HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Im, S. S. Articles by Berman, M. L. Search for Related Content PubMed PubMed Citation Articles by Im, S. S. Articles by Berman, M. L. Related Collections Gynecologic oncology Gynecologic surgery

Validation of Referral Guidelines for Women With Pelvic Masses

Samuel S. Im, MD^*, Alan N. Gordon, MD, Barbara M. Buttin, MD, Charles A. Leath, III, MD, Bobbie S. Gostout, MD^¶, Chirag Shah, MD^||, Kenneth D. Hatch, MD^**, Jianmin Wang, MD and Michael L. Berman, MD^*

From the *University of California Irvine Medical Center, Orange, California; Baylor University Medical Center, Dallas, Texas; Barnes-Jewish Hospital, St. Louis, Missouri; University of Alabama Birmingham Hospital, Birmingham, Alabama; ¶Mayo Clinic, Rochester, Minnesota; ||University of Washington Medical Center, Seattle, Washington; **Arizona Health Sciences Center, Tucson, Arizona; and Gynecologic Oncology Group Statistical Office, Buffalo, New York.

	ABSTRACT

TOP ABSTRACT SGO and ACOG Referral... MATERIALS AND METHODS RESULTS DISCUSSION Modified Referral Criteria Based... REFERENCES

 
OBJECTIVE: Guidelines for referring women with pelvic masses suspicious for ovarian cancers to gynecologic oncologists have been published jointly by Society of Gynecologic Oncologists (SGO) and the American College of Obstetricians and Gynecologists (ACOG). They are based on patient age, CA 125 level, physical findings, imaging study results, and family history. Although the guidelines are evidence-based, their predictive value in distinguishing cancers from benign masses is unknown.

METHODS: Chart review for factors included in the guidelines of surgically evaluated women with pelvic masses at 7 tertiary care centers during a 12-month interval was performed. This information was used to estimate the predictive values of the SGO and ACOG guidelines in identifying patients with malignant pelvic masses.

RESULTS: A total of 1,035 patients were identified, including 318 (30.7%) with primary malignancies of the ovary, fallopian tube, or peritoneum. Seventy-seven were younger than 50 years old (premenopausal group), and 240 were 50 years old or older (postmenopausal group). Fifty additional patients (4.8%) had cancers metastatic to the ovaries, and the remaining 667 (64.4%) had benign masses. The referral guidelines captured 70% of the ovarian cancers in the premenopausal group and 94% of the ovarian cancers in the postmenopausal group. The positive predictive value was 33.8% for the premenopausal group and 59.5% for the postmenopausal group, whereas the negative predictive values were more than 90% for both groups. Elevated CA 125 level was the single best predictor of malignancy in both groups.

CONCLUSION: The SGO and ACOG referral guidelines effectively separate women with pelvic masses into 2 risk categories for malignancy. This distinction permits a rational approach for referring high-risk patients to a gynecologic oncologist for management.

LEVEL OF EVIDENCE: III

Guidelines for referral of women with pelvic masses by gynecologists to gynecologic oncologists have not been universally recognized. This issue was addressed recently by the American College of Obstetricians and Gynecologists (ACOG) and the Society of Gynecologic Oncologists (SGO) with the joint publication of referral guidelines to assist the gynecologist in triaging women with pelvic masses.¹³ These referral guidelines are based on patient age, CA 125 level, physical findings, imaging study results, and family history of breast or ovarian cancer in a first-degree relative (Referral Guidelines box). The distinction between CA 125 levels in premenopausal compared with postmenopausal women is based on the common observation of elevated CA 125 levels, sometimes up to 200 units or more, associated with some benign masses, including endometriomas. Similarly, the findings of a nodular or fixed mass are included as a referral criterion for postmenopausal women only, because pelvic infections and endometriosis often induce such changes in the premenopausal years. Only 1 criterion from the respective premenopausal or postmenopausal list is required to recommend referral to a gynecologic oncologist. Although the guidelines are specific, they should be used in conjunction with careful medical judgment when evaluating patients with clinically suspect findings that do not necessarily meet the criteria.

	SGO and ACOG Referral Guidelines for a Newly Diagnosed Pelvic Mass

TOP ABSTRACT SGO and ACOG Referral... MATERIALS AND METHODS RESULTS DISCUSSION Modified Referral Criteria Based... REFERENCES

 
 Premenopausal (< 50 Years Old)

 CA-125 > 200 U/mL

Ascites

Evidence of abdominal or distant metastasis (by exam or imaging study)

Family history of breast or ovarian cancer (in a first-degree relative)

 Postmenopausal ( 50 Years Old)

 CA-125 > 35 U/mL

Ascites

Nodular or fixed pelvic mass

Evidence of abdominal or distant metastasis (by exam or imaging study)

Family history of breast or ovarian cancer (in a first-degree relative)

Although the guidelines are both intuitive and evidence-based, their positive and negative predictive values in distinguishing ovarian cancers from benign pelvic masses are uncertain. Also, it is not known whether some modifications in these criteria might be warranted to maximize referral of women with cancers while minimizing referral of women with benign masses. To answer these questions, a multicenter retrospective chart review of patients who were admitted to the 7 respective hospitals for surgical management of pelvic masses was performed.

	MATERIALS AND METHODS

TOP ABSTRACT SGO and ACOG Referral... MATERIALS AND METHODS RESULTS DISCUSSION Modified Referral Criteria Based... REFERENCES

 
A review of all women who underwent surgical exploration for a pelvic mass between July 2000 and June 2001 at 6 university-based academic institutions and 1 large tertiary community hospital was performed. The participating facilities included the University of California, Irvine Medical Center, the University of Alabama at Birmingham, the University of Arizona, Tucson, Arizona, the Mayo Clinic, Rochester, Minnesota, the University of Washington, Seattle, Washington, Washington University, St. Louis, Missouri, and Baylor University Medical Center, Dallas, Texas. All women 18 years or older who underwent definitive surgical management of pelvic masses were identified at each site. The only exclusion criterion was a prior history of an invasive gynecologic malignancy.

The International Classification of Diseases, 9th revision (ICD-9) discharge diagnosis codes for benign and malignant neoplasms of the ovary, fallopian tube, and peritoneum were used to identify the study population (Table 1). Other ICD-9 codes for ascites, low malignant potential tumor of the ovary, and ovarian and tubal cysts, were also used as a cross-check; however, our initial list of ICD-9 codes identified all patients explored for a pelvic mass at the 7 institutions. In addition, separate lists were generated from the tumor registries and pathology tissue logs from each hospital to cross-check for these diagnoses. These were compared with the list from the hospital database to ensure complete capture of all qualified patients.

Once the patient list was generated from the discharge diagnoses, tumor registries, and tissue logs, we reviewed the patient medical records, imaging studies (computed tomography [CT], magnetic resonance imaging, and ultrasound reports), pathology reports, and CA 125 levels. The pelvic examination documented in the admitting history and physical was evaluated for finding of tumor fixation in the pelvis or nodularity. Pertinent physical findings and imaging studies were reviewed to determine preoperative evidence of abdominal or distant metastasis. The presence of first-degree relatives with breast or ovarian cancer was determined from the admitting history and physical examination. The specialty of the primary surgeon was identified by the site investigator at each site from the operative reports. Tumor histology and stage of disease were determined from review of the pathology reports.

The study was approved by the institutional review boards at all institutions in accordance with assurances filed with and approved by the U.S. Department of Health and Human Services. The referral guidelines in the ACOG Committee Opinion (Number 280) published in the December 2002 issue of Obstetrics & Gynecology¹³ were applied to our patient list to estimate positive and negative predictive values (PPV and NPV) of the individual criteria and the guidelines taken as a group.

The guidelines divide patients into premenopausal and postmenopausal groups. For the purpose of our analysis, women aged younger than 50 years were considered to be premenopausal and those 50 years and older were considered to be postmenopausal. For the purpose of this study, women with invasive ovarian cancer, low malignant potential tumors of the ovary, and primary peritoneal tumors were all categorized as "ovarian cancer patients."

The individual factors defined within the referral guidelines were analyzed for their PPV and NPV using the SAS statistics program (SAS Institute, Cary, NC). For the categorical variables, the Fisher exact test was used for the univariate analysis. Logistic regression was used when the variables were continuous. From the statistically significant variables identified from the univariate analysis, multiple logistic regression models were developed. For the determination of optimal cutoff point for the CA 125 level, discriminant analysis was performed by using the linear discriminant function in the SAS statistics program.

	RESULTS

TOP ABSTRACT SGO and ACOG Referral... MATERIALS AND METHODS RESULTS DISCUSSION Modified Referral Criteria Based... REFERENCES

 
A total of 1,035 patients were identified, including 318 (30.7%) with primary malignancies, 50 (4.8%) with cancers metastatic to the ovary(ies), and 667 benign masses. One of the 318 patients with a primary cancer was operated on by a general surgeon and was excluded from our analysis. The age distribution ranged from 18 to 91 years with a median of 51 years. Preoperative CA 125 levels were measured in 700 women (68%) and preoperative imaging was recorded in 988 (95%), including 701 ultrasound procedures and 541 CT scans. Two hundred fifty-four patients had both studies performed. In addition, 31 patients had undergone preoperative magnetic resonance imaging studies and 3 underwent positron emission tomography scans before surgery. All 34 of these patients also underwent ultrasound or CT scans or both preoperatively.

Menopausal status was difficult to ascertain from the clinical records. We elected to use age as a surrogate for menopausal status, assigning women aged 50 years and younger to the premenopausal group and women aged older than 50 years to the postmenopausal group. Of 317 women with cancers included in our analysis, 77 were thus classified as premenopausal (24.3%) and 240, postmenopausal (75.7%). At the 7 participating centers, 81% of premenopausal women with cancers and 91% of postmenopausal women with cancers were operated on by gynecologic oncologists, with the remainder operated on by obstetrician–gynecologists. The subset of 37 women who were operated on by generalists rather than gynecologic oncologists included 13 with low malignant potential tumors, 22 with invasive epithelial cancers, and 2 with germ cell cancers. Among the 22 invasive epithelial cancers, 5 were stage I, 4 were stage II, 9 were stage III, 1 was stage IV, and 3 were unstaged. Nine of the 22 cancer patients operated on by the generalists did not have preoperative CA 125 levels drawn. Also, when the SGO and ACOG referral guidelines were applied to these patients, 11 had at least 1 criterion for referral to a gynecologic oncologist. In contrast, at these 7 tertiary care centers, 58% of women with benign masses (389 of 667) underwent operation by gynecologic oncologists, including 77% of postmenopausal women (224/290) and 44% of premenopausal women (165/377). Among the 224 postmenopausal and 165 premenopausal women with benign masses who were operated on by gynecologic oncologists, 108 (48%) and 64 (39%) women had at least 1 referral criterion, respectively.

The positive and negative predictive value of each referral guideline individually and for the guidelines considered as a group were calculated both for premenopausal and postmenopausal women (Table 2). In the premenopausal group, the PPV ranged from 58% to 70% for elevated CA 125 level, presence of ascites, and evidence of metastasis by examination or imaging study. However, the PPV for family history of breast or ovarian cancer in a first-degree relative was only 19%. By contrast, the NPV for all the guidelines were above 80%, and the overall NPV was 92% in the premenopausal group. For postmenopausal patients, the overall PPV was 60%, and the overall NPV was 91%. The criterion, which correlated least well with cancer risk in the postmenopausal group, was family history of breast or ovarian cancer with a PPV of 42% and a NPV of 56%. The presence of ascites, evidence of distant metastases, and elevated CA 125 had the highest positive predictive value in this group,

To estimate how accurately the SGO and ACOG referral guidelines would have distinguished women with benign masses from those with cancers, we applied them to our patient population (Table 3). Because not all patients had recorded data for each criterion in the guidelines, the subset of 267 cancer patients (61 premenopausal and 206 postmenopausal women) and 397 benign mass patients (180 premenopausal and 217 postmenopausal women) with complete information was used for this analysis. In this select group the guidelines correctly identified 70% of ovarian cancer patients in the premenopausal and 94% in the postmenopausal group. Absence of any referral criterion was found in 69% of women with benign masses in premenopausal and 58% in the postmenopausal group. Thus, 31% of premenopausal women and 42% of postmenopausal women with benign masses would have been referred to a gynecologic oncologist if the primary physician adhered strictly to the referral guidelines.

To estimate the significance of each variable from the referral guidelines in distinguishing ovarian cancers from benign masses, a univariate analysis was performed. The analysis showed that family history of breast and ovarian cancer was a relatively poor predictor for ovarian cancer (Table 4). However, postmenopausal state, high abnormal preoperative CA 125 level, presence of ascites, and suspicious pelvic examination or suspicious imaging study results for abdominal and distant metastasis were all shown to be pertinent factors in predicting ovarian malignancy. Based on these univariate analyses, multiple logistic regression models were established, and 4 variables were identified as the most significant factors in predicting ovarian cancer. These factors were postmenopausal state, abnormal preoperative CA 125 level, presence of ascites, and evidence of abdominal or distant metastasis by examination or imaging study.

	DISCUSSION

TOP ABSTRACT SGO and ACOG Referral... MATERIALS AND METHODS RESULTS DISCUSSION Modified Referral Criteria Based... REFERENCES

 
A woman with a pelvic mass can present a difficult dilemma for the gynecologic surgeon because of the uncertainty about whether she should be referred for specialty surgical care. If the mass proves to be malignant, it will typically require an extensive operation, including formal staging, which is best performed by a gynecologic oncologist. Specific referral guidelines have been recommended jointly by the SGO and ACOG to help guide the generalist in deciding whom to operate and whom to refer. The usefulness of these guidelines in distinguishing benign from malignant masses, however, has not been thoroughly examined.

In our study, we reviewed the medical records of more than 1,000 women who presented with pelvic masses requiring operative intervention from 6 academic and 1 large tertiary community hospital. Based on the review of these records, we assessed whether the referral guidelines were able to correctly distinguish ovarian cancer patients from those with benign masses based on the positive and negative predictive values associated with each of the criteria and the group as a whole. We also determined the pattern of referral of women with pelvic masses cared for mainly in the tertiary care setting of academic institutions. Overall, the high NPV for both the premenopausal and postmenopausal women in our study group demonstrated that the referral guidelines are useful in identifying noncancer patients. Furthermore, the data suggest that the universal application of the guidelines would result in referral of 70% of the ovarian cancers in premenopausal women and 94% in postmenopausal women.

The guidelines do, however, tend to over-refer women with benign masses with a PPV of 33.8% in premenopausal women and 59.5% in postmenopausal women. This excessive capturing of benign masses by the criteria was due in part to the large number of women with benign masses and a positive family history of breast and ovarian cancer in our study population. Thus, the family history of ovarian and breast cancer in first-degree relatives as the only referral criterion may capture an occasional patient with an ovarian cancer; however, much more commonly this criterion alone identifies a woman who is most likely to have a benign mass. It should be remembered, however, that in the context of a strong history of breast or ovarian cancer the possible finding of an unsuspected malignancy is one of several reasons for obtaining consultation with a gynecologic oncologist. Counseling regarding the role of genetic testing, the impact of surgery on both breast and ovarian cancer risk, and possible enrollment in clinical trials can all be accomplished through such a consultation.

A shortcoming of this study, which is based on chart reviews, is the absence of recorded data for each criterion for every patient. For example, 32% of patients had no measurement of CA 125 level and 5% had no imaging studies. This absence of data likely would have affected the PPV and NPV values. Accordingly, the referral criteria would benefit from an analysis done in patients prospectively enrolled with a trial designed to assess them. The missing data points reveal another potential advantage of the guidelines. As generalists become increasingly familiar with the guidelines, they will be able to reflect on the criteria in a manner similar to a checklist, prompting them to measure a CA 125 level and to consider the possibility of ascites, fixation, and nodularity as they perform the patient assessment.

In this study, a small subgroup of patients with primary cancers was identified who had been operated on by general gynecologists. This group comprised 37 patients, 60% of whom had invasive cancer and more than half of whom had advanced stage disease. Many of these patients did not have preoperative CA 125 levels drawn, which if done, might have helped correctly triage these patients to gynecologic oncologists. Indeed, when the referral guidelines were applied to these patients, even without the benefit of a CA 125 level, half of them had at least 1 other criterion for triage to a gynecologic oncologist.

In an effort to explore models that might increase the positive predictive value of the referral criteria we reanalyzed the data from our patient population excluding the criterion of family history of breast or ovarian cancer in all subjects and that of nodularity and fixation of the tumor on pelvic examination in the postmenopausal group. Based on discriminant analysis for various CA 125 levels at presentation we found that a CA 125 of greater than 50 units/mL was a better discriminator of cancer compared with a benign mass than the 200 units/mL value. Accordingly we decreased the threshold for referral based on CA 125 level in the premenopausal group of women to 50 units/mL in our revised and simplified criteria (Modified Referral Criteria box). When the modified criteria were applied to our data set, 85% of the premenopausal and 90% of the postmenopausal women with ovarian cancers would have been correctly identified for referral to gynecologic oncologists (Table 5). In addition, only 27% of the premenopausal and 24% of the postmenopausal women with benign pelvic masses would have met the criteria for referral. Clearly these adjustments should be evaluated further when refinements of the 2002 SGO and ACOG referral guidelines are being considered.

	Modified Referral Criteria Based on Multiple Regression Models and Discriminant Analysis

TOP ABSTRACT SGO and ACOG Referral... MATERIALS AND METHODS RESULTS DISCUSSION Modified Referral Criteria Based... REFERENCES

 CA-125 > 50 U/mL

 Ascites

 Evidence of abdominal or distant metastasis

 Postmenopausal ( 50 Years Old)

 CA-125 > 35 U/mL

 Ascites

 Evidence of abdominal or distant metastasis

For the current study, 6 of the 7 participating hospitals are academic centers with established gynecologic oncology divisions. The seventh hospital, Baylor University Medical Center in Dallas, TX, is a large tertiary care center in the private setting. The gynecologic oncologists at all 7 centers provide support for the obstetric and gynecologic physicians who practice there, allowing easy referral access for women who present with pelvic masses. It is likely that patterns of referral in such institutions do not reflect that of many physicians treating similar patients at smaller community-based hospitals. The authors also acknowledge that positive and negative predictive values are affected by the incidence of ovarian cancer in the population. This study population from tertiary care centers probably has a higher incidence of malignant masses than that seen in the national mix of all patients diagnosed with pelvic masses. As such, a true population-based PPV and NPV of the referral guidelines cannot be gleaned from this study. Accordingly, the impact of the referral guidelines on patient referral in the community setting remains unknown. Nevertheless, our findings strongly suggest that, if the guidelines are applied, most malignant pelvic masses will be referred to gynecologic oncologists, whereas some benign masses will also be referred. In the absence of a perfect screening test that distinguishes women with benign masses from those with cancers, the SGO and ACOG have constructed the guidelines to allow the referral of a minority of patients with benign pelvic masses to insure referral of the majority of patients with malignant pelvic masses. With future refinement, these guidelines may reduce the referral of women with benign masses without compromising care to women with cancers. But for today's practitioner, the SGO and ACOG referral guidelines perform far better than the former method of guessing.

	Footnotes

 
The authors thank the individuals who were instrumental in completing this project: Drs. J. Max Austin Jr, Ronald Alvarez, and Rod Rocconi from University of Alabama, Birmingham, Alabama; Dr. Karl Podratz from Mayo Clinic, Rochester, Minnesota; Dr. Barbara Goff from University of Washington, Seattle, Washington; Dr. David Mutch from Washington University, St. Louis, Missouri; Dr. Pamela Edillon from University of California, Irvine, Orange, California; Dr. Jessy Joykutty from Baylor University Medical Center, Dallas, Texas; and Dr. Wendi Carns from University of Arizona, Tucson, Arizona.

Reprints are not available. Address correspondence to: Michael L. Berman, MD, The Chao Family Comprehensive Cancer Center, Division of Gynecologic Oncology, Irvine Medical Center, University of California, 101 The City Drive, Building 23, Room 107, Orange, CA 92668; e-mail: mberman{at}uci.edu.

Received May 18, 2004. Received in revised form September 17, 2004. Accepted October 7, 2004.

doi:10.1097/01.AOG.0000149159.69560.ef

	REFERENCES

TOP ABSTRACT SGO and ACOG Referral... MATERIALS AND METHODS RESULTS DISCUSSION Modified Referral Criteria Based... REFERENCES

 
1. Griffiths CT. Surgical resection of tumor bulk in the primary treatment of ovarian carcinoma. Natl Cancer Inst Monogr 1975;42:101–4.

2. Piver MS, Lele SB, Marchetti DL, Baker TR, Tsukada Y, Emrich LJ. The impact of aggressive debulking surgery and cisplatin-based chemotherapy on progression-free survival in stage III and IV ovarian carcinoma. J Clin Oncol 1988;6:983–9.[Abstract/Free Full Text]

3. Bertelsen K. Tumor reduction surgery and long-term survival in advanced ovarian cancer: a DACOVA study. Gynecol Oncol 1990;38:203–9.[Medline]

4. Hoskins WJ, Bundy BN, Thigpen JT, Omura GA. The influence of cytoreductive surgery on recurrence-free interval and survival in small-volume stage III epithelial ovarian cancer: a Gynecologic Oncology Group study. Gynecol Oncol 1992;47:159–66.[Medline]

5. Chen SS, Bochner R. Assessment of morbidity and mortality in primary cytoreductive surgery for advanced ovarian carcinoma. Gynecol Oncol 1985;20:190–5.[Medline]

6. Eisenkop SM, Spirtos NM, Montag TW, Nalick RH, Wang HJ. The impact of subspecialty training on the management of advanced ovarian cancer. Gynecol Oncol 1992;47:203–9.[Medline]

7. Mayer AR, Chambers SK, Graves E, Holm C, Tseng PC, Nelson BE, Schwartz PE. Ovarian cancer staging: does it require a gynecologic oncologist? Gynecol Oncol 1992;47:223–7.[Medline]

8. Nguyen HN, Averette HE, Hoskins W, Penalver M, Sevin BU, Steren A. National survey of ovarian carcinoma. Part V. The impact of physician's specialty on patients’ survival. Cancer 1993;72:3663–70.[Medline]

9. Kehoe S, Powell J, Wilson S, Woodman C. The influence of the operating surgeon's specialisation on patient survival in ovarian carcinoma. Br J Cancer 1994;70:1014–7.[Medline]

10. Woodman C, Baghdady A, Collins S, Clyma JA. What changes in the organisation of cancer services will improve the outcome for women with ovarian cancer? Br J Obstet Gynaecol 1997;104:135–9.[Medline]

11. Junor EJ, Hole DJ, McNulty L, Mason M, Young J. Specialist gynaecologists and survival outcome in ovarian cancer: a Scottish national study of 1866 patients. Br J Obstet Gynaecol 1999;106:1130–6.[Medline]

12. Bristow RE, Tomacruz RS, Armstrong DK, Trimble EL, Montz FJ. Survival effect of maximal cytoreductive surgery for advanced ovarian carcinoma during the platinum era: a meta-analysis. J Clin Oncol 2002;20:1248–59.[Abstract/Free Full Text]

13. ACOG Committee Opinion: number 280, December 2002. The role of the generalist obstetrician-gynecologist in the early detection of ovarian cancer. Obstet Gynecol 2002;100:1413–6.[Medline]

This article has been cited by other articles:

				A. C. Dearking, G. D. Aletti, M. E. McGree, A. L. Weaver, M.-K. Sommerfield, and W. A. Cliby How Relevant Are ACOG and SGO Guidelines for Referral of Adnexal Mass? Obstet. Gynecol., October 1, 2007; 110(4): 841 - 848. [Abstract] [Full Text] [PDF]

				A. N. Fader and P. G. Rose Role of Surgery in Ovarian Carcinoma J. Clin. Oncol., July 10, 2007; 25(20): 2873 - 2883. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Im, S. S. Articles by Berman, M. L. Search for Related Content PubMed PubMed Citation Articles by Im, S. S. Articles by Berman, M. L. Related Collections Gynecologic oncology Gynecologic surgery