- Email: [email protected]
Guidelines to determine the route of oophorectomy with hysterectomy
Guidelines to determine the route of oophorectomy with hysterectomy S. Robert Kovac, MD, and Stephen H. Cruikshank, MD Dayton, Ohio, and St. Louis, Missouri OBJECTIVES: Our purpose was to determine whether there is adequate visibility and access for transvaginal oophorectomy in most patients and the success rate of the transvaginal approach. The final goal was to establish objective guidelines for choosing the route of oophorectomy with hysterectomy. STUDY DESIGN: Patients underwent laparoscopy-assisted vaginal hysterectomy (n = 91) or vaginal hysterectomy (n = 875). Ovarian removal, either unilateral (n = 97) or bilateral (n = 187), was carried out for clinical or prophylactic reasons. The accessibility of the ovaries for transvaginal removal was assessed by stretching the infundibulopelvic ligament and grading the position of the ovaries from 0 (no descent) to III (descent past the hymenal ring with traction). RESULTS- In 158 patients transvaginal bilateral oophorectomy was performed without laparoscopic assistance. In another 29 patients bilateral transvaginal oophorectomy was performed with laparoscopyassisted vaginal hysterectomy, and prophylactic bilateral oophorectomy by the transvaginal route was successful in all but 1 of 143 patients with ovaries of grade I or higher. In 20 patients laparoscopic lysis of adhesions was necessary to permit transvaginal oophorectomy. Ninety-seven patients underwent transvaginal unilateral oophorectomy, 74 with conventional vaginal hysterectomy and 23 with laparoscopy-assisted vaginal hysterectomy. Among the patients not having oophorectomy, all ovaries had sufficient mobility to have been removed transvaginally. CONCLUSION: Good surgical practice dictates that visibility and accessibility be the primary criteria for selecting the route of oophorectomy with hysterectomy. In most patients the ovaries are visible and accessible to transvaginal removal. (Am J Obstet Gynecol 1996;175:1483-8.)
Key words" Endometriosis, ovariectomy, peritoneoscopy, surgery, gynecologic
According to the National Center for Health Statistics, 1 the frequency with which bilateral oophorectomy accompanies hysterectomy has increased from 36% for the years 1965 through 1984 to 50% for the years 1988 through 1990. Thirty-seven percent of women under the age of 45 who undergo hysterectomy now have bilateral oophorectomy, and for women over 45 years of age, the figure increases to 68%. For all age groups, bilateral oophorectomy is significantly more likely during abdominal than vaginal surgery, a The choice of vaginal vs. abdominal hysterectomy appears d e p e n d e n t on both the preoperative diagnosis and the physician's training. 2 However, guidelines have been established to determine the appropriateness of the vaginal vs. the laparoscopy-assisted and abdominal techniques s and should eventually increase the number of From the Division of Pelvic Reconstructive Surgery and Umgynecology, Department of Obstetrics and Gynecology, Wright State University School of Medicine, and St. John's Mercy Medical Center. Presented at the Twenty-second Annual Meeting of the Society of Gynecologic Surgeons, Albuquerque, New Mexico, March 4-6, 1996. Repent requests: S. Robert Kovac, MD, Department of Obstetrics and Gynecology, Wright State University School of Medicine, 128E. Apple Street, Suite 3800 CHE, Dayton, OH 45409-2793 Copyright © 1996 by Mosby-Year Book, Inc. 0002-9378/96 $5.00 + 0 6/6/76817
salpingo-oophorectomies as well as hysterectomies perf o r m e d transvaginally. Certainly, the indications for ovarian removal should be similar regardless of whether an abdominal, laparoscopic, or vaginal hysterectomy is being performed. Also, as Thompson 4 stated, "If the ovaries are accessible, the same guidelines for prophylactic oophorectomy should apply to the menopausal and postmenopausal patient at the time of vaginal hysterectomy." There appears to be some reluctance to combine vaginal hysterectomy with oophorectomy, because this is thought to be a risky and difficult procedure. Two factors seem to foster this perception: (1) fear of restricted access to the ovaries and (2) inadequate visibility of the adnexa during conventional vaginal surgery. To address these concerns, guidelines to determine the surgical route for oophorectomy are needed. The objectives of this study were as follows: (1) to determine whether there is adequate visibility and access for transvaginal oophorectomy in most patients, (2) to ascertain the number of oophorectomies performed by vaginal and laparoscopy-assisted vaginal techniques, (3) to determine the success rate of the transvaginal approach as an intended operation for prophylactic ovarian removal, and (4) to calculate the percentage of ovaries that could be removed transvaginally. The final 1483
1484
Kovac and Cruikshank
December 1996 AmJ Obstet Gynecol
Table I. Indications for hysterectomy and type of procedure in 966 patients
Indication
Lapmvscopy-assisted vaginal hysterectomy
Vaginal
Total
2 42 33 0 0 0 0 0 0 12 2
262 0 0 112 120 24 244 91 13 0 7
264 42 33 112 120 24 244 91 13 12 9
0 91
2 875
2 966
Leiomyoma Adnexal mass Endometriosis Adenomyosis Abnormal bleeding Carcinoma in situ of cervix Pelvic relaxation Urinary incontinence Atypical hyperplasia Chronic pelvic pain Adenocarcinoma of endometrium (stage IA, grade 1) Family history of ovarian cancer TOTAL
goal was to establish objective guidelines for choosing the route of oophorectomy. "
Material and methods Patients. From January 1989 to Dec. 31, 1994, 966 patients aged 24 to 84 (mean 42) years were assigned to laparoscopy-assisted vaginal hysterectomy or vaginal hysterectomy on the basis of previously described guidelines (Table I).S Eight h u n d r e d seven procedures were performed at St.John's Mercy Medical Center, St. Louis, and 159 were done at Wright State University, Dayton, Ohio. Ovarian removal, either unilateral (n = 97) or bilateral (n = 187), was carried out.for clinical or prophylactic reasons. The clinical indications, as d e t e r m i n e d preoperatively or intraoperatively, were diseases of the ovaries m a n d a t i n g either unilateral or bilateral oophorectomy. Prophylactic oophorectomy was considered for all women aged ->50 years and for women <50 years old who were d e t e r m i n e d to be menopausal either clinically o r by laboratory testing. Prophylactic oophorectomy also was carried out if there was a strong familial history of ovarian cancer or a familial history of breast or colon cancer or of Peutz-Jeghers syndrome. Informed written consent for indicated or prophylactic oophorectomy was obtained preoperatively from all patients. Senior resident physicians performed 99% of all operative procedures. Eight h u n d r e d seventy-five patients (90.6%) underwent standard vaginal hysterectomy as described, 4'5 whereas 91 (9.4%) u n d e r w e n t laparoscopy-assisted vaginal hysterectomy as previously reported. 6 After the uterus had been removed, the accessibility of the ovaries for transvaginal removal was assessed by stretching the infundibulopelvic ligament and grading the position of the ovaries in relation to the long axis of the vagina. For those patients assigned to laparoscopyassisted vaginal hysterectomy, the accessibility of the
ovaries was determined at laparoscopy by a scoring system described previously.7 The need for and extent of laparoscopic surgery for transvaginal adnexal removal was determined by the mobility of the adnexa. When oophorectomy was not indicated, the same technique of mobilizing and grading ovarian mobility and accessibility was carried out by placing a clamp across the infundibulopelvic ligament or mesovarium without the actual clamping. The degree of ovarian descent was graded by means of a systems previously used to classify pelvic organ prolapse. It was modified in an attempt to define more clearly the probability of ovarian removal by the vaginal route. The grade corresponded to the m i n i m u m degree of descent of either ovary (Fig. 1) : Grade 0, No descent. Infundibulopelvic ligament has little or no stretchability. Ovaries are positioned at the lateral pelvic wall at or above the ischial spines and cannot be brought with traction into the long axis of the vagina. Grade I, Infundibulopelvic ligament stretchability brings the descent of the ovaries into a long-axis plane of the vagina with traction halfway between the ischial spines and midvagina. Grade II, Infundibulopelvic ligament stretchability brings the ovaries into the long axis of the vagina with traction between the midportion of the vagina and the hymenal ring. Grade III, Infundibulopelvic stretchability brings the ovaries into the longitudinal plane of the vagina with traction past the hymenal ring. To determine what grade would be acceptable for transvaginal oophorectomy, the experience of other surgical specialties was considered. The distance from the hymenal ring to the ischial spine is approximately 8 cm. In dentistry the distance from the front teeth to the last molars is 6 cm, and in otolaryngology the distance from the front teeth to the tonsils for tonsillectomy is 10 cm. Therefore we postulated that any ovary that was grade I or higher should be visible and accessible for transvaginal removal by most gynecologic surgeons.
Volume 175, Number 6 AmJ Obstet Gynecol
Surgical technique. The uterus was removed before removal of a n y adnexal structure. The mobility and position of the ovary with traction were graded as described. Traction was used in a down-and-outward fashion by a ligature previously placed around the uteroovarian and r o u n d ligaments and fallopian tube. The fimbriated end of the fallopian tube was folded back toward the uteroovarian ligament and r o u n d ligaments. A curved clamp was placed across a broad end of the infundibulopelvic ligament between the r o u n d ligament and fallopian tube in Arey's space for a salpingo-oophorectomy or across the mesovarium for an oophorectomy. A suture ligature was placed around the end of the tip of the clamp, and a single throw was used to transfix the ovarian vessels. The suture was then placed around the clamp, transfixed to the mesovarium, and tied. A metal clip was placed at the apex of the ligature to mark the upper margin of the surgical dissection. This clip marks the relation between the ligature and the ureter if this knowledge is desired postoperatively. Results The 284 patients (29.4%) undergoing unilateral or bilateral oophorectomy for any indication included 52 (57.1%) of the 91 patients having laparoscopy-assisted vaginal hysterectomy. These patients were selected for laparoscopy-assisted vaginal hysterectomy because the preoperative evaluation suggested serious disease that might contraindicate the vaginal route. Twenty of these women (38.5%) required laparoscopic techniques to reduce the contraindications to transvaginal hysterectomy and oophorectomy. Of the 42 women assigned to laparoscopic evaluation because of benign ovarian disease, 29 (69%) had documented intrinsic disease of the ovary; 15 of them had bilateral and 14 unilateral ovarian removal. The other 13 (31%) had extrinsic lesions, such as adhesions or paratubal cysts, suggestive of ovarian abnormality indicating unilateral oophorectomy in 9 cases. Six (14%) required laparoscopic surgery to improve the mobility of the ovaries for transvaginal removal. Moderate mobility with moderate adhesions was found in 5 patients and poor mobility with severe adhesions in 1 patient. At transvaginal unilateral or bilateral oophorectomy in these 38 women assigned to laparoscopy-assisted vaginal hysterectomy the ovaries were j u d g e d grade II. Of the 33 patients assigned to laparoscopic evaluation because of a clinical suspicion or history of endometriosis, 16 (48%) had documented disease stage II, 5; stage III, 4; stage IV, 7). Laparoscopic techniques were required in 14 of them for ablation of endometriosis that was inaccessible by the vaginal route. In addition, laparoscopic lysis of adhesions was required to increase the mobility of the ovaries for transvaginal removal. Laparoscopic surgery was required in only 3 of.the 5
Kovac and Cruikshank
1485
Is)hial5pine I~-,ii2 id Vagina
A J
Hymenal IRin9 l
Fig. 1. Grading system according to minimum descent of either ovary. patients with stage II endometriosis. These three patients had moderate degrees of mobility and adhesions; however, at vaginal oophorectomy the ovaries were grade II. All four patients with stage III endometriosis required laparoscopic surgery to reduce the moderate adhesions with moderate mobility. At vaginal hysterectomy these ovaries were grade II. Seven patients with stage IV endometriosis had severe adhesions, whereas 4 also had poor mobility and 3 had moderate mobility. At vaginal surgery 6 o f the patients had grade II ovaries, and the other one had grade I ovaries. Had laparoscopic surgery not been performed on these 20 patients, the grade of ovarian descent at the time of vaginal hysterectOmy would have been between 0 and I. Thus laparoscopic techniques enabled transvaginal oophorectomy in these 20 patients by improving the mobility of the ovaries through lysis of adhesions, Prophylactic bilateral oophorectomy with vaginal hysterectomy was successful in all but one of the 143 patients whose ovaries were grade I or higher. The ages of these patients and the grades of the ovaries are seen in Table II. One hundred nineteen (83%) were assigned to grade II, whereas 15 (10.5%) were grade I, 8 (5.6%) were grade III, and 1 (0.7%) was grade 0. In 44 patients the disease was such that intraoperative decisions were made to remove the ovaries bilaterally. Most of these patients (n = 39, 89%) had grade II ovaries, with 5 (11%) being grade III. Ninety-seven patients underwent unilateral oophorectomy, 74 (76.3%) with conventional vaginal hysterectomy (Table III). Fifty-four of these ovaries (73%) were grade II, 17 (22.9%) were grade III, and 2 (2.7%) were grade I. One patient, aged 67, had grade 0 ovaries, and unilateral oophorectomy was performed. The left ovary was inac-
1486
Kovac and Cruikshank
December 1996 Am J Obstet Gynecol
T a b l e II. G r a d e o f ovaries in relation to p a t h o l o g i c diagnosis a n d p a t i e n t ' s age in w o m e n u n d e r g o i n g prophylactic bilamral o o p h o r e c t o m y with vaginal h y s t e r e c t o m y (n = 143) 30-39 yr
50-59 yr
60-69 yr
70-79 yr
80-84 yr
0 2 0
4 63 6
1" 3 41 2
5 13 0
3 0 0
1
1
• Total series Grade 0 Grade I Grade II Grade III Carcinoma in situ Grade II Bleeding Grade 0 Grade I Grade II Fibroids Grade I Grade II Grade III Hyperplasia Grade II Endometrial adenocarcinoma Grade I Grade II Ovarian carcinomat Grade II Incontinence Grade I Grade II Prolapse Grade I Grade II Grade I II
1
2 22
1 1" 1 3
2 11 6
2 6
5
8
3 2
7
9
17
11 2
*Unilateral oophorectomy. tBy family history.
T a b l e III. Ovarian g r a d e in w o m e n having unilateral
T a b l e IV. G r a d e o f ovaries in relation to p a t h o l o g i c
o o p h o r e c t o m y with vaginal hysterectomy (n = 74)
diagnosis a n d p a t i e n t ' s age in w o m e n n o t u n d e r g o i n g
Grade
0
24-29 yr
30-39 yr
40-49 yr
60-69 yr
0
0
0
1"
I
0
0
2
0
II III TOTAL
3 1 4
35 7 42
16 9 27
0 0 1
*One ovary in this patient was inaccessible; see text for discussion.
cessible to the vaginal r o u t e even t h o u g h it was visible with l o n g Briesky-Navratil retractors. This ovary was atrop h i c a n d a d h e r e n t to the lateral pelvic wall. Because o f the a p p e a r a n c e o f this ovary a decision was m a d e to leave it in situ. H a d t h e r e b e e n a significant familial history that m a n d a t e d its removal, laparoscopic o o p h o r e c t o m y w o u l d have b e e n p e r f o r m e d after c o m p l e t i o n o f the vaginal hysterectomy. T h e r e m a i n i n g 23 (23.7%) w o m e n u n d e r g o i n g unilateral o o p h o r e c t o m y with laparoscopyassisted vaginal h y s t e r e c t o m y h a d g r a d e II ovaries as f o u n d d u r i n g transvaginal removal. T h e patients' ages a n d the grades o f the ovaries that were n o t r e m o v e d are s e e n in Table IV. Six (0.8%) were
o o p h o r e c t o m y with vaginal h y s t e r e c t o m y (n = 688)
Total series Grade I Grade II Grade III Carcinoma in situ Grade II Bleeding Grade I Grade I! Grade I I I Fibroids Grade I Grade II Grade III Adenomyosis Grade I Grade II Grade III Incontinence Grade II Grade III Prolapse Grade II Grade III
24-29 yr
30-39 yr
40-49 yr
0 16 7
0 285 14
6 354 6
3
17
4
24
5 4
101 7
2 103 3
4 1
80
2 22
2
2 43 1
33 3
26
30 4
160 2
Kovac and Cruikshank
Volume 175, Number 6 AmJ Obstet Gynecol
grade I, 655 (95.2%) were grade II, and 27 (3.9%) were grade III. There were no complications associated with either unilateral or bilateral transvaginal oophorectomy. Comment
Routine use of the laparoscope to perform an oophorectomy or salpingo-oophorectomy is being heralded as commonplace. In addition, "to be sure we can get the ovaries" is a phrase too often heard worldwide to justify the use of the abdominal or laparoscopic hysterectomy. Is this attitude appropriate? As shown in this study the visibility and accessibility provided by vaginal hysterectomy make it possible to grade the position of the ovaries accurately and to determine whether they can be removed transvaginally. This is the first attempt in the literature to provide definitive evidence that most ovaries are not only visible but also accessible for removal at simple vaginal hysterectomy. Wilcox et al. 9 found that 49.6% of hysterectomies included oophorectomy for all age groups. Interestingly, only 10.3% of the same patients who underwent vaginal hysterectomy also had oophorectomy in the United States during a 9-year period. In the current study of 966 patients the vaginal oophorectomy rate was 29.4%. Moreover, according to the guidelines presented, 99.9% of all ovaries were or could have been removed transvaginally for disease or prophylactic reasons. Among the 875 patients who underwent vaginal hysterectomy without laparoscopic assistance, the ovaries were grade II in 813 (92.9%) and grade III in 40 (4.6%). These figures correlate well with recent demonstrations that planned vaginal oophorectomy is successful in 94% 1° to 97% 3 of cases. However, these results conflict with the frequently stated "presumptive evidence" that the ovaries are not accessible for vaginal removal. In fact, most ovaries are accessible; they are not "high" in the pelvis. Therefore, before selecting other techniques for hysterectomy and oophorectomy that may have greater morbidity, surgeons should document the inability to remove the ovary transvaginally. A particular consideration is the opportunity to reduce urologic morbidity. The ureters lie close to the infundibulopelvic ligaments, and the reported incidence of ureteral injury, with its potentially serious morbidity, during major gynecologic surgery ranges from 0.3% to 7%. 11 Ureteral injury is most common with abdominal procedures; however, as the use of operative laparoscopy has increased, transection, thermal injury, and stapling damage all have been inflicted on the uretersJ 1, 12 The ureter has a mechanism to protect it during vaginal hysterectomy, 1~ and although the activity of this protective mechanism during transvaginal oophorectomy has not been documented, an extensive search of the literature revealed no reports of ureteral injury during trans-
1487
vaginal oophorectomy. The same cannot be said of the abdominal and laparoscopic techniques) 2' ~4 The laparoscopic scoring system previously described 7 has been valuable in determining not only which patients actually require laparoscopic surgery but also to what extent. Lysis of adhesions was the only type of laparoscopic surgery performed on the ovaries in this series. Although laparoscopic oophorectomy could have been carried out in these cases, there is no evidence that this technique is safer than transvaginal operation. As gynecologic surgeons and resident teachers, it is our responsibility, in the interests of our patients, to base our surgical decisions on factual evidence. For too many years "presumptive evidence" has guided the performance and selection of the route for hysterectomy and oophorectomy in most gynecologic conditions. "Presumptive evidence" of a lack of visibility and accessibility for transvaginal oophorectomy has been accepted without the application of systematic guidelines to determine the most appropriate surgical route. The use of such guidelines may have a dramatic impact on surgical practice. For example, after the recent introduction of guidelines for the route of hysterectomy in the United States, s France, 15 and England, 16 the vaginal route without laparoscopic or abdominal assistance has proved successful in 89%, 77%, and 95% of women, respectively. These figures differ substantially from the national practice in the absence of guidelines, in which approximately 75% of patients undergo abdominal hysterectomy. Good surgical practice dictates that visibility and accessibility be the primary criteria for selecting the route of oophorectomy. The evidence presented here demonstrates that the ovaries are visible and accessible to transvaginal removal in most patients. Application of the guidelines described could reverse past and current trends toward selecting the abdominal or laparoscopic technique. This change in practice would have farreaching effects on patient morbidity and medical costs. We thank Steven Conover, MD, for his medical illustration.
REFERENCES
1. Dicker RC, GreenspanJR, Strauss LT, Cowart MR, ScallyMJ, Peterson HB, et al. Hysterectomy among women of reproductive age: trends in the United States, 1970-1978.JAMA 1982;248:323-7. 2. Kovac SR, Christie SJ, Bindbeutel GA. Abdominal versus vaginal hysterectomy: a statistical model for determining physician decision making and patient outcome. J Med Decision Making 1991;11:19-28. 3. Kovac SR. Guidelines to determine the route of hysterectomy. Obstet Gynecol 1995;85:18-23. 4. Thompson JD. Hysterectomy. In: Thompson JD, Rock JA, editors. TeLinde's operative gynecology. Philadelphia: JB Lippincott, 1992:663-732. 5. Nichols DH, Randall CL. Vaginal surgery. 2nd ed. Baltimore: Williams & Wilkins, 1983:548-60.
1488
Kovac and Cruikshank
6. Kovac SR. tntramyometrial coring as an adjunct to vaginal hysterectomy. Obstet Gynecol 1986;67:161-6. 7. Kovac SR, Cruikshank SH, Retto HF. Laparoscopy-assisted vaginal hysterectomy. J Gynecol Surg 1990;6:185-93. 8. American College of Obstetricians and Gynecologists Committee on Technical Bulletins. Pelvic organ prolapse. Washington (DC):The College, 1995:1-8.ACOGTechnical Bulletin No.: 214. 9. Wilcox LS, Koonin LM, Pokras R, Strauss LT, Xia Z, Peterson HB. Hysterectomy in the United States, 1988-1990. Obstet Gynecol 1994;83:549-55. 1O. Sheth SS. The place of oophorectomy at vaginal hysterectomy. BrJ Obstet Gynaecol 1991;98:662-6. 11. Mann WJ, Ko0nings PP. Ureteral injuries in gynecologic surgery. Int Urogynecol J 1993;4:361-5. 12. Woodland MB. Ureter injury during laparoscopy-assisted vaginal hysterectomy with the endoscopic linear stapler. AmJ Obstet Gynecol 1992;167:756-7. 13. Cruikshank SH, Kovac SR. Role of the uterosacral-cardinal ligament complex in protecting the ureter during vaginal hysterectomy. IntJ Gynecol Obstet 1993;40:141-4. 14. Laparoscopic injuries raise caution flags with doctors, insurers. Adv Technol Surgi Care 1995 Mar:29-33. 15. Querleu D, Cosson M, Parmentier D, DeBondinance P. The impact of laparoscopic surgery on vaginal hysterectomy. Gynecol Endosc 1993;2:89-91. 16. Richardson RE, Bournas N, Magos AL. Is laparoscopic hysterectomy a waste of time? Lancet 1995;345:36-41.
Discussion DR. JoHN KNAUS, Evanston, Illinois. This report and others document the high success rate of bilateral salpingo-oophorectomy when performed with vaginal hysterectomy. Nevertheless, only as few as 2% of gynecologists routinely perform salpingo-oophorectomy with vaginal hysterectomy. The grading system proposed by the authors may or may not prove practical. First, many variables affect the consistency of such a grading system (endometriosis, adhesions, prior surgery, the option of laparoscopy). Second, I do not think there is a proficient vaginal surgeon who feels intimidated by removing any adnexa when doing a vaginal hysterectomy, stretch or no stretch.
December 1996 AmJ Obstet Gynecol
The value of this report is that it may encourage capable gynecologists who currently do not remove adnexa vaginally to do so. These persons may progress from a more careful inspection of the adnexa to mobilizing them to removing them. In the end salpingo-oophorectomy performed with vaginal hysterectomy is not for the purpose of surgical exercise or to determine a success rate. It is to prevent some women from needing an operation in the future for benign disease or, more important, potentially to reduce ovarian cancer deaths. DR. KOVAC (Closing). I thank Dr. Knaus for his thoughtful comments. We do not include the round ligament in the clamp that is placed across the infundibulopelvic ligament. We place our clamp between the round ligament and the infundibulopelvic ligament in Arey's space, as previously described by Dr. Kermit Krantz. The strategy that was used to increase the numbers of transvaginal oophorectomies was simple. Quite frankly, most ovaries are accessible and visible for transvaginal removal. We were surprised, when we began to record the grades of each ovary, how many of them fell in the grade II to III range. Ninety-five to ninety-seven percent of the ovaries in this series were grade II or grade III. Therefore it was not difficult for residents to place a clamp across the infundibulopelvic ligament on ovaries that were so accessible. I think that the ability to successfully remove ovaries transvaginally in this study dispels the long-standing belief that the ovaries are difficult or impossible to remove transvaginally. Grading the accessibility of the ovaries may be just as important as grading pelvic organ prolapse, especially if this affects the surgical route of hysterectomy. Surgeons who select potentially more morbid and costly approaches to hysterectomy (abdominal and laparoscopic) solely on the basis of a need to remove the ovaries without any other contraindications to the vaginal approach may need to rethink their surgical decisionmaking process.
Key words" Endometriosis, ovariectomy, peritoneoscopy, surgery, gynecologic
According to the National Center for Health Statistics, 1 the frequency with which bilateral oophorectomy accompanies hysterectomy has increased from 36% for the years 1965 through 1984 to 50% for the years 1988 through 1990. Thirty-seven percent of women under the age of 45 who undergo hysterectomy now have bilateral oophorectomy, and for women over 45 years of age, the figure increases to 68%. For all age groups, bilateral oophorectomy is significantly more likely during abdominal than vaginal surgery, a The choice of vaginal vs. abdominal hysterectomy appears d e p e n d e n t on both the preoperative diagnosis and the physician's training. 2 However, guidelines have been established to determine the appropriateness of the vaginal vs. the laparoscopy-assisted and abdominal techniques s and should eventually increase the number of From the Division of Pelvic Reconstructive Surgery and Umgynecology, Department of Obstetrics and Gynecology, Wright State University School of Medicine, and St. John's Mercy Medical Center. Presented at the Twenty-second Annual Meeting of the Society of Gynecologic Surgeons, Albuquerque, New Mexico, March 4-6, 1996. Repent requests: S. Robert Kovac, MD, Department of Obstetrics and Gynecology, Wright State University School of Medicine, 128E. Apple Street, Suite 3800 CHE, Dayton, OH 45409-2793 Copyright © 1996 by Mosby-Year Book, Inc. 0002-9378/96 $5.00 + 0 6/6/76817
salpingo-oophorectomies as well as hysterectomies perf o r m e d transvaginally. Certainly, the indications for ovarian removal should be similar regardless of whether an abdominal, laparoscopic, or vaginal hysterectomy is being performed. Also, as Thompson 4 stated, "If the ovaries are accessible, the same guidelines for prophylactic oophorectomy should apply to the menopausal and postmenopausal patient at the time of vaginal hysterectomy." There appears to be some reluctance to combine vaginal hysterectomy with oophorectomy, because this is thought to be a risky and difficult procedure. Two factors seem to foster this perception: (1) fear of restricted access to the ovaries and (2) inadequate visibility of the adnexa during conventional vaginal surgery. To address these concerns, guidelines to determine the surgical route for oophorectomy are needed. The objectives of this study were as follows: (1) to determine whether there is adequate visibility and access for transvaginal oophorectomy in most patients, (2) to ascertain the number of oophorectomies performed by vaginal and laparoscopy-assisted vaginal techniques, (3) to determine the success rate of the transvaginal approach as an intended operation for prophylactic ovarian removal, and (4) to calculate the percentage of ovaries that could be removed transvaginally. The final 1483
1484
Kovac and Cruikshank
December 1996 AmJ Obstet Gynecol
Table I. Indications for hysterectomy and type of procedure in 966 patients
Indication
Lapmvscopy-assisted vaginal hysterectomy
Vaginal
Total
2 42 33 0 0 0 0 0 0 12 2
262 0 0 112 120 24 244 91 13 0 7
264 42 33 112 120 24 244 91 13 12 9
0 91
2 875
2 966
Leiomyoma Adnexal mass Endometriosis Adenomyosis Abnormal bleeding Carcinoma in situ of cervix Pelvic relaxation Urinary incontinence Atypical hyperplasia Chronic pelvic pain Adenocarcinoma of endometrium (stage IA, grade 1) Family history of ovarian cancer TOTAL
goal was to establish objective guidelines for choosing the route of oophorectomy. "
Material and methods Patients. From January 1989 to Dec. 31, 1994, 966 patients aged 24 to 84 (mean 42) years were assigned to laparoscopy-assisted vaginal hysterectomy or vaginal hysterectomy on the basis of previously described guidelines (Table I).S Eight h u n d r e d seven procedures were performed at St.John's Mercy Medical Center, St. Louis, and 159 were done at Wright State University, Dayton, Ohio. Ovarian removal, either unilateral (n = 97) or bilateral (n = 187), was carried out.for clinical or prophylactic reasons. The clinical indications, as d e t e r m i n e d preoperatively or intraoperatively, were diseases of the ovaries m a n d a t i n g either unilateral or bilateral oophorectomy. Prophylactic oophorectomy was considered for all women aged ->50 years and for women <50 years old who were d e t e r m i n e d to be menopausal either clinically o r by laboratory testing. Prophylactic oophorectomy also was carried out if there was a strong familial history of ovarian cancer or a familial history of breast or colon cancer or of Peutz-Jeghers syndrome. Informed written consent for indicated or prophylactic oophorectomy was obtained preoperatively from all patients. Senior resident physicians performed 99% of all operative procedures. Eight h u n d r e d seventy-five patients (90.6%) underwent standard vaginal hysterectomy as described, 4'5 whereas 91 (9.4%) u n d e r w e n t laparoscopy-assisted vaginal hysterectomy as previously reported. 6 After the uterus had been removed, the accessibility of the ovaries for transvaginal removal was assessed by stretching the infundibulopelvic ligament and grading the position of the ovaries in relation to the long axis of the vagina. For those patients assigned to laparoscopyassisted vaginal hysterectomy, the accessibility of the
ovaries was determined at laparoscopy by a scoring system described previously.7 The need for and extent of laparoscopic surgery for transvaginal adnexal removal was determined by the mobility of the adnexa. When oophorectomy was not indicated, the same technique of mobilizing and grading ovarian mobility and accessibility was carried out by placing a clamp across the infundibulopelvic ligament or mesovarium without the actual clamping. The degree of ovarian descent was graded by means of a systems previously used to classify pelvic organ prolapse. It was modified in an attempt to define more clearly the probability of ovarian removal by the vaginal route. The grade corresponded to the m i n i m u m degree of descent of either ovary (Fig. 1) : Grade 0, No descent. Infundibulopelvic ligament has little or no stretchability. Ovaries are positioned at the lateral pelvic wall at or above the ischial spines and cannot be brought with traction into the long axis of the vagina. Grade I, Infundibulopelvic ligament stretchability brings the descent of the ovaries into a long-axis plane of the vagina with traction halfway between the ischial spines and midvagina. Grade II, Infundibulopelvic ligament stretchability brings the ovaries into the long axis of the vagina with traction between the midportion of the vagina and the hymenal ring. Grade III, Infundibulopelvic stretchability brings the ovaries into the longitudinal plane of the vagina with traction past the hymenal ring. To determine what grade would be acceptable for transvaginal oophorectomy, the experience of other surgical specialties was considered. The distance from the hymenal ring to the ischial spine is approximately 8 cm. In dentistry the distance from the front teeth to the last molars is 6 cm, and in otolaryngology the distance from the front teeth to the tonsils for tonsillectomy is 10 cm. Therefore we postulated that any ovary that was grade I or higher should be visible and accessible for transvaginal removal by most gynecologic surgeons.
Volume 175, Number 6 AmJ Obstet Gynecol
Surgical technique. The uterus was removed before removal of a n y adnexal structure. The mobility and position of the ovary with traction were graded as described. Traction was used in a down-and-outward fashion by a ligature previously placed around the uteroovarian and r o u n d ligaments and fallopian tube. The fimbriated end of the fallopian tube was folded back toward the uteroovarian ligament and r o u n d ligaments. A curved clamp was placed across a broad end of the infundibulopelvic ligament between the r o u n d ligament and fallopian tube in Arey's space for a salpingo-oophorectomy or across the mesovarium for an oophorectomy. A suture ligature was placed around the end of the tip of the clamp, and a single throw was used to transfix the ovarian vessels. The suture was then placed around the clamp, transfixed to the mesovarium, and tied. A metal clip was placed at the apex of the ligature to mark the upper margin of the surgical dissection. This clip marks the relation between the ligature and the ureter if this knowledge is desired postoperatively. Results The 284 patients (29.4%) undergoing unilateral or bilateral oophorectomy for any indication included 52 (57.1%) of the 91 patients having laparoscopy-assisted vaginal hysterectomy. These patients were selected for laparoscopy-assisted vaginal hysterectomy because the preoperative evaluation suggested serious disease that might contraindicate the vaginal route. Twenty of these women (38.5%) required laparoscopic techniques to reduce the contraindications to transvaginal hysterectomy and oophorectomy. Of the 42 women assigned to laparoscopic evaluation because of benign ovarian disease, 29 (69%) had documented intrinsic disease of the ovary; 15 of them had bilateral and 14 unilateral ovarian removal. The other 13 (31%) had extrinsic lesions, such as adhesions or paratubal cysts, suggestive of ovarian abnormality indicating unilateral oophorectomy in 9 cases. Six (14%) required laparoscopic surgery to improve the mobility of the ovaries for transvaginal removal. Moderate mobility with moderate adhesions was found in 5 patients and poor mobility with severe adhesions in 1 patient. At transvaginal unilateral or bilateral oophorectomy in these 38 women assigned to laparoscopy-assisted vaginal hysterectomy the ovaries were j u d g e d grade II. Of the 33 patients assigned to laparoscopic evaluation because of a clinical suspicion or history of endometriosis, 16 (48%) had documented disease stage II, 5; stage III, 4; stage IV, 7). Laparoscopic techniques were required in 14 of them for ablation of endometriosis that was inaccessible by the vaginal route. In addition, laparoscopic lysis of adhesions was required to increase the mobility of the ovaries for transvaginal removal. Laparoscopic surgery was required in only 3 of.the 5
Kovac and Cruikshank
1485
Is)hial5pine I~-,ii2 id Vagina
A J
Hymenal IRin9 l
Fig. 1. Grading system according to minimum descent of either ovary. patients with stage II endometriosis. These three patients had moderate degrees of mobility and adhesions; however, at vaginal oophorectomy the ovaries were grade II. All four patients with stage III endometriosis required laparoscopic surgery to reduce the moderate adhesions with moderate mobility. At vaginal hysterectomy these ovaries were grade II. Seven patients with stage IV endometriosis had severe adhesions, whereas 4 also had poor mobility and 3 had moderate mobility. At vaginal surgery 6 o f the patients had grade II ovaries, and the other one had grade I ovaries. Had laparoscopic surgery not been performed on these 20 patients, the grade of ovarian descent at the time of vaginal hysterectOmy would have been between 0 and I. Thus laparoscopic techniques enabled transvaginal oophorectomy in these 20 patients by improving the mobility of the ovaries through lysis of adhesions, Prophylactic bilateral oophorectomy with vaginal hysterectomy was successful in all but one of the 143 patients whose ovaries were grade I or higher. The ages of these patients and the grades of the ovaries are seen in Table II. One hundred nineteen (83%) were assigned to grade II, whereas 15 (10.5%) were grade I, 8 (5.6%) were grade III, and 1 (0.7%) was grade 0. In 44 patients the disease was such that intraoperative decisions were made to remove the ovaries bilaterally. Most of these patients (n = 39, 89%) had grade II ovaries, with 5 (11%) being grade III. Ninety-seven patients underwent unilateral oophorectomy, 74 (76.3%) with conventional vaginal hysterectomy (Table III). Fifty-four of these ovaries (73%) were grade II, 17 (22.9%) were grade III, and 2 (2.7%) were grade I. One patient, aged 67, had grade 0 ovaries, and unilateral oophorectomy was performed. The left ovary was inac-
1486
Kovac and Cruikshank
December 1996 Am J Obstet Gynecol
T a b l e II. G r a d e o f ovaries in relation to p a t h o l o g i c diagnosis a n d p a t i e n t ' s age in w o m e n u n d e r g o i n g prophylactic bilamral o o p h o r e c t o m y with vaginal h y s t e r e c t o m y (n = 143) 30-39 yr
50-59 yr
60-69 yr
70-79 yr
80-84 yr
0 2 0
4 63 6
1" 3 41 2
5 13 0
3 0 0
1
1
• Total series Grade 0 Grade I Grade II Grade III Carcinoma in situ Grade II Bleeding Grade 0 Grade I Grade II Fibroids Grade I Grade II Grade III Hyperplasia Grade II Endometrial adenocarcinoma Grade I Grade II Ovarian carcinomat Grade II Incontinence Grade I Grade II Prolapse Grade I Grade II Grade I II
1
2 22
1 1" 1 3
2 11 6
2 6
5
8
3 2
7
9
17
11 2
*Unilateral oophorectomy. tBy family history.
T a b l e III. Ovarian g r a d e in w o m e n having unilateral
T a b l e IV. G r a d e o f ovaries in relation to p a t h o l o g i c
o o p h o r e c t o m y with vaginal hysterectomy (n = 74)
diagnosis a n d p a t i e n t ' s age in w o m e n n o t u n d e r g o i n g
Grade
0
24-29 yr
30-39 yr
40-49 yr
60-69 yr
0
0
0
1"
I
0
0
2
0
II III TOTAL
3 1 4
35 7 42
16 9 27
0 0 1
*One ovary in this patient was inaccessible; see text for discussion.
cessible to the vaginal r o u t e even t h o u g h it was visible with l o n g Briesky-Navratil retractors. This ovary was atrop h i c a n d a d h e r e n t to the lateral pelvic wall. Because o f the a p p e a r a n c e o f this ovary a decision was m a d e to leave it in situ. H a d t h e r e b e e n a significant familial history that m a n d a t e d its removal, laparoscopic o o p h o r e c t o m y w o u l d have b e e n p e r f o r m e d after c o m p l e t i o n o f the vaginal hysterectomy. T h e r e m a i n i n g 23 (23.7%) w o m e n u n d e r g o i n g unilateral o o p h o r e c t o m y with laparoscopyassisted vaginal h y s t e r e c t o m y h a d g r a d e II ovaries as f o u n d d u r i n g transvaginal removal. T h e patients' ages a n d the grades o f the ovaries that were n o t r e m o v e d are s e e n in Table IV. Six (0.8%) were
o o p h o r e c t o m y with vaginal h y s t e r e c t o m y (n = 688)
Total series Grade I Grade II Grade III Carcinoma in situ Grade II Bleeding Grade I Grade I! Grade I I I Fibroids Grade I Grade II Grade III Adenomyosis Grade I Grade II Grade III Incontinence Grade II Grade III Prolapse Grade II Grade III
24-29 yr
30-39 yr
40-49 yr
0 16 7
0 285 14
6 354 6
3
17
4
24
5 4
101 7
2 103 3
4 1
80
2 22
2
2 43 1
33 3
26
30 4
160 2
Kovac and Cruikshank
Volume 175, Number 6 AmJ Obstet Gynecol
grade I, 655 (95.2%) were grade II, and 27 (3.9%) were grade III. There were no complications associated with either unilateral or bilateral transvaginal oophorectomy. Comment
Routine use of the laparoscope to perform an oophorectomy or salpingo-oophorectomy is being heralded as commonplace. In addition, "to be sure we can get the ovaries" is a phrase too often heard worldwide to justify the use of the abdominal or laparoscopic hysterectomy. Is this attitude appropriate? As shown in this study the visibility and accessibility provided by vaginal hysterectomy make it possible to grade the position of the ovaries accurately and to determine whether they can be removed transvaginally. This is the first attempt in the literature to provide definitive evidence that most ovaries are not only visible but also accessible for removal at simple vaginal hysterectomy. Wilcox et al. 9 found that 49.6% of hysterectomies included oophorectomy for all age groups. Interestingly, only 10.3% of the same patients who underwent vaginal hysterectomy also had oophorectomy in the United States during a 9-year period. In the current study of 966 patients the vaginal oophorectomy rate was 29.4%. Moreover, according to the guidelines presented, 99.9% of all ovaries were or could have been removed transvaginally for disease or prophylactic reasons. Among the 875 patients who underwent vaginal hysterectomy without laparoscopic assistance, the ovaries were grade II in 813 (92.9%) and grade III in 40 (4.6%). These figures correlate well with recent demonstrations that planned vaginal oophorectomy is successful in 94% 1° to 97% 3 of cases. However, these results conflict with the frequently stated "presumptive evidence" that the ovaries are not accessible for vaginal removal. In fact, most ovaries are accessible; they are not "high" in the pelvis. Therefore, before selecting other techniques for hysterectomy and oophorectomy that may have greater morbidity, surgeons should document the inability to remove the ovary transvaginally. A particular consideration is the opportunity to reduce urologic morbidity. The ureters lie close to the infundibulopelvic ligaments, and the reported incidence of ureteral injury, with its potentially serious morbidity, during major gynecologic surgery ranges from 0.3% to 7%. 11 Ureteral injury is most common with abdominal procedures; however, as the use of operative laparoscopy has increased, transection, thermal injury, and stapling damage all have been inflicted on the uretersJ 1, 12 The ureter has a mechanism to protect it during vaginal hysterectomy, 1~ and although the activity of this protective mechanism during transvaginal oophorectomy has not been documented, an extensive search of the literature revealed no reports of ureteral injury during trans-
1487
vaginal oophorectomy. The same cannot be said of the abdominal and laparoscopic techniques) 2' ~4 The laparoscopic scoring system previously described 7 has been valuable in determining not only which patients actually require laparoscopic surgery but also to what extent. Lysis of adhesions was the only type of laparoscopic surgery performed on the ovaries in this series. Although laparoscopic oophorectomy could have been carried out in these cases, there is no evidence that this technique is safer than transvaginal operation. As gynecologic surgeons and resident teachers, it is our responsibility, in the interests of our patients, to base our surgical decisions on factual evidence. For too many years "presumptive evidence" has guided the performance and selection of the route for hysterectomy and oophorectomy in most gynecologic conditions. "Presumptive evidence" of a lack of visibility and accessibility for transvaginal oophorectomy has been accepted without the application of systematic guidelines to determine the most appropriate surgical route. The use of such guidelines may have a dramatic impact on surgical practice. For example, after the recent introduction of guidelines for the route of hysterectomy in the United States, s France, 15 and England, 16 the vaginal route without laparoscopic or abdominal assistance has proved successful in 89%, 77%, and 95% of women, respectively. These figures differ substantially from the national practice in the absence of guidelines, in which approximately 75% of patients undergo abdominal hysterectomy. Good surgical practice dictates that visibility and accessibility be the primary criteria for selecting the route of oophorectomy. The evidence presented here demonstrates that the ovaries are visible and accessible to transvaginal removal in most patients. Application of the guidelines described could reverse past and current trends toward selecting the abdominal or laparoscopic technique. This change in practice would have farreaching effects on patient morbidity and medical costs. We thank Steven Conover, MD, for his medical illustration.
REFERENCES
1. Dicker RC, GreenspanJR, Strauss LT, Cowart MR, ScallyMJ, Peterson HB, et al. Hysterectomy among women of reproductive age: trends in the United States, 1970-1978.JAMA 1982;248:323-7. 2. Kovac SR, Christie SJ, Bindbeutel GA. Abdominal versus vaginal hysterectomy: a statistical model for determining physician decision making and patient outcome. J Med Decision Making 1991;11:19-28. 3. Kovac SR. Guidelines to determine the route of hysterectomy. Obstet Gynecol 1995;85:18-23. 4. Thompson JD. Hysterectomy. In: Thompson JD, Rock JA, editors. TeLinde's operative gynecology. Philadelphia: JB Lippincott, 1992:663-732. 5. Nichols DH, Randall CL. Vaginal surgery. 2nd ed. Baltimore: Williams & Wilkins, 1983:548-60.
1488
Kovac and Cruikshank
6. Kovac SR. tntramyometrial coring as an adjunct to vaginal hysterectomy. Obstet Gynecol 1986;67:161-6. 7. Kovac SR, Cruikshank SH, Retto HF. Laparoscopy-assisted vaginal hysterectomy. J Gynecol Surg 1990;6:185-93. 8. American College of Obstetricians and Gynecologists Committee on Technical Bulletins. Pelvic organ prolapse. Washington (DC):The College, 1995:1-8.ACOGTechnical Bulletin No.: 214. 9. Wilcox LS, Koonin LM, Pokras R, Strauss LT, Xia Z, Peterson HB. Hysterectomy in the United States, 1988-1990. Obstet Gynecol 1994;83:549-55. 1O. Sheth SS. The place of oophorectomy at vaginal hysterectomy. BrJ Obstet Gynaecol 1991;98:662-6. 11. Mann WJ, Ko0nings PP. Ureteral injuries in gynecologic surgery. Int Urogynecol J 1993;4:361-5. 12. Woodland MB. Ureter injury during laparoscopy-assisted vaginal hysterectomy with the endoscopic linear stapler. AmJ Obstet Gynecol 1992;167:756-7. 13. Cruikshank SH, Kovac SR. Role of the uterosacral-cardinal ligament complex in protecting the ureter during vaginal hysterectomy. IntJ Gynecol Obstet 1993;40:141-4. 14. Laparoscopic injuries raise caution flags with doctors, insurers. Adv Technol Surgi Care 1995 Mar:29-33. 15. Querleu D, Cosson M, Parmentier D, DeBondinance P. The impact of laparoscopic surgery on vaginal hysterectomy. Gynecol Endosc 1993;2:89-91. 16. Richardson RE, Bournas N, Magos AL. Is laparoscopic hysterectomy a waste of time? Lancet 1995;345:36-41.
Discussion DR. JoHN KNAUS, Evanston, Illinois. This report and others document the high success rate of bilateral salpingo-oophorectomy when performed with vaginal hysterectomy. Nevertheless, only as few as 2% of gynecologists routinely perform salpingo-oophorectomy with vaginal hysterectomy. The grading system proposed by the authors may or may not prove practical. First, many variables affect the consistency of such a grading system (endometriosis, adhesions, prior surgery, the option of laparoscopy). Second, I do not think there is a proficient vaginal surgeon who feels intimidated by removing any adnexa when doing a vaginal hysterectomy, stretch or no stretch.
December 1996 AmJ Obstet Gynecol
The value of this report is that it may encourage capable gynecologists who currently do not remove adnexa vaginally to do so. These persons may progress from a more careful inspection of the adnexa to mobilizing them to removing them. In the end salpingo-oophorectomy performed with vaginal hysterectomy is not for the purpose of surgical exercise or to determine a success rate. It is to prevent some women from needing an operation in the future for benign disease or, more important, potentially to reduce ovarian cancer deaths. DR. KOVAC (Closing). I thank Dr. Knaus for his thoughtful comments. We do not include the round ligament in the clamp that is placed across the infundibulopelvic ligament. We place our clamp between the round ligament and the infundibulopelvic ligament in Arey's space, as previously described by Dr. Kermit Krantz. The strategy that was used to increase the numbers of transvaginal oophorectomies was simple. Quite frankly, most ovaries are accessible and visible for transvaginal removal. We were surprised, when we began to record the grades of each ovary, how many of them fell in the grade II to III range. Ninety-five to ninety-seven percent of the ovaries in this series were grade II or grade III. Therefore it was not difficult for residents to place a clamp across the infundibulopelvic ligament on ovaries that were so accessible. I think that the ability to successfully remove ovaries transvaginally in this study dispels the long-standing belief that the ovaries are difficult or impossible to remove transvaginally. Grading the accessibility of the ovaries may be just as important as grading pelvic organ prolapse, especially if this affects the surgical route of hysterectomy. Surgeons who select potentially more morbid and costly approaches to hysterectomy (abdominal and laparoscopic) solely on the basis of a need to remove the ovaries without any other contraindications to the vaginal approach may need to rethink their surgical decisionmaking process.