The prevention and control of infection within the nursery, once the major clinical concern of those supervising and working in neonatal units, has commanded less attention in recent years. This is due in part to the decline in frequency of epidemics, which once ravaged many nurseries. When infections now occur, they tend to be sporadic and more easily controlled. Increased understanding of modes of transmission and the availability of recently developed antibiotics to treat specific infections have made control of infections easier. In recent years, newly developed cardiopulmonary techniques have allowed the survival of many infants who would otherwise have expired shortly after birth. These neonates, surviving for extended periods with major cardiorespiratory support, are at greatly increased risk of bacterial infection. It might therefore be expected that as early mortality from ventilatory insufficiency drops, later deaths attributable to infectious diseases would sharply increase. In the term nursery as well, healthy newborns must be protected from the risk of overwhelming infection. Because of the newborn's vulnerability, it is essential that we not become complacent about infection surveillance and control. This article will attempt to summarize current thinking on prevention and control of neonatal bacterial infections and will discuss some new aspects of this changing problem.
PREVENTION OF INFECTION
The control of neonatal staphylococcal disease still determines much of our thinking in the prevention of infection. In the past, staphylococci were a leading cause of neonatal sepsis, meningitis, and pneumonia, as weil as of local or disseminated skin disease. Although systemic staphylococcal disease has been controlled to a large extent, staphylococcal skin disease persists as a major source of morbidity in many nurseries. As long as colonization rates are allowed to remain high, the potential for systemic dissemination exists.
Understanding of the epidemiology of staphylococcal disease is relatively recent. As late as 1961, the Committee on the Control of Infectious Diseases for the American Academy of Pediatrics emphasized that air was the major route of spread for neonatal staphylococcal disease.1 Once it was learned, however, that the umbilical stump and skin were colonized before the nose and nasopharynx,2-7 it was demonstrated that treatment of the umbilical stump with a variety of agents could reduce nasal colonization and control the spread of staphylococcal infection. Later, elegant epidemiologic studies showed conclusively that physical.
especially hand, contact was usually necessary for cross-contamination and infection.7-10 Based on these findings, suggestions for infection control included:
1. Use of separate fresh sterile gloves and gowns for each infant. Although this would theoretically be the most effective plan, it was thought totally unfeasible because of the number of handlings of infants and the requisite number of changes of this protective barrier.
2. Application of antibacterial ointments to the hands of personnel. Although some studies suggested that this was of help in the control of colonization,3,11 repeated ointment application was thought to be as unworkable as repeated gloving.
3. Application of a substance to the umbilical cord to prevent or reduce colonization. Many nurseries utilize this technique, and representative studies have reported reductions in staphylococcal colonization with hexachlorophene (1 to 3 per cent),3 antibacterial powders and ointments,6,12-14 and triple dye (brilliant green, proflavine hemisulfate, crystal violet in water).2,15,16
4. Washing of infants with soap. Skin cleansing in the newborn is performed both for prevention of infection and for esthetic purposes. Since the newborn skin is not colonized with protective flora, is breached by an open umbilicus susceptible to colonization, often has a surgical wound (circumcision), and is exposed to a variety of environmental pathogens, it is no surprise that skin care has been the focus of so much attention. Hexachlorophene, a chlorinated bisphenol, has been widely employed since 1961, when it was demonstrated that its use greatly reduced the incidence of staphylococcal skin colonization.8,16-20 In 1971, however, evidence began to accumulate that the use of a 3 per cent hexachlorophene solution in the neonatal period was associated with myelin vacuolation in the central nervous system. These changes were found in both human beings, especially small premature infants,21,22 and experimental animals.23-26 The published data prompted the Food and Drug Administration in 197227 to greatly restrict the use of hexachlorophene for bathing newborns. After hexachlorophene bathing was discontinued, the Center for Disease Control, in a random survey of nurseries that had discontinued its use, found an increased incidence of staphylococcal disease compared with nurseries that had not abandoned hexachlorophene bathing.28 Most of the increase was due to skin disease, but there was also an increase in systemic staphylococcal disease. In balancing risk versus benefit, the following recommendations were promulgated by the Committee on Fetus and Newborn of the American Academy of Pediatrics:29 1. Any skin cleansing of the newborn should be delayed until the infant's temperature has stabilized in the normal range. 2. Soft cotton sponges soaked with sterile water should be used to remove blood from the face and head and meconium from the perineum. A mild nonmedicated soap may also be used, followed by careful water rinsing. 3. Since vernix caseosa is likely to aid in the prevention of infection, the rest of the body should not be washed unless it is grossly soiled. 4. For the rest of the baby's hospitalization, the buttocks and perianal area should be kept clean with sterile water and cotton. 5. Some form of cord care may be helpful, but no specific agent is recommended. The use of hexachlorophene soaps is believed to have a place in the control of serious staphylococcal outbreaks despite the cited dangers. Risk versus benefit should be weighed in each circumstance, but use in premature infants should be avoided. A hexachlorophene soap remains an excellent agent for hand washing by nursery personnel.
5. Strict attention to hand washing by all personnel. Hand washing has become the mainstay of infection control in most nurseries. Effective hand washing will keep staphylococcal colonization at a low level; breaks in technique will be reflected almost immediately in a rise in rates of colonization and infection. Both an iodophor (water-soluble complex of iodine and surface active agents) and an antiseptic solution containing 3 per cent hexachlorophene meet the criteria for use in the nursery area: they do not stain or irritate in most people, are generally nonsensitizing, have persistent local action, and kill pathogenic bacteria. Whereas hexachlorophene is bactericidal for gram-positive organisms, such as staphylococci, the iodophors have bactericidal activity against both gram-positive and gram-negative bacteria. Upon entry to the nursery area, all jewelry to the elbows is removed. Sleeves should be rolled to the elbows and a two-to-three-minute scrub performed without a brush. Washing with a stiñ brush will drive the surface batería into the deeper dermal layers, from which they soon re-emerge. Foot-controlled faucets aid greatly in control of contamination. In their absence, hand faucets should be turned off with a towel rather than by hand. A sterile gown is then put on and sleeves are pushed to the elbows before any baby is handled. After a baby is touched, the hands are rewashed to the elbow for 15 to 30 seconds. After a clean gown has been put on, it is not necessary to change that gown if the baby is housed in an isolette or open bassinet and does not come into physical contact with the gown, and the sleeves are pushed above the elbows. If contact with the gown is made, the gown should be discarded and a new one taken for the handling of the next infant. Masks and caps are not necessary, since their use does not affect the spread of infection,30*31 serving only to make physician and parental contact with the baby more difficult. Nurses wearing scrub dresses need not wear cover gowns if they handle babies without contact to the dress; if such contact does occur, as in feeding an infant while sitting, a cover gown must be worn. Surveillance over these measures is absolutely critical, although at times difficult. The general activity of a busy neonatal intensive-care unit, with its diverse personnel (physicians, nurses, students, technicians, researchers, housekeepers, etc.), makes this surveillance all the more important. Repeated washing by nurses and technicians between infant contacts may be necessary 100 or more times each shift, and this must be stressed despite common resentment or refusal to wash so frequently. Any nursery personnel with infections of the skin, eyes, or respiratory and gastrointestinal tracts should not work in contact with newborns until such infections have cleared. Most units have screening programs that include annual physical examinations, chest x-rays, serologic tests for syphilis, and tuberculin tests. Some neonatal units may choose to regularly screen personnel for nasopharyngeal carriage of pathogenic staphylococci and treat identified carriers, but evidence for its benefit is lacking.
6. Many nurseries utilize the "cohort system" to control the spread of contamination. This system is a means of minimizing contacts among babies. As much as possible, a group of newborns stay together within the same physical environment and are cared for by the same nursing staff until their discharge. When babies leave the room - e.g., for radiologic studies - they are transported in closed isolettes and return to the same room. After the room is emptied, it is completely cleaned with a strong detergentdisinfectant and prepared for the next cohort group. All equipment must be similarly scrubbed. The "cohort system" does not apply to a neonatal intensive-care unit, in which maintenance of cohort groups is usually impossible. It is also important to provide adequate space between babies to prevent accidental physical contamination. Current recommendations of the Academy of Pediatrics are 20 square square feet per baby in a regular nursery, 30 square feet in a transitional nursery, and between 50 and 150 square feet in a neonatal intensive-care unit.32
7. Physical environment and equipment. Close attention to the physical environment and equipment is of paramount importance. As much disposable equipment as possible should be used. Nondisposable equipment should be cleaned by autoclave or gas sterilization, with an appropriate airing period of 24 hours or more after the latter. A disinfectant - such as an iodophor, chlorine compound, phenolic compound, or glutaraldehyde - should be employed in the nursery area. Because of the potential toxicity of some of these agents, great care should be taken in their use to avoid patient contact. Cleaning of floors, walls, and windows should avert dispersal of dust into the air. This may be accomplished by dry followed by wet vacuuming or by clean wet mopping with a disinfectant. The order of cleaning should be from patient areas to adjacent areas to hallways. Cleaning must include telephones, door handles, and other surfaces where hand contact is made.
Infant incubators, bassinets, and radiant warmers should be thoroughly cleaned and disinfected after each use. If infants remain in the unit for long, they should be transferred to clean housing weekly. All detachable parts of the housing should be removed and cleaned according to the manufacturer's instructions. Areas of hand contact, such as portholes and their cuffs, are a source of heavy contamination and should be cleaned thoroughly. Sterile distilled water with an added iodophor (2 drops/500 cc. of water) should be changed daily if maintenance of humidity in incubators is desired. Nebulizers and tubing should be changed daily and cleaned immediately. Each infant should have his own stethoscope, linen supply, medications, etc. Resuscitation equipment should be thoroughly cleaned by autoclaving, gas sterilization, or immersion for 15 minutes in a strong disinfectant, followed by careful washing and drying. Bottle warmers and other standing water pots should be eliminated; formula need not be warmed. A program of linen handling is also essential and should be developed among medical, nursing, housekeeping, and administrative personnel. Disposable diapers have found wide acceptance in many nurseries. Nondisposable linen should be placed in containers lined with disposable plastic and operated by foot pedals. Dirty linen should be removed three times a day and promptly taken from the nursery area. Garbage pails must be covered and should have either flip tops or foot-operated lift tops.
It may be useful to do spot cultures on a random basis to evaluate technique and remind personnel of the need for surveillance. Visual inspection and constant observation of technique cannot be overemphasized. The services of an epidemiologist may be of considerable help in this regard.
Despite these many precautions, we believe that it is extremely important to allow parents maximal contact with their infants. It has been suggested that disruption or prevention of parent-infant interaction in the critical postpartum days may lead to abnormal parenting relationships after discharge.33,34 This is of particular concern in the neonatal intensive- care unit, where hospitalization may be quite lengthy. We think it is extremely important, when parents may feel helpless to aid their critically ill newborn, that they feel that they have offered all their physical and emotional support. In this regard, parents are encouraged to touch and handle their infants and to visit at any hour that might be convenient to them. Of course, washing and gowning precautions apply to the parents. When physical facilities allow, rooming-in should be encouraged, with the same precautions.
In spite of many preventive measures, outbreaks of staphylococcal disease do occasionally occur. If this should happen, the following measures should be instituted immediately: 1. All babies should be cultured to determine the extent of the problem. All cultured organisms should be phage typed as an aid in epidemiologic investigation. 2. All recently discharged babies should be contacted to determine the frequency of disease in this group. 3. The nares and hands of all personnel and visitors entering the unit should be cultured with appropriate phage typing. 4. Aseptic technique, especially hand washing, should be reinforced and carefully monitored. 5. Consideration should be given to application of antibacterial substances to the babies' umbilical cords or pHisoHex® washes to the skin if the outbreak is severe. 6. If it is severe, the number of visitors and other personnel should be limited and the nursery closed to admissions until the outbreak is controlled. 7. Surveillance should be maintained for at least two to three months to be sure that the epidemic is over.
BACTERIAL DIARRHEAL OUTBREAKS
If an outbreak of bacterial diarrhea, most commonly due to gramnegative conforms, should occur in the nursery, the following measures are recommended: 1. Determine the extent of the problem by culturing all babies and possible adult carriers, such as parents and staff, in contact with the infants. 2. Discharge all babies who can go home, and close the nursery. 3. To determine the full extent of the problem, survey recently discharged babies. 4. Culture the nursery equipment, especially that containing water - e.g., handwashing sinks, respirators, nebulizers, humidifiers. 5. Reclean the nursery and equipment. 6. Restress technique, especially hand washing and stool handling. This is also true for control of Salmonella infections in the nursery. 7. If the outbreak is severe, treat all exposed babies with Colistin, 5-10 mg./kg./day, in four divided doses by mouth while they are in the hospital and for 48 hours after they get home. 8. Maintain surveillance of those infants discharged, and maintain surveillance in the unit with cultures after the nursery is reopened.
A postpartum woman with shigellosis should be isolated from the general ward. Since neonatal shigellosis is a fulminant disease, the infant should not enter the general nursery. Strict hand washing and stool precautions should be observed.
Control of gonococcal ophthalmia should be stressed, since complacency may result from a decline in its incidence. Until recently topical penicillin, neosporin, or other antibiotic ointments were used for prophylaxis of this potentially serious eye infection. Concern over topical antibiotic sensitization has led to replacement in some centers with 1 per cent silver nitrate. Silver nitrate, which is packaged in individual wax ampules that can be easily broken and applied in the delivery room before eyelid edema is apparent, should be liberally applied to the conjunctival sac, one eye at a time, and should remain in contact with the sac for 15 to 60 seconds. This application should be followed by careful washing with sterile water. Some neonatal services repeat the sterile-water rinse twice more during the first day of life in an attempt to minimize the not uncommon chemical conjunctivitis. It is extremely important to observe the infants' eyes carefully during the nursery stay, both for infectious disease and for silver nitrate reaction. This also applies to babies being treated with phototherapy for hyperbilirubinemia, in whom patches must be removed at least once a day and the eyes inspected.
Syphilis is not acquired in the hospital under ordinary circumstances. It may, however, be transmitted from a weeping mucocutaneous patch to an open lesion, usually on the hands of a staff member who is in close contact with the infant.
CHANGES IN NEONATAL BACTERIAL ECOLOGY
The need for constant surveillance of neonatal colonization and infection is graphically illustrated by the broad ecologie shifts in neonatal bacteriology in the past 40 years. In the late 1940s, coliform organisms replaced group B beta-hemolytic streptococci (GBS) as the leading neonatal pathogens. Staphylococcal epidemics were common in the 1950s but subsided in the 1960s. With this decline, coliform organisms re-emerged as the major infectious agents. In the past 10 years and especially the past two to three years, data from geographically separated centers have shown the increasingly important role played by group B Streptococcus. A typical such outbreak occurred at our medical center several years ago.35 In "early onset" GBS disease, in which the mortality has been approximately 50 per cent, the mode of transmission is from mother to neonate, in utero or during passage through a contaminated vaginal canal. Since maternal vaginal carriage rates are reported to be between 5 and 25 per cent, monitoring of these flora is essential. Treatment of the mother carrying GBS, as well as of the neonate born to such a mother, is highly controversial. It has been suggested that the "late onset" GBS disease, occurring between one and 21 weeks of age, may be acquired from the mother, nursery staff, or nursery mates in the neonatal period.36-38 This observation has considerable relevance in the surveillance of the nursery environment.
The prevention and control of bacterial infection in the nursery are clearly difficult and complicated. Close cooperation of medical, nursing, housekeeping, laboratory, and administrative personnel who are aware of the problems peculiar to the neonatal period is essential. An ongoing in-service educational program is of considerable value in the dissemination of new information to the nursery staff. Regular meetings should be held to discuss surveillance, current practices, possible changes in technique, and problems applicable to the specific neonatal unit. Knowledge can then assure effective prevention and control of neonatal bacterial infection.
1 Report of the Committee on the Control of Infectious Diseases, 13th Edition. American Academy of Pediatrics, 1961.
2. Jellard, J. Umbilical cord as a reservoir of infection in a maternity hospital. Br. Med. J. 1 (1957), 925-928.
3. Gillespie, W. A.. Simpson, K., and Tozer, R. C. Staphylococcal infection in a maternity hospital: Epidemiology and control. Lancet 2 (1958), 1075-1080.
4. Fairchild, J. P., et al. Flora of the umbilical stump. J. Pediatr. 53 (1958). 538-546.
5. Hurst, V. Transmission of hospital staphylococci among newborn infants. II: Colonization of the skin and mucous membranes of the infants. Pediatrics 25 (1960), 204-214.
6. Huntingford, P. J., et al. The problem of the neonatal umbilicus and its relation to the incidence of sepsis in a maternity unit. J. Obstet. Gynecol. Br. Commonw. 68 (1961), 179-187.
7. Love, G. J., et al. Relation of intensity of staphylococcal infection in newborn infants to contamination of nurses' hands and surrounding equipment. Pediatrics 32 (1963), 956-965.
8. Baldwin, J. N" et al. Staphylococcal infections in newborn infants. Am. J. Dis. Child. 94 (1957). 107116.
9. Wolinsky, E., et al. Acquisition of staphylococci by newborns: Direct versus indirect transmission. Lancet 2 (1960), 620-622.
10. Mortimer, E. A., Jr., et al. Transmission of staphylococci between newborns: Importance of the hands of personnel. Am. J. Dis. Child 104 (1962), 289-295.
11. Murray, J., and Caiman, R. M. Control of crossinfection by means of an antiseptic hand cream. Br. Med. J. 1 (1955). 81-83.
12. Klainer, L. M., et al. Bacitracin ointment and neonatal staphylococci. Am. J. Dis. Child 103 (1962), 564-568.
13. Laursen, H. Bacteriological colonisation of infants and mothers m a maternity unit. Acra Obsfef. Gynecol Scand. 42 (1963), 43-64.
14. Tötterman, L. E., and Autio, S. Treatment of the umbilical cord of the newborn. Acra Obstet. Gynecol. Scand. 49 (1970). 57-59.
15. Cook, J., Parrish, J. A., and Shooter, R. A. Acquisition of Staph, aureus by newborn babies in a hospital maternity department. Br. Med. J. 1 (1958), 74-76.
16. Pildes, R. S., Ramamurthy. R. S.. and Vidyasagar, D. Effect of triple dye on staphylococcal colonization in the newborn infant. J. Pediatr. 82 (1973). 987-990.
17. Simon. H. J., Yaffe, S. J., and Gluck. L. Effective control of staphylococci in a nursery. N. Engl. J. Med. 265 (1961), 1171-1176.
18. Gluck, L., and Wood, H. F. Effect of an antiseptic skin-care regimen in reducing staphylococcal colonization in newborn infants. N. Engl. J. Med. 265 (1961). 1177-1181.
19. Gezon, H. M., et al. Hexachlorophene bathing in early infancy: Effect on staphylococcal disease and infection. N. Engl. J. Med. 270 (1964). 379-386.
20. Kwong, M. S., et al. The effect of hexachlorophene on staphylococcal colonization rates in the newborn infant: A controlled study using a singlebath method. J. Pediatr. 82 (1973). 982-986.
21. Powell, H., et al. Hexachlorophene myelinopathy in premature infants. J. Pediatr. 82 (1973). 976981.
22. Shuman, R. M., Leech, R. W., and Alvord. E. C, Jr. Neurotoxicity of hexachlorophene in the human. Pediatrics 54 (1974). 689-695.
23. Kimbrough, R. D., and Gaines. T. B. Hexachlorophene effects on the rat brain. Arch. Environ. Health 23 (1971), 114-118.
24. Lockhart, J. D. How toxic is hexachlorophene? Pediatrics 50 (1972). 229-235.
25. Lampert. P. W., et al. Hexachlorophene encephalopathy. J. Neuropathol. Exp. Neurol. 32 (1973), 178.
26. Lampert. P.. O'Brien. J., and Garrett, R. Hexachlorophene encephalopathy. Acta Neuropathol. 23 (1973). 326-333.
27. Federal Register 37 (1972), 219.
28. Center for Disease Control. Morbidity Mortality Weekly Report 21 (1972), 37.
29. Standards and Recommendations for Hospital Care of Infants, Fifth Edition. American Academy of Pediatrics, 1972, pp. 67-73.
30. Forfar, J. 0., and Maccabe, A. F, Masking and gowning in nurseries for the newborn infant: Effect on staphylococcal carriage and infection. Br. Med. J. 1 (1958). 76-79.
31. Silverman, W. A., and Sinclair, J. C. Evaluation of precautions before entering a neonatal unit. Pediatrics 40 (1967), 900-901.
32. Skin care of newborns. Committee on Fetus and Newborn, American Academy of Pediatrics. Pediatrics 54 (1974), 682-683.
33. Klaus, M. H., et al. Maternal attachment. N. Engl. J. Med. 286 (1972), 460-463.
34. Kennell, J. H., et al. Maternal behavior one year after early and extended post-partum contact. Dev. Med. Child Neurol. 16 (1974), 172-179.
35. Tseng, P. I.. and Kandall, S. R. Group B streptococcal disease in neonates and infants. N. Y. State J. Med. 74 (1974). 2169-2173.
36. Franciosi. R. ?., Knostman, J. D., and Zimmerman, R. A. Group B streptococcal neonatal and infant infections. J. Pediatr. 82 (1973). 707-718.
37. Baker, C J., et al. Suppurative meningitis due to streptococci of Lancefield group B: A study of 33 infants. J. Pediatr. 82 (1973), 724-729.
38. Baker, C J., and Barrett, F. F. Transmission of group B streptococci among parturient women and their neonates. J. Pediatr. 83 (1973), 919-925.