THE NEW ENGLAND JOURNAL OF MEDICINE, Volume 22, Number 18, Pages 1312-1315,
May 3, 1990.



Ronald L. Poland, M. D.

      ROUTINE neonatal circumcision is the subject of much debate in the United States. The practice has attracted passionate advocates and critics. The analysis presented here examines the available medical evidence relating to routine neonatal circumcision in order to weigh its risks, costs, and benefits. This assessment is meant to help clinicians formulate their own recommendations about the advisability of prophylactic neonatal circumcision.

      Of the approximately 1.95 million boys born in the United States during 1987, 1.19 million, or 61 percent were circumcised, according to the National Center for Health Statistics. Fees paid to physicians range between $75 and $100, and most hospitals do not charge separately for circumcision. In states that use the Medicaid prospective payment system, the facility's charges for circumcision are part of a single payment to hospitals for the care of a normal newborn infant. Some private insurance plans reimburse physicians for the procedure and others do not. The reimbursement of physicians for circumcisions performed on patients covered by medical assistance or Medicaid in Pennsylvania or Michigan ranges from $20 to $28, and each state has its own payment structure. About 30 percent of all newborns were insured though Medicaid programs in 1988, according to state reports. The cost of universal circumcision in the United States would therefore be about $140 million a year.

      Relatively few medical procedures are routinely recommended for the care of infants and children. In general, the risks that stem from not undergoing a procedure must be substantial, and the risks and costs of undergoing it relatively small. Procedures that satisfy these requirements include the intramuscular administration of vitamin K for the prevention of hemorrhagic disease of the newborn and immunizations against polio, diphtheria, and measles. Influenza vaccine on the other hand, is recommended only for children in high risk groups, such as those with chronic pulmonary disease, since the risk of influenza in otherwise healthy children is not considered high enough to justify the minimal risks of vaccination.1 A good general rule is to withhold the routine application of procedures to large groups of people unless it is clear that the benefits far outweigh the risks and costs. Is this principle satisfied for routine neonatal circumcision?

      The immediate risks of neonatal circumcision are small. Any operative procedure, including circumcision, may be accompanied by infection of the wound or excessive bleeding. Wiswell and Geschke2 found a rate of early complications (mostly local infection and bleeding) or 0.19 percent for routine infant circumcision of 136,086 boys. Since their data were collected retrospectively and included only diagnoses entered on inpatient records, they recognized, that they might have underestimated the rate of complications. Sporadic cases of operative misadventure have been reported in which the glans penis was damaged or partially amputated, or too much skin was removed from the shaft of the penis.3 Meatal ulceration and meatal stenosis have also been described consequences of neonatal circumcision.4 Most of these complications are treatable and have no long term consequences, although there have been rare cases of gender reassignment due to surgical error. Pain and physiologic evidence of stress accompany neonatal circumcision when no anesthesia is administered. These problems can be alleviated by the use of regional anesthesia such as dorsal penile nerve block.5 However we have no reliable estimate of the risks, either short- or long-term, of administering dorsal nerve block to large numbers of newborns. There is a need for more information about the safety of penile nerve block and perhaps safer techniques that alleviate the pain of circumcision. In summary there is little controversy about the magnitude of the operative risk: in experienced hands, circumcision involves a very low but not completely negligible risk.

      The controversy resides instead in estimating the benefits that may accrue from routine neonatal circumcision. Many benefits have been postulated. For example, it has been said that neonatal circumcision facilitates the maintenance of genital hygiene6 and may prevent local infections such as balanoposthitis (inflammation of the glans and foreskin),6 urinary tract infections,7-9 and venereal diseases.10 In addition, it has been claimed that in circumcised men and their female partners, carcinoma of the penis and uterine cervix are less prevalent.6,11 What is the evidence for such reported benefits.?

      The facilitation of cleanliness has been proposed as an argument for routine infant circumcision. It is difficult to understand, however, how maintaining cleanliness in an infant could be enhanced by routine circumcision. The booklet of the American Academy of Pediatrics on the care of the uncircumcised penis12 recommends no special cleaning of the foreskin other than than involved in regular bathing. It states "The uncircumcised penis is easy to keep clean. No special care is required!" The foreskin is seldom retractable at birth, and full spontaneous separation of the foreskin may not occur until the age of five. Retracting the foreskin should not be painful, and forced retraction should be avoided. Once full retraction can be carried out with ease, only occasional retraction and cleaning are recommended until puberty. At puberty boys should be instructed to clean under the foreskin as part of each bath or shower.12,13

      Of course, neither posthitis (inflammation of the foreskin) nor paraphimosis (constriction of the glans by retraction and swelling of an inflamed foreskin can occur in circumcised men and boys. Balanitis (inflammation of the glans penis can occur in those who are circumcised as well as those who are not. It has been reported that 5 to 10 percent of those not circumcised at birth eventually undergo circumcision for one of these problems,6,14 but it is difficult to confirm this contention with recent data. In a prospective study covering the eight years of life, 146 boys in New Zealand had an 11.1 percent incidence of penile problems, such as inflammation or inadequate circumcision, as compared with an 18.8 percent incidence of similar problems (e. g., inflammation or phimosis) among 445 uncircumcised boys (difference not significant).15 Thus, circumcision may reduce the incidence of penile inflammatory disorders but proper hygiene will probably have a similar benefit.

      Several recent reports have linked an increased incidence of urinary tract infection with the uncircumcised state.7-9 In retrospective studies, Wiswell et al. reported that boys who were not circumcised had a 10- to 20- fold higher risk of urinary tract infections.7,8 Herzog's study excluded almost half the original cohort from the data analysis, and this may have biased the findings. It would be very difficult to perform a randomized, controlled trial of the effects of neonatal circumcision on the incidence of urinary tract infection. However, these retrospective studies depended heavily on the coding of inpatient medical records which did not count urinary tract infection treated outside the hospital,7,8 and did not consider potentially important confounding variables such as physician bias concerning the association between the uncircumcised state and urinary tract infection,7-9 leaving us in doubt as to whether all important sources of bias were examined. Their main conclusion - that circumcision helps prevent urinary tract infection is also in doubt. These problems underline the need for a well planned, prospective, population based surveillance study to document whether the protective effect of circumcision against urinary tract infection is genuine.

      It has been proposed that circumcision may protect men from infection with the human immunodeficiency virus (HIV). Two recent reports16,17 concluded that uncircumcised men exposed to HIV during heterosexual intercourse appeared to be more susceptible to infection. The presence of a penile ulcer was found to be as important, or more important, as a risk factor for seroconversion. The data collected were collected in a clinic for venereal disease in Kenya, where the heterosexual transmission of HIV appears to be more prevalent than in North America. In these studies, the uncircumcised men came from areas of the country in which HIV was endemic, and the circumcised men came from areas of lower risk. Which is cause and which effect remains unknown. Many of the studies' conclusions depended upon retrospective data, and again, important sources of bias may have obscured the true contribution of circumcision to the risk of the transmission of HIV. We must remember that circumcision is not performed randomly; it is more common in certain socioeconomic, ethnic, and religious groups. Jews and Muslims are circumcised for religious reasons. In the United States circumcision is less common among the poor and among Hispanic Americans. In Japan and Sweden, neonatal circumcision is not performed routinely. These demographic factors may in turn be associated with sexual practices and exposures that have a significant effect on the incidence of HIV infection. Nevertheless, the studies cited do suggest that HIV may be more infective during heterosexual intercourse if the male partner is uncircumcised and has a mucosal or cutaneous ulcer. In preventing the transmission of HIV, however, circumcision is unlikely to be as effective as limiting the number of sexual contacts or using condoms properly.

      It has been suggested that circumcision prevents cancer of the penis and of the uterine cervix.6,11 Both kinds of cancer are associated with genital infection with specific types of human papillamoviruses.18 It has been alleged that cancer of the penis is virtually unknown among men circumcised in infancy. However penile carcinomas have in fact been reported among circumcised men.19 The incidence of cancer of the penis in the United States, where most men have been circumcised at birth, is reported to be between 0 and 2.1 per 100,000, representing less than 1 percent of all cancers in men.20 This incidence of penile cancer is similar to the rates in Denmark (1.1 per 100,0000 and Japan (0.3 per 100,0000 where neonatal circumcision is not routinely performed.20 One study, reporting contradictory data, claimed that the lifetime risk in American men is 1 in 600.21 This estimate was extrapolated from a study of the incidence of male circumcision22 and depended on three incorrect assumptions: that all penile carcinomas occur in men who are not circumcised at birth, that the circumcision rate between 1890 and 1905 was the same as that in 1959, and that men presenting at a cancer detection center in Los Angeles, 40 percent of whom were classified as professionals or managers, were representative of the U.S. male population. Thus the lifetime risk of cancer reported in this study is probably incorrect. The evidence that penile cancer has a viral cause and the disease's similar incidence rates in countries with very different rates of circumcision cast doubt on the proposition that circumcision is the most effective way to prevent cancer of the penis.

      Cervical cancer presents epidemiological similarities to venereal diseases. It is more prevalent among the poor, blacks, and Hispanic Americans. In addition to socioeconomic factors, sexual behavior (sexual activity at an early age and multiple partners) smoking, the low intake of certain micronutrients, the use of hormonal contraceptives, and infection with sexually transmitted diseases have been identified as risk factors for cervical cancer.23 Among the sexually transmitted diseases that may be risk factors, human papillomaviruses, herpes simplex type 2 virus, cytomegalovirus, Trichomona vaginalis, Chlamydia trachomatis, spirochetes, Neisseria gonorrhoae and Mycoplasma hominis have all been suspected.11,24 Venereal transmission is further supported by the observation that the wives of men with penile cancer have a significantly increased risk of cervical cancer. Also the subsequent wives of men whose previous wives had cervical cancer have significantly higher rates than women in the general population.23 At present, human papillomaviruses, especially types 16 and 18, appear to play a part in the pathogenesis of the disease.23,24 Once again, studies have correlated exposure to uncircumcised sexual partners with the incidence of cervical cancer, but the circumcised state is also associated with the presence of human papillomavirus and other possibly oncogenic viruses. It is thus difficult to sort out a cause-and-effect relation.11

      In summary, the risks of circumcision are very small when the procedure is done by an experienced practitioner. The cost of routine infant circumcision is substantial and must be considered in making public health recommendations. The benefits of routine circumcision may include a reduced rate of urinary tract infection in infancy. As noted the studies that considered this issue were retrospective and may not have eliminated important sources of bias. These studies are important and very useful in generating hypotheses, but they cannot be considered to offer proof of benefit. Confirming the effect of circumcision on the incidence of urinary tract infection will require well designed studies that follow entire cohorts of circumcised and uncircumcised subjects from birth.

      Although the risks of routine neonatal circumcision are small, the benefits appear to be uncertain. It therefore seems prudent to consider neonatal circumcision a procedure to be performed at the discretion of the parents not as a part of routine medical care. Omitting circumcision in the newborn period should not be considered to be medical neglect. Parents should be informed of the current state of medical knowledge regarding the risks and benefits of the procedure. Their ultimate decision may hinge on nonmedical considerations.

Ronald L. Poland, M. D.

Pennsylvania State University
College of Medicine
Hershey, PA 17033


  1. Peter G. ed Report of the Committee on Infectious Diseases, 21st ed. Elk Grove Village, Ill.: American Academy of Pediatrics. 1988: 243-51.
  2. Wiswell TE, Geschke DW. Risks from circumcision during the first month of life compared with those for uncircumcised boys. Pediatrics 1989: 83:1011-5.
  3. Kunz HV. Circumcision and meatotomy. Prim Care 1986: 13: 513-25.
  4. Kaplan GW. Complications of circumcision.Urol Clin North AM 1983. 10: 543-9.
  5. Anand KJS, Hickey PR. Pain and its effects on the human neonate and fetus. N Engl J Med 1987: 317: 1321-8.
  6. Warner E, Strashin E. Benefits and risks of circumcision. Can Med Assoc J 1981: 125: 967-76, 992.
  7. Wiswell TE, Smith FR, Bass JW. Decreased incidence of urinary tract infections in circumcised male infants. Pediatrics 1985: 75: 901-3.
  8. Wiswell TE, Smith FR, Holton ME, Cornish JD, Hankins CT. Declining frequency of circumcision: implications for change in the absolute incidence and male to female sex ratio of urinary tract infections in early infancy. Pediatric 1987: 79:338-42.
  9. Herzog LW. Urinary tract infections and circumcision: a case-control study. Am J Dis Child 1989: 143: 348-50.
  10. Wilson RA. Circumcision and venereal disease. Can Med Assoc J 1947: 56: 54-6.
  11. Kesseler II. Etiological concepts in cervical carcinogens. Gynecol Oncol 1981: 12: Suppl 2:57-524.
  12. Care of the uncircumcised penis. Elk Grove Village, Ill.: American Academy of Pediatrics. 1986.
  13. Nasrullah PF. Circumcision Pro and Cons. Prim Car 1985. 12: 593-605.
  14. McKim JS. Neonatal Circumcision. Can Med Assoc J 1981: 125:955.
  15. Fergusson DM, Lawson JW, Shannon FT. Neonatal circumcision and penile problems: an 8 year longitudinal study. Pediatrics 1988: 81: 537:41.
  16. Simonsen JN, Cameron DW, Gakinya MN, et al. Human Immunodeficiency virus infection among men with sexually transmitted diseases: experience from a center in Africa. N Engl J Med 1988:319:274-8.
  17. Cameron DW, Simonsen JN, D'Costa LJ et al. Female to male transmission of human immunodeficiency virus type 1: risk factors for seroconversion in men. Lancet 1989: 2: 403-7.
  18. Barrasso R, De Brux J, Croissant O, Orth G. High prevalence of papillomavirus-associated penile intraepithelial neoplasma in sexual partners of women with cervical intraepithelial neoplasia. N Engl J Med 1987: 317:916-23.
  19. Boczko S. Freed S. Penile carcinoma in circumcised males. N. Y. State J Med. 1979: 79: 1903-4.
  20. Waterhouse J. Shanmugaramam K. Muir C. Powell J. Cancer incidence in five continents. IARC Sci Publ 1982:
  21. Kochen, M. McCurdy S. Circumcision and the risk of cancer of the penis: a life-table analysis. Am J Dis Child 1980: 134: 484-6.
  22. Stern E. Lachenbruch PA. Circumcision information in a cancer detection center population. J Chronic Dis 1965: 21: 117-24.
  23. Reeves WC, Rawls WE, Brinton LA. Epidemiology of genital papillomaviruses and cervical cancer. Rev Infec Dis 1989: 11: 416-39.
  24. Chang AR. Human papillomaviruses and cervical neoplasia: a review. N Z Med J 1989: 102:47-50.

[CIRP note: Professor Poland served on the American Academy of Pediatrics 1989 task force on circumcision. Professor Poland's views were rejected by the AAP 1989 Task Force on Circumcision. The Canadian Paediatric Association, however, adopted his position in its 1996 policy statement, Neonatal Circumcision Revisited. The American Academy of Pediatrics subsequently reversed formed a new task force, reversed its position, and adopted his 1990 position in its 1999 Circumcision Policy Statement.

(File revised 7 May 2005)