Horizontal transmission of group B streptococcus in a neonatal intensive care unit

Abstract

The incidence of early-onset group B streptococcal (GBS) sepsis in the neonatal population has decreased substantially since the introduction of maternal intrapartum antibiotic prophylaxis and routine prenatal screening. However, these strategies have not reduced the incidence of late-onset GBS infections. Additional research pertaining to the transmission of late-onset GBS infections is required to develop effective preventive methods. The present report describes probable horizontal transmission of late-onset GBS infection among three infants in a neonatal intensive care unit. GBS strain confirmation was based on the microbiological picture, antibiogram and pulsed-field gel electrophoresis. These cases highlight the morbidity associated with late-onset GBS disease and the importance of considering horizontal transmission as an etiological factor in GBS infection in the newborn period. Further studies assessing horizontal transmission in late-onset GBS disease may improve prevention and early intervention.

Group B streptococcal (GBS) sepsis is a leading cause of morbidity and mortality in the neonatal population (1). A significant reduction has been seen in the incidence of early-onset GBS sepsis (defined as zero to seven days of age) from 1.7 cases per 1000 live births in 1993 to 0.31 cases per 1000 live births in 2005 (2). This reduction was due to the introduction of routine prenatal screening and maternal intrapartum antibiotic prophylaxis (2). However, the incidence of late-onset GBS disease (defined as seven to 89 days of age) has remained static and is now higher than early-onset GBS disease, averaging 0.35 cases per 1000 live births (1,2). The transmission of late-onset GBS sepsis is hypothesized to occur through multiple mechanisms such as persistent neonatal mucous membrane colonization, contamination of shared equipment via other infected neonates and through infected maternal breast milk (3–10). Currently, no preventive strategy has been shown to effectively reduce the incidence of late-onset GBS sepsis. The present case report describes probable acquisition of GBS disease through horizontal transmission in a neonatal intensive care unit based on the clinical and microbiological picture as well as the confirmation of strain based on antibiotic resistance patterns and pulsed-field gel electrophoresis.

CASE PRESENTATIONS

Infant 1

Infant 1 was born at 27 weeks and two days to a 29-year-old woman with unknown GBS status. The infant’s birth weight was 1140 g. There was no history of GBS infection in any of the mother’s previous pregnancies. The current pregnancy was uncomplicated until 27 weeks’ gestation, when the mother presented with placental abruption and spontaneous onset of labour. Infant 1 was admitted to the neonatal intensive care unit (NICU) due to prematurity. On day 25 of life, the infant showed signs of septic deterioration, and a subsequent blood culture grew GBS bacteria at 9 h. Infant 1 was treated with ampicillin for 14 days, but experienced subsequent recurrent infections, with blood culture positivity for GBS infection at six weeks and nine weeks of age.

Further investigation was undertaken to determine a possible source of the recurrent infection. As part of the investigation, fresh and frozen breast milk was sent for culture and sensitivity testing. Subsequently, both the frozen expressed breast milk and the fresh breast milk grew GBS bacteria. On further investigation, no alternative sources of infection were found. Furthermore, once breast-feeding stopped and infant 1 was fed formula, no further GBS septic episodes occurred. In addition, infant 1 was fed expressed breast milk exclusively via nasogastric tube; therefore, it was less likely that the initial infection was caused by mucous membrane colonization, suggesting that maternal breast milk contamination was the source.

Infant 2

Infant 2 was born at 26 weeks and three days to a 39-year-old woman. The infant’s birth weight was 1066 g. The mother presented with placental abruption, and a stat caesarean section was performed due to fetal distress. The pregnancy was otherwise uncomplicated, and the maternal GBS status was unknown. There was no history of GBS infection in any previous pregnancies. Infant 2 was also admitted to the NICU on the same day as infant 1. Both of these infants had cots beside one another in the NICU for the first 26 days of life. Infant 2 experienced three septic deteriorations within the first month of life; however, no organism or source was identified.

On day 40 of life, 10 days after infant 1 was diagnosed with GBS sepsis, infant 2 demonstrated clinical evidence of septic deterioration. Blood cultures grew GBS bacteria at 8 h. Lumbar puncture and urinalysis results were negative. Despite a 14-day course of ampicillin, infant 2 experienced two further septic deteriorations with blood cultures positive for GBS infection. Further investigation for a possible source of persistent infection revealed multiple liver abscesses, likely due to GBS infection. There was no evidence of maternal mastitis, and maternal breast milk was negative. The infant was discharged home on long-term antibiotics administered via a peripherally inserted central catheter line until regression of liver abscesses was seen.

Infant 3

Infant 3 was born at 25 weeks and two days to a primigravida 34-year-old woman. This was a dichorionic-diamniotic pregnancy by in vitro fertilization. The pregnancy was otherwise uncomplicated. The maternal GBS status was unknown at the time of delivery. At 25 weeks, premature rupture of membranes from the other twin occurred, resulting in a caesarean section. The birth weight of infant 3 was 720 g. On day 17 of life, six weeks after infant 1 was diagnosed with GBS sepsis, infant 3 demonstrated signs of septic deterioration and grew GBS bacteria from two blood cultures. A lumbar puncture was not performed due to clinical instability. Infant 3 was treated with ampicillin for 14 days and had no subsequent infection. No etiological mechanism was identified on further investigation. The twin of infant 3 was otherwise well, with no signs of GBS sepsis.

Further investigation

Further investigation was undertaken to determine the underlying etiology of GBS disease in these three infants. Specific concern regarding possible horizontal transmission was warranted due to the unusual occurrence of three infants diagnosed with late-onset GBS infection within a short period of time. Confirmation of horizontal transmission of GBS infection among the three infants was based on the microbiological picture of GBS disease, an antibiogram and pulsed-field gel electrophoresis. The antibiogram, which refers to the sensitivity and resistance pattern of an organism, revealed that the GBS strain infecting all three infants had 100% similarity in resistance and sensitivity patterns (Table 1). All three strains displayed resistance to both erythromycin and clindamycin (7). To further confirm the concern regarding possible horizontal transmission, pulsed-field gel electrophoresis was conducted. This is a method for separating DNA molecules and is used for genotyping (11). All three infants (as well as the breast milk of infant 1’s mother) displayed an identical band pattern, providing further evidence that all three infants were infected with an identical strain of GBS bacteria (Figure 1) (12).

TABLE 1.

Antibiogram sensitivity and resistance patterns

Identification	Source	Similarity, %
Infant 1	Blood	100
Infant 1	Milk	100
Infant 2	Blood	100
Infant 2	Blood	100
Infant 3	Blood	100

Figure 1).

Pulsed-field gel electrophoresis of DNA digested with Sma1 of isolates from group B streptococcus-positive infants and from two samples of maternal breast milk. The image shows the molecular weight standard (MW), freshly expressed milk from the mother of infant 1 (lane 1), blood cultures for infant 1 (lane 2), infant 2 (lane 3) and infant 3 (lane 4), group B streptococcal strain from blood culture of an infant in the neonatal intensive care unit unrelated to the present report’s cluster (lane 5) and frozen maternal milk – infant 1 (lane 6). Lane 7 is an unrelated strain from a different hospital that was tested at the same time

Eradication of GBS mucous membrane colonization with the addition of rifampin was discussed for infants 1 and 2 due to their recurrent GBS disease. However, rifampin was not indicated because infant 1 likely had recurrent disease due to re-exposure to infected maternal breast milk, while infant 2 had documented liver lesions as a source of persistent infection (13). Maternal treatment with amoxicillin to eradicate maternal carrier state has been documented; however, infant 1’s mother had no evidence of mastitis and previously experienced an anaphylactic reaction to penicillin (8). The GBS strain was resistant to clindamycin (the second line of therapy), the mother was unwilling to consider intravenous therapy with vancomycin, and she could not afford treatment with linezolid. Therefore, no treatment was immediately initiated for infant 1’s mother. Infant 1 was restricted to exclusive formula feeding because of the recurrent episodes of GBS sepsis. Before a treatment strategy could be devised for the mother, her milk production stopped.

DISCUSSION

Case reports of neonatal late and recurrent GBS disease associated with maternal milk, and documentation of strain similarities among NICU outbreaks is relatively rare (6–10). In our case, although 100% confirmation was not possible, we hypothesize that the cause of infection for infant 1 was vertical transmission of GBS through contaminated maternal breast milk. For infant 2, due to the physical proximity to infant 1, horizontal transmission between infants was the presumed etiological mechanism – most likely via the hands of a shared caregiver. The transmission to infant 3 in a separate pod of the neonatal nursery was hypothesized to also be via transmission though contaminated equipment or through transmission via health care professionals. Previous case reports have documented iatrogenic horizontal transmission via health care workers by demonstrating that the strain types found in health care professionals was identical to the infant strains (3). Health care worker strain typing and colonization testing were not conducted in the present case. In addition, although previous reports often site maternal mastitis as an etiological mechanism for late-onset GBS disease (6–9), it is possible to have asymptomatic excretion (10), as demonstrated in the present case. Olver et al (11) also described cases of GBS infection in preterm infants fed with maternal milk via nasogastric tube alone, with no evidence of maternal mastitis or colonization.

Screening protocols for infants with recurrent GBS disease, as well as isolation and infection control principles for GBS-positive infants, must be considered as possible mechanisms for reducing disease transmission risk.

CONCLUSION

Horizontal transmission of GBS disease is a preventable cause of morbidity and mortality in the neonatal population. The present case report highlights the probable role of nosocomial horizontal transmission of GBS disease in the neonates, specifically highlighting the potential role of health care workers and horizontal transmission among infants within an NICU. Hand washing, repeated instrument and equipment cleaning (including breast milk pumps), and careful limitation of other vectors for horizontal transmission may play a role in preventing similar cases in the future. Additional research regarding the transmission of late-onset GBS infection is required to develop effective preventive methods and decrease the significant morbidity and mortality seen in this high-risk population.

CLINICAL PEARLS

• Horizontal transmission of GBS infection is a preventable cause of morbidity and mortality in the neonatal population.

• Maternal breast milk and poor infection control measures can lead to neonatal unit outbreaks of GBS infection.

• Investigation of horizontal transmission of GBS infection includes maternal breast milk culture and sensitivity, an antibiogram and pulsed-field gel electrophoresis.