JOURNAL OF CLINICAL MICROBIOLOGY, May 1990, p. 941-943 0095-1137/90/050941-03$02.00/0 Copyright © 1990, American Society for Microbiology
Vol. 28, No. 5
Swab-Wash Method for Quantitation of Cutaneous Microflora N. KEYWORTH,* M. R. MILLAR, AND K. T. HOLLAND Department of Microbiology, University of Leeds, Leeds LS2 9JT, United Kingdom Received 13 September 1989/Accepted 22 January 1990
We describe a comparison of the scrub-wash method of Williamson and Kligman and a swab-wash method for the enumeration of cutaneous microflora. The swab-wash method provides a less traumatic alternative to the scrub-wash method and can be used to sample the cutaneous microflora of premature neonates.
Antibiotic-resistant coagulase-negative staphylococci are the most common blood culture isolates from premature neonates in many neonatal units (1, 6). Premature neonates have a reduced level of serum opsonic activity for coagulasenegative staphylococci (5), and the numbers of coagulasenegative staphylococci associated with bacteremia may exceed 1,000 CFU ml of anti-coagulated whole blood' (8). It is likely that strains of coagulase-negative staphylococci causing bacterial sepsis in premature neonates originate on the skin surface of the patient. To quantify the skin microflora of premature neonates, a method of sampling which would not traumatize the patient was sought. Skin biopsy, while probably the most efficient sampling technique, is obviously unsuitable. The established scrub-wash method of Williamson and Kligman (11) (developed primarily for use on adults) is too aggressive for use on the fragile skin of premature neonates. We describe a comparison of the standard scrub-wash method and a less traumatic swab-wash method for the sampling and enumeration of cutaneous microflora.
before use. A sterile Dacron swab (No. MW 104; Medical Wire, Corsham, England) was moistened by immersion in 1 ml of sterile wash fluid (composition as described above for the scrub-wash method) in a bijou. Excess moisture was removed by compression of the swab against the inner wall of the bijou (otherwise, fluid on the swab was lost under the template as a result of capillary action). The swab was rubbed while being rotated between thumb and forefinger across the entire sampling site for 20 s. The swab tip was then broken off into the bijou containing the wash fluid. The bijou was capped and mixed vigorously for 20 s before quantitation of wash fluid microbial content. Quantitation of microorganisms in wash fluids. Samples were serially diluted 10-fold in half-strength wash fluid (in distilled water). Each dilution was vigorously mixed, and 100-,ll volumes of diluted and undiluted wash fluid were spread over the surface of the following selective and nonselective media: heated 5% (vol/vol) horse blood agar (CM 55; Oxoid, Basingstoke, England) incubated at 37°C in air for 4 days; Sabouraud dextrose agar (Oxoid CM 41) with chloramphenicol (50 mg liter') incubated at 37°C in air for 7 days; 5% (vol/vol) horse blood agar with neomycin (final concentration, 28 mg liter ') incubated at 37°C anaerobically for 7 days; aerobic coryneform agar with and without furazolidone (6 mg liter') (2) incubated at 37°C aerobically for 2 days; reinforced clostridial agar with and without furazolidone (2) incubated at 37°C for 7 days anaerobically; and Pityrosporum agar (10) incubated at 34°C aerobically in a moist chamber for 14 days. After incubation, colonies were enumerated with a colony counter (Gallenkamp CNW32501OX; Loughborough, United Kingdom) to give a total viable count expressed as CFU centimeter -2. Identification of organisms. Preliminary identification of staphylococcal species was based on colonial morphology, positive catalase test, slide coagulase test, and plate DNase test (3). Ten colonies of each colonial type were identified as to species by the scheme of Kloos and Schleifer (9). Propionibacteria were identified on the basis of colony and Gram stain morphology and classified as to species as described by Holland (7). Analysis of quantitative data. The correlation coefficient of log1o CFU cm 2 of modified scrub-wash and swab-wash method results was determined, using the OXSTAT statistics program (Microsoft Version 4; Holmes, Jones, Wiggins & Walter) on an Amstrad 1512 HD10 PC computer.
now
MATERIALS AND METHODS
Sampling of cutaneous microflora. Informed consent was obtained from the parents of each of the infants sampled. Local ethical committee approval was obtained for this study. Scrub-wash method. The steel ring described in the scrubwash method of Williamson and Kligman (11) is physically too large for sampling skin microflora in neonates. Therefore, a smaller ring enclosing 1.8 cm2 was used. After the ring was placed securely against the skin, 1.0 ml of sterile wash fluid (0.075 mol liter-' of phosphate buffer [pH 7.9] containing 0.1% [vol/vol] Triton X-100 [Sigma, Poole, England]) was added to cover the sampling site. The skin enclosed within the ring was scrubbed with a flat-ended Teflon (E. I. duPont de Nemours & Co., Inc.) rod (diameter, 1.0 cm). To minimize trauma to the skin of neonates, the scrubbing time was reduced from two 1-min scrubs to one 20-s scrub. Extending the scrubbing time beyond 20 s did not increase the number of microorganisms recovered from a variety of skin sites from infants sampled within 12 h of death. The wash fluid was collected for quantitation of microorganisms. For ethical reasons, extended scrubbing times were not used on living neonates. Swab-wash method. Templates of flexible polyethylene sheet (thickness, 1.0 mm) were used to delineate a sampling area of 3 cm2. The polyethylene templates were stored in 70% (vol/vol) aqueous ethyl alcohol and allowed to dry *
RESULTS The standard scrub-wash method of Williamson and Kligman (11) was compared with the modified scrub-wash method and the swab-wash sampling method at adjacent sites on adult volar forearm skin. This site was chosen
Corresponding author. 941
942
KEYWORTH ET AL.
J. CLIN. MICROBIOL.
TABLE 1. Comparison of skin flora results, using scrub-wash and swab-wash methods in full-term neonates Neonate
1 2 3 4 5 6 7 8 9 10
Postnatal age (days)
5 6 9 5 10 6 7 6 10 9
Mean SD
No. of aerobic bacteria (log1o CPU cm-2) on:
Periumbilicus
Back
Forearm
Foreleg
Chest
Scrub
Swab
Scrub
Swab
Scrub
Swab
Scrub
Swab
Scrub
Swab
2.9 0.8 1.9 2.5 2.6 1.5 1.2 1.8 1.8 2.3
3.0 1.3 1.9 3.0 2.5 1.6 1.5 2.2 1.7 2.2
2.5 1.7
2.8 0.8 0.8 1.7 2.0 0 1.5 2.4 1.6 2.8
3.4 1.2 1.9 2.0 2.1 1.7 0.2 1.0 1.5 1.8
2.0 0.8 2.5 3.1 2.5 2.2 0 1.7 3.6 1.4
2.2 1.0 1.9 2.9 3.0 2.6 1.8 1.8 2.2 2.0
3.3 0 1.35
3.4 1.8 1.33
1.0 2.0 1.7 3.6 1.4
2.5 1.5 1.5 2.5 2.2 1.9 2.3 1.7 0.8 2.6
1.1 1.0 1.5 2.5 2.0 2.6
2.2 1.5 1.7 2.9 2.0 2.5
1.93 0.65
2.1 0.6
1.9 0.72
2.0 0.6
1.6 0.91
1.7 0.8
2.0 1.1
2.1 0.6
1.7 0.9
2.1 0.7
1.7 2.0 1.5
a
a-, Not sampled.
because of the absence of large pore structures, the predominance of vellus hair, and the low level of sebaceous activity (approximating the skin surface of premature neonates). The volar forearms of seven adult volunteers were sampled on three consecutive days. The mean results of the three samplings from each adult ranged from 19 to 485 CFU cm 2 for the standard scrub-wash method, 4 to 71 CFU cm-2 for the modified scrub-wash method, and 18 to 440 CFU cm-2 for the swab-wash method. Repeated sampling of the same volar forearm site of an additional seven adult volunteers with the swab-wash method showed that a mean of 53.3% (with a range of 29.7 to 78.4%) of the total count of aerobic microflora removed after 10 swab-wash samplings was removed with the first sampling. A comparison of skin flora results (Table 1) from 10 full-term neonates on postnatal days 5 to 10, using small ring scrub-wash and swab-wash methods to sample adjacent sites, revealed a high level of correlation of the results from the two techniques (correlation coefficient, r = 0.718). The neonates sampled were healthy, and while all reacted well to the swab-wash method, they were generally disturbed by the greater interference involved in the scrub-wash technique. The use of the swab-wash method for sampling cutaneous microflora at various sites in a premature neonate with a gestational age of 25 to 40 weeks and a birth weight of 660 g (postnatal age, 12 days) is illustrated in Table 2. In this
neonate, the periumbilical site had been treated with 0.33% hexachlorophane powder (Hough Hoseasons, Manchester, England), which probably accounts for the absence of staphylococci. Other sites showed a range of bacterial species and numbers. Bacterial counts at all sites were similar and, by comparison with adults (for whom counts of 106 cm-2 are not uncommon [11]), were relatively low. DISCUSSION A method other than biopsy for sampling cutaneous microflora that effectively removes all of the bacteria from a skin site has yet to be described. Williamson and Kligman (11) reported 97 to 98% recovery of aerobic microflora, using a scrub-wash technique. The epidermis of premature neonates lacks a cornified layer and well-developed secretory structures and may be only three cells thick (4). A major consideration in sampling the skin of premature neonates is the damage that might be caused by the sampling procedure. Consequently, a compromise which, while removing the majority of organisms, does not traumatize the skin of premature neonates was sought. The swab-wash method described above provides a less traumatic alternative to the scrub-wash method of Williamson and Kligman (11) for the sampling of cutaneous microflora of premature neonates. For ethical reasons, it is not possible to compare the
TABLE 2. Comparison of skin flora of a Sitea
1 2 3 4 5
6 7 8
Staphylococcus
CPU cm-2 43.3 13.3 6.6
586.6 16.6 918.3 153.3
premature neonate at various sites
sppb Species
PropCiFoerfum acnes cm2
S. epidermidis, S. warneri, S. hominis 2 S. epidermidis, S. simulans S. epidermidis, S. saprophyticus S. aureus, S. epidermidis, S. haemolyticus 1, S. warneri, S. hominis 1, S. hominis 2, S. saprophyticus, S. capitis S. epidermidis S. aureus, S. epidermidis, S. haemolyticus 1 S. aureus, S. epidermidis, S. haemolyticus 1
1, Forearm; 2, upper arm; 3, back; 4, periumbilicus; 5, upper leg; 6, lower leg; 7, chest; 8, b In addition, there was one Micrococcus sp. at site 1. a
CFU of Candida albicans cm-
Total count (CFU cm-)
6.6 3.3 16.6 20.0
1.6
43.3 19.9 11.5 16.6 606.6
3.3 3.3 16.6
3.3
19.9 924.9 169.9
forehead.
VOL. 28, 1990
SWAB-WASH METHOD FOR SAMPLING CUTANEOUS MICROFLORA
swab-wash method with a scrub-wash method for the sampling of skin of premature neonates, but the results from mature neonates and adult volar forearm skin suggest that the two methods give comparable results and that a large proportion of cutaneous microflora is removed by one swabwash sampling. LITERATURE CITED 1. Battisti, O., R. Mitchison, and P. A. Davis. 1981. Changing blood culture isolates in a referral neonatal intensive care unit. Arch. Dis. Child. 56:775-778. 2. Cove, J. H., and E. A. Eady. 1982. A note on a selective medium for the isolation of cutaneous propionibacteria. J. Appl. Bacteriol. 53:289-292. 3. Disalvo, J. W. 1958. Deoxyribonuclease and coagulase acting micrococci. U.S. Armed Forces M. J. Tech. Bull. 9:191-196. 4. Evans, N. J., and N. Rutter. 1986. Development of the epidermis in the newborn. Biol. Neonate 49:74-80. 5. Fleer, A., L. J. Gerards, P. Aerts, A. C. Westerdaal, R. C.
6.
7.
8. 9. 10. 11.
943
Senders, H. VanDik, and J. Verhoef. 1985. Opsonic defence to Staphylococcus epidermidis in the premature neonate. J. Infect. Dis. 152:930-937. Freedman, R. M., D. L. Ingram, I. Goss, A. Ehrenkranz, J. B. Warshaw, and R. S. Baltimore. 1981. A half century of neonatal sepsis at York 1928-1978. Am. J. Dis. Child. 135:140-144. Holiand, K. T. 1989. Microbiology of acne, p. 178-218. In W. J. Cunliffe (ed.), Acne. Martin Dunitz, London. Kite, P., V. Langdale, N. Todd, M. R. Millar, and P. Mackay. 1989. Direct isolation of coagulase negative staphylococci from neonatal blood samples. J. Hosp. Infect. 14:135-140. Kloos, W. E., and K. H. Schleifer. 1975. Simplified scheme for routine identification of human Staphylococcus species. J. Clin. Microbiol. 1:82-88. Leeming, J. P., and F. H. Notman. 1987. Improved methods for isolation and enumeration of Malassezia furfur from human skin. J. Clin. Microbiol. 25:2017-2019. Williamson, P., and A. M. Kligman. 1965. A new method for the quantitative investigation of cutaneous bacteria. J. Invest. Dermatol. 45:498-530.
Vol. 28, No. 5
Swab-Wash Method for Quantitation of Cutaneous Microflora N. KEYWORTH,* M. R. MILLAR, AND K. T. HOLLAND Department of Microbiology, University of Leeds, Leeds LS2 9JT, United Kingdom Received 13 September 1989/Accepted 22 January 1990
We describe a comparison of the scrub-wash method of Williamson and Kligman and a swab-wash method for the enumeration of cutaneous microflora. The swab-wash method provides a less traumatic alternative to the scrub-wash method and can be used to sample the cutaneous microflora of premature neonates.
Antibiotic-resistant coagulase-negative staphylococci are the most common blood culture isolates from premature neonates in many neonatal units (1, 6). Premature neonates have a reduced level of serum opsonic activity for coagulasenegative staphylococci (5), and the numbers of coagulasenegative staphylococci associated with bacteremia may exceed 1,000 CFU ml of anti-coagulated whole blood' (8). It is likely that strains of coagulase-negative staphylococci causing bacterial sepsis in premature neonates originate on the skin surface of the patient. To quantify the skin microflora of premature neonates, a method of sampling which would not traumatize the patient was sought. Skin biopsy, while probably the most efficient sampling technique, is obviously unsuitable. The established scrub-wash method of Williamson and Kligman (11) (developed primarily for use on adults) is too aggressive for use on the fragile skin of premature neonates. We describe a comparison of the standard scrub-wash method and a less traumatic swab-wash method for the sampling and enumeration of cutaneous microflora.
before use. A sterile Dacron swab (No. MW 104; Medical Wire, Corsham, England) was moistened by immersion in 1 ml of sterile wash fluid (composition as described above for the scrub-wash method) in a bijou. Excess moisture was removed by compression of the swab against the inner wall of the bijou (otherwise, fluid on the swab was lost under the template as a result of capillary action). The swab was rubbed while being rotated between thumb and forefinger across the entire sampling site for 20 s. The swab tip was then broken off into the bijou containing the wash fluid. The bijou was capped and mixed vigorously for 20 s before quantitation of wash fluid microbial content. Quantitation of microorganisms in wash fluids. Samples were serially diluted 10-fold in half-strength wash fluid (in distilled water). Each dilution was vigorously mixed, and 100-,ll volumes of diluted and undiluted wash fluid were spread over the surface of the following selective and nonselective media: heated 5% (vol/vol) horse blood agar (CM 55; Oxoid, Basingstoke, England) incubated at 37°C in air for 4 days; Sabouraud dextrose agar (Oxoid CM 41) with chloramphenicol (50 mg liter') incubated at 37°C in air for 7 days; 5% (vol/vol) horse blood agar with neomycin (final concentration, 28 mg liter ') incubated at 37°C anaerobically for 7 days; aerobic coryneform agar with and without furazolidone (6 mg liter') (2) incubated at 37°C aerobically for 2 days; reinforced clostridial agar with and without furazolidone (2) incubated at 37°C for 7 days anaerobically; and Pityrosporum agar (10) incubated at 34°C aerobically in a moist chamber for 14 days. After incubation, colonies were enumerated with a colony counter (Gallenkamp CNW32501OX; Loughborough, United Kingdom) to give a total viable count expressed as CFU centimeter -2. Identification of organisms. Preliminary identification of staphylococcal species was based on colonial morphology, positive catalase test, slide coagulase test, and plate DNase test (3). Ten colonies of each colonial type were identified as to species by the scheme of Kloos and Schleifer (9). Propionibacteria were identified on the basis of colony and Gram stain morphology and classified as to species as described by Holland (7). Analysis of quantitative data. The correlation coefficient of log1o CFU cm 2 of modified scrub-wash and swab-wash method results was determined, using the OXSTAT statistics program (Microsoft Version 4; Holmes, Jones, Wiggins & Walter) on an Amstrad 1512 HD10 PC computer.
now
MATERIALS AND METHODS
Sampling of cutaneous microflora. Informed consent was obtained from the parents of each of the infants sampled. Local ethical committee approval was obtained for this study. Scrub-wash method. The steel ring described in the scrubwash method of Williamson and Kligman (11) is physically too large for sampling skin microflora in neonates. Therefore, a smaller ring enclosing 1.8 cm2 was used. After the ring was placed securely against the skin, 1.0 ml of sterile wash fluid (0.075 mol liter-' of phosphate buffer [pH 7.9] containing 0.1% [vol/vol] Triton X-100 [Sigma, Poole, England]) was added to cover the sampling site. The skin enclosed within the ring was scrubbed with a flat-ended Teflon (E. I. duPont de Nemours & Co., Inc.) rod (diameter, 1.0 cm). To minimize trauma to the skin of neonates, the scrubbing time was reduced from two 1-min scrubs to one 20-s scrub. Extending the scrubbing time beyond 20 s did not increase the number of microorganisms recovered from a variety of skin sites from infants sampled within 12 h of death. The wash fluid was collected for quantitation of microorganisms. For ethical reasons, extended scrubbing times were not used on living neonates. Swab-wash method. Templates of flexible polyethylene sheet (thickness, 1.0 mm) were used to delineate a sampling area of 3 cm2. The polyethylene templates were stored in 70% (vol/vol) aqueous ethyl alcohol and allowed to dry *
RESULTS The standard scrub-wash method of Williamson and Kligman (11) was compared with the modified scrub-wash method and the swab-wash sampling method at adjacent sites on adult volar forearm skin. This site was chosen
Corresponding author. 941
942
KEYWORTH ET AL.
J. CLIN. MICROBIOL.
TABLE 1. Comparison of skin flora results, using scrub-wash and swab-wash methods in full-term neonates Neonate
1 2 3 4 5 6 7 8 9 10
Postnatal age (days)
5 6 9 5 10 6 7 6 10 9
Mean SD
No. of aerobic bacteria (log1o CPU cm-2) on:
Periumbilicus
Back
Forearm
Foreleg
Chest
Scrub
Swab
Scrub
Swab
Scrub
Swab
Scrub
Swab
Scrub
Swab
2.9 0.8 1.9 2.5 2.6 1.5 1.2 1.8 1.8 2.3
3.0 1.3 1.9 3.0 2.5 1.6 1.5 2.2 1.7 2.2
2.5 1.7
2.8 0.8 0.8 1.7 2.0 0 1.5 2.4 1.6 2.8
3.4 1.2 1.9 2.0 2.1 1.7 0.2 1.0 1.5 1.8
2.0 0.8 2.5 3.1 2.5 2.2 0 1.7 3.6 1.4
2.2 1.0 1.9 2.9 3.0 2.6 1.8 1.8 2.2 2.0
3.3 0 1.35
3.4 1.8 1.33
1.0 2.0 1.7 3.6 1.4
2.5 1.5 1.5 2.5 2.2 1.9 2.3 1.7 0.8 2.6
1.1 1.0 1.5 2.5 2.0 2.6
2.2 1.5 1.7 2.9 2.0 2.5
1.93 0.65
2.1 0.6
1.9 0.72
2.0 0.6
1.6 0.91
1.7 0.8
2.0 1.1
2.1 0.6
1.7 0.9
2.1 0.7
1.7 2.0 1.5
a
a-, Not sampled.
because of the absence of large pore structures, the predominance of vellus hair, and the low level of sebaceous activity (approximating the skin surface of premature neonates). The volar forearms of seven adult volunteers were sampled on three consecutive days. The mean results of the three samplings from each adult ranged from 19 to 485 CFU cm 2 for the standard scrub-wash method, 4 to 71 CFU cm-2 for the modified scrub-wash method, and 18 to 440 CFU cm-2 for the swab-wash method. Repeated sampling of the same volar forearm site of an additional seven adult volunteers with the swab-wash method showed that a mean of 53.3% (with a range of 29.7 to 78.4%) of the total count of aerobic microflora removed after 10 swab-wash samplings was removed with the first sampling. A comparison of skin flora results (Table 1) from 10 full-term neonates on postnatal days 5 to 10, using small ring scrub-wash and swab-wash methods to sample adjacent sites, revealed a high level of correlation of the results from the two techniques (correlation coefficient, r = 0.718). The neonates sampled were healthy, and while all reacted well to the swab-wash method, they were generally disturbed by the greater interference involved in the scrub-wash technique. The use of the swab-wash method for sampling cutaneous microflora at various sites in a premature neonate with a gestational age of 25 to 40 weeks and a birth weight of 660 g (postnatal age, 12 days) is illustrated in Table 2. In this
neonate, the periumbilical site had been treated with 0.33% hexachlorophane powder (Hough Hoseasons, Manchester, England), which probably accounts for the absence of staphylococci. Other sites showed a range of bacterial species and numbers. Bacterial counts at all sites were similar and, by comparison with adults (for whom counts of 106 cm-2 are not uncommon [11]), were relatively low. DISCUSSION A method other than biopsy for sampling cutaneous microflora that effectively removes all of the bacteria from a skin site has yet to be described. Williamson and Kligman (11) reported 97 to 98% recovery of aerobic microflora, using a scrub-wash technique. The epidermis of premature neonates lacks a cornified layer and well-developed secretory structures and may be only three cells thick (4). A major consideration in sampling the skin of premature neonates is the damage that might be caused by the sampling procedure. Consequently, a compromise which, while removing the majority of organisms, does not traumatize the skin of premature neonates was sought. The swab-wash method described above provides a less traumatic alternative to the scrub-wash method of Williamson and Kligman (11) for the sampling of cutaneous microflora of premature neonates. For ethical reasons, it is not possible to compare the
TABLE 2. Comparison of skin flora of a Sitea
1 2 3 4 5
6 7 8
Staphylococcus
CPU cm-2 43.3 13.3 6.6
586.6 16.6 918.3 153.3
premature neonate at various sites
sppb Species
PropCiFoerfum acnes cm2
S. epidermidis, S. warneri, S. hominis 2 S. epidermidis, S. simulans S. epidermidis, S. saprophyticus S. aureus, S. epidermidis, S. haemolyticus 1, S. warneri, S. hominis 1, S. hominis 2, S. saprophyticus, S. capitis S. epidermidis S. aureus, S. epidermidis, S. haemolyticus 1 S. aureus, S. epidermidis, S. haemolyticus 1
1, Forearm; 2, upper arm; 3, back; 4, periumbilicus; 5, upper leg; 6, lower leg; 7, chest; 8, b In addition, there was one Micrococcus sp. at site 1. a
CFU of Candida albicans cm-
Total count (CFU cm-)
6.6 3.3 16.6 20.0
1.6
43.3 19.9 11.5 16.6 606.6
3.3 3.3 16.6
3.3
19.9 924.9 169.9
forehead.
VOL. 28, 1990
SWAB-WASH METHOD FOR SAMPLING CUTANEOUS MICROFLORA
swab-wash method with a scrub-wash method for the sampling of skin of premature neonates, but the results from mature neonates and adult volar forearm skin suggest that the two methods give comparable results and that a large proportion of cutaneous microflora is removed by one swabwash sampling. LITERATURE CITED 1. Battisti, O., R. Mitchison, and P. A. Davis. 1981. Changing blood culture isolates in a referral neonatal intensive care unit. Arch. Dis. Child. 56:775-778. 2. Cove, J. H., and E. A. Eady. 1982. A note on a selective medium for the isolation of cutaneous propionibacteria. J. Appl. Bacteriol. 53:289-292. 3. Disalvo, J. W. 1958. Deoxyribonuclease and coagulase acting micrococci. U.S. Armed Forces M. J. Tech. Bull. 9:191-196. 4. Evans, N. J., and N. Rutter. 1986. Development of the epidermis in the newborn. Biol. Neonate 49:74-80. 5. Fleer, A., L. J. Gerards, P. Aerts, A. C. Westerdaal, R. C.
6.
7.
8. 9. 10. 11.
943
Senders, H. VanDik, and J. Verhoef. 1985. Opsonic defence to Staphylococcus epidermidis in the premature neonate. J. Infect. Dis. 152:930-937. Freedman, R. M., D. L. Ingram, I. Goss, A. Ehrenkranz, J. B. Warshaw, and R. S. Baltimore. 1981. A half century of neonatal sepsis at York 1928-1978. Am. J. Dis. Child. 135:140-144. Holiand, K. T. 1989. Microbiology of acne, p. 178-218. In W. J. Cunliffe (ed.), Acne. Martin Dunitz, London. Kite, P., V. Langdale, N. Todd, M. R. Millar, and P. Mackay. 1989. Direct isolation of coagulase negative staphylococci from neonatal blood samples. J. Hosp. Infect. 14:135-140. Kloos, W. E., and K. H. Schleifer. 1975. Simplified scheme for routine identification of human Staphylococcus species. J. Clin. Microbiol. 1:82-88. Leeming, J. P., and F. H. Notman. 1987. Improved methods for isolation and enumeration of Malassezia furfur from human skin. J. Clin. Microbiol. 25:2017-2019. Williamson, P., and A. M. Kligman. 1965. A new method for the quantitative investigation of cutaneous bacteria. J. Invest. Dermatol. 45:498-530.
