Relationship Between Bacterial Colonization and Infection
The role of bacterial colonization and how it relates to neonatal infection has been disputed greatly by researchers. Some researchers report a positive relationship between colonization and infection.[5,710,15,17,24,25,27] Others report an unclear relationship between colonization and infection.[6,13,28] Lastly, some researchers report that there is no significant relationship between colonization and infection.[1,2,4,16,21]
Research findings have shown no relationship between a maternal carrier state and an increased colonization of newborns. Research findings also demonstrate no relationship between staff carrier state, maternal carrier state, or environmental contamination with increased infant colonization. In fact the research that demonstrated this was done during a transition into a newly built nursery and was able to present findings that indicated that the cross-contamination was coming from the infants themselves. Another study was also able to relate an increase in colonization with an increase in length of maternal or neonatal hospital stay.
Three studies of particular interest discussed bacterial colonization and the development of disease as they attempted to adopt cord care techniques that did not treat the cord at all. Curiously, these studies were not cited in the current and most recent literature that recommends no cord care or natural healing. Although there are several pathogens such as S. aureus, S. epidermidis , and B-hemolytic streptococcus that may colonize the umbilical stump, S. aureus was the major pathogen associated with the infectious outbreaks and infant mortality in maternity and neonatal units in the 1950s and 1960s when antimicrobial treatment of umbilical cords was initially instituted.[21,27] Due to this knowledge, the studies of interest represented in this article are those that examined S. aureus colonization in particular.[25,27]
The first study was conducted in 1992 due to concern over reported drug toxicity to nervous tissue such as nerves and nerve cells, prolonged cord separation and healing, and staff questioning the need to treat the cord at all (based on recent literature recommendations). This study was undertaken to explore the possibility of abandoning antimicrobial treatment of umbilical cords. The study was a prospective study using a convenience sample of 102 neonates born in a UK hospital excluding those with early premature rupture of membranes (>24 hours) and those admitted to NICU. The study compared the colonization rate and length of cord separation between treatment with alcohol and 0.33% hexachlorophane powder with tap water and dry care. Cultures were taken from the umbilical stump on the fifth day of life and colonization and length of time for cord separation were compared using odds ratio (OR) and 95% confidence interval (CI). The study showed that those babies whose umbilical cords were untreated had a colonization rate 1.75 times greater than those treated (OR = 1.75; 95% CI, 1.08, 2.85). It also showed that cord separation time is shorter with no cord treatment (OR = 1.78; 95% CI, 1.04, 3.05). The study was conducted under the premise that prevention of infection is the primary objective of cord care. Although prolonged cord attachment is also an issue, in this study it was not the primary purpose of cord care because there are fewer complications with prolonged cord attachment than the complications associated with umbilical and neonatal infection. The study demonstrated that colonization quickly increased once antiseptic practices were modified. The researchers feel that those advocating no treatment have not presented a balanced equation and argument to include the considerable cost and consequences of sporadic cases and outbreaks of staphylococcal infection. They suggest that this perspective may be related to the fact that this new generation of health care providers have not worked or experienced the decades when S. aureus was a major perinatal pathogen. They feel that, without further evidence, antiseptic use should be continued in cord treatment. They regard limiting staphylococcal colonization as the single most important aim in cord care.
The second study was conducted in 1992 in a hospital where routine cord care consisted of leaving the umbilical area dry without applications of any kind. The convenience sample consisted of 370 newborn infants born at a UK hospital. The incidence of S. aureus colonization and its relationship to infection was examined. Cultures of the umbilical stump were obtained regardless of infant location at 48 hours and at 8 to 9 days after birth. The level of colonization was compared with the incidence of infection using the Chi-square test. Differences in infant location at time of collection were compared using ANOVA. Changes of colonization and differences between locations were compared using ANCOVA. The findings demonstrated that the incidence of colonization was significantly higher with no cord treatment (68% of infants at 48 hours). Forty-nine percent (171 neonates) of the colonized infants were heavily colonized. They were able to significantly show that neonates with heavy colonization at 48 hours were at a greater risk to develop an infection ( P < 0.001). Twenty-eight percent of this group (44 infants) developed a staphylococcal infection (12% of the entire sample size). In view of these findings, the researchers believe that reducing colonization rates from time of delivery might also lead to a reduction in postdischarge infection.
The third study was conducted due to doubts regarding the need for antiseptic treatment of the cord by some researchers. The 1992 six-month prospective study was conducted in a UK hospital that had been using dry cord care. The study compared colonization rate and length of cord separation (using Chi-square) with cords treated with hexachlorophane, chlorhexidine, and early bath or dry cord care. A convenience sample of 630 newborns was cultured at the umbilicus on the morning of the 3rd day of life and a second swab was obtained at discharge if the infant stayed longer than 3 days. A significant decrease in colonization was seen with chlorhexidine ( P < 0.001), a modest significant decrease was seen with chlorohexaphane ( P < 0.025), and no significant decrease was seen with an early bath ( P = 0.9). The study revealed a high colonization rate along with intermittent outbreaks of pemphigus neonatorum in association with dry cord care. The investigators feel the study confirms that the use of dry cord care alone will lead to an unacceptably high colonization rate with S. aureus . They also feel that the relationship between cord colonization and infection is well established and that many health care providers that are promoting dry cord care are not aware of the actual occurrences of postdischarge infections. They recommend the use of antiseptic treatment for umbilical cords because cross-infection happened principally when no antiseptic was used on cords.
Other studies exploring natural healing and the effects of antimicrobials on the length of time for cord separation are not convinced that colonization correlates with infection. Some researchers believe that cord separation is most likely mediated through leukocyte infiltration and subsequent digestion of the umbilical cord. Interventions such as antimicrobials will impede or inhibit migration and activity of leukocytes. The presence of bacteria in wounds does not necessarily have to be associated with detrimental outcomes. The stimulatory effect of bacteria on wound healing has long been recognized. All wounds are colonized, but not all wounds are infected. Therefore, colonization does not necessarily lead to infection. Bacteria are believed to help initiate the inflammatory or first stage of wound healing. Some of the major functions of this stage include removal of cellular debris and the attack and removal of infectious agents. Chemical mediators and chemoattractants are considered important to guide the neutrophils and macrophages to the site.
Based on this body of knowledge the cessation of routine antimicrobial treatment of the umbilical cord can be recommended. One study conducted from 1995 to 1996 compared alcohol cleaning with natural drying. The study consisted of 1,811 newborns. The study found that there were statistically significant differences in cord separation time. The natural drying group time was shorter by 1.7 days. No infections occurred with either treatment. Based on this study, the researchers recommend discontinuing the use of alcohol for newborn cord care.
Another study conducted in 1996 was a randomized trial to evaluate the difference in alcohol and water in promoting umbilical cord separation. The sample size consisted of 148 term, healthy newborns. The findings supported the hypothesis that treating the cord prolonged the time for cord separation. The time was lengthened by a mean of 2 to 3 days. Colonization studies supported the hypothesis that there was an insignificant difference in colonization with the two methods of treatment. The authors speculate that there is no reason to suspect that this would not be the case with any antimicrobial solution. They suggest that there is no evidence to continue the use of alcohol as routine umbilical cord treatment.
The research proposing the benefits of natural healing is limited because it has only examined the use of alcohol compared with natural healing and did not include other agents. It is difficult to surmise if similar findings would be attainable if comparisons were made with the other antimicrobials currently used. These studies were also conducted in Canada. It is unclear whether these findings could be duplicated in the United States. Evidence has shown an unexplained lengthened cord separation time for infants in this country compared with other developed countries. The time of separation ranging from 9.8 days to 15 days has been attributed to variances in frequency of cord care, length of hospital stay, and environmental factors such as humidity. These explanations have been based on speculation and are yet to be proven.
Before generalizing the findings of the Canadian studies, more research comparing outcomes with natural healing and antimicrobials other than alcohol would be beneficial. It is also imperative to conduct this research using a population that demonstrates characteristics that would allow it to be generalized to infants in the United States.
The hospital costs of umbilical cord treatment vary with the method of application and length of hospital stay. One study estimates the monetary costs of 10 minutes to teach cord care at $20/hr to be approximately $16,000 at a hospital with 4,800 births per year. Another study cites the specific costs of alcohol application to be from $0.52 to $2.74 per newborn depending on application method and length of hospital stay. No references were found estimating the cost of other antimicrobials.
Although there was no cost of natural drying, costs could be increased due to expensive cultures obtained because of concerns of health care providers. More than three times more cultures were done on infants attempting natural healing than those using alcohol. The use of cultures could potentially be decreased with proper education and increased awareness on the natural process of cord separation involving odor and an infected appearance.
NAINR. 2004;4(4):211-222. © 2004 W.B. Saunders
Cite this: Trends in Umbilical Cord Care: Scientific Evidence for Practice - Medscape - Dec 01, 2004.