Edgar J Schoen. The New England Journal of Medicine. Volume 322, Issue 18. May 1990.
From the early 1940s until the mid-1970s, circumcision of newborn boys was accepted in the United States as a simple procedure that promoted hygiene and prevented genital disease. Educated middle-class parents almost always had their newborn sons circumcised. The opinion that circumcision prevented cancer of the penis and was associated with a lower incidence of cervical cancer in sexual partners was generally accepted. It helped prevent phimosis and associated local infection and resulted in genital cleanliness. These middle-class parents were willing to pay for the procedure (there were few insurance programs in the 1940s and 1950s). The infant sons of poor parents were usually uncircumcised because their parents were unaware of the benefits and could not afford the cost. By the late 1950s and early 1960s, as the cost began to be covered by third-party payers, including welfare programs, circumcision became the American standard.
In the late 1960s, however, the indications for the circumcision of newborns began to be questioned. It was claimed that penile cancer could be prevented simply by proper cleaning; circumcision had complications such as bleeding and infection, and the procedure was painful and psychologically traumatic. In 1971 and again in 1975, the Committee on the Fetus and Newborn of the American Academy of Pediatrics (AAP) took a stand against the routine circumcision of newborns. This position was reiterated in 1983 by both the AAP and the American College of Obstetricians and Gynecologists in their joint publication Guidelines for Perinatal Care. In response to these statements there has been a strong anticircumcision movement led by more affluent suburban parents (the original supporters of circumcision), some physicians, and third-party insurers, which increasingly refuse to pay for the procedure. U.S. Army studies show that the rate of circumcision fell from 85 to 70 percent between 1978 and 1984. Recent data indicate that the current rate of circumcision among newborns in the United States is about 60 percent, with the lowest rates on the West Coast (less than 50 percent) and the highest in the Midwest (about 70 percent) (National Center for Health Statistics, National Hospital Discharge Survey: unpublished data).
It is ironic that in the past few years, as the anticircumcision movement has grown and the rate of circumcision has decreased, evidence has mounted of the medical benefits of the procedure, particularly in the prevention of urinary tract infections in infant boys and sexually transmitted diseases in young men. This evidence will be cited below.
To update its stand on the circumcision of newborns in the light of these recent data, the AAP appointed a task force on circumcision with multidisciplinary representation. I served as chairman of the task force. The consensus report of this task force, as modified by the executive board of the AAP, has been released. The report summarized the evidence for and against the routine circumcision of newborns, but made no final recommendations. In the present discussion I offer a more detailed and personal examination of current evidence concerning the medical basis for routine neonatal circumcision. I also include material that was omitted from the task force report because a consensus could not be reached, as well as relevant information that has become available since the report was submitted.
Genital Hygiene, Balanoposthitis, and Phimosis
The circumcision of newborns facilitates genital hygiene throughout life under varying environmental conditions. It prevents preputial colonization with uropathic bacteria in infancy and childhood. Pyelonephritogenic fimbriated Escherichia coli bind avidly to the inner lining of the prepuce within the first few days of life. Other bacteria that attach preferentially to the mucosal surface of the foreskin include fimbriated strains of Proteus mirabilis and nonfimbriated pseudomonas, klebsiella, and serrada. The bacteria most commonly isolated from the prepuce of adults are group B streptococci, which may be sexually transmitted and cause balanitis.
In uncircumcised men, the importance of optimal genital hygiene in avoiding penile cancer was emphasized by the 1975 AAP task force. Maintaining good hygiene is difficult not only in underdeveloped countries but also in a large, advanced country like the United States, which has a multiracial population with wide socioeconomic and cultural differences. Even in smaller highly developed countries with more homogeneous populations there is evidence that genital hygiene is not optimal. A 1964 survey of uncircumcised British schoolboys showed that 70 percent had poor genital hygiene. A study in Denmark showed that 63 percent of 6-year-old boys had preputial adhesions, and the level decreased to 3 percent only at the age of 17 years; the incidence of phimosis decreased from 8 percent at the age of 6 to 1 percent at the age of 17. Thus, the goal of good genital hygiene in uncircumcised boys may be difficult to achieve throughout most of childhood.
A recent British report on balanoposthitis estimated that the incidence was 4 percent, with a peak between the ages of two and five years. Of the 100 cases reported, the foreskin was fully retractable in only 13 percent, and 1 patient had phimosis.
Urinary Tract Infections in Young Infants
In 1982 Ginsburg and McCracken reviewed data on 109 infants between the ages of five days and eight months who had urinary tract infections. They found that male infants predominated, contrary to the situation later in life, and that 95 percent of the boys with urinary tract infections were uncircumcised. In 1985 Wiswell et al. found a markedly increased incidence of urinary tract infections among uncircumcised male infants in a series of 5261 infants born at an army hospital. Wiswell et al. supplemented this work in 1987 by reviewing the records of 427,698 infants (219,755 boys) born in U.S. armed forces hospitals. They confirmed the earlier findings by showing a 10-fold increase in the incidence of urinary tract infections among uncircumcised boys and found that as the rate of circumcision decreased, the number of these infections in male infants increased. More recent reports from this army study have demonstrated increased periurethral flora of uropathic bacteria in uncircumcised boys in the first six months of life as well as a higher risk of serious sequelae (bacteremia and meningitis) during the first month of life. Their work has been criticized as a retrospective study of inpatients; however, more recently at Children’s Hospital in Boston, Herzog observed febrile infants in an outpatient clinic and found a markedly increased incidence of urinary tract infections among uncircumcised boys. Also, prospective studies of uncircumcised boys in Sweden have shown a 0.6 to 0.8 percent incidence of infection in the first year of life. Although there was no circumcised control group, this figure is six to eight times as high as the incidence among circumcised boys in the United States reported by Wiswell et al.
In 1986, Roberts hypothesized that circumcision protects male infants against urinary tract infections by preventing bacterial colonization of the prepuce. The foreskin often does not separate naturally within the first few years of life, and genital hygiene is usually inadequate in uncircumcised male infants. Roberts believed that bacterial adherence and colonization of the prepuce was followed by periurethral colonization, leading to ascending urinary tract infection. Acute pyelonephritis early in life involves a greater risk of renal parenchymal damage than infection at a later age. Two recent reviews concluded that circumcision of newborns prevented urinary tract infections, although the authors of the studies were not in favor of the procedure.
Cancer of the Penis
Routine circumcision of newborns almost completely eliminates the possibility of cancer of the penis. In 1932 Wolbarst reviewed 1103 cases of carcinoma of the penis; none of the patients had been circumcised in infancy. The annual incidence of penile cancer among men in the United States is less than 1 per 100,000—similar to the incidence in highly developed countries such as Norway and Sweden—and the mortality rate is as high as 25 percent. The low incidence figure in the United States is misleading, because it represents a mixture of uncircumcised and circumcised men, whereas in Scandinavian countries virtually all men are uncircumcised. According to Kochen and McCurdy, the incidence of penile cancer in the United States is essentially zero among circumcised men, and 2.2 per 100,000 among uncircumcised men. These authors estimated that if all men in the United States were uncircumcised, the number of cases of penile cancer would increase to more than 3000 a year. Using life-table analysis, they estimated that an uncircumcised man in the United States has a lifetime risk of penile cancer of 1 in 600. There are 750 to 1000 cases of penile cancer annually in the United States; however, in the past 20 years only 3 cases of penile cancer have been reported in men who had been circumcised as newborns. There have been six major studies of cancer of the penis in the United States since 1932, and none of the more than 1600 patients studied had been circumcised in infancy.
In 1947 Plaut and Kohn-Speyer produced local tumors in mice by applying smegma to the skin; they expressed a belief in a carcinogenic factor in smegma but were unable to identify it. Retrospective speculation about the role of human papillomavirus is interesting because the skin tumors at the site of smegma application included four cases of condyloma acuminatum (one of which was “carcinoma-like”) and two cases of squamous-cell carcinomas. The etiologic agent of condyloma acuminatum, human papilloma-virus, has been found in Bowen’s disease, another premalignant lesion of the penis, as well as cancer of the penis and vulva. In 1986 McCance et al. studied 53 Brazilian men with carcinoma of the penis and found DNA sequences of human papillomavirus type 16 in 49 percent, and sequences of type 18 in 9 percent. In women from the same social group who had cancer of the cervix, sequences of type 16 were found in 40 percent.
Cancer or the Cervix
Epidemiologic studies have shown that sexual activity at an early age, multiple sexual partners, and frequent sexual intercourse predispose women to cervical carcinoma, suggesting that the cancer is sexually transmitted. Kessler presented evidence that there was an increased risk of cervical cancer among the wives of men who had previously been married to women with cervical cancer and concluded that “certain males may venereally transmit a risk factor for cervical cancer to coital contacts.” In Puerto Rico, where most men are uncircumcised and there is a high incidence of both penile and cervical cancer, Martinez found eight cases of cervical cancer in the wives of 889 men with penile cancer, and none among the wives of a control group of men. Studies in the 1970s suggested that herpes simplex virus type 2 was the important etiologic agent in cervical cancer, but more recent evidence centers on human papillomavirus types 16 and 18, the same agents implicated in penile cancer.
Sexually Transmitted Diseases
During World War II and in the wars in Korea and Vietnam, the U.S. armed forces generally believed that circumcision, by promoting genital hygiene, helped prevent balanitis under unsanitary field conditions (and unpublished data) and helped protect against sexually transmitted diseases. A 1947 Canadian Army study supported this opinion, showing an increased incidence of these diseases in uncircumcised men. More recently, Parker et al. analyzed the findings in 1350 men seen in an Australian clinic that treated sexually transmitted diseases and found a statistically significant increase in the incidence of four diseases in uncircumcised men: genital herpes, candidiasis, gonorrhea, and syphilis. On the other hand, Smith et al. found no difference in the incidence of gonococcal urethritis in uncircumcised as compared with circumcised American servicemen, but found an increased risk of nongonococcal urethritis in circumcised men.
Uncircumcised men appear to be more susceptible to sexually transmitted diseases that disrupt epithelial surfaces—genital herpes, syphilis, condyloma acuminatum (caused by human papillomavirus), and chancroid. In a 1975 study Taylor and Rodin found a significantly lower rate of circumcision in a group of 214 men with genital herpes than in a control group of 410 men who did not have the disease. Recent studies have suggested that diseases causing genital ulceration in men are important risk factors in the transmission of the human immunodeficiency virus.
Pain and Anesthesia
Infants who are circumcised without anesthesia feel pain. The anatomical and functional pathways necessary for the perception of pain are present in newborns of all viable gestational ages. Infants respond to painful stimuli with behavioral, cardiovascular, and hormonal changes. Behavioral responses include crying, irritability, and varied sleep patterns. The physiologic responses consist of increased heart rate and blood pressure, decreased oxygen saturation, and increased serum Cortisol levels. These responses are short-lived, lasting only minutes to hours, and there is no evidence of long-term sequelae. The transitory nature of these responses to pain has resulted in a reluctance to risk the use of general anesthesia and narcotic analgesia for circumcision, a procedure with negligible mortality and low morbidity. Recently, there has been interest in the use of dorsal penile-nerve block with lidocaine to relieve the pain of circumcision. Dorsal penile-nerve block has been shown to reduce the immediate behavioral changes and evidence of stress in newborns after circumcision, but experience with this procedure is limited. Caution is warranted before the widespread use of dorsal penile-nerve block can be recommended, because lidocaine has been shown to affect the auditory evoked response of the brain stem and can also have disturbing local effects.
The benefits of neonatal circumcision lie in its ability to prevent certain diseases. As with other public health measures such as immunization, its disadvantages are short-term—any untoward effects occur during or shortly after the procedure—and its advantage is long-term—protection against future disease. The potential medical benefits of circumcision of newborns are seen over a lifetime and involve reducing the incidence of a number of diseases, ranging from urinary tract infections in early infancy to penile cancer in middle and old age, and the continued ease of genital hygiene and avoidance of balanoposthitis and phimosis. Physicians’ attitudes toward the circumcision of newborns have reflected their specialties, and problems involving the foreskin have been complicated by this multidisciplinary interest. Neonatologists, seeing only the pain and complications of the procedure, have tended to be against circumcision, whereas urologists treating penile cancer in men have generally favored the procedure, looking on it as a simple and effective method of cancer prevention.
These interspecialty disagreements were reflected in the unreferenced report of the 1975 AAP task force, in which the members were sharply divided and the chairman objected to the wording of the final statement that there was “no absolute medical indication for routine circumcision of the newborn.” The 1988 AAP task force on circumcision was more broadly representative, documented its findings with a bibliography, and reached a broad consensus. It considered protection against cancer of the penis, balanoposthitis, and the complications of phimosis to be conclusive advantages of the circumcision of newborns, and believed that there was evidence of protection against urinary tract infections in male infants. Pain and immediate postoperative complications were considered the main disadvantages. The rate of complications was low, and adverse effects consisted mainly of local infection and hemorrhage (incidence, 0.2 to 0.6 percent); serious complications were rare, and the number of deaths from circumcision was negligible. Although the AAP did not make a recommendation for or against the routine circumcision of newborns, I believe that the advantages of the procedure outweigh the disadvantages. To keep the complication rate low, well-trained operators need to use proper surgical techniques. Lay anticircumcision groups (e.g., National Organization of Circumcision Resource Centers [NOCIRC], and the International Organization against Circumcision Trauma [INTACT]) have concentrated on making the uncircumcised (or intact) state a matter of “genital chic.” These increasingly powerful organizations have used three methods in their attempt to discourage circumcision: they have put pressure on third-party payers, used the media to air their views, and used lawsuits to intimidate the medical profession. Blue Shield of Pennsylvania, Prudential Insurance Company, other health insurers, and welfare programs are refusing to pay for circumcision. The views of those who are opposed to circumcision predominate on radio and television programs and in newspaper features. Their attempt to intimidate the medical profession is exemplified by the case of a California mother whose infant was circumcised uneventfully after she had given written informed consent. Citing the 1975 AAP statement that there is no absolute indication for routine circumcision, she subsequently sued the physician who performed the procedure for “battery, cruelty, infliction of pain, child abuse, false imprisonment, kidnaping and mayhem.” The fact that California has the lowest rate of circumcision of newborns in the country is attributed to the high level of activity of anticircumcision groups in the state, combined with the influx of Asian and Hispanic immigrants who traditionally remain uncircumcised.
The status of circumcision is changing as new evidence appears. At the time of the 1975 AAP recommendations, the relation between the circumcision of newborns and urinary tract infections had not yet been described. Preliminary information about this relation suggests new potential advantages of circumcision. Group B streptococcus, the cause of severe neonatal sepsis, is known to be sexually transmitted and colonizes the prepuce in many uncircumcised men. Also, recent studies from Africa indicated that uncircumcised men were more likely to become infected with the human immunodeficiency virus after heterosexual exposure. This information is preliminary, but it emphasizes the need for additional data to determine the role of circumcision in serious sexually transmitted diseases.
Attempts to assess the cost—benefit ratio of neonatal circumcision have yielded conflicting results. Warner and Strashin calculated that the routine circumcision of newborns in Canada would result in an annual saving by preventing penile cancer and obviating the need for later circumcision for phimosis and balanoposthitis, whereas Cadman et al. believed that the costs of neonatal circumcision exceeded the benefits. Neither of these studies was a complete economic analysis, and both were performed before the information on urinary tract infections became available. A more extensive study of cost effectiveness would have to consider the potential ability of circumcision to prevent infant urinary tract infections, balanoposthitis, and cancer of the penis. Such analyses would have to be updated regularly as data accumulate on such factors as sexually transmitted diseases and possible long-term renal sequelae of urinary tract infections in infants.
The benefits of routine circumcision of newborns as a preventive health measure far exceed the risks of the procedure, although some may question its cost effectiveness and priority in the delivery of health care.