Risk Factors for Neonatal Methicillin‐Resistant Staphylococcus aureus Infection in a Well‐Infant Nursery
Objective. To determine risk factors for neonatal methicillin‐resistant Staphylococcus aureus (MRSA) skin and soft‐tissue infection in a well‐infant nursery.
Design. Case‐control studies.
Setting. A well‐infant nursery in a nonteaching, community hospital.
Methods. Case infants were newborns in the nursery who were born in the period November 2003 through June 2004 and had onset of MRSA skin and soft‐tissue infection within 21 days after discharge from the nursery. Site inspections were conducted. Control infants were randomly selected male infants in the nursery during the outbreak periods. MRSA isolates were characterized with pulsed‐field gel electrophoresis.
Results. Eleven case infants were identified in 2 outbreaks: outbreak 1 occurred from November 18 through December 24, 2003, and outbreak 2 occurred from May 26 through June 5, 2004. All were full‐term male infants with pustular‐vesicular lesions in the groin. Inspection revealed uncovered circumcision equipment, multiple‐dose lidocaine vials, and inadequate hand hygiene practices. In outbreak 1, case infants (
) had a significantly higher mean length of stay than control infants (3.7 vs 2.5 days; 
). In outbreak 2, case infants (
) were more likely to have been circumcised in the nursery (OR, undefined [95% CI, 1.7 to undefined]) and to have received lidocaine injections (OR, undefined [95% CI, 2.6 to undefined]). Controlling for length of stay, case infants were more likely to have been circumcised in the nursery (OR, 12.2 [95% CI, 1.5 to undefined]). Pulsed‐field gel electrophoresis showed that 7 available isolates were indistinguishable from a community‐associated MRSA strain (USA300‐0114).
Conclusions. Newborns in well‐infant nurseries are at risk for nosocomial infection with community‐associated MRSA strains. Reducing length of stay, improving circumcision and hand hygiene practices, and eliminating use of multiple‐dose lidocaine vials should decrease transmission of community‐associated MRSA strains in nurseries.
Received April 6, 2006; accepted June 14, 2006; electronically published March 15, 2007.
Methicillin‐resistant Staphylococcus aureus (MRSA) has been a cause of outbreaks that affect premature and ill infants in neonatal intensive care units.1 Reports have indicated that community‐associated MRSA (CA‐MRSA) strains are capable of infecting otherwise healthy infants in hospital nurseries.2 Neonatal pustulosis or colonization due to S. aureus has been associated with prolonged hospital length of stay (LOS), delivery by cesarean section, and history of plastic‐bell circumcision.3‐5 However, data are limited on risk factors for nosocomially acquired infection due to CA‐MRSA strains in well‐infant nurseries.
In 2004, pediatricians reported to the Los Angeles County Department of Health Services 2 clusters of neonatal MRSA skin and soft‐tissue infection (SSTI). All the SSTIs occurred in male infants who had been discharged from a 36‐bed well‐infant nursery at a nonteaching community hospital. The hospital averaged 250 deliveries monthly and had no previous SSTI outbreak in the nursery. We present the findings of our investigation and potential risk factors for infection.
Methods
Case Definitions and Findings
Case infants were newborns born at the hospital and admitted to the nursery during November 1, 2003, to June 14, 2004, who had onset of culture‐confirmed MRSA SSTI within 21 days after discharge from the nursery and who had had no interim hospitalizations or hospital visits. Case infants were identified through staff obstetricians, pediatricians, and women who had given birth at the hospital during the outbreak periods.
Colonization and Laboratory Studies
Presence of MRSA colonization was determined on the basis of culture of specimens obtained from newborns at discharge from the nursery during the periods January‐February 2004 (nasal specimen cultures) and July‐December 2004 (paired cultures of nasal and axilla specimens, and of umbilical and groin specimens). The hospital decided to obtain nasal specimens for culture from a convenience sample of available nurses on recognition of the first outbreak. On June 14, 2004, a point‐prevalence nasal carriage study was conducted.
Available MRSA isolates were retrieved for pulsed‐field gel electrophoresis (PFGE) characterization by SmaI digestion to determine genotypic relatedness and strain type.6,7 Antimicrobial susceptibility data were determined by the hospital laboratory.
Case‐Control Studies
Unmatched case‐control studies were conducted. Control infants were randomly selected from asymptomatic male infants born at the hospital and admitted to the nursery during the outbreak periods. Every third male newborn was chosen from the delivery logs, so that at least 3 control infants would be enrolled per case infant. Physicians were contacted to document any outpatient circumcisions and SSTIs that occurred within 21 days after discharge.
Prepartum, intrapartum, and postpartum risk factors were considered. Data were abstracted from the medical records of case infants, control infants, and their mothers. We collected information on demographic characteristics, pregnancy, hospital course, procedures, and staff contacts. Maternal factors included age, group B Streptococcus status, prepartum antimicrobial exposure, parity and gravidity, smoking history, underlying diseases, prenatal infections, medications, and prepartum and delivery complications. Neonatal factors included prolonged rupture of membranes, multiple gestations, Apgar scores, delivery mode, presentation, presence of meconium, birth weight and size, breast‐feeding status, history of phlebotomy, history of jaundice, and history of circumcision and type of device used for the procedure (plastic bell, Gomco clamp, or Mogen clamp). We defined prolonged LOS in the nursery as more than 2 days.
Using the Fisher exact test, we calculated odds ratios (ORs), 95% confidence intervals (95% CIs), and P values with Epi Info statistical software.8 P values less than .05 were considered statistically significant. Data from each outbreak were analyzed separately and together for comparison. We compared mean values using the Student t test. Multivariable models were analyzed by using exact conditional logistic regression with SAS statistical software, version 9.1 (SAS Institute).
Results
Outbreak 1
Outbreak 1 involved 6 case infants (infants 1‐6) born between November 18 and December 24, 2003 (Figure 1A). Four were delivered by cesarean section. All case infants were full‐term male infants of appropriate size for gestational age without reports of fever or other signs of infection. Illness onset occurred at a median of 4 days (range, −1 to 17 days) after discharge from the nursery and at a mean age of 9.5 days (range, 3‐19 days). Case infant 2 developed lesions on the day before discharge (day −1). His lesions were mistaken for a normal newborn rash but were later confirmed to be due to MRSA. No cases of maternal infection occurred.
Figure 1. Time lines for outbreak 1 (A) and outbreak 2 (B) of skin and soft‐tissue infection due to a community‐associated methicillin‐resistant Staphylococcus aureus strain among case infants discharged from the well‐infant nursery during the period November 2003 through June 2004. “Rooming‐in” refers to the housing of newborns with their mothers rather than in the common nursery; “public health” refers to the public health department.
All case infants were circumcised. Case infants 1 and 5 had lesions in the groin before the procedure. The median interval from circumcision to illness onset (
) was 4.5 days (range, 2‐17 days). Case infants 1, 4, and 5 were circumcised after discharge by 3 different mohels (Jewish ritual circumcisors). Infections did not begin at lidocaine injection sites or excision sites. The only clinical finding was a pustular‐vesicular rash. Infections began as fewer than 5 lesions in the anterior diaper area and along the edges of the diaper area with no involvement of the skin‐folds, penis, perianal area. Case infant 1 developed a facial abscess after developing a groin pustule. Case infant 4 developed additional lesions near the umbilicus, thighs, and scrotum. Five case infants were hospitalized again for a mean of 4.8 days so that a diagnostic work‐up for sepsis could be done. None had bacteremia, and all recovered without adverse sequelae. The median interval from the onset of lesions to the confirmatory culture result was 11.5 days (range, 3‐59 days).
Outbreak 2
The second outbreak consisted of 5 case infants (infants 7‐11) born during the period from May 26 through June 5, 2004 (Figure 1B). All had presentations similar to those in outbreak 1. Illness onset occurred a median of 2 days (range, −1 to 6 days) after discharge from the nursery and at a mean age of 5.6 days (range, 4‐8 days). Case infant 8 developed lesions before discharge. No concurrent mother‐newborn infections occurred.
Three case infants were rehospitalized for a median of 3 days (range, 2‐10 days). As in outbreak 1, nonhospitalized case infants were observed clinically or received only topical ointment. No case infant developed any adverse sequelae. Three case infants had been delivered by cesarean section; all were circumcised in the nursery before discharge. The median interval from circumcision to illness onset was 4 days (range, 2‐6 days). Compared with outbreak 1, the delay in diagnosis was significantly reduced to a median of 2 days (range, 0‐5 days) (
). As of July 2005, no new cases were reported.
Colonization and Laboratory Studies
Surveillance cultures performed for newborns (
) discharged from the nursery during January through February 2004 identified 2 (1.5%) with nasal carriage of methicillin‐susceptible S. aureus (MSSA) and 1 (0.7%) with nasal carriage of MRSA. The pediatrician caring for the MRSA‐colonized newborn was notified. The infant was not decolonized. There was no reported progression to infection. One of 17 cultures of nasal specimens from nurses (that for nurse A) was positive for MRSA.
On June 14, 2004, we obtained specimens for culture from 35 women in labor and postpartum women for the point prevalence study. Four (11%) were nasal carriers of MSSA; none was a carrier of MRSA. Of 33 newborns with cultures performed, none had cultures positive for either MSSA or MRSA. During the period July 1 to December 31, 2004, paired samples for culture were obtained from all newborns (
) at discharge. None of the newborns had MSSA detected, but 7 (0.4%) were colonized with MRSA. Six of these newborns had MRSA detected only in paired cultures of umbilical and groin specimens; 1 had MRSA detected only in a paired cultures of nasal and axilla specimens. All isolates had antimicrobial susceptibility patterns consistent with CA‐MRSA strains. Four of these 7 infants were male; 3 were female. All were observed clinically, and none developed infection.
MRSA isolates from case infants 1, 3, 6, 7, 8, 9, and 11 and from nurse A were available for PFGE analysis (Figure 2). All had a pattern indistinguishable from that of a CA‐MRSA strain (USA300‐0114) responsible for other local SSTI outbreaks.9 All isolates were resistant to β‐lactam antimicrobials, oxacillin, and erythromycin. The isolates were susceptible to trimethoprim‐sulfamethoxazole, clindamycin, gentamicin, and vancomycin and had intermediate‐to‐full resistance to the fluoroquinolones.
Figure 2. Pulsed‐field gel electrophoresis patterns after SmaI digestion for 7 available methicillin‐resistant Staphylococcus aureus (MRSA) isolates from case infants. Lanes 1 and 10, NCTC 8325 molecular weight standard. Lanes 2‐4 and 6‐9, MRSA isolates from case infants. Lane 5, a community‐associated MRSA strain (USA300‐0114) responsible for other outbreaks of skin and soft‐tissue infection in Los Angeles County, California.
Site Inspections and Infection Control Interventions
In January 2004, the hospital reinforced adequate hand hygiene techniques, increased the frequency of terminal cleaning of the nursery, and queried women on admission regarding SSTIs by means of a nurse‐administered questionnaire. Patients and household members were asked about skin lesions and any history of MRSA infection or SSTI. A woman in labor was identified as having an axillary MRSA lesion. We distributed educational materials about CA‐MRSA strains to hospital pediatricians and obstetricians.10 Interviews with case infants’ mothers revealed no common factors (ie, diapers, wipes, ointment, or lotion), hospital exposures, or contacts with SSTI or MRSA infection. However, the case infants' mothers reported that certain staff members had not washed their hands when handling their newborns.
On June 15, 2004, we conducted a subsequent site inspection. We focused on circumcision practices. Other than preparation and postcircumcision wound care with sterile gauze and petroleum jelly, no differences were found in handling between male and female newborns. Direct observation identified lapses in hygiene practices. The circumcision tray with the sterile equipment intended for use was left open for an extended period, which is standard practice to expedite the procedure. A physician was observed performing a circumcision without hand washing. We noted that lidocaine was drawn from an uncovered, previously opened multiple‐dose vial. The vial was used exclusively for circumcisions and typically was in use for 1‐2 weeks. Environmental samples were obtained from surfaces and fomites in the nursery, circumcision equipment, and the open lidocaine vial. None of the samples tested positive for MRSA.
Review of staff schedules indicated that each case infant was cared for by a median of 7 nurses (range, 5‐10 nurses). Medical record reviews were unable to identify any single physician, physician group, nurse (including nurse A), or staff member who had documented contact with all case infants. In addition, no individual had been absent during the 5 months between outbreaks and returned at the onset of outbreak 2. Because of the lack of evidence suggesting a possible source individual, we did not recommend further MRSA screening of staff members.
During these outbreaks, no other nurseries in the county reported similar infections. Notably, most facilities opted to house newborns with their mothers rather than in a well‐infant nursery. In June 2004, the hospital began housing newborns in their mother's rooms and instituted bathing of infants with 2% chlorhexidine at discharge.
Case‐Control Studies
In outbreak 1, there were 6 case infants and 24 control infants enrolled in the study. No differences were found between mean maternal age for case infants and control infants (31.8 vs 30.5 years; 
). No significant risk factors were identified on bivariate analysis (Table). Case infants had a significantly higher mean LOS than control infants (3.7 vs 2.5 days; 
).
In outbreak 2, there were 5 case infants and 22 control infants enrolled. The LOS was not significantly different between case infants and control infants (3.2 vs 2.4 days; 
). On bivariate analysis, newborns circumcised in the nursery and infants who received lidocaine injections were more likely to develop infection (Table). Mean maternal age was significantly higher for case infants than for control infants (34 vs 28.2 years; 
). Maternal age of more than 30 years was a significant risk factor for MRSA infection. Controlling for LOS, case infants were 12 times more likely than control infants to have been circumcised in the nursery (OR, 12.2 [95% CI, 1.5 to undefined]; 
). No association was found between the type of circumcision device used and infection.
Analysis of the combined data indicated that circumcision was significantly associated with infection (OR, 7.0 [95% CI, 1.21‐71.4]; 
), as was receipt of a lidocaine injection (OR, 5.5 [95% CI, 1.09‐35.7]; 
), prolonged LOS (OR, 5.5 [95% CI, 1.09‐35.7]; 
), and maternal age of more than 30 years (OR, 6.4 [95% CI, 1.11‐65.3]; 
). Mean maternal age was significantly higher among infants with a prolonged LOS than among those without this risk factor (32.2 vs 28.6 years; 
). Controlling for maternal age and LOS, receipt of lidocaine injection remained an independent risk factor for infection (OR, 5.5 [95% CI, 1.08‐27.6]; 
).
Discussion
Infants who were circumcised, particularly those circumcised in the nursery, were at risk for infection with CA‐MRSA strains. We postulated that the handling required for preparation and execution of the procedure presented opportunities for bacterial acquisition. MRSA has been shown to survive on fomites for months in neonatal hospital settings.11‐14 In outbreak 2, uncovered equipment or a multiple‐dose lidocaine vial could have become externally contaminated with MRSA, which may have facilitated bacterial transmission to the surgical field (groin), consistent with the clinical findings. If the vial were internally contaminated, infections would be expected to occur at injection sites. Since we were unable to retrieve the vials used during the outbreaks, the inability to culture MRSA from the vial that was retrieved was anticipated. The analysis indicated that lidocaine injection increased the risk of infection to a value greater than the risk posed by circumcision. Therefore, the use of multiple‐dose lidocaine vials should be discouraged.
Uncircumcised male infants were also at risk for SSTI due to CA‐MRSA strains, because not all case infants were circumcised before illness. Research has found that male infants are more susceptible to neonatal skin infections.15 We found that maternal age of more than 30 years was associated with prolonged LOS, which consequently presented additional opportunities for infection. Hospital transmission of CA‐MRSA strains has been reported among postpartum women.2,16 Multiparity and delivery by cesarean section have been associated with maternal SSTI due to CA‐MRSA strains.17,18 These findings indicate that the maternity‐neonatal wards are high‐risk areas for outbreaks of nosocomial infection due to CA‐MRSA strains.
Screening of staff for MRSA by culture was not recommended, because the data did not indicate a potential source person. Retrospectively, traveling nurses were no longer available to provide samples for culture. The positive culture result for nurse A likely represented the background prevalence of CA‐MRSA strains, rather than a source of infection. Because of the limited number of circulating CA‐MRSA strains, PFGE was not useful in identifying a source. The Centers for Disease Control and Prevention concurred that culture should be performed only for implicated personnel.19 A screening culture positive for a CA‐MRSA strain and/or a matching PFGE pattern without other implicating epidemiologic evidence would be uninterpretable.
The hospital conducted one of the largest MRSA colonization studies done to date, to our knowledge, which involved more than 1,600 newborns in the nursery. No MSSA colonization was detected, which contrasts with the 10%‐68% colonization rate expected among newborns sampled for culture at discharge.20,21 In contrast, the MRSA colonization rate (0.4%) was consistent with data that reported a MRSA prevalence of 0.8% in the community.22 It was unclear how ongoing infection control measures, such as chlorhexidine baths at discharge, affected these findings. The findings demonstrated that male and female newborns are capable of acquiring MRSA in the nursery, but not all developed SSTI. Guidelines for managing newborns with asymptomatic CA‐MRSA colonization should be developed.
MRSA transmission was apparently facilitated by inadequate hand hygiene practices. Efforts to minimize contacts between newborns and staff were pursued. Universal gloving and gowning was impractical. The hospital began housing newborns with their mothers. This approach had been reported to reduce the incidence of nosocomial infection by fostering neonatal colonization with maternal flora, which might prevent colonization with more pathogenic organisms.23
Underreporting may have occurred because of misdiagnoses by clinicians and nonreporting by patients and staff. The small number of case infants limited our ability to detect other significant risk factors. Because of a lack of evidence, we could not determine a definitive source of infection. Although the outbreaks had similar characteristics, in part due to the common pathogen, the precise clustering of case infants was consistent with 2 distinct outbreaks. The identified risk factors had varying significance, depending on the conditions under which each outbreak occurred. This finding may explain why certain risk factors were associated with one but not the other outbreak.
As the prevalence of CA‐MRSA strains increases in the community, it will become more difficult to prevent outbreaks such as the one we describe. Efforts to limit introduction of CA‐MRSA strains into the nursery should include routine verbal queries to identify potential sources. Maintaining adequate hygiene practices, especially during circumcisions, should limit the dissemination of CA‐MRSA strains. Educating staff, patients, visitors, and mohels about infection with CA‐MRSA strains will improve disease recognition and outbreak detection.
References
- 1. Healy CM, Hulten KG, Palazzi DL, Campbell JR, Baker CJ. Emergence of new strains of methicillin‐resistant Staphylococcus aureus in a neonatal intensive care unit. Clin Infect Dis 2004; 39:1460‐1466.
- 2. Bratu S, Eramo A, Kopec R, et al. Community‐associated methicillin‐resistant Staphylococcus aureus in hospital nursery and maternity units. Emerg Infect Dis 2005; 11:808‐813.
- 3. Rush J. Does routine newborn bathing reduce Staphylococcus aureus colonization rates? A randomized controlled trial. Birth 1986; 13:176‐180.
- 4. Rush J, Fiorino‐Chiovitti R, Kaufman K, Mitchell A. A randomized controlled trial of a nursery ritual: wearing cover gowns to care for healthy newborns. Birth 1990; 17:25‐30.
- 5. Stranko J, Ryan ME, Bowman AM. Impetigo in newborn infants associated with a plastic bell clamp circumcision. Pediatr Infect Dis J 1986; 5:597‐599.
- 6. Tenover FC, Arbeit RD, Goering RV, et al. Interpreting chromosomal DNA restriction patterns produced by pulsed‐field gel electrophoresis: criteria for bacterial strain typing. J Clin Microbiol 1995; 33:2233‐2239.
- 7. McDougal LK, Steward CD, Killgore GE, Chaitram JM, McAllister SK, Tenover FC. Pulsed‐field gel electrophoresis typing of oxacillin‐resistant Staphylococcus aureus isolates from the United States: establishing a national database. J Clin Microbiol 2003; 41:5113‐5120.
- 8. Centers for Disease Control and Prevention, Atlanta, GA, and US Department of Health and Human Services, CDC. 2004. Epi Info version 3.3, released October 5, 2004. Available at: http://www.cdc.gov/epiinfo/. Accessed December 1, 2004.
- 9. Centers for Disease Control and Prevention. Public health dispatch: outbreaks of community‐associated methicillin‐resistant Staphylococcus aureus skin infections—Los Angeles County, California, 2002‐2003. MMWR Morb Mortal Wkly Rep 2003; 52:88.
- 10. Community‐Associated Methicillin‐Resistant Staphylococcus aureus. Los Angeles County Department of Health Services’ Web Information Page. Available at: http://lapublichealth.org/acd/MRSA.htm. Accessed December 1, 2004.
- 11. Hota B. Contamination, disinfection, and cross‐colonization: are hospital surfaces reservoirs for nosocomial infection? Clin Infect Dis 2004; 39:1182‐1189.
- 12. Oie S, Kamiya A. Survival of methicillin‐resistant Staphylococcus aureus (MRSA) on naturally contaminated dry mops. J Hosp Infect 1996; 34:145‐149.
- 13. Dietze B, Rath A, Wendt C, Martiny H. Survival of MRSA on sterile goods packaging [review]. J Hosp Infect 2001; 49:255‐261.
- 14. Fujimura S, Kato S, Hashimoto M, Takeda H, Maki F, Watanabe A. Survey of methicillin‐resistant Staphylococcus aureus from neonates and the environment in the NICU. J Infect Chemother 2004; 10:131‐132.
- 15. Sinha A, Yokoe D, Platt R. Epidemiology of neonatal infections: experience during and after hospitalization. Pediatr Infect Dis J 2003; 22:244‐251.
- 16. Saiman L, O’Keefe M, Graham PL III, et al. Hospital transmission of community‐acquired methicillin‐resistant Staphylococcus aureus among postpartum women. Clin Infect Dis 2003; 37:1313‐1319.
- 17. King MD, Humphrey BJ, Wang YF, Kourbatova EV, Ray SM, Blumberg HM. Emergence of community‐acquired methicillin‐resistant Staphylococcus aureus USA 300 clone as the predominant cause of skin and soft‐tissue infections. Ann Intern Med 2006; 144:309‐317.
- 18. Laibl VR, Sheffield JS, Roberts S, McIntire DD, Trevino S, Wendel GD Jr. Clinical presentation of community‐acquired methicillin‐resistant Staphylococcus aureus in pregnancy. Obstet Gynecol 2005; 106:461‐465.
- 19. Information About MRSA for Healthcare Personnel. Centers for Disease Control and Prevention Web Information Page. Available at: http://www.cdc.gov/ncidod/dhqp/ar_mrsa_healthcareFS.html. Accessed September 4, 2004.
- 20. Stark V. Harrison SP. Staphylococcus aureus colonization of the newborn in a Darlington hospital. J Hosp Infect 1992; 21:205‐211.
- 21. Hargiss C, Larson E. The epidemiology of Staphylococcus aureus in a newborn nursery from 1970 through 1976. Pediatrics 1978; 61:348‐353.
- 22. Kuehnert MJ, Kruszon‐Moran D, Hill HA, et al. Prevalence of Staphylococcus aureus nasal colonization in the United States, 2001‐2002. J Infect Dis 2006; 193:172‐179.
- 23. Kitajima H. Prevention of methicillin‐resistant Staphylococcus aureus infections in neonates. Pediatr Int 2003; 45:238‐245.