MEDICAL JOURNAL OF AUSTRALIA, Volume 1, Number 10: Pages 332-4,
March 5, 1977.

Reports of Cases


The Neonatal Unit, Princess Margaret Hospital for Children, Perth

Med. J. Aust., 1977, 1: 332-334.

This paper presents four cases of fulminating neonatal sepsis with meningitis. In each infant, there was evidence of an infected circumcision wound. Two infants had Escherichia coli and two had Group B haemolytic streptococcus cultured from the cerebrospinal fluid. One infant died. The risk of introducing infection through iatrogenic portals of entry is a definite problem in the neonate. Circumcision is an unnecessary routine procedure, which puts the infant at risk.

CIRCUMCISION is a very common procedure, called by some a “social ritual”,1 carried out on the male neonate in the Australian community. Doctors for some years have been discussing the advantages and disadvantages of this procedure. In the literature, especially in Australia, the various complications listed are haemorrhage, meatal ulcer, meatal stenosis, urethral fistula, phimosis, (if there is incomplete removal of the prepuce), and infection (mainly local). In the past four years at Princess Margaret Hospital, there have been four patients admitted for fulminating neonatal sepsis with meningitis after circumcision. (Three of the circumcisions were carried out in the hospital of delivery, and one was done in a general practitioner’s surgery.) In all cases there was evidence of infected circumcision wounds with the same organism found in the cerebrospinal fluid.


A male infant was circumcised on the fourth day of life. On admission to the hospital on the sixth day, he had three episodes of apnoea and stiffness consistent with a tonic convulsion. On examination, he was a pale irritable infant, his temperature was 38·4°C, his pulse rate was 160/min, and his respiration was 72/min. He had “twitching” movements and increased tone in all limbs. The circumcision site looked infected. A lumbar puncture yielded turbid cerebrospinal fluid, of which the white cell count was 25,000/mm3 (polymorphs, 100%) the protein level was 5·28 g/l. and the sugar level was zero. Gram-positive cocci were present on a smear. Later, beta haemolytic streptococcus Group B was grown from the cerebrospinal fluid. This organism was also cultured from the blood, the circumcision site and the umbilicus. Penicillin, chloramphenicol and sulphadimidine were administered. The infant had seizures for the next three days and was given anticonvulsants, phenobarb and paraldehyde. At follow-up, at the age of 31/2 years, his development was with normal limits.


At the age of five days, a male infant was circumcised. The next day he was pale and not feeding. On the seventh day of life he presented febrile, fitting and with a poor peripheral circulation. On admission to hospital he was a desperately ill infant with a temperature of more than 45°C, convulsions and cyanosis, a pulse rate of 180/min, and a respiration rate of 100/min. He was given diazepam (Valium) intravenously and cooled. His temperature fell to 38.5°C After the fits had ceased the child had prolonged apnoea, and required assisted ventilation. A lumbar puncture yielded turbid cerebrospinal fluid, whose protein level was 7 g/l., glucose level was 1·4 g/l., white cell count was 9,000/mm3. Gram-negative bacilli were seen and culture yielded Escherichia coli. E. coli was also was also isolated from penile, nasal and umbilical swabs. Despite gentamicin administered intrathecally and parenteral administration of ampicillin, gentamicin and anticonvulsants, the child continued to have fits. His condition deteriorated, and he died three days after admission to hospital.


A male infant was circumcised on the eighth day of life. Within 48 hours, he was lethargic and had increasing feeding problems. On the day of admission to hospital he was irritable and his mother had noticed a bulging fontanelle. On examination, he was found to be a slightly irritable infant, with a very full fontanelle, and the sutures were all splayed. The circumcision site was not healed and looked infected. A lumbar puncture yielded cloudy, cerebrospinal fluid of which the protein level was 1·2 g/l., the glucose level was 2·1 mmol/l., the white cell count was 3,600/mm3, of which 84% were polymorphs) the red cell count was 20/mm3, and the culture yielded Escherichia coli. A urine speciment taken soon after treatment was commenced, revealed a white cell count of 3,700/mm3, and a red cell count of 10/mm3. Gram-negative rods were seen, but the culture did not yield any growth. Gentamicin was administered both intrathecally and intramuscularly and ampicillin was administered intravenously. The infant made good progress. When he was last reviewed at the age of seven months, his development was normal.


A male infant was circumcised on the eleventh day of life, and the next day his parents noticed he was febrile, not feeding and crying. When he was examined, no abnormality was detected and no treatment was given. On the fourteenth day the infant presented with a persistence of poor feeding, held his head back and had a strange cry. On examination he was found to be an irritable staring child, with a high-pitched cry and full fontanelle, lying stiffly with head retracted. The circumcision site was swollen and red and looked infected. A lumbar puncture yielded cloudy cerebrospinal fluid of which the white cell count was 312/mm3 (of which 38% were polymorphs), the red cell count was 235/mm3, the protein count was 4·8 g/l. and the sugar level was 0.3 mmol/l. Gram-positive cocci were present on a smear. Counter immune electrophoresis of the cerebrospinal fluid, urine, throat, umbilicus and circumcision site yielded beta haemolytic streptococcus Group B. The infant was treated initially with penicillin and gentamicin and then just penicillin. On the evening of admission to hospital he had tonic and focal convulsions. Despite administration of phenobarb and diazepam (Valium), he had fits for four days. When he was reviewed at the age of five months, he was found to be an irritable baby with severe feeding problems, and showed signs of cerebral palsy.


These four cases demonstrate a close association between the event of circumcision followed by local clinical infection resulting in septicaemia, and meningitis. In Case 3, although no organism was grown from the urine and peripheral sites, on microscopy examination the urine had a white cell count of 3,800-mm3, ad Gram-negative rods were seen. We believe this is circumstantial evidence that E. coli was present at the circumcision site which on admission to hospital was clinically infected. Although infection is a complication of circumcision, there is poor documentation in the literature of this topic. Most general practitioners are familiar with local infection, but not with the severe septicaemia that can result. Two cases of septicaemia associated with circumcision were reported by Birrell1 in 1965. Fredman2 states, from the commonwealth Bureau of Census and Statistics and the Victorian Department of Health’s neonatal death survey, that two babies in a 10-year period had died as a direct result of neonatal circumcision with post operative infection and septicaemia.

In Melbourne, in 1969, Fredman found that about 70% of the male infants were circumcised. In Perth we checked with infant health centres to try to determine the frequency of circumcision. One survey was carried out in a suburb of relative socioeconomic affluence, and the other was of poorer socio-economic standing. Over the past two years about 60% and 30% respectively of the male infants were found to be circumcised. In view of the large number of infants circumcised, infection is therefore a rare complication. However, any elective procedure that may have such severe complications should be carefully considered. Of our cases, one infant died, one infant shows evidence of psychomotor retardation, and two infants appear normal, the elder being 3 ½ years old.

At various times it has been stated that circumcision prevents penile and cervical carcinoma. However, it appears that as long as adequate penile hygiene is carried out circumcision confers little protection. 3 4 5 The report of the Ad-Hoc Task Force on Circumcision in the U.S.A. states: “Circumcision is a surgical procedure that requires careful aseptic techniques, systematized postoperative observation and evaluation after discharge from the hospital. Prematurity, neonatal illness, any congenital anomaly (especially hypospadias) or bleeding problems are absolute contraindications to neonatal circumcision.” These are principles with which we would concur.

The late James Spence6 wrote to a colleague: “The anatomists have not studied the form and evolution of the preputial orifice . . . They do not understand that Nature does not intend it to be stretched or retracted. What looks like a pin point opening at 7 months will become a wide channel of communication at 17 . . . Nature is a possessive mistress and whatever mistakes she makes about the structure of the less essential organs such as the brain and stomach in which she is not much interested, you can be sure she knows best of the genital organs.”

Circumcision should not be the routine normal management for male newborn infants. Our cases demonstrate one of the rare, but severe problems that may result.


1 BIRRELL, R., A case against circumcision. MED. J. AUST., 1965; 2: 393.

2 FREDMAN, R. M., Neonatal circumcision: A general practitioners survey. MED. J. AUST., 1969, I: 117.

3 LEITCH, I. O. W., Circumcision: A continuing enigma. Aust. Paediatr. J., 1970; 6: 59.

4 Report of the Ad-Hoc Task Force on Circumcision, Pediatrics, 1975; 56: 610.

5 WRIGHT, J. E. Non-therapeutic circumcision, MED. J. AUST., 1967, I: 1083.

6 SPENCE, J., quoted in Spence on Circumcision, Lancet, 1964; 2: 902.

* Senior Registrar.
† Neonatal Pediatrician.

Address for reprints: Dr J. M. Scurlock, Neonatal Unit, Princess Margaret Hospital for Children, G.P.O. Box D184, Perth, W.A. 6001.

(File revised 28 January 2007)

Return to CIRP library