can prevent infection or clear B. pertussis from the nasopharynx and thus reduce secondary spread.16 However, defining “close contacts” in a healthcare setting is difficult. In general, HCWs caring for pediatric patients are considered at high risk for exposure to pertussis. Performing a physical examination, feeding, bathing, suctioning, intubation, and other activities that require close or prolonged contact with a patient with pertussis constitute significant exposure.17
For patients, exposure is usually defined as sharing a room or common living space with a patient who has pertussis or receiving direct care from an HCW with pertussis. This investigation highlights some of the pitfalls in establishing who provided care in a neonatal care setting. HCW A was considered a possible source of infection for the infant case patient only because she recalled one close contact in which she held the infant during a feeding. According to daily work logs, she was never assigned to care for the index patient, and HCWs B, C and D were likewise not assigned to care for the infant and recalled no specific contact. In this ICN, as in others, HCWs worked collaboratively and assist with the care of infants throughout the unit. Which HCW performed tasks such as feeding, bathing, and diaper changing was often not recorded.
This investigation also illustrates that pertussis transmission from an HCW to a susceptible infant may be unpredictable, occurring after a relatively brief exposure. The number of contact hours documented did not correlate with transmission, as HCWs A, B, C and D had prolonged contact with a number of other infants in the ICN, none of whom developed pertussis.
The criteria relating to exposure of one HCW to another are not well‐defined.18,19 Transmission of pertussis has been documented after direct face‐to‐face contact with a symptomatic case patient, with shared confined space in close proximity for 1 hour or more, and after contact with respiratory, oral, and/or nasal secretions. HCWs B, C, and D each reported direct face‐to‐face contact with HCW A, as did other HCWs who did not develop apparent B. pertussis infection. The retrospective nature of this investigation limited our ability to draw conclusions about risk factors associated with transmission between HCWs. We considered all HCWs in the ICN exposed and therefore offered prophylaxis, because the type of exposure that results in infection is often unpredictable and because HCWs can transmit pertussis to susceptible infants who are at risk of developing severe disease.7
A course of therapy with a macrolide antibiotic (erythromycin, azithromycin, or clarithromycin) is the standard treatment for and prophylactic regimen for pertussis.20 In neonates, erythromycin therapy has been associated with idiopathic infantile hypertrophic pyloric stenosis,21 and use of azithromycin for prophylaxis against pertussis in exposed infants less than 6 months of age currently is not approved.22,p2682 In this outbreak, azithromycin was used for treatment and chemoprophylaxis for both infants and adults because of anticipated benefits of improved compliance and fewer gastrointestinal side effects.23 The risk of infantile hypertrophic pyloric stenosis after azithromycin therapy is unknown. Education of parents and HCWs about the potential risks of infantile hypertrophic pyloric stenosis was an important component of this outbreak investigation, and azithromycin was well tolerated, as in previous outbreaks.24,25
In outbreaks, vaccination with diphtheria and tetanus toxoids and acellular pertussis vaccine can be started at 6 weeks of age, with the 2 subsequent doses of the primary series given at intervals of 4 weeks or more.26 Most of the exposed infants in this outbreak had been discharged from the hospital by the time the exposure was identified.
Reimmunization of HCWs against pertussis is an attractive concept.27 In June 2005, the US Food and Drug Administration licensed a tetanus and diphtheria toxoids and acellular pertussis vaccine with indications for adolescents and adults.28 Booster immunization against pertussis could be a valuable tool for preventing disease among HCWs and reducing the number of healthcare‐associated outbreaks of pertussis.
Nosocomial outbreaks of pertussis result in substantial morbidity and cost. In this outbreak, we assessed only a few of the direct costs; the real costs were much greater.29 Pertussis is common in HCWs, particularly those who work with pediatric patients, and HCWs can serve as the source of pertussis in nosocomial outbreaks.1,25 Estimates of the annual incidence of pertussis among HCWs range from 1% to 3.6%.30 Early recognition of the signs and symptoms of pertussis in HCWs is essential to prevent ongoing nosocomial transmission.
A diagnosis of pertussis should be an early consideration for hospital employees with cough illness, particularly those with paroxysms or posttussive emesis. Current employee health policies exclude from work only HCWs with febrile respiratory illness and will not effectively identify HCWs with pertussis, in whom fever is uncommon. HCW A was afebrile and continued to work in the ICN despite her cough illness. Earlier recognition of signs and symptoms compatible with pertussis could have reduced the opportunities for nosocomial transmission and the number of exposed individuals. Noncompliance with recommended prophylaxis by HCW E resulted in pertussis in the HCW, and as a consequence a new investigation and administration of prophylaxis to additional contacts were required. Employees and employee health personnel would benefit from education about the signs and symptoms of pertussis in HCWs, as well as the importance of prophylaxis both for personal protection and to control the outbreak. HCWs have a responsibility to their patients and their coworkers to seek medical evaluation for prolonged cough illness and should consider booster vaccination with a tetanus and diphtheria toxoids and acellular pertussis vaccine to prevent pertussis. Revised policies for evaluating potential cases are needed.
References
- 1. Kurt TL, Yeager AS, Guenette S, et al. Spread of pertussis by hospital staff. JAMA 1972; 221:264‐267.
- 2. Shefer A, Dales L, Melson M, Wener B, Baron R, Jackson R. Use and safety of acellular pertussis vaccine among adult hospital staff during an outbreak of pertussis. J Infect Dis 1995; 171:1053‐1056.
- 3. McCall BJ, Tilse M, Burt B, Watt P, Barnett M, McCormack JG. Infection control and public health aspects of a case of pertussis infection in a maternity health worker. Commun Dis Intell 2002; 26:584‐586.
- 4. Valenti WM, Pincus PH, Messner MK. Nosocomial pertussis: possible spread by a hospital visitor. Am J Dis Child 1980; 134:520‐521.
- 5. Spearing M, Horvath RL, McCormack JG. Pertussis: adults as a source in healthcare settings. Med J Australia 2002; 177:568‐569.
- 6. Alles S, Watson B, Dash G. Role of healthcare workers in a pertussis outbreak in a neonatal intensive care unit. Paper presented at: Neonatal Vaccination Workshop; 2004; McClean, VA.
- 7. Centers for Disease Control and Prevention. Guidelines for the control of pertussis in outbreaks. Centers for Disease Control and Prevention: Atlanta, GA, 2000. Available at: http://www.cdc.gov/nip/publications/pertussis/guide.htm. Accessed May 12, 2006.
- 8. Yih KW, Lett SM, des Bignes FN, Garrison KM, Sipe PL, Marchant CD. The increasing incidence of pertussis in Massachusetts adolescents and adults, 1989‐1998. J Infect Dis 2000; 182:1409‐1416.
- 9. Cooper WO, Griffin MR, Arbogast P, Hickson GB, Gautam S, Ray WA. Very early exposure to erythromycin and infantile hypertrophic pyloric stenosis. Arch Pediatr Adolesc Med 2002; 156:647‐650.
- 10. Mahon BE, Rosenman MB, Kleinman MB. Maternal and infant use of erythromycin and other macrolide antibiotics as risk factors for infantile hypertrophic pyloric stenosis. J Pediatr 2001; 139:380‐384.
- 11. Honein MA, Paulozzi LJ, Himelright IM. Infantile hypertrophic pyloric stenosis after pertussis prophylaxis with erythromycin: a case review and cohort study. Lancet 1999; 354:2101‐2105.
- 12. Mertens PLJM, Stals FS, Schellekens JFP, Houben AW, Huisman J. An epidemic of pertussis among elderly people in a religious institution in the Netherlands. Eur J Clin Micro Infect Dis 1999; 18:242‐247.
- 13. Keital WA, Edwards KM. Pertussis in adolescents and adults: time to reimmunize? Semin Respir Infect 1995; 10:51‐57.
- 14. Dworkin MS. Adults are whooping, but are internists listening? Ann Intern Med 2005; 142:832‐835.
- 15. Couzigou C, Flahault A. Is pertussis being considered as a cause of persistent cough among adults? Eur J Epidemiol 2003; 18:1013‐1015.
- 16. Halperin SA, Botolussi R, Langley JM, et al. A randomized, placebo‐controlled trial of erythromycin estolate prophylaxis four household contacts of children with culture‐positive Bordetella pertussis infection. Pediatrics 1999; 104:e42.
- 17. Weber D, Rutala W. Management of healthcare workers exposed to pertussis. Infect Control Hosp Epidemiol 1994; 15:411‐415.
- 18. Statement on management of persons exposed to pertussis and pertussis outbreak control. National Advisory Committee on Immunization, the Advisory Committee on Epidemiology, and the Canadian Pediatric Society. Can Commun Dis Rep 1994; 20:193‐200.
- 19. Guris D, Streble PM, Wharton M. Pertussis: In: Centers for Disease Control and Prevention. Manual for the surveillance of vaccine‐preventable diseases. CDC: Atlanta, GA 30333, 1997: Chapter 8.
- 20. Centers for Disease Control and Prevention. January 2005 Updates—Guidelines for the Control of pertussis outbreaks, 2000. Available at: http://www.cdc.gov/nip/publications/pertussis/guide/htm. Accessed May 12, 2006.
- 21. US Department of Health and Human Services. Hypertrophic pyloric stenosis in infants following pertussis prophylaxis with erythromycin—Knoxville, Tennessee, 1999. MMWR Morb Mortal Wkly Rep 1999; 48:1117‐1120.
- 22. Physicians’ Desk Reference. 58th ed. Montvale, NJ: Thompson PDR; 2004.
- 23. Langley JM, Halperin SA, Boucher FD, Smith B. Pediatric Investigators Collaborative Network on Infections in Canada (PICNIC). Azithromycin is as effective as and better tolerated than erythromycin estolate for the treatment of pertussis. Pediatrics 2004; 114:e96‐101.
- 24. Bace A, Kuzmanovic N, Zrnic T. Pertussis: simple and short therapy with azithromycin. In: Program and abstracts of the 4th International Conference on Macrolides, Azalides, Streptogramins, and Ketolides, 1998; Barcelona, Spain;. Abstract 6.04.
- 25. Friedman DS, Curtis R, Schauer SL et al. Surveillance for transmission and antibiotic adverse events among neonates and adults exposed to a healthcare worker with pertussis. Infect Control Hosp Epidemiol 2004; 25:967‐973.
- 26. American Academy of Pediatics. Pertussis. In: Pickering LK, ed. Red Book: 2003 Report of the Committee on Infectious Diseases. 26th ed. Elk Grove Village, IL: American Academy of Pediatrics; 2003.
- 27. Linnemann CC, Ramundo N, Perlstein PH, Minton SD, Englender GS. Use of pertussis vaccine in a epidemic involving hospital staff. Lancet 1975; 2:540‐543.
- 28. United States Food and Drug Administration. Tetanus toxoid, reduced diphtheria toxoid, and acellular pertussis vaccine adsorbed. Available at: http://www.fda.gov/cber/approvltr/tdapave061005L.htm. Accessed May 23, 2006.
- 29. Calugar A, Orega‐Sanchez I, Jahre I, et al. Economic evaluation of a nosocomial pertussis outbreak in Pennsylvania, 2003. In: Program and abstracts of the 42nd Annual Meeting of the Infectious Disease Society of America; September 30–October 3, 2004; Boston, MA. Abstract 196.
- 30. Wright SW, Decker MD, Edwards KM. Incidence of pertussis in healthcare workers. Infect Control Hosp Epidemiol 1999; 20:120‐123.
-
K.A.B. has received research and speakers' bureau support from Sanofi Pasteur.