AMERICAN FAMILY PHYSICIAN, Volume 41, Number 3: Pages 817-820,
March 1990.


The Circumcision Controversy

Science is a wonderful thing. One gets such wholesale return of conjecture for such a trifling investment of fact.

--Mark Twain
Life on the Mississippi

Until 1985, the pendulum of public and scientific opinion appeared to be swinging away from routine newborn circumcision. In that year, Wiswell and colleagues' retrospective study of infant urinary tract infection1 demonstrated a 10-fold difference in incidence between uncircumcised and circumcised males. This article sparked a lively controversy about the relationship between the foreskin and urinary tract infections in infant boys.

The controversy must now be examined in the context of the long- running dispute about the benefits and risks of circumcision. This dispute has been characterized by significant discrepancies in the statistics to support various points of view.

For example, in the article published in this issue, Wiswell2 indicates that 10 to 15 percent of uncircumcised boys will require circumcision in adulthood. But Wallerstein3 estimates that in Finland the rate of adult circumcision is six per 100,000. Herzog and Alverez4 found that penile problems are more common in uncircumcised boys, but others have reported opposite findings and have noted that meatitis seems to occur exclusively in circumcised boys.5

Wiswell and associates1,6 have reported rates of urinary tract infection ranging from 1.1 percent to 4.1 percent in uncircumcised male infants. But my data from an unpublished retrospective study at Northern California Kaiser Permanente) show much lower rates, ranging from 0.09 to 0.16 percent. Finally, results of studies linking circumcision to sexual transmitted disease have varied widely.

This welter of confusing statistics is not at all surprising. Most of the numbers are taken from retrospective studies that suffer from one of three flaws: uncontrolled variables, subjectivity, or questionable interpretation.

Uncontrolled variables are a major stumbling block in determining infection rates. In my study of urinary tract infection, the circumcised population was predominantly white, while the uncircumcised population was predominantly Asian and Latino. It is hoped (but no means certain) that these differences had little effect on the incidence of urinary tract infections. If the object of the study had been sexually transmitted disease, the cultural differences would have completely invalidated the data. Yet, these are exactly the kinds of data that are used to link the intact foreskin with sexually transmitted disease.3

Subjectivity is the next major stumbling block, particularly in determining the incidence of "penile problems," such as phimosis and meatitis, and the need for circumcision after the neonatal period. The typical infantile foreskin is a long, narrow and unretractable nozzle of tissue. Many physicians in the United States use the pathologic term "phimosis" to describe this normal anatomy. As a practical matter, the presence of the infantile prepuce precludes the diagnosis of meatitis - if you can't see the meatus, you can't diagnose meatitis.

The rate of circumcision after the neonatal period depends on whether physicians and patients view the foreskin as dispensable. This perspective is subject to cultural biases. Having acquired a reputation as a "foreskin consultant," I see numerous dissatisfied patients who request either circumcision or recircumcision for cosmetic reasons. Invariably these patients either come from families who lacked insurance coverage at birth or who believe that the original surgeon "didn't take enough off." I have never seen a child who had a true objective indication for circumcision. I have seen only one child who had an objective indication for recircumcision (ischemic glans).

Finally, there is the question of interpretation. Many physicians were alarmed by Hertzog's report7 that eight of 31 uncircumcised males with urinary tract infections had genitourinary abnormalities, which implied that abnormalities were a consequence of infection. But Rockney and Caldamone8 point out that Herzog's eight abnormalities "were almost certainly present from birth and are likely to have been factors that contributed to the development of urinary tract infection rather than the result."

In my series, only 48 of 96,000 normal male infants (excluding 16 patients with congenital genitourinary anomalies) had urinary tract infections. Only three of the 48 had acquired genitourinary problems (reflux), and one child who had persistent reflux and infection was circumcised.

From my data, I have estimated that the cost of using routine circumcision to prevent infantile urinary tract infection is $60,000 per infection prevented. The cost of preventing one urethral reimplantation is estimated at $3 million. Infants with occult congenital obstructive uropathy (the "few" who need to be "protected") may actually be harmed by routine circumcision. If circumcision does indeed prevent urinary tract infection in these patients, then it also delays diagnosis and treatment of the underlying lesion.

We in the United States are culturally acclimated to regard the foreskin as non-essential and even pathologic. Therefore, forsaking the ancient cautionary primum non nocere becomes easy. We must not forget that the burden of proof is on the circumcision advocates. Showing disease association is not sufficient. They must prove cause and effect. Furthermore, they must prove (not conjecture) that the advantages of circumcision outweigh the risks.

Salem, Oregon

  1. Wiswell TE, Smith FR, Bass JW. Decreased incidence of urinary tract infections in circumcised male infants. Pediatrics 1985;75:91-3.
  2. Wiswell TE, Routine neonatal circumcision: a reappraisal. Am Fam Physician 1990;41:859-863.
  3. Wallerstein E. Circumcision: an American Health Fallacy. New York: Springer, 1980.
  4. Herzog LW, Alvarez SL. The frequency of foreskin problems in uncircumcised children. Am J Dis Child 1986; 140:254-6.
  5. Marino LJ. An emphatic vote against circumcision. Contemp Pediatr 1989;6(11):11-6.
  6. Wiswell TE, Roscelli JD. Corroborative evidence for the decreased incidence of urinary tract infections in circumcised male infants. Pediatrics 1986;78:96-9.
  7. Herzog LW. Urinary tract infections and circumcisions--a case control study. Am J Dis Child 1989;143:348-50.
  8. Rockney RM, Caldamone AA. Circumcision and urinary tract abnormalities [Letter]. Am J Dis Child 1989; 143:1261-2.

(File revised 20 February 2005)