Journal
of Hospital
Infection
Staphylococcus
(1992) 21, 205-211
aureus colonization in a Darlington V. Stark
of the newborn
hospital
and S. P. Harrisson”
Department of Clinical Practice, Darlington Memorial Hospital, Darlington, Co. Durham DL3 6HX, UK and *Health Care Research Unit, University of Durham, Durham DHl 3JT, UK Accepted for publication
20 March
1992
Summary:
Evidence from research studies suggests a relationship between neonatal infection with Staphylococcus amem and the level of umbilical colonization. During a 3-month prospective study (September-December 1990) the incidence and levels of S. LZUWUScolonization were determined for all 370 live births in the Darlington Unit by taking swabs at 48 h and S/9 days from the base of the umbilical cord. Infants were situated in one of four locations (The Special Care Unit, one of two wards or home) and the location at the time of swabbing was recorded. The overall percentages colonized at 48 h and S/9 d were 68% and 65% respectively. Forty-eight hours after delivery, 49% showed a high level of S. aureus colonization. Although the percentage of infants colonized with S. aureu~ was almost identical at each sampling, only 62% were culture-positive on both occasions. Between 48 h and 8/9days, 12% (44) of infants developed S. aureus infections of whom 35 showed heavy growth. Statistical analysis showed a significant relationship between levels of colonization at the two sampling times but no relationship between location and levels of colonization over the time period. Keywords:
Staphylococcus
aureus; umbilical
colonization;
newborn
infant.
Introduction
Staphylococcus
aweus is a common commensal organism carried by 25% of the population and also is a cause of skin pustules, pemphigus neonatorum, conjunctivitis, umbilical infection and breast abscess.’ The umbilicus is usually the first site of colonization.2 Evidence suggests a strong correlation between neonatal infection and the level of umbilical colonization.3 The efficacy of hexachlorophane in the treatment of S. aweus skin colonization was shown as early as 1952.4 Its use declined following reports adverse reactions were of its toxicity and absorption in animals, ‘A although not found when emulsions and talcum powder preparations were used correctly.7,8 Clinical studies of alternative cord-care treatments show that dry cord Correspondence DHl
3JT,
UK.
to: Dr S. P. Harrisson,
Health Care Research Unit,
University
of Durham,
Durham
206
V. Stark
and
S. P. Harrisson
care is not effective in reducing S. UUY~US colonization3,9,‘0 and that colonization increased with the use of alcohol.” Treatment with 4% chlorhexidine (‘Hibiscrub’ ICI Pharmaceuticals) reduced S. ~UY~US colonization12~‘3 and was safe in terms of levels of absorption.‘4,‘5 Other clinical studies of treatments using various combinations of triple dye, alcohol and sulphadiazinei6 and differing concentrations of chlorhexidine and benzine17 found them to be equally effective in reducing infection rates. The purpose of this study was to determine the incidence and level of S. aUreUS colonization in newborn infants in the Darlington Unit. Methods
All live infants born in the Darlington Maternity Unit between September and December 1990 who were discharged into the community within the Darlington area were included in the study (N= 370). Routine cord care consisted of leaving the umbilical area dry and no applications of any kind were used. Two dry swabs were taken from the base of the umbilical cord of every infant, one at 48 h after delivery and the second on the eighth or ninth day; the ninth day was used if the eighth fell on a Sunday, when laboratory facilities were not available. Swabs were taken by hospital or community midwifery staff who had been instructed in the method, and were placed in transport medium. Most infants had been discharged home before the second swab was taken. The location of the infant at the time each swab was taken was recorded (Special Care Baby Unit [SCBU], Wards A or B or infants’ homes). Additional swabs from appropriate sites were taken from infants with clinical signs of infection and the site of infection and the laboratory results were recorded. A 2-week pilot study was carried out to test the recording methods and to enable midwives to familiarize themselves with the required procedures; the results of the pilot study were not analysed further. Swabs were plated on blood agar and mannitol salt agar and incubated for 48 h. Levels of S. ~UY~UScolonization were recorded on a scale of 1-5 (as used in the Darlington laboratory): Level 1: no growth Level 2: < 20 colonies (scanty growth) Level 3: 20-100 colonies ( + ) Level 4: > 100 colonies but less than confluent growth ( + + ) Level 5: confluent growth (+ + +) Statistical analysis The relationship between the level of colonization and the location of infants at 48 h and S/9 days was tested using two separate one-way analysis of variance (ANOVA). Changes in levels of colonization and differences between locations were tested using analysis of covariance (ANCOVA).
Staphylococcus
aureus
colonization
of the newborn
207
Differences between the colonization levels at 48 h of those infants with clinical signs of infection were analysed using the x2 test. Values of I’< 0.05 were taken as significant. Results
Forty-eight hours after delivery, 314 (85%) infants were located in one of the two wards (40%, Ward A,; 45%, Ward B), 26 (7%) were in the SCBU and 23 (6%) had been discharged home; the location of the remaining seven (a’/) was unrecorded. Although all infants had swabs taken at 48 h and S/9 days the results for 22 (6%) samples at 48 h and 11 (3’/ o ) samples at S/9 days were not recorded. The percentages of neonates colonized with 5’. aUreUS relate to samples where a result was recorded (N=348, 1st swab; N=359, 2nd swab). All percentages were rounded to the nearest whole number. Level of S. aureus colonization The overall percentages of neonates colonized by S. aweus at 48 h and S/9 days were 68% and 65% respectively (see Figure 1). The levels of colonization at 48 hours and S/9 days are shown in Figure 2. Although these percentages are very similar, 97 infants were not colonized on both occasions. Forty-eight per cent (47), not colonized at 48 h were colonized at S/9 days, and conversely, 52% (50), colonized at 48 h showed no
60
50 7 e 2 .-t a .-F 1
40
30
20
IO
0 Negative
Positive Umbilical
swabs.
lst=48
h. 2nd=8/9
Figure 1. Staphylococcus CZUY~US colonization. Overall percentage and S/9 d. Negative = no colonization, positive = total colonized. 17 = first swab, II0 = second swab.
days
of infants
colonized
at 48 h
V. Stark
208
and
S. P. Harrisson 49
501
I
2 Colonization
(I=nil,
3 5= heavy
growth)
5 1st swab=48
h. 2nd
swab=819
days
Figure 2. Staphylococcus aweus colonization. Level of colonization at 48 h and S/9 days. (Level 1 = no growth, level 2 = scanty growth, ( < 20 colonies per agar plate), level 3 = 20-100 colonies, level 4= > 100 colonies, level 5 =confluent growth). 0 = first swab (N= 348), •@= second swab (N= 359).
colonization at S/9 days. Twenty-eight per cent (14) of the latter group developed a staphylococcal infection. Forty-eight hours after delivery 49% (171) neonates showed a heavy growth of S. UUY~US(level 5). Staphylococcal and other infections Between 48 h and S/9 days 12% (44) of infants developed S. aureus infections; of these, results were recorded for 41 samples. The infections were septic spots/blisters (27) or eye (9) and umbilical infections (5). Of the 41 infants, 35 (85%) were colonized at level 5 (heavy growth), three (7%) at level 4 and two (5%) at level 3; one was not colonized. There was a significant difference in infection rate (x2 test; P~O.001) between those neonates with level 5 colonization and those with either no colonization (level 1) or with levels 2, 3 or 4 colonization. Table I shows the level of colonization at 48 h of the 41 infants for whom samples were available. Five per cent (20) of infants developed other infections with Group B haemolytic streptococci (17), enterococci (2) or Pseudomonas (1). Relationship between level of colonization and location of infants at 48 h and 819 d Two separate one-way analysis of variance (ANOVA) for the first and second measurements (variables x and y respectively) showed no significant relationship between the level of colonization and the location of infants at
Staphylococcus
aureus
colonization
of the newborn
209
Table I. Level of colonization at 48 h of 41 of the 44 infants who developed Staphylococcus aureus infections between 48 h and S/9 days Level
of colonization
No. infants
5
35 (85)
4
3 (7)
3
2 (9
(%)
either 48 h or S/9 d (ANOVA for x: F=0.8793; P=O.4250. y: F= 1.6146; P= 0.1856). Changes in levels of colonization, between 48 h and S/9 d, were looked for using an analysis of covariance (ANCOVA) to allow for differences in location; this showed a significant relationship between the two measurements (P
of Hospital
Infection
Staphylococcus
(1992) 21, 205-211
aureus colonization in a Darlington V. Stark
of the newborn
hospital
and S. P. Harrisson”
Department of Clinical Practice, Darlington Memorial Hospital, Darlington, Co. Durham DL3 6HX, UK and *Health Care Research Unit, University of Durham, Durham DHl 3JT, UK Accepted for publication
20 March
1992
Summary:
Evidence from research studies suggests a relationship between neonatal infection with Staphylococcus amem and the level of umbilical colonization. During a 3-month prospective study (September-December 1990) the incidence and levels of S. LZUWUScolonization were determined for all 370 live births in the Darlington Unit by taking swabs at 48 h and S/9 days from the base of the umbilical cord. Infants were situated in one of four locations (The Special Care Unit, one of two wards or home) and the location at the time of swabbing was recorded. The overall percentages colonized at 48 h and S/9 d were 68% and 65% respectively. Forty-eight hours after delivery, 49% showed a high level of S. aureus colonization. Although the percentage of infants colonized with S. aureu~ was almost identical at each sampling, only 62% were culture-positive on both occasions. Between 48 h and 8/9days, 12% (44) of infants developed S. aureus infections of whom 35 showed heavy growth. Statistical analysis showed a significant relationship between levels of colonization at the two sampling times but no relationship between location and levels of colonization over the time period. Keywords:
Staphylococcus
aureus; umbilical
colonization;
newborn
infant.
Introduction
Staphylococcus
aweus is a common commensal organism carried by 25% of the population and also is a cause of skin pustules, pemphigus neonatorum, conjunctivitis, umbilical infection and breast abscess.’ The umbilicus is usually the first site of colonization.2 Evidence suggests a strong correlation between neonatal infection and the level of umbilical colonization.3 The efficacy of hexachlorophane in the treatment of S. aweus skin colonization was shown as early as 1952.4 Its use declined following reports adverse reactions were of its toxicity and absorption in animals, ‘A although not found when emulsions and talcum powder preparations were used correctly.7,8 Clinical studies of alternative cord-care treatments show that dry cord Correspondence DHl
3JT,
UK.
to: Dr S. P. Harrisson,
Health Care Research Unit,
University
of Durham,
Durham
206
V. Stark
and
S. P. Harrisson
care is not effective in reducing S. UUY~US colonization3,9,‘0 and that colonization increased with the use of alcohol.” Treatment with 4% chlorhexidine (‘Hibiscrub’ ICI Pharmaceuticals) reduced S. ~UY~US colonization12~‘3 and was safe in terms of levels of absorption.‘4,‘5 Other clinical studies of treatments using various combinations of triple dye, alcohol and sulphadiazinei6 and differing concentrations of chlorhexidine and benzine17 found them to be equally effective in reducing infection rates. The purpose of this study was to determine the incidence and level of S. aUreUS colonization in newborn infants in the Darlington Unit. Methods
All live infants born in the Darlington Maternity Unit between September and December 1990 who were discharged into the community within the Darlington area were included in the study (N= 370). Routine cord care consisted of leaving the umbilical area dry and no applications of any kind were used. Two dry swabs were taken from the base of the umbilical cord of every infant, one at 48 h after delivery and the second on the eighth or ninth day; the ninth day was used if the eighth fell on a Sunday, when laboratory facilities were not available. Swabs were taken by hospital or community midwifery staff who had been instructed in the method, and were placed in transport medium. Most infants had been discharged home before the second swab was taken. The location of the infant at the time each swab was taken was recorded (Special Care Baby Unit [SCBU], Wards A or B or infants’ homes). Additional swabs from appropriate sites were taken from infants with clinical signs of infection and the site of infection and the laboratory results were recorded. A 2-week pilot study was carried out to test the recording methods and to enable midwives to familiarize themselves with the required procedures; the results of the pilot study were not analysed further. Swabs were plated on blood agar and mannitol salt agar and incubated for 48 h. Levels of S. ~UY~UScolonization were recorded on a scale of 1-5 (as used in the Darlington laboratory): Level 1: no growth Level 2: < 20 colonies (scanty growth) Level 3: 20-100 colonies ( + ) Level 4: > 100 colonies but less than confluent growth ( + + ) Level 5: confluent growth (+ + +) Statistical analysis The relationship between the level of colonization and the location of infants at 48 h and S/9 days was tested using two separate one-way analysis of variance (ANOVA). Changes in levels of colonization and differences between locations were tested using analysis of covariance (ANCOVA).
Staphylococcus
aureus
colonization
of the newborn
207
Differences between the colonization levels at 48 h of those infants with clinical signs of infection were analysed using the x2 test. Values of I’< 0.05 were taken as significant. Results
Forty-eight hours after delivery, 314 (85%) infants were located in one of the two wards (40%, Ward A,; 45%, Ward B), 26 (7%) were in the SCBU and 23 (6%) had been discharged home; the location of the remaining seven (a’/) was unrecorded. Although all infants had swabs taken at 48 h and S/9 days the results for 22 (6%) samples at 48 h and 11 (3’/ o ) samples at S/9 days were not recorded. The percentages of neonates colonized with 5’. aUreUS relate to samples where a result was recorded (N=348, 1st swab; N=359, 2nd swab). All percentages were rounded to the nearest whole number. Level of S. aureus colonization The overall percentages of neonates colonized by S. aweus at 48 h and S/9 days were 68% and 65% respectively (see Figure 1). The levels of colonization at 48 hours and S/9 days are shown in Figure 2. Although these percentages are very similar, 97 infants were not colonized on both occasions. Forty-eight per cent (47), not colonized at 48 h were colonized at S/9 days, and conversely, 52% (50), colonized at 48 h showed no
60
50 7 e 2 .-t a .-F 1
40
30
20
IO
0 Negative
Positive Umbilical
swabs.
lst=48
h. 2nd=8/9
Figure 1. Staphylococcus CZUY~US colonization. Overall percentage and S/9 d. Negative = no colonization, positive = total colonized. 17 = first swab, II0 = second swab.
days
of infants
colonized
at 48 h
V. Stark
208
and
S. P. Harrisson 49
501
I
2 Colonization
(I=nil,
3 5= heavy
growth)
5 1st swab=48
h. 2nd
swab=819
days
Figure 2. Staphylococcus aweus colonization. Level of colonization at 48 h and S/9 days. (Level 1 = no growth, level 2 = scanty growth, ( < 20 colonies per agar plate), level 3 = 20-100 colonies, level 4= > 100 colonies, level 5 =confluent growth). 0 = first swab (N= 348), •@= second swab (N= 359).
colonization at S/9 days. Twenty-eight per cent (14) of the latter group developed a staphylococcal infection. Forty-eight hours after delivery 49% (171) neonates showed a heavy growth of S. UUY~US(level 5). Staphylococcal and other infections Between 48 h and S/9 days 12% (44) of infants developed S. aureus infections; of these, results were recorded for 41 samples. The infections were septic spots/blisters (27) or eye (9) and umbilical infections (5). Of the 41 infants, 35 (85%) were colonized at level 5 (heavy growth), three (7%) at level 4 and two (5%) at level 3; one was not colonized. There was a significant difference in infection rate (x2 test; P~O.001) between those neonates with level 5 colonization and those with either no colonization (level 1) or with levels 2, 3 or 4 colonization. Table I shows the level of colonization at 48 h of the 41 infants for whom samples were available. Five per cent (20) of infants developed other infections with Group B haemolytic streptococci (17), enterococci (2) or Pseudomonas (1). Relationship between level of colonization and location of infants at 48 h and 819 d Two separate one-way analysis of variance (ANOVA) for the first and second measurements (variables x and y respectively) showed no significant relationship between the level of colonization and the location of infants at
Staphylococcus
aureus
colonization
of the newborn
209
Table I. Level of colonization at 48 h of 41 of the 44 infants who developed Staphylococcus aureus infections between 48 h and S/9 days Level
of colonization
No. infants
5
35 (85)
4
3 (7)
3
2 (9
(%)
either 48 h or S/9 d (ANOVA for x: F=0.8793; P=O.4250. y: F= 1.6146; P= 0.1856). Changes in levels of colonization, between 48 h and S/9 d, were looked for using an analysis of covariance (ANCOVA) to allow for differences in location; this showed a significant relationship between the two measurements (P
