HomeAbout Our StudyAbout the AuthorsNormal OvariesOvarian CystsOvarian CancerRisksScreeningCA-125SurgeryShould I Have My Ovaries Removed?Professional and Media Responses to our StudyProfessionalMedia



News Review From Harvard Medical School

Keep Ovaries, Researchers Say

Removing the ovaries as well as the uterus during a hysterectomy does not help most women and might do harm, concluded a study published August 1 in the journal Obstetrics & Gynecology. Many women now have their ovaries removed at the same time as a hysterectomy to prevent ovarian cancer, but the ovaries do have a function after menopause since they produce hormones, the Associated Press (AP) reported. Using a mathematical model based on previous studies, researchers concluded that ovary removal may increase the risk of death from heart disease for women under 65, AP reported.

By Howard LeWine, M.D.
Harvard Medical School

What Is the Doctor's Reaction?

Of the thousands of medical studies reported each year, only a handful have an immediate and very significant impact on the daily practice of medicine and surgery.

One such study has just been published in the August issue of the medical journal Obstetrics and Gynecology. The study examined the overall benefit of routine removal of the ovaries when a woman has a hysterectomy (surgical removal of the uterus) for a disease other than cancer, such as prolapse of the uterus, fibroids or endometriosis.

Until now, the official position regarding surgical removal of the ovaries (called oophorectomy) at the time of hysterectomy is to individualize the decision, assuming the woman is at the usual low risk of ovarian cancer. However, the most common practice is to perform oophorectomy routinely for women who have completed menopause.

In women younger than 40, the general recommendation is not to remove the ovaries. For women between 40 and 54, gynecologists vary in how strongly they will recommend routine oophorectomy.

Although not everyone has agreed, the general dogma has been that once a woman reaches menopause, ovarian production of hormones falls to such a low level that the ovaries offer little if any health benefits. Removing them almost completely eliminates cancer and other diseases of the ovary. Complication rates from hysterectomy with and without oophorectomy are nearly equal.

Now we have new compelling data that refutes the dogma. According to the results in this study, women under the age of 65 actually have better outcomes by not having oophorectomy at the time of hysterectomy, when the procedure is performed for diseases other than cancer. Women in the 65 and older group did not show a survival advantage either with or without routine oophorectomy.

What Changes Can I Make Now?

If you have a family history of ovarian cancer or a genetic predisposition to ovarian cancer-related syndromes (such as multiple family members with breast cancer or known positive BRCA genes), the results of this study are not relevant to you. Women with these characteristics were excluded from the study.

For the vast majority of women, this new information is very helpful. Even after menopause, the ovaries make small amounts of estrogen, and often make relatively significant quantities of certain male-type hormones that may have important health benefits. Interestingly, women who had routine oophorectomy under age 65 had higher death rates from heart disease than women who kept their ovaries.

Ovaries that continue to produce even small amounts of female and male hormones will help to maintain bone strength, preventing osteoporosis and fractures, a major cause of disability and death in older women.

Despite the apparent advantages of ovary conservation, it is quite reasonable for a woman to make the choice to have oophorectomy so she will not worry about developing ovarian cancer.

What Can I Expect Looking to the Future?

Like all studies, this study has potential pitfalls that the researchers recognize and state. Other questions about the methods used and conclusions drawn probably will arise. However, the results of this study will change how gynecologists present the pros and cons of routine oophorectomy with hysterectomy for benign disease. The decisions by patients and their doctors can now be based upon better facts rather than just "what is done."




David L. Olive, M.D.
Professor of Obstetrics and Gynecology
University of Wisconsin School of Medicine

Dogma is a well entrenched concept in the medical world, playing a critical role in both education and everyday clinical practice. Not one of us has progressed through our training and careers immune to the dogmatic hand-me-downs of an earlier, more experienced generation. Walking alongside tradition, dogma is an important foundation upon which much of allopathic medicine is constructed.

The sources of dogma may be multiple and varied. Clinical and basic research, clinical experience, logic, opinion, and critical debate all contribute to the development of our dogmatic precepts. The common interpretation of dogma is the uncritical assertion of a belief, but in fact dogmatism is a form of explicit expression of the sum total of past data, leaving little room for ambiguity of stance.

The new evidence-based approach to medical thinking brings with it an inherent attack upon the dogmas of the past, a skeptic's approach to analysis of the available data. Such skepsis is in fact an important co-conspirator with dogma to produce optimization of patient care, for it serves to periodically challenge and either ultimately support or refute the prevailing wisdom. As suggested by Edmond Murphy, "its relationship to dogma is like that of a grindstone to a blade: for the grindstone may destroy the blade of poor metal, but is what turns a lump of chilled steel into a sword" (1).

The role of prophylactic oophorectomy at the time of hysterectomy has long been a controversial issue. Proponents have argued that the procedure is justified due to the elimination of future ovarian cancer risk, as well as the decrease in future surgery or treatment of symptomatology related to retained ovaries. They suggest that the role of the ovary is for reproduction and hormone production, and when each of these functions cease (or are expected to terminate soon) the ovary serves no purpose other than generating mischief (2). Critics of this view point out that even post-menopausal ovaries are hormonally active, albeit not principally for the production of estrogen, and the importance of ovarian androgen production in later life has not been entirely delineated (3). Moreover, the psychological impact of castration can be profound in some individuals (4).

The result has been an uneasy dogma. Despite the official position by the American College of Obstetricians and Gynecologists that the decision should be individualized (5), the predominant teaching is that prophylactic oophorectomy in the low-risk patient should be avoided under the age of 40, should be routinely performed over age 55, and should be considered and discussed in the interval between. Most of us practice according to this dogma, and many utilize the cutoff of age 45 as a practical delineator for when to strongly advise the procedure.

In this issue of Obstetrics and Gynecology, Parker and colleagues perform decision analysis in an attempt to evaluate whether prophylactic oophorectomy at the time of hysterectomy for benign disease, in the woman with no apparent risk factors for ovarian malignancy, is in fact advantageous (6). Their calculations draw upon the best available evidence regarding mortality risks for a variety of disorders that have been linked to the presence or absence of ovaries. These data are then subjected to Markov modeling to estimate age-specific mortality for four situations: ovarian conservation with and without subsequent estrogen therapy, and oophorectomy with and without subsequent estrogen therapy. Their findings suggest that women undergoing oophorectomy are at greater mortality risk when the surgery is performed prior to the age of 65; furthermore, removal of the ovaries could not be shown to produce a survival benefit at any age.

To be sure, there are several shortcomings in this study. A decision analysis model is only as good as the data placed into it. In this case, while several elements of the input are drawn from high-quality data sources, a number of probability estimates are from trohoc studies with substantial potential bias. Still other data are derived from calculation, and some pieces of the puzzle are missing entirely resulting in assumptions of unknown validity.

More importantly, the calculated increase in mortality associated with oophorectomy (in the absence of estrogen therapy) derives almost entirely from enhancement of coronary heart disease risk. The magnitude of this risk derives from the Nurses' Health Study, a prospective cohort investigation (7). Given the design, the study is certainly subject to allocation and ascertainment bias, and other findings from the study have proven not to be congruent with randomized data from the Women's Health Initiative (8). Thus, the validity of the findings can easily be questioned.

Nevertheless, the authors have demonstrated via sensitivity analysis that even if no coronary heart disease increase is seen with oophorectomy, there remains no demonstrable advantage to the procedure in terms of longevity! This finding raises a multitude of follow-up questions worthy of investigation. Can these findings be supported as higher quality data are accumulated for inclusion into the model? If the results of the study are supported, can we define subgroups at low risk for ovarian cancer that may still benefit from oophorectomy? Can we utilize other outcome measures (quality of life, cost) to help weigh the relative value of this procedure? These are important considerations for future researchers.

While this study is certainly not definitive regarding the issue of prophylactic oophorectomy, it is sure to provide significant impact upon clinical practice. It is likely that the discussion regarding oophorectomy in the post-menopausal woman undergoing hysterectomy was brief and pointed; this is no longer likely to be the case. In the already complex discussion regarding oophorectomy among peri-menopausal women, the issue of potential effect upon mortality must now be brought to the table. The very topic of life expectancy as a result of a surgical procedure is probably one that is too infrequently broached. The result is likely to be a more lively, informative, and considered discussion with many women. Such enhanced complexity of interaction, while time-consuming to many and annoying to some, will almost certainly add to the quality of information utilized by the consenting patient.

As we tear down old dogmas with skepsis and subsequent analysis, we will inevitably create newer, shinier, fancier ones. It is important that we realize such pronouncements are rarely definitive or unassailable; that shortcomings virtually always exist in any collected packet of evidence, no matter how scrupulous the designers. It is our responsibility as physicians and patient advocates to continuously challenge these new truths with a critical eye, a reasoned perspective, and whenever possible a smattering of imagination.


  1. Murphy EA. Skepsis, Dogma, and Belief. Uses and Abuses in Medicine. Jo hns Hopkins University Press, Baltimore, 1981, p. 17.
  2. Mattingly RF. TeLinde's Operative Gynecology. Fifth edition. J. B. Lippincott Co., Philadelphia , 1977, pp. 198-199.
  3. Guzick DS, Hoeger K. Sex, hormones, and hysterectomies. New Engl J Med 2000;343:730-731.
  4. Taylor M. Psychological consequences of surgical menopause. J Reprod Med 2001;46:317-324.
  5. ACOG Practice Bulletin. Prophylactic Oophorectomy. 1999.
  6. Parker WH, Broder MS, Liu Z, Shoupe D, Farquhar C, Berek JS. Ovarian conservation at the time of hysterectomy for benign disease. Obstet Gynecol 2005; in press.
  7. Colditz G, Willett W, Stampfer M, Rosner B, Speizer F, Hennekens C. Menopause and the risk of coronary heart disease in women. N Engl J Med 1987;316:1105-1110.
  8. Writing Group for the Women's Health Initiative Investigators. Risks and benefits of estrogen plus progestin in healthy postmenopausal women. Principal results from the Women's Health Initiative randomized controlled trial. JAMA 2002;288:321.



To the Editor:

We read with interest the decision analysis model developed by Parker et al regarding ovarian conservation at the time of hysterectomy for benign disease. The previous model developed by Armstrong et al2 suggested that prophylactic oophorectomy after completion of childbearing lengthened life expectancy in women with inherited BRCA1/2 mutations. In contrast, current study reports that prophylactic oophorectomy at the time of benign pelvic surgery at less than 65 years of age increases the risk of dying, primarily due to increased deaths from coronary heart disease. The basis for the heart disease data is the Nurses Health Study,3 an observational study that reported twice the risk of dying from coronary disease after surgical menopause compared with premenopausal women. There is a well-known discrepancy between the results of observational studies and experimental trials with regard to cardio-protective effect of hormone replacement therapy. The Women's Health Initiative study4 reported an increased rate of coronary events in women on estrogen with a relative risk of 1.29. In light of this, one could question the cardio-protective effect of the minute amount of endogenous estrogen (if any) secreted by postmenopausal ovaries and, thereby, the role of ovarian conservation in cardiovascular mortality. Additionally, the predominant hormone secreted by the postmenopausal ovary is testosterone, which has an adverse affect on lipid profile.

The validity of the results derived from the decision analysis models is dependent on the data that are entered into the model. The current model looked only at the risk of mortality from ovarian cancer but does not take into account the morbidity related to ovarian preservation, such as the 5% risk of reoperation for benign ovarian disease, the technical difficulty associated with a repeat operation, and the anxiety incurred by the patient until malignancy is excluded. In contrast, Pell et al5 included both life expectancy, quality adjusted life expectancy, and individual patient characteristics in their decision model for prophylactic oophorectomy, to provide better guidance in the decision-making process.

It is unlikely that many of the practicing physicians will change their practices based on the current decision model. As new evidence emerges, the conclusions from models such as these would need to be updated. Ultimately, a randomized trial comparing prophylactic oophorectomy with ovarian conservation at the time of pelvic surgery for benign disease in low-risk women will provide the final answer to this widely debated question.

Vani Dandolu , MD
Department of Obstetrics and Gynecology,
Division of Urogynecology,
Temple University Hospital ,
Philadelphia , Pennsylvania

Enrique Hernandez, MD
Department of Obstetrics and Gynecology,
Division of Gynecologic Oncology,
Temple University Hospital ,
Philadelphia , Pennsylvania


  1. Parker WH, Broder MS, Liu Z, Shoupe D, Farquhar C, Berek JS. Ovarian conservation at the time of hysterectomy for benign disease. Obstet Gynecol 2005;106:219-26.
  2. Armstrong K, Sanford Schwartz, J, Randall T, Rubin SC , Weber B. Hormone replacement therapy and life expectancy after prophylactic oophorectomy in women with BRCA1/2 mutations: a decision analysis. J Clin Oncol 2004;22: 1045-54.
  3. Colditz GA, Willett WC, Stampfer MJ, Rosner B, Speizer FE, Hennekens CH. Menopause and the risk of coronary heart disease in women. N Engl J Med 1987;316:1105-10.
  4. Rossouw JE, Anderson GL, Prentice RL, LaCroix AZ, Kooperberg C, Stefanick ML, et al . Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women's Health Initiative randomized controlled trial. JAMA 2002;288:321-33.
  5. Pell I, Dowie J, Clarke A, Kennedy A, Bhavnani V. Development and preliminary evaluation of a clinical guidance programme for the decision about prophylactic oophorectomy in women undergoing a hysterectomy. Qual Saf Health Care 2002;11:32-9.


To the Editor:

The recent article by Parker et al1 and the editorial by Olive2 may add to the controversy concerning ovarian ablation at the time of surgery for benign disease. It is striking that neither the article nor the editorial consider the impact of cholesterol-lowering medication (statins) on coronary artery disease and mortality or the impact of the bisphosphonates on bone health in women at risk for osteoporosis. The collected data cited by Parker et al apparently overlook these factors, and both Parker and Olive acknowledge the benefits of endogenous estrogen with retained ovaries, but do not consider the benefits of estrogen replacement or bisphosphonate treatment.

In the clinical arena, both the article and the editorial may deter women and physicians from making the correct decision regarding ovarian conservation or ablation at the time of surgery for benign gynecologic or other pelvic disease. The increased risk for future surgery related to retained ovaries should be known to the patient. Hormone replacement and osteoporosis prevention and treatment should be evaluated for all women, but especially for those whose age raises these issues. Ovariana blation may not protect against peritoneal carcinoma, but it will certainly reduce the risk of death from ovarian cancer. The cited benefits for ovarian conservation are offset by the availability of statins and bisphosphonates. The protection from ovarian carcinoma offered by ovarian ablation should be emphasized to women who are candidates for pelvic surgery when they are beyond childbearing interest or age.

Robert C. Wallach, MD
Department of Obstetrics and Gynecology,
New York University School of Medicine ,
Clinical Cancer Center ,
New York , New York


  1. 1. Parker WH, Broder MS, Zhimei L, Shoupe D, Farquhar C, Berek JS. Ovarian conservation at the time of hysterectomy for benign disease. Obstet Gynecol 2005;106:219-26.
  2. Olive DL. Dogma, skepsis, and the analytic method: the role of prophylactic oophorectomy at the time of hysterectomy. Obstet Gynecol 2005;106:Part 1, November214-5.


In Reply:

We are pleased that our article is encouraging debate over prophylactic oophorectomy at the time of hysterectomy for benign disease. Drs. Dandolu and Hernandez correctly state that the Nurse's Health Study was observational. Nevertheless, it followed 121,700 women over 6 years and remains the largest database available.1 The findings of the Women's Health Initiative stated in their letter are incorrect; the estrogen-only arm found no evidence of increased risk of coronary heart disease (hazard ratio 0.91, confidence interval 0.75-1.12).2

The effect of testosterone on lipids is a surrogate outcome. Instead, we used coronary heart disease mortality as a reliable outcome measure. No morbidity data could be found in quantitative form. The study by Pell et al3 used only 10 women to assess quality-of-life parameters. Surgical mortality is extremely low, and including reoperation rates would not influence our results. Also not included are significant morbidities from nonfatal heart disease, crippling hip fracture, or quality-of-life concerns following abrupt surgical menopause.

It is unlikely that a randomized trial will be ever performed to answer the questions raised by our article. It would not be feasible to randomize women for oophorectomy/ovarian conservation if the outcome was the long-term mortality rate.

Dr. Wallach suggests that statins and bisphosphonates be given to treat heart disease and osteoporosis caused by oophorectomy. However, these drugs (also estrogen) are taken by less than 25% of patients for whom they are prescribed, making these strategies a poor substitute for ovarian conservation. 4,5

We hope that women and their physicians will discuss the long-term implications of oophorectomy and that many physicians will recommend ovarianconservation for women at average risk of developing ovarian cancer.

William H. Parker, MD
Michael S. Broder, MD, MPH
Jonathan S. Berek, MD, MMSc

David Geffen School of Medicine,
University of California ,
Los Angeles , California

Zhimei Liu, PhD
Cerner Health Insights,
Beverly Hills , California

Donna Shoupe, MD
School of Medicine ,
University of Southern California ,
Los Angeles , California

Cindy Farquhar, MD
School of Medicine ,
University of Auckland ,
Auckland , New Zealand


  1. Colditz G, Willett W, Stampfer M, Rosner B, Speizer F, Hennekens C. Menopause and the risk of coronary heart disease in women. N Engl J Med 1987; 316:1105-10.
  2. Women's Health Initiative Steering Committee. Effects of conjugated equine estrogen in postmenopausal women with hysterectomy: the Women's Health Initiative randomized controlled trial. JAMA 2004;291:1701-12.
  3. Pell I, Dowie J, Clarke A. Kennedy A, Bhavnani V. Development and preliminary evaluation of a clinical guidance programme for the decision about prophylactic oophorectomy in women undergoing a hysterectomy. Qual Saf Health Care 2002;11:32-8.
  4. McCombs J, Thiebaud P, McLaughlin- Miley C, Shi J. Compliance with drug therapies for the treatment and prevention of osteoporosis. Maturitas 2004;48: 271-87.
  5. Huser M, Evans T, Berger V. Medication adherence trends with statins. Adv Ther 2005;22:163-71.

North American Menopause Society

First To Know
February 15, 2006

Women undergoing hysterectomy before age 65 derive survival benefit from ovarian conservation

Parker WH, Broder MS, Liu Z, Shoupe D, Farquhar C, Berek JS.
Ovarian conservation at the time of hysterectomy for benign disease. Obstet Gynecol 2005;106:219-226. Level of evidence: III.

For women who are at average risk for ovarian cancer, conserving the ovaries during hysterectomy for benign disease benefits long-term survival until at least age 65, according to this study that used a Markov decision analytic model to determine the optimal strategy for maximizing survival in a hypothetical group of healthy women aged 40 to 80 years. Similar women were followed in the model from the age of surgery (40 or older) until age 80.

The model simulated possible outcomes and estimated the benefits or risk of each of four treatment strategies: ovarian conservation with or without estrogen therapy, and bilateral oophorectomy with or without estrogen therapy. Five conditions were identified with reported risk related to the presence of postmenopausal ovaries or estrogen therapy-ovarian cancer, coronary heart disease, hip fracture, breast cancer, and stroke. Based upon a review of the English literature, the model considered the risk of mortality from these conditions after ovarian conservation versus oophorectomy. The outcome measured was the proportion of women who would be alive at age 80 years for each the four treatment strategies. The model determined that ovarian conservation until at least age 65 benefits long-term survival in women with benign disease: women with oophorectomy before age 55 have 8.5% excess mortality by age 80 compared with those in whom ovaries are spared. Women with oophorectomy before age 59 have 3.92% excess mortality; by age 75, excess mortality for oophorectomy is less than 1%. For women at average risk of ovarian cancer, coronary heart disease, osteoporosis, breast cancer, and stroke, the probability of surviving to age 80 after hysterectomy at ages 50 to 54 varies from 62.46% for ovarian conservation without estrogen therapy to 53.88% for oophorectomy without estrogen therapy, according to the model. Reductions in death from coronary heart disease and hip fracture for those in whom ovaries were conserved far outweighed the number who would have died from ovarian cancer.

According to the authors, oophorectomy is routinely recommended for women undergoing hysterectomy to reduce the risk of ovarian cancer. But after menopause, the ovaries continue to produce significant amounts of testosterone and androstenedione, which are 5 converted to estrogen peripherally. Although younger women with BRCA1 or BRCA2 mutations who have significant risk for breast or ovarian cancer may derive a survival benefit from oophorectomy, women at average risk for these cancers experience increased risk of osteoporotic fracture and coronary heart disease. The study found that women younger than age 65 clearly benefit from ovarian conservation, and at no age is there a clear benefit from oophorectomy.

Comment. This study was designed to provide guidance regarding the choice of including a concomitant bilateral oophorectomy at the time of hysterectomy, for clinicians who perform hysterectomy and for women who are about to undergo hysterectomy. For years, prophylactic oophorectomy has been routinely recommended to women older than 40 years of age in order to prevent the future occurrence of ovarian cancer. The rationale was that ovarian cancer is often a silent disease, generally coming to clinical attention at a late stage associated with a poor prognosis of survival, and that highly sensitive means of screening and improved therapies do not yet exist for ovarian cancer. This dogma, which results in an estimated 300,000 prophylactic bilateral oophorectomies each year, is now called into question.

This study utilized a Markov decision analytic model, a technique that simulates large hypothetical groups of patients that are followed through the model over time. By recreating possible outcomes for these hypothetical patients, the benefits or risks for each treatment strategy may be estimated. The endpoint studied in this particular model was maximal survival.

The important finding from this analysis is that ovarian preservation was associated with a lower mortality at all ages from 40 to 80. The increased risk of coronary heart disease (CHD) after oophorectomy was an important driver of the final results; however, even when the model assumed no harmful effects of oophorectomy on subsequent occurrence of CHD, an advantage of ovarian preservation on mortality was still observed. A limitation of decision analysis is that the results are derived solely from risk estimates that are incorporated into the model; thus, the final analysis is only as reliable as the risk estimates available. In this study, the probability estimates were obtained mostly from case-control studies with potential for bias and incomplete information. Much of the data used to establish the risk estimates was derived from studies of predominantly white women, another limitation, since a substantial number of hysterectomies are performed on nonwhite women, and the probability estimates utilized need to be verified for these other groups of women. In addition, other important endpoints such as quality of life, cancer phobia, persistent pelvic pain, and need for reoperation for adnexal pathology were not included in this model, as few data exist regarding these outcomes or the contribution of these variables to mortality was estimated to be negligible by the investigators.

The importance of this study is the demonstration that ovarian conservation at the time of hysterectomy (in women who are not at increased risk for ovarian cancer) may lead to longer survival, making ovarian conservation a viable option rather than an extreme measure, even in the postmenopausal woman. Clearly, the decision for oophorectomy or ovarian conservation should balance risks specific to the patient as well as potential benefits and patient preferences. Certainly, there are still women who will benefit from concomitant oophorectomy at the time of hysterectomy, notably, those with intractable endometriosis. However, for most women, the scales are now tipping towards ovarian conservation.

Esther Eisenberg, MD, MPH
Professor of Obstetrics and Gynecology
Vanderbilt University Medical Center
Nashville , TN


Response to Letters to the Editor in Climacteric

To the Editor,

We welcome the opportunity to respond to the three invited commentaries in this journal regarding our recent article on ovarian conservation. 1-3 4

Dr. Studd expresses his opinion that we should not have included the oophorectomy and no estrogen therapy (ET) arm in our study, since this choice is unacceptable to him. However, this is clearly the most common situation found clinically. Studies show that actual continuation rates of ET are very low and, despite Dr. Davy's comments, continuation rates for bisphosphanates or statins are no higher.

In 1999, prior to publication of the Women's Health Initiative (WHI), and "at a time when medical support for ET and publicity for ET were high", only 31% of women continued to use ET for 5 years or longer following hysterectomy and oophorectomy. 5 In the 6 months following publication of the WHI continuation rates of ET decreased from 12.6% to 9.1% and new starts also decreased significantly. 6

Even women initiating therapy with either ET, estrogen and progestin, bisphosphantes or raloxifene for treatment of documented osteoporosis (n=58,109) had medication continuation rates less than 25% at 12 months. 7 Statin continuation rates are 18% at the end of one year. 8 It should be noted that these studies do not account for women who never see a doctor, who never get a prescription, or who get a prescription but choose not to fill it and, therefore, underestimate the risk of disease for large numbers of women.

Despite Dr. Studd's feeling that physicians should appropriately counsel women about long-term hormone therapy after oophorectomy, women have side-effects, fears and belief systems about medications which may preclude use despite a physician's recommendation. The mortality data presented in our paper for oophorectomy with ET is calculated for 100% continuation rates to age 80. Lesser continuation rates, as noted in real life, would proportionally increase mortality and, therefore, oophorectomy is a disservice for the vast majority of these women.

There is no evidence to support oophorectomy as prophylaxis for the future possibility of pelvic pain, residual ovary syndrome, mild endometriosis or ovarian cyst formation. Among 2561 women having a hysterectomy without oophorectomy for any indication and followed for 20 years, subsequent oophorectomy was performed in just 2.8%. 9 Furthermore, while oophorectomy may be indicated in the rare situation of severe PMS with a woman's full understanding and consent, recommending this for women with a few bad days a month, as implied by Dr. Studd, would not be acceptable to most women.

It is now clear that benign ovarian cysts are common among postmenopausal women and do not require surgical intervention. The study inappropriately quoted by Drs. Davy and Oehler reported that 19% of 234 women over age 45 had cysts, all 5 cm or smaller, found at autopsy. 10 None were malignant. Earlier studies of postmenopausal women subjected to surgery for adnexal masses with benign sonographic appearance and normal CA-125 levels found no malignant tumors. 11 Sonographic screening of 7705 asymptomatic postmenopausal women found unilocular cysts in 3.3%. No woman was found to have cancer. 12 Many gynecologic oncologists would recommend expectant management for postmenopausal women with no increase in cyst size or increase of CA-125. 13 14

Dr. Studd takes issue with data "that hysterectomy reduces risk of ovarian cancer by 40%". However, as cited in our study, virtually all epidemiologic data support this view. In addition, our sensitivity analysis using no decrease in the risk of ovarian cancer following hysterectomy did not alter outcomes. Regarding Dr. Shapiro's comment, we did note that, as used by convention in many other studies, average risk of ovarian cancer included women without germ-line mutations. Dr. Davy comments that oophorectomy reduces the risk of subsequent breast cancer, but such protection occurs only when oophorectomy is performed before age 50 and only persists for the first 10 years following surgery. This is not a very convincing reason to perform oophorectomy.

Dr. Studd states the WHI found ET before age 60 decreases the risk of MI and breast cancer. He interestingly uses relative risks here (rather than absolute risk) and fails to mention that the confidence intervals are wide and the results are not statistically significant. Apparently it is Dr. Studd's analysis, rather than our own, that approximates meta-physics.

We agree with Dr. Shapiro's statement that there are many risk factors (menopausal status, age, lipoprotein patterns, etc) for mortality that our study did not take into account. We purposely constructed the model without too much complexity by chosing the 5 conditions most commonly associated with oophorectomy. This does not invalidate the model or results. Dr. Shapiro criticizes exclusion of venous thromboembolism and colon cancer from our analysis, but we are not aware of evidence that these outcomes are influenced by oophorectomy.

The authors collectively fail to understand the difference between meta-analysis and decision analysis. Meta-analyses require a systematic review of the literature with specific inclusion and exclusion criteria to answer a specific clinical question and entails statistical pooling of the numeric data from more than one study. We did not perform a meta-analysis. Decision analysis, as performed in our study, while based on a comprehensive literature review, uses probabilities derived from epidemiological data and creates a Markov model where all possible outcomes are considered. In our study the outcomes were mortality following hysterectomy with and without oophorectomy, and with and without HRT.  

The evidence grading system used in our study is one of the most commonly used and has been validated in health services research and epidemiology. 15 We used the best sources of mortality data available; SEER and National Vital Statistics. Relative risk estimates were derived from each of the selected best quality studies. Histories of hysterectomy with or without oophorectomy were adequately precise and had been validated by study authors. Variations among individual studies were captured by our sensitivity analyses. Confidence intervals were also captured by our extensive sensitivity analyses, including worst-case scenarios. While Dr. Studd sees our findings as politically correct, the results of the analysis were not known until the computer generated the survival data. No scenario for oophorectomy was found to confer any survival advantage.

Dr. Studd recommends a randomized trial to answer this important question. As discussed in our paper, there are major obstacles to this type of study. Women are not very likely to accept randomization to oophorectomy or ovarian conservation, large groups of women would need to be studied, and a 30-year wait for true outcome data (rather than surrogate) is not practical. Even if such a study were begun, it is likely that other research (such as a reliable proteiomics blood test for early ovarian cancer) might make the eventual findings irrelevant.

We agree that the public health consequences regarding incidental oophorectomy are major, in that approximately 300,000 US women are subjected to this procedure annually. Up until now, this decision was based on one factor alone - the risk and fear of ovarian cancer. Dr. Studd states that "it is hard to believe that estrogens are more dangerous than conserved ovaries" and then pleads for castration. What ever happened to "evidence" and the principle "first, do no harm"?

Our study was not meant to replace a woman's decision as to whether or not she should have an oophorectomy. We chose to model a typical woman's survival with regards to the impact of oophorectomy and allow individual women to determine their own particular risks and concerns regarding the related conditions. Further stratifying the data will be the goal of future research, pending availability of additional studies. We are presently seeking funding for a study in collaboration with the Nurse's Health Study group in order to apply data from those 122,000 women to our model. Hopefully, our present study will raise awareness among women and their gynecologists regarding the public health consequences of routinely performed oophorectomy.

William H. Parker, MD
Michael S. Broder, MD, MPH
Zhimei Liu, PhD
Donna Shoupe, MD
Cindy Farquhar, MD
Jonathan S. Berek, MD, MMSc

Reference List

  1. Studd J. Does retention of the ovaries improve long-term survival after hysterectomy? Prophylactic oophorectomy. Climacteric 2006;9:164-6.
  2. Davy M, Oehler MK. Does retention of the ovaries improve long-term survival after hysterectomy? A gynecological oncological perspective. Climacteric 2006;9:167-8.
  3. Shapiro S. Does retention of the ovaries improve long-term survival after hysterectomy? The validity of the epidemiological evidence. Climacteric 2006;9:161-3.
  4. Parker WH, Broder MS, Liu Z, Shoupe D, Farquhar C, Berek JS. Ovarian conservation at the time of hysterectomy for benign disease. Obstet Gynecol 2005;106:219-26.
  5. Brett KM, Reuben CA. Prevalence of estrogen or estrogen-progestin hormone therapy use. Obstet Gynecol 2003;102:1240-9.
  6. Buist DS, Newton KM, Miglioretti DL, Beverly K, Connelly MT, Andrade S et al. Hormone therapy prescribing patterns in the United States. Obstet Gynecol 2004;104:1042-50.
  7. McCombs JS, Thiebaud P, McLaughlin-Miley C, Shi J. Compliance with drug therapies for the treatment and prevention of osteoporosis. Maturitas 2004;48:271-87.
  8. Huser MA, Evans TS, Berger V. Medication adherence trends with statins. Adv Ther 2005;22:163-71.
  9. Dekel A, Efrat Z, Orvieto R, Levy T, Dicker D, Gal R et al. The residual ovary syndrome: a 20-year experience. Eur J Obstet Gynecol Reprod Biol 1996;68:159-64.
  10. Dorum A, Blom GP, Ekerhovd E, Granberg S. Prevalence and histologic diagnosis of adnexal cysts in postmenopausal women: an autopsy study. Am J Obstet Gynecol 2005;192:48-54.
  11. Parker WH, Levine RL, Howard FM, Sansone B, Berek JS. A multicenter study of laparoscopic management of selected cystic adnexal masses in postmenopausal women. J Am Coll Surg 1994;179:733-7.
  12. Bailey CL, Ueland FR, Land GL, DePriest PD, Gallion HH, Kryscio RJ et al. The malignant potential of small cystic ovarian tumors in women over 50 years of age. Gynecol Oncol 1998;69:3-7.
  13. van Nagell JR, DePriest PD. Management of adnexal masses in postmenopausal women. Am J Obstet Gynecol 2005;193:30-5.
  14. Nardo LG, Kroon ND , Reginald PW. Persistent unilocular ovarian cysts in a general population of postmenopausal women: is there a place for expectant management? Obstet Gynecol 2003;102:589-93.
  15. Harris RP, Helfand M, Woolf SH, Lohr KN, Mulrow CD, Teutsch SM et al. Current methods of the US Preventive Services Task Force: a review of the process. Am J Prev Med 2001;20:21-35.




Copyright © 2005 OvaryResearch.com