Percutaneous absorption of chlorhexidine in neonatal cord care

Short reports Percutaneous absorption of chlorhexidine in neonatal cord care P J AGGETT, L V COOPER, S H ELLIS, AND J McAINSH Queen Charlotte's Maternity Hospital, London, and ICI Limited, Pharmaceuticals Division, Alderley Park, Macclesfield SUMMARY The percutaneous absorption of chlorhexidine during its routine use in topical antiseptic preparations used in umbilical cord care was investigated by determining plasma chlorhexidine concentrations at ages 5 and 9 days. These showed that percutaneous absorption of chlorhexidine occurred in preterm neonates treated with a 1 % solution of chlorhexidine in ethanol, but not in term infants similarly treated, or in preterm infants treated only with a dusting powder containing 1 % chlorhexidine and 3 % zinc oxide. Chlorhexidine (CHD) (1, 6 -di (4-chlorophenyldiguanido) hexane) (Hibitane ICI Ltd UK) is a topical antiseptic which has a wide activity against Gram-positive and Gram-negative bacteria and some yeasts.1 Its ability to disinfect skin2 and reduce newborn staphylococcal infection3 is comparable with that of hexachlorophane (HCP). The demonstration of percutaneous absorption of HCP, of its neurotoxicity in man and animals, and the subsequent realisation of the risk incurred by its excessive application to injured skin or to the intact skin of preterm newborn babies has been reviewed.4 5 Restrictions on the routine use of HCP led to an increased incidence of staphylococcal infections in neonatal intensive care units which indicated the efficacy of topical antisepsis in routine neonatal care, and necessitated the cautious reintroduction of HCP.5 CHD is an alternative antiseptic agent to HCP. However, an earlier investigation suggested that the daily use of a CHD preparation for bathing newborn infants might result in percutaneous absorption of the antiseptic.6 The two studies described in this paper were undertaken to investigate plasma CHD levels in term and preterm newborn babies who were exposed to applications of CHD preparations for umbilical cord care. Patients and methods In the first study (Study A) 25 term (mean birthweight Arch Dis Child: first published as 10.1136/adc.56.11.878 on 1 November 1981. Downloaded from http://adc.bmj.com/ on January 22, 2022 by guest. Protected by copyright. Archives of Disease in Childhood, 1981, 56, 878-891 3 * 34 kg; range 2 * 13-4 22) and 23 preterm babies with gestational ages between 31 and 36-5 weeks (mean birthweight 2 * 08 kg; range 1 * 29-2*79) were investigated. Three term and 4 preterm infants were black, one preterm infant was Asian, and the remainder were white. At delivery, the umbilical cord was cut and ligated 4 cm from the skin and treated with a solution of 1 % CHD in ethanol, and a 1 % CHD and 3 % zinc oxide dusting powder. This treatment was repeated 4-hourly thereafter for at least 9 days. With the sole exception of venepunctures the CHD solution was also used for topical antisepsis during all invasive procedures, such as lumbar punctures and umbilical catheterisation. A second study (Study B) was performed because Study A had shown detectable plasma concentrations of CHD in preterm infants; in this, only the dusting powder was used for cord care but otherwise the procedure was identical with that used in the first study. The alcoholic CHD solution was used exclusively for invasive procedures as described above. The group comprised 29 preterm infants admitted consecutively to the intensive care nursery. Their mean gestational age was 32-5 weeks (range 26-36 5) and their mean birthweight was 1 870 kg (range I -02 53). These infants included one Asian and two black babies, and one set of triplets. Gestational age was estimated by neurological and morphological examination7 8 and, where possible, from serial antenatal ultrasonography. One preterm infant in Study A and 9 in Study B had umbilical catheters inserted during their initial management; in addition in Study B, one 4-day-old infant had an exchange transfusion and 4 others had phototherapy for at least 3 days. Venous blood was taken from an antecubital vein on days 5 and 9 after birth. Samples were taken with others needed either for clinical care or for screening for inborn errors of metabolism. The skin site was cleansed twice with alcohol, once with water, and then dried before venepuncture. In Study A cord blood was also collected. The plasma CHD content 878 was determined by gas liquid chromatography.9 Calibration curves were fitted by the method of least squares and 95 % confidence limits were determined. The data from Study A were non-normally distributed and plasma CHD values from term and preterm infants were compared using the nonparametric Wilcoxon's rank sum test (single sided); the median value being used as a measure of location rather than the mean. Both these studies were approved by the hospital's ethical committee and the informed consent of both parents was obtained whenever possible. Results The results of Study A (Table 1) show no significant difference in the cord blood plasma CHD levels of term and preterm neonates at birth. Whereas plasma CHD levels were constant in the term infants, there was a progressive increase in plasma CHD concentrations in the preterm neonates and their values on days 5 and 9 were significantly higher than those in term infants. However, no correlation between birthweight or gestational age and plasma CHD was apparent in either study. In Study B, 13 samples of plasma from 10 infants produced measurable peaks of gas liquid chromatography and 5 of these (from 4 infants) exceeded the limit of detection as judged by peak ratios (which could not be allied to specific values due to variations Table I Median plasma chlorhexidine concentrations (ng/ml) in term and preterm infants in Study A Sample Time taken Median CHD concentration Significance Wilcoxon's (ng/ml) rank sum Term Preterm test Cord blood Day 0 -9 1 Venous blood Day 5 -9 10 Not significant P < 0-0)1 Venous blood Day 9 -9 32 P < 0-0001 The estimated negative value is included rather than setting this to zero to avoid bias in the statistical comparisons being made. Conversion: traditional to S I units-CHD 1 ng/ml = 1-98 nmol/l. in blood volumes measured) (Table 2). Three of these infants (of 30 weeks' gestation, or less) had had umbilical catheters and in 2, both plasma samples gave signals. The fourth infant of 33 weeks' gestational age had had phototherapy. Discussion These results confirm the earlier observation that percutaneous absorption of CHD may occur in the newborn6 and demonstrate that it does so to a greater extent in preterm than in term infants. Studies in animals indicate that the epidermal barrier matures during the last quarter of gestation ;10 the cutaneous permeability of the human newborn decreases with gestational age.1" The principal cutaneous permeability barrier is the stratum corneum but organic solvents (especially those-such as ethanol-which are miscible in both lipid and water) increase its porosity and greatly reduce its efficiency.12 The preterm infant would therefore be at an even greater risk of increased cutaneous permeability if its skin were exposed to ethanol. The increasing plasma CHD levels in the preterm infants of Study A and the absence of this feature in those infants from Study B, who were only exposed to CHD zinc oxide dusting powder for routine cord care, suggests that the repeated use of the ethanol CHD solution was responsible for percutaneous absorption of CHD by the preterm babies in the first study. The humidity within an incubator and exposure to phototherapy may increase cutaneous vascular perfusion thereby enhancing skin permeability. These factors were more prevalent in the second study and were probably therefore of less significance than ethanol in influencing percutaneous absorption of CHD, but we cannot exclude the possibility that such factors might have contributed to the presence of detectable plasma CHD in one of the infants from Study B. The significant plasma CHD concentrations in the other 3 infants from the second Table 2 Plasma chlorhexidine concentrations in Study B infants Case 1 2 3 4 5 6 7 8 9 10 Sex M M F F M M M M F F Birthweight Gestational age (g) (weeks)s 1500 1560 1740 1900 2100 1320 2210 1830 1565 1240 33 30 33 33 34 29 34 33 29 28 Chlorhexidine peak ratiot and concentrations (ng/ml) Day 5 Day 9 (0*0049) 23 (0-0199) *23 ND (0-0185)*48 (0-0095) 16 (.0-148) *16 (0*0039) 6 ND (0-0055) 9 ND ND (0-0176) *92 (0*0065) 17 ND (0-0104) 15 ND ND (0*0591) 9 (0-0053) 7 (0.0121) *45 Procedures Nil Umbilical catheter Nil Nil Nil Umbilical cath-ter Phototherapy Phototherapy Umbilical catheter Umbilical catheter, parenteral nutrition, ventilated tLimit of detection for peak ratio: 0.0106. * Ratio above limit of detection. ND none detected. Arch Dis Child: first published as 10.1136/adc.56.11.878 on 1 November 1981. Downloaded from http://adc.bmj.com/ on January 22, 2022 by guest. Protected by copyright. Percutaneous absorption of chlorhexidine in neonatal cord care 879 study could have resulted from inadvertent introduction of CHD during umbilical catheterisation. However, the absence of such levels in other infants from Study B, who underwent invasive procedures, suggests that accidental introduction of CHD can be avoided and that the occasional use of an ethanol CHD solution for topical antisepsis in preterm infants need not result in high plasma CHD concentrations such as those encountered with prophylactic use. There is no evidence that the presence of CHD in plasma is indicative of possible toxicity. In one study, designed to match one which demonstrated HCP neurotoxicity, CHD was detected in the liver, kidneys, and adipose tissue of newborn rhesus monkeys bathed regularly for 90 days in an 80% CHD detergent-based solution, but there was no clinical or histological evidence of toxicity and there was no detectable CHD in their brains.13 Four cases of accidental intravenous administration have been reported in human adults14 but with the exception of extensive haemolysis which occurred in 3 of them there were no toxic effects. CHD would therefore appear to be of fairly low toxicity and of proved efficacy in reducing umbilical colonisation in the neonate.3 15 This study shows that significant percutaneous absorption of CHD is unlikely in term infants but may occur in preterm neonates treated with an ethanol solution of the antiseptic for cord care. This may however, be considerably reduced by avoiding the use of ethanol in the routine cord care of preterm infants. Miss Lesley Cooper died on 15 April 1977. We thank Mr J L Honigman for advice and help, Dr D R Harvey and Dr A P Norman for permission to study their patients, and Mr B Holmes for measurement of chlorhexidine levels. L V C was supported by a grant from Birthright and P J A by grants from the Medical Research Council and Rank Prize Fund. References Beeuwkes H. The use of chlorhexidine. Antonie van Leeuwenhoek 1958; 24: 49-62. Lowbury E J, Lilly H A. Use of 4% chlorhexidine detergent solution (Hibiscrub) and other methods of skin disinfection. Br Med J 1973; i: 510-5. 3Alden V G, Burman D, Simpson R A, Fysh J, Gillespie W A. Comparison of hexachlorophene and chlorhexidine powders in prevention of neonatal infection. Arch Dis Child 1980; 55: 277-80. 4Powell H C, Lampert P W. Hexachlorophene toxicity. In: Vinken P J, Bruyn G W, eds. Handbook of clinical neurology. Vol. 37. Amsterdam: North Holland, 1979: 479-509. 5 Tyrala E E, Hillman L S, Hillman R E, Dodson W E. Clinical pharmacology of hexachlorophene in newborn infants. J Pediatr 1977; 91: 481-6. 6 Cowen J, Ellis S H, McAinsh J. Absorption of chlorhexidine from the intact skin of newborn infants. Arch Dis Child 1979; 54: 379-83. 7Robinson R J. Assessment of gestational age by neurological examination. Arch Dis Child 1966; 41: 437-47. 8 Dubowitz L M S, Dubowitz V, Goldberg C. Clinical assessment of gestational age in the newborn infant. JPediatr 1970; 77: 1-10. 9Holmes B F. A note on hibitane assay with final iodination. Blood drugs and other analytical challenges. In: Reid E, ed. Methodological surveys in biochemistry. Vol. 7. Chichester: Ellis Horwood, 1978: 328-30. 10 Singer E J, Wegmann P C, Lehman M D, Christensen M S, Vinson L J. Barrier development ultrastructure and sulfhydryl content of the fetal epidermis. J Soc Cosmet Chem 1971; 22: 119-37. Nachman R L, Esterly N B. Increased skin permeability in preterm infants. J Pediatr 1971; 79: 628-32. 12 Scheuplien R. Site variations in diffusion and permeability. In: Jarrett A, ed. The phvsiology and pathology of the skiin. Vol. 5. The sweat glands, skin permeation, lymphatics, the nails. New York: Academic Press, 1978: 1731-52. 13 Case D E, McAinsh J, Rushton A, Winrow M J. Chlorhexidine: attempts to detect percutaneous absorption in man. In: Williams J D, Geddes A M, eds. Special problems in chemotherapy. New York: Plenum, 1976: 367-74. 14 ICI Limited, Pharmaceuticals Division. Unpublished reports 1981. 15 George R H. The effect of antimicrobial agents on the umbilical cord. In: Williams J D, Geddes A M, eds. Special problems in chemotherapy. New York: Plenum, 1976: 415-9. 2 Correspondence to Dr P J Aggett, Department of Physiology, Marischal College, Aberdeen AB9 lAS. Received 12 December 1980 Device for continuous urine collection in the newborn R J LUND, H B VALMAN, AND ADAM PLATT Department of Paediatrics, Northwick Park Hospital, Harrow SUMMARY A device for the continuous collection of urine from newborn infants is described. This apparatus replaces the nursing tray of the Vickers 59 incubator. Arch Dis Child: first published as 10.1136/adc.56.11.878 on 1 November 1981. Downloaded from http://adc.bmj.com/ on January 22, 2022 by guest. Protected by copyright. 880 Aggett, Cooper, Ellis, and McAinsh Collection of urine in the newborn infant is notoriously difficult, especially if prolonged collection is required for either diagnostic or research purposes. Various methods previously used depend on modifications of urine or colostomy bags, which tend to