Estrogens & cervical cancer

Commentary: Although the most critical end-point of HRT’s effect on bone is fracture frequency, studies such as this one on calcium turnover in various bones help us understand the magnitude of the effect on bone of various hormone replacement regimens. This study is convincing in its findings supporting either combined or sequential HRT. The authors found no differences between the two protocols of HRT but found striking differences from placebo. Note that lumbar spine density increased with hormones but that not all bones were affected equally: there seemed to be a greater effect on axial bones than on peripheral ones. These findings were thought by the authors to be explained by the mainly trabecular bone of the spine compared with the predominantly cortical bone of the forearm and femur. There were two other interesting aspects of the study: 1) After cessation of therapy of 2 years duration, women lost bone at the same rate as the untreated group; this suggests that to prevent fractures women might need to take hormones forever. 2) After 8 years there was no difference in weight gain between the women receiving HRT and those receiving placebo. They all gained about 6 lb. Weight gain is one of the most frequent concerns expressed by patients considering HRT. This study gives us hard data to cite in refuting their fear about weight gain due to hormones.

Estrogens & Cervical Cancer Parazzini F, La Vecchia C, Negri E, Franceschi S, Moroni S, Chatenoud L, et al. Casecontrol study of oestrogen replacement therapy and risk of cervical cancer. BMJ 1997; 315:85-g.

Synopsis: In this study, the authors looked at the use of estrogen by 645 northern Italian women over age 40 years who had cervical cancer and compared them with case controls. They found that women who took exogenous estrogens were not at an increased risk of cervical cancer. Furthermore, their findings suggest that the risk of cervical cancer may even be reduced in women taking estrogens. . . . 01998 by the American College of Obstetricians and Gynecologists PublIshed by Elsewr Science lnc 1085-6862198155 50

Commentary: These authors pursued a theory that estrogens, such as those found in oral contraceptives, may increase the risk of cervical cancer by oncogenic transformation of papillomavirus. They found no such tendency but instead found a tendency for reduced risk of cervical cancer by hormone users. In addition, their data indicated that the risk of cervical cancer decreased with duration of estrogen use. The most likely explanation for their finding is that women taking estrogens are more likely to be screened regularly with Papanicolaou smears. However, use of the number of Papanicolaou smears recalled by the patients in a multivariate analysis did not alter the risk. I still believe that more frequent screening with detection and treatment of earlier precancerous lesions was a major factor underlying the authors’ findings. A weakness of the study is that the authors did not separate their analyses by tissue type, lumping together those with squamous cell, adenocarcinoma, and undefined histology. They also did not control for presence of papillomavirus. Thus, differences due to the effect of hormones on viral proliferation may have been obscured by the larger population studied. Nonetheless, this study is good news. It gives us data to pass on to our patients who may fear that hormones increase the risk of cervical cancer. We can tell our patients that not only do hormones not increase the risk of cervical cancer but that they may even reduce the risk.

Subtotal Hysterectomy Scott JR, Sharp HT, Dodson MK, Norton PA, Warnr HR. Subtotal hysterectomy in modern gynecology: A decision analysis. Am J Obstet Gynecol 1997; 176:118692.

Synopsis: A decision analysis was performed on studies published in English between 1980 and 1996 to compare total hysterectomy with subtotal hysterectomy for benign disease. The authors examined rates of surgical death, infection, hemorrhage, adjacent organ injury, impaired sexual function, vesicourethral symptoms, and vaginal or cervical cancer deaths January/February

in the articles reviewed. The rate of complications was fairly similar for the two groups. The effects on sexual and vesicourethral function were similar. Using the evidence in the literature, the authors concluded that the purported benefits of subtotal hysterectomy were not proven. . . . Commentary: This study is an attempt to use the principle of evidencebased medicine to make decisions about medical management. This is a worthwhile and laudable goal. The problem is that decision analysis is itself subject to interpretation and bias, and the desirable outcome may not be agreed to by the interested parties: physicians, payers, patients. This study was presented at the 64th Annual Meeting of the Central Association of Obstetricians and Gynecologists held in Houston, October 17-19, 1996, and was discussed by two members. In the response to questions posed by the members, one of the authors admitted that after independent review of the literature by the authors, there was considerable disagreement and that the authors “wrestled back and forth,” eventually changing their conclusions from those preliminarily submitted. This difficulty experienced by the authors reflects the difficulty of finding truly objective data that are comparable from study to study on which to base a final decision. The authors concluded that “potential bias in’ observational studies makes judgments about probabilities of clinical events imprecise and subjective.” But we should keep trying.

Induced Abortion & Ectopic Pregnancy Skjeldestad FE, Gargiullo PM, Kendrick JS. Multiple induced abortions as risk factor for ectopic pregnancy: A prospective study. Acta Obstet Gynecol &and 1997; 76:691-6.

Synopsis: To assess the risk of ectopic pregnancy associated with induced abortion, 3751 women in Norway were followed prospectively for l-6 years. The number of prior abortions was related to incidence of subsequent 1398

l

ACOG

CLINICAL

KEVIEW

l

7