Correspondence

Benign Anal Lesions and Anal Cancer

N Engl J Med 1995; 332:190-191January 19, 1995

Article

To the Editor:

Frisch et al. (Aug. 4 issue)1 examined the association between benign anal lesions and the risk of anal cancer. To evaluate this association further in different racial groups, we used a data base generated from computerized discharge records for inpatient visits at Veterans Affairs (VA) hospitals across the United States, as described elsewhere.2 Benign anal lesions were identified on the basis of the following discharge codes in the International Classification of Diseases (8th Revision, Adapted, and 9th Revision, Clinical Modification): anal fissure (code 565.0), anal fistula (code 565.1), perianal abscess (code 566), and hemorrhoids (code 455). Patients with known risk factors for anal cancer, such as Crohn's disease, ulcerative colitis, and condylomas,3 were excluded. The study cohort consisted of 221,073 male patients (174,789 whites and 46,284 blacks) with benign anal lesions diagnosed from July 1, 1969, to September 30, 1990. There were 1,929,535 person-years of follow-up. Ninety-one anal cancers (codes 154.2, 154.3, and 154.8) occurred in this period.

We used the standardized incidence ratio (the ratio of the number of cancers observed to the number expected) as a measure of the relative risk of cancer. The cancer-incidence rates for VA hospitals as a group were used to estimate the expected number of cancers.2 The number of cancers observed was assumed to have a Poisson distribution, and 95 percent confidence intervals were computed by the exact procedure of Liddell.4

The overall relative risks of anal cancer were significantly elevated in the first four years after hospitalization for both whites and blacks (Table 1Table 1Relative Risk of Anal Cancer after Hospitalization for a Benign Anal Lesion, According to Race.). The rapid decline in the risk of anal cancer among both whites and blacks after the first year after diagnosis was consistent with the findings of Frisch et al. The persistently elevated risk of anal cancer up to nine years after hospitalization among blacks may reflect the small number of cases and unstable estimates.

The strengths of our study include its large size, national scale, and ability to link records and to follow patients for more than 20 years. However, veterans who use VA hospitals may differ from other men in the United States in terms of risk factors for cancer such as tobacco and alcohol use, occupational exposures, and socioeconomic status. To compensate for these potential differences and the incompleteness of follow-up and ascertainment, we used the cancer rates in VA hospitals to estimate the number of cancers expected. Harris et al.5 have demonstrated that the cancer-incidence rates for VA hospitals as a group are comparable to those for the United States as a whole.

Our analysis of a large cohort of blacks and whites supports the finding of Frisch et al. that these benign anal lesions do not appear to be risk factors for anal cancer but may represent the early symptoms of anal cancer and may be helpful in the diagnosis of such cancers.

Albert Y. Lin, M.D., M.P.H.
Gloria Gridley, M.S.
Margaret Tucker, M.D.
National Cancer Institute, Bethesda, MD 20892

5 References

1
Frisch M, Olsen JH, Bautz A, Melbye M. Benign anal lesions and the risk of anal cancer. N Engl J Med 1994;331:300-302
Full Text | Web of Science | Medline

2
Gridley G, Klippel JH, Hoover RN, Fraumeni JF Jr. Incidence of cancer among men with the Felty syndrome. Ann Intern Med 1994;120:35-39
Web of Science | Medline

3
Holly EA, Whittemore AS, Aston DA, Ahn DK, Nickoloff BJ, Kristiansen JJ. Anal cancer incidence: genital warts, anal fissure or fistula, hemorrhoids, and smoking. J Natl Cancer Inst 1989;81:1726-1731
CrossRef | Web of Science | Medline

4
Liddell FD. Simple exact analysis of the standardised mortality ratio. J Epidemiol Community Health 1984;38:85-88
CrossRef | Web of Science | Medline

5
Harris RE, Hebert JR, Wynder EL. Cancer risk in male veterans utilizing the Veterans Administration medical system. Cancer 1989;64:1160-1168
CrossRef | Web of Science | Medline

Citing Articles (1)