on the Use of Male Circumcision to Prevent
San Anselmo, California
Male circumcision is an operation for which innumerable claims of medical or prophylactic benefit have been advanced through the years.1 The claims invariably have been proven to be without substance.1
Amongst these is the claim that male circumcision can reduce the incidence of cervical cancer in the female sexual partner, which first appeared in 1954 when a study by Ernst L. Wynder et al. appeared in the medical literature.2 Wynder claimed that Jewish women with circumcised partners had fewer cases of cervical cancer than non-Jewish women. The true cause of cervical cancer was unknown at the time so almost any claim could be made. Wynder obtained his information on the circumcision status of the husband by asking the female patient.2 Many were unable to provide the circumcision status, so Wynder’s results are suspect.
Jones, Macdonald, and Breslow also studied the etiology of cervical cancer. The report of their findings was published in 1958. In contrast to that of Wynder’s group, this study found no relationship between cervical cancer and the circumcision status of the male partner.
Circumcision of the marital partner has been claimed by some workers to be the most important difference between cervix cancer cases and controls. Our interviews showed that circumcision of the first husband, or partner of longest duration, was equally frequent in cases and controls when the Jewish patients were eliminated from the study (Table III). The data was enough different from previously published reports that our colloborating statisticians rematched cases and controls by various techniques of regrouping. All approaches verified the initial conclusion, namely, that circumcision, or lack of circumcision, was as frequent in partners of cases with cancer as in the partners of controls [emphasis added].3
Wynder then attempted to validate his 1954 study by a second study to determine the accuracy of reported circumcision status.4 Wynder, however, found that fully 25% of the men in his study did not know their circumcision status and only 62% of wives accurately knew their husband’s circumcision status. He concluded that only direct examination by a physician was accurate.4 Therefore, the results of his 1954 study, which relied on a report by the female patient to assign circumcision status, were cast into doubt.
Stern and Neely also investigated the etiology of cervical cancer using multiple regression analysis:
Circumcision status did not emerge as a factor. Multiple marriages, however, did emerge as a risk factor.5The discovery rate for cancer of the cervix among non-Jewish women whose marital partners were circumcised was no different from the rate among non-Jewish women with non-circumcised husbands5
Boyd and Doll addressed the question of the cause of cervical cancer in a study published in 1964.6 Boyd and Doll also did not confirm Wynder’s 1954 conclusion that lack of male circumcision was a cause of cervical cancer.2 They did point to some factor associated with coitus as a cause of cancer but they were unable to identify that factor.6 They also discovered that low socioeconomic status was a risk factor for cervical cancer.6
The matter rested there until 1965 when a study by Aitken-Swan and Baird was published in the medical literature.7 Aitken-Swan and Baird conducted yet another study in which the circumcision status of the husband of patients and controls was determined by physical examination by a physician. Aitken-Swan found that the husband stated his circumcision status correctly in 84 percent of the cases, and the wife’s opinion was correct in 62 percent of the cases. They found 54 percent of husbands were not circumcised, 22 percent were “partially circumcised” and 24 percent were completely circumcised. Aitken-Swan and Baird found that “there was no significant difference in these proportions between patients and controls.”7
The year 1970 saw the appearance of two articles that reviewed the alleged health benefits of male neonatal circumcision. Leitch in Australia reviewed the evidence for circumcision’s value in prevention of cervical cancer and concluded that low socio-economic status and low personal hygiene, not lack of circumcision, are the predominant etiological factors.8 Noel Preston in the United States independently reviewed the evidence for circumcision’s value in prevention of cervical cancer. He concluded that coitus, not lack of circumcision, is a factor in the etiology of cervical cancer and proposed the use of barrier contraceptives.9
Terris, Wilson, and Nelson (1973) undertook a study of the relationship, if any, between cervical cancer and circumcision of the male partner.10 Terris et al. found no significant difference in the circumcision status of the partners of cervical cancer cases and controls.
Sumithran investigated the incidence of cervical cancer amongst the aboriginal Orang Isli people of Malaysia, who do not practice male circumcision. They have a strict moral code, however, that limits extra-marital sexual activity. The rate of cervical cancer was found to be very low despite the lack of circumcision of the males.11
Megafu investigated the incidence of cervical cancer amongst the Ibo people of Nigeria. The males in that population are ritually circumcised. The rate of cervical cancer was found to be high although the males are circumcised.12
Smegma, a natural substance, is normally found under the foreskin in both males and females, human and animal. Reddy investigated smegma to see if it could possibly be carcinogenic, and concluded that smegma is not a carcinogen.13
After all of this investigation, it became clear that lack of male circumcision is not a risk factor for cervical cancer, however, the main etiological factor still had not been identified. Whatever the cause, it was associated with low socio-economic status, early and frequent sexual activity, and multiple partners. Cervical cancer was seen to be a sexually transmitted disease. Discovery of the true cause would have to await advances in the science of molecular biology.
The True Cause
Studies carried out in the 1980s reported finding DNA from human papillomavirus (HPV) types 16 and 18 in human cervical cancer cells.14 Thus it became clear by 1990 that HPV infection, which is contracted by coital activity, is a risk factor for cervical cancer.15 Other risk factors include malnutrition, smoking, use of oral contraceptives, sexual behavior, infection with sexually transmitted diseases, having a husband whose previous wife had cervical cancer, a husband who has penile cancer,15 or a husband with genital warts.16
Later research further confirmed the role of HPV in cervical cancer. Walboomers et al. now state that HPV is found in 99.7 percent of cervical cancers,17 and HPV infection is seen to be a necessary condition for the development of cervical cancer.17 Some researchers believe that another factor, in addition to HPV infection, is required to start the cancer process. Smoking is one such factor.18,19
Walboomers et al. suggest that it is time to supplement or replace the venerable Pap smear with testing for HPV.17 Also, it now is clear that a woman may put herself at increased risk of cervical cancer by her own behavior.
The Centre International de Recherche sur le Cancer (CIRC) [International Agency for Research on Cancer (IARC)], with headquarters in Lyon, has been carrying out studies into the risk factors for cervical cancer in several nations for some years. Recently, CIRC doctors prepared three reports. Two were published in The Lancet.20,21 The third report, however, was not published by The Lancet, but was published in The New England Journal of Medicine.22
The New England Journal of Medicine gave the article great fanfare with an accompanying editorial23 and an inaccurate press release that was carried in most major newspapers and electronic media. In actuality, Castellsaqué and colleagues concluded that circumcision was associated with a modest non-significant decrease in cervical cancer in the female partners of circumcised men.22 In the special case of a male partner who was a “high risk” partner—that is, one who had had intercourse before the age of 17, had had six or more sexual partners, or a history of contact with prostitutes—Castellsaqué et al. reported that there was a statistically significant increase in risk to the female partner.22
Clinicians determined circumcision status by physical examination in Brazil, Thailand, and the Philippine Islands.22 Circumcision status in the other countries apparently was not determined by examination. The authors report an error rate in self-determined circumcision status of 5.3 percent.22 This is substantially lower than that found by Wynder and Licklider4 who reported an error rate of 24 percent for self-determination and, therefore, is suspiciously low. The authors did not question their findings although only 41.5 percent of the men were examined.22 In the end, the authors used self-reported status, not status determined by physical examination. This is surprising because several studies have shown self-reported circumcision status to be highly unreliable.4,7,25 This flaw alone may be enough to invalidate the study.
Clinicians detected HPV in the males by taking swabs from the distal urethra and the coronal sulcus.22 It is difficult to understand how or why this could be done without simultaneously noting circumcision status, but that apparently is the case.
The report combines data from five previous studies that were carried out in five different countries,22 which is spurious because it attempts to compare data from widely diverse regions of the earth and from population groups with different levels of HPV infection, different levels of cigarette smoking, different sexual habits, and other differences. Poland quite accurately observes that circumcision does not occur randomly.16 This is true in the present study because 65 percent of the circumcised men were concentrated in the Philippine Islands, where circumcision of male children is an engrained social practice. This study did not consider variations in incidence of HPV infection amongst the various populations studied, nor does it consider the incidence of cigarette smoking, use of oral contraceptives, and parity amongst the female patients and controls and their variance between population groups.
Castellsaqué et al. only considered women who reported that they had had only one sexual partner. Since, in some cultures, virginity at marriage is highly prized, women may have been unwilling to admit to a previous sexual partner. The study focuses on the present partners of women with cervical cancer. Cervical cancer has a long incubation period that may span several decades.23 Many women in the study could have become infected with HPV by a previous partner earlier in their sexually active years. Aitken-Swan and Baird7 and Adami and Trichopoulos23 note that extra-marital partners cannot be ruled out.
Castellsaqué et al. report that non-circumcised men have a greater rate of HPV infection.22 Aynaud et al., however, found that there is no significant difference in HPV infection between circumcised and non-circumcised males.26,27
Castellsaqué et al. studied the HPV infection incidence of men whose wives were patients with cervical cancer.22 It is highly probable that at least some of these men were infected by their wives.27 Apparently, the possibility that men were infected by their wives was not considered. This could seriously distort the outcome of the study.
Castellsaqué et al. made “adjustments” to their data to get to their conclusion.22 However, they did not publish the factors used to make these adjustments, so their validity cannot be independently verified.
The three CIRC articles used “unconditional logistic regression” statistical analysis to control for confounding factors.20-22 The circumcision study included the male factors of level of education, age of first sexual intercourse, lifetime number of sexual partners, and self-reported frequency of genital washing; and female factors of age, lifetime number of sexual partners, and age of first sexual intercourse. Factors of smoking, parity, and use of oral contraceptives were not controlled. Skegg noted that
The failure of Castellsaqué et al. to control for parity and use of oral contraceptives22 is astonishing because CIRC simultaneously published two companion studies of those factors.20,21“Logistic regression analyses can become unstable when there is adjustment for many variables in relation to the number of events.”28
Scientific studies of etiology of disease must adequately control for known confounding factors or their results are inconclusive. It now appears that, although HPV infection is a necessary condition for the development of ano-genital cancers, including cervical cancer, some other factor may be necessary to trigger the development of neoplasia. Known factors include low socio-economic status,7 smoking,18,19 use of hormonal contraceptives,20 and multiparity.21 The study by Castellsaqué et al. fails to control for these factors.22
As noted above, Castellsaqué et al. found a non-significant association of male circumcision with a decline in the risk of cervical cancer.22 However, when they considered only “high-risk” males, they reported a significant reduction in the risk of cervical cancer.22
The findings of Castellsaqué’s team in regard to self-determination of circumcision status is sharply at variance with a number of studies that were carried out in the 1950s and 1960s.3,4,7 Their finding that circumcision status generally is not significant confirms previous findings.5-7,10-12 Their finding, however, that non-circumcised males who are “high-risk” pose an additional danger to their marital partners greater than circumcised “high-risk” males is new. There are many methodological flaws in their study, so the study must be regarded as non-conclusive. The weight of medical evidence shows that male circumcision is not of value in preventing HPV infection in women. The Castellsaqué study does little or nothing to change that. No change in NOCIRC’s longstanding evidence-based position that male circumcision is not of value in preventing cervical cancer can be made on the basis of the slight and dubious evidence provided by the Castellsague study.
Cervical cancer is caused by a sexually transmitted virus, and it may be activated by certain factors that are under the control of the individual. Early sexual activity, multiple sexual partners, and failure to use condoms increase the risk of contracting HPV infection. Smoking, use of oral contraceptives, and multiple numbers of births also increase the risk of cervical cancer. Women have a duty to protect themselves by healthful behavior from cervical cancer. Education is needed to enable women to better protect themselves from HPV infection.
The medical profession also has a duty to make regular screening by pap smears30 and/or the newer, better HPV test31 available to women everywhere. Wiping with vinegar turns HPV infected tissue white. The cervix may be wiped with vinegar and then visually inspected to determine if HPV infection exists.32 A better or less expensive test could hardly be imagined.
Male circumcision does not insure protection from HPV infection.6-14 The possible reduction in risk is slight at best. Even if it were to be proven effective, the difficulties of instituting worldwide mass male circumcision seem insurmountable.
There are also legal and ethical obstacles. HPV infection occurs in sexually active older persons. It is not a disease of childhood. Childhood circumcision should be used only for the very rare instances of penile maladies in children that are not responsive to conservative treatment. With regard to prevention of adult diseases, the decision should be postponed until the young man may make a decision for himself.32,33
Male circumcision excises irreplacable functioning human tissue from the genital organs, so it violates a person’s legal right to bodily integrity. Circumcision would not be performed for the benefit of the individual but for the benefit of some other person. Legally, parents actually lack the power to consent to non-therapeutic excision of human tissue from children.34-36 Consent is necessary before a circumcision can be performed. This may be difficult to obtain even in the case of adult males. The foreskin is erogenous tissue.37 Adult males may be unwilling to give up documented sexual pleasure38,39 in the slight hope of reducing incidence of HPV infection in others.
- Gollaher DL. From ritual to science: the medical transformation of circumcision in america. Journal of Social History 1994;28(1):5-36.
- Wynder EL, Cornfield J, Schroff PD, Doraiswami KR. A study of environmental factors in carcinoma of the cervix. Am J Obstet Gynecol 1954;68:1016-52.
- Jones EG, MacDonald I, Breslow L. A study of epidemiologic factors in carcinoma of the uterine cervix. Am J Obstet Gynecol 1958;76(1):1-10.
- Wynder EL, Licklider SD. The question of circumcision. Cancer 1960;13(3):442-445.
- Stern E, Neely PM. Cancer of the cervix in reference to circumcision and marital history. J Am Med Womens Assoc 1962;17(9):739-70.
- Boyd JT, Doll R. A study of the aetiology of carcinoma of the cervix uteri. Brit J Cancer 1964;XVIII(3):419-28.
- Aitken-Swan J, Baird D. Circumcision and cancer of the cervix. Brit J Cancer 1965; XIX(2):217-227.
- Leitch IOW. Circumcision - a continuing enigma. Aust Paediatr J 1970;6:59-65.
- Preston EN. Whither the foreskin. JAMA 1970; 213(11):1853-1858.
- Terris M, Wilson F, Nelson JH. Relation of circumcision to cancer of the cervix. Am J Obstet Gynecol 1973;117(8):1056-66.
- Sumithran E. Rarity of cancer of the cervix in the Malaysian Orang Asli despite the presence of known risk factors. Cancer 1977;39(4):1570-2.
- Megafu U. Cancer of the genital tract among the Ibo women in Nigeria. Cancer 1979;44(5):1875-8.
- Reddy DG, Baruah IK. Carcinogenic action of human smegma. Archives of Pathology 1963;75(4):414-420.
- Reeves WC, Rawls WE, Brinton LA. Epidemiology of genital papillomaviruses and cervical cancer. Rev Infect Dis 1989;11(3):426-39.
- Poland RL. The question of routine neonatal circumcision. N Eng J Med 1990; 322:1312-1315.
- Kjaer SK, de Villiers EM, Dahl C, et al. Case-control study of risk factors for cervical neoplasia in Denmark. I: Role of the "male factor" in women with one lifetime sexual partner. Int J Cancer 1991;48(1):39-44.
- Walboomers JM, Jacobs MV, Manos MM, et al. Human papillomavirus is a necessary cause of invasive cervical cancer worldwide. J Pathol 1999;189(1):12-9.
- Ho GY, Kadish AS, Burk RD, et al. HPV 16 and cigarette smoking as risk factors for high-grade cervical intra-epithelial neoplasia. Int J Cancer 1998;78(3):281-5.
- Wyatt SW, Lancaster M, Bottorff D, Ross F. History of tobacco use among Kentucky women diagnosed with invasive cervical cancer: 1997-1998. J Ky Med Assoc 2001;99(12):537-9.
- Moreno V, Bosch FX, Muñoz N, et al. Effect of oral contraceptives on risk of cervical cancer in women with human papillomavirus infection: the IARC multicentric case-control studyPDF. Lancet 2002; 359:1085-92.
- Muñoz N, Fransceschi, Bossetti C, et al. Role of parity and human papillomavirus in cervical cancer: the IARC multicentric case-control studyPDF. Lancet 2002; 359:1093-101.
- Castellsaqué X, Bosch FX, Muñoz, et al. Male circumcision, penile human papillomavirus infection, and cervical cancer in female partners. New Engl J Med 2002;346(15):1105-12.
- Adami H, Trichopoulos. Cervical cancer and the elusive male factor. New Engl J Med 2002; 346(15):1160-1.
- Dunn Jr JE, Buell P. Association of cervical cancer with circumcision of sexual partner. J Nat Cancer Inst 1959;22(4):749-64.
- Aynaud O, Ionesco M, Barrasso R. Penile intraepithelial neoplasia. Specific clinical features correlate with histologic and virologic findings. Cancer 1994;74(6):1762-7.
- Aynaud O, Piron D, Bijaoui G, Casanova JM. Developmental factors of urethral human papillomavirus lesions: correlation with circumcision. BJU Int 1999;84(1):57-60.
- Barrasso R, De Brux J, Croissant O, et al. High prevalence of papillomavirus-associated penile intraepithelial neoplasia in sexual partners of women with cervical intraepithelial neoplasia. N Engl J Med 1987;317(15):916-23.
- Skegg, DCG. Oral contraceptives, parity, and cervical cancerPDF. Lancet 2002;359:1080-2 [citing Peduzzi P, Concato J, Kemper E, Holford TR, Feinstein AR. A simulation study of the number of events per variable in logistic regression analysis. J Clin Epidemiol 1996;12:1373-79].
- Austoker J. Cancer Prevention in Primary Care: Screening for cervical cancer. BMJ 1994;309:241-248.
- Lytwyn A, Sellors JW, Mahony JB, et al. Comparison of human papillomavirus DNA testing and repeat Papanicolaou test in women with low-grade cervical cytologic abnormalities: a randomized trial. CMAJ 2000; 163 (6):701-7.
- Kmietowicz Z. Vinegar and visual inspection used to detect cervical cancer. BMJ 1999;318:757.
- Bollgren I, Winberg J. Letter. Acta Paediatrica Scandinavia 1991; 80: 575-7.
- American Academy of Pediatrics Committee on Bioethics. Informed consent, parental permission, and assent in pediatric practice. Pediatrics 1995;95(2):314-317.
- Somerville MA. Therapeutic and Non-Therapeutic Medical Procedures -- What are the Distinctions? Health Law in Canada 1981;2(4):85-90.
- Gregory J Boyle, J Steven Svoboda, Christopher P Price, J Neville Turner. Circumcision of Healthy Boys: Criminal Assault? 7 J Law Med 301 (2000).
- J. Steven Svoboda, Robert S. Van Howe, James G. Dwyer. Informed Consent for Neonatal Circumcision: An Ethical and Legal Conundrum. 17 J Contemporary Health Law Policy 61 (2000).
- Taylor JR, Lockwood AP, Taylor AJ. The prepuce: specialized mucosa of the penis and its loss to circumcision. Br J Urol 1996;77:291-295.
- Pang MG, Kim DS. Extraordinarily high rates of male circumcision in South Korea: history and underlying causes. BJU Int 2002;89:48-54.
- Fink KS, Carson CC, DeVellis RF. Adult Circumcision Outcomes Study: Effect on Erectile Function, Penile Sensitivity, Sexual Activity and Satisfaction. J Urol 2002;167(5):2113-2116.
- Josefson D. Vaccine against cervical cancer virus passes phase 1 trials. BMJ 2001;322:510.
- Franco EL, Duarte-Franco E, Ferenczy A. Cervical cancer: epidemiology, prevention and the role of human papillomavirus infection. CMAJ 2001;164(7):1017-25.
- Lehtinen M, Dillner J. Preventive human papillomavirus vaccination. Sex Transm Inf 2002;78:4-6.