Vaginal hysterectomy has less perioperative morbidity and shorter postoperative recovery time compared with abdominal hysterectomy, although three of four hysterectomies in the United States and abroad are done abdominally.1,2 Belief that nulliparity, previous pelvic surgery, and uterine enlargement greatly impede vaginal hysterectomy have supported a high frequency of abdominal and laparoscopically assisted hysterectomies in women without prolapse.3,4 The objectives of this prospective study were to assess the effectiveness and safety of vaginal hysterectomy for nonprolapsed uteri and to identify factors associated with success and failure.
Three hundred consecutive women without uterine prolapse, suspected adnexal disease, or suspected malignancy, who required hysterectomy for benign gynecologic diseases (no condition was suspicious for malignancy) were scheduled for vaginal hysterectomy. Indications for hysterectomy included symptomatic leiomyomata uteri, abnormal bleeding, biopsy-proven adenomyosis, or cervical dysplasia. Some women had more than one indication. Absence of prolapse was defined as a cervix that did not reach the lower third of the vagina on traction with a tenaculum. Uterine mobility was assessed by bimanual pelvic examination. Uterine immobility and narrow pelvimetry were not exclusion criteria. Each woman had endometrial biopsy and transvaginal sonographic adnexal and uterine evaluation preoperatively. No woman in the study had laparoscopy before hysterectomy.
The mean patient age was 45.9 years (range 36–58 years), mean body mass index (BMI) was 27.2 kg/m2 (range 17.2–46.7 kg/m2), and mean parity was 3.1 (range 0–6). Sixty-nine women had BMI greater than 30 kg/m2, 21 (7%) were nulliparous, and 219 (73%) had histories of pelvic surgery. Prior operations included cesarean delivery (150 cases, 50%), myomectomy, adnexal surgery, colporrhaphies, and combined procedures.
Each woman received antibiotic prophylaxis before surgery. Procedures used were the classic Heaney techniques5 or the Pelosi technique6 (in cases of suboptimal vaginal exposure), with or without morcellation (hemisection, intramyometrial coring, myomectomy, or wedge resection individually or in combination). The Heaney technique is well known. The Pelosi technique consists of pericervical vasoconstrictor injection followed by posterior cervicocolpotomy, sutureless division of the uterosacral-cardinal ligament complex, and conventional mobilization of the uterine arteries and upper uterine attachments (Figures 1 and 2). The vaginal cuff is repaired in a conventional fashion.
Operating time, estimated blood loss, surgical technique (Heaney, Pelosi, morcellation), operative complications, conversion to laparoscopy or laparotomy, and length of hospital stay were recorded for each case. Hospital discharge criteria included ability to tolerate a normal diet and to void and defecate comfortably without assistance, and the absence of fever or excess discomfort. Postoperatively, women were examined twice in the first month, then semiannually if they had no clinical problems.
Vaginal hysterectomy was successful in 297 of 300 cases (99%). The mean operating time was 51 minutes (range 20–130 minutes), and mean estimated blood loss was 180 mL (range 50–1050 mL). The Heaney technique was used in 260 cases (86.7%), the Pelosi technique in 40 cases (13.3%), and morcellation in 170 cases (56.7%). Of three intraoperative complications (1%), two were sharp bladder perforations and one was a sharp anterior rectal wall perforation. All were repaired transvaginally and were not associated with postoperative complications. The mean uterine weight was 186 ± 105 g (range 30–1160 g). Bilateral oophorectomy was done in 50 women. No pathologic evidence of unsuspected malignancy was found in any specimen.
Three vaginal hysterectomies failed (1%), two of which were laparotomy conversions for control of hemostasis at the level of the infundibulopelvic ligaments. One of those women received blood transfusion intraoperatively. One hysterectomy required a laparoscopic conversion for removal of an adherent ovary with unsuspected endometriosis.
The mean postoperative hospitalization was 22 hours. Two hundred four women (68%) were discharged within 24 hours of surgery. Twelve women's (4%) postoperative stays exceeded 48 hours. Postoperative complications were limited to 11 urinary tract infections (3.7%).
A total of 379 hysterectomies were done during the study, of which 341 (90.0%) were vaginal. Seventy-nine women were excluded from the study because they had vaginal hysterectomy and pelvic floor repairs for prolapse (44 cases) or laparoscopy or laparotomy for suspected malignancy or adnexal disease (35 cases).
Nulliparity, previous pelvic surgery, uterine enlargement, or obesity did not impede successful vaginal hysterectomy when it was applied uniformly to a group of women without uterine prolapse. Despite a 50% prevalence of previous cesareans in our subjects, the incidence of surgical bladder injury (0.67%) was comparable to the range of 0.3–1.0% commonly quoted for total abdominal hysterectomy,1,3 comparable to the range of 0.5–1.5% commonly quoted for vaginal hysterectomies done without conventional contraindications,1,3 and less than the 1.03–1.8% rate quoted for laparoscopically assisted hysterectomies.7,8 Our data indicate that vaginal hysterectomy despite traditional contraindications did not result in increased surgical morbidity to the bladder. In terms of the type ofbladder injury that incurred, vaginal hysterectomy was associated only with sharp injuries, which are easily repaired and unlikely to prolong hospitalization or compromise normal healing.
The rate of rectal injury was 0.33%, which compares favorably with overall rectal injury rates of 0.4%, 0.3%, and 0.19% for vaginal, abdominal, and laparoscopic hysterectomies, respectively, in a large literature review.7 The 0.33% transfusion rate in our study was lower than recent rates of 1–5% reported for abdominal and vaginal hysterectomies, and a rate of 1.58% for laparoscopic hysterectomy.7 It emphasizes that a high morcellation rate does not necessarily result in a high transfusion rate.9 The 3.7% rate of symptomatic urinary tract infection was within the range of 1–5% recently reported for abdominal and vaginal hysterectomies.3
The high frequency of morcellation (56.7%) in this series shows a combination of operator preference and surgical necessity for a high percentage of successful transvaginal uterine extirpations under the conditions studied. In our experience, when uterine enlargement is the only impediment to removal of a relatively mobile uterus, the critical uterine size that predicates morcellation ranges from 10–12 weeks' gestational equivalent, depending on the breadth of the pubic arch. However, if the uterus is relatively immobile, morcellation is needed at a smaller uterine size.10 The advantage of morcellation in both settings is improved access to the broad ligament attachments and proximal adnexal pedicles.9–11
The morcellation technique most commonly used in this study was simple hemisection (ie, bivalving). The maneuver is simple and well described in the literature.12 Few uteri needed additional or alternate maneuvers, such as coring or wedge morcellation, techniques that demand more experience but which are readily learned. Supported by our findings, we believe that if hemisection alone were taught and practiced widely and then applied to the majority of contraindicated vaginal hysterectomies, when conventional intact uterine removal was not possible, most surgeons would be able to complete vaginal hysterectomies more safely and effectively. In our experience, the learning curve for morcellation techniques is rapid compared with that for advanced laparoscopic hysterectomy techniques and retains the efficiency of classic surgery with simple, durable, inexpensive, and readily available equipment.
The three vaginal hysterectomy failures occurred secondary to difficulties with adnexectomy after uterine removal. In a sense, the hysterectomies were successful, but the secondary adnexal procedures were not. A 1% failure rate related to adnexectomy, when one of three failures was managed laparoscopically in women traditionally treated by laparotomy, is acceptable. It would be rare to find a woman who would not consent to attempted vaginal hysterectomy with a 99% chance of success and a 99.3% chance of avoiding laparotomy. Also, it is hard to argue in favor of preliminary laparoscopy to offset a 0.67% laparotomy risk in women without suspected adnexal or extiapelvic disease. This study suggests that with a mean operating time of 51 minutes for vaginal hysterectomy, the best strategy for an indicated laparoscopically assisted vaginal hysterectomy might be treatment of extrauterine disease and laparoscopic mobilization of the adnexa followed by vaginal hysterectomy.
1. Dicker RC, Greenspan JR, Strauss LT, Scally MJ, Peterson HB, DeStefano F, et al. Complications of abdominal and vaginal hysterectomy among women of reproductive age in the United States. The Collaborative Review of Sterilization. Am J Obstet Gynecol 1982;144:841–8.
2. Kovac RS, Cruikshank SH. Guidelines to determine the route of hysterectomy. Obstet Gynecol 1995;85:18–23.
3. Harris WJ. Early complications of abdominal and vaginal hysterectomy. Obstet Gynecol Surv 1995;50:795–805.
4. Reich HA. Laparoscopic hysterectomy. In: Hulka JF, Reich H, eds. Textbook of laparoscopy. Philadelphia, Pennsylvania: WB Saunders. 1998:363–79.
5. Heaney NS. A report of 565 vaginal hysterectomies performed for benign pelvic disease. Am J Obstet Gynecol 1934;28:751–5.
6. Pelosi M A 3rd, Pelosi MA. Simplified vaginal hysterectomy: A pioneer approach. N J Med 1997;34:33–9.
7. Harris WJ, Daniell JF. Early complications of laparoscopic hysterectomy. Obstet Gynecol Surv 1996;51:559–67.
8. Meikle SF, Nugent EW, Orleans M. Complications and recovery from laparoscopy-assisted vaginal hysterectomy compared with abdominal and vaginal hysterectomy. Obstet Gynecol 1997;89:304–11.
9. Kovac SR. Intramyometrial coring as an adjunct to vaginal hysterectomy. Obstet Gynecol 1986;67:131–6.
10. Pelosi MA 3rd, Pelosi MA. Vaginal hysterectomy for benign uterine disease in the laparoscopically confirmed frozen pelvis. J Laparoendosc Adv Surg Tech A 1997;7:345–51.
11. Pelosi MA, Pelosi MA 3rd. Should uterin size alone require laparoscopic assistance? Vaginal hysterectomy for a 2,003-g uterus. J Laparoendosc Adv Surg Tech A 1998;8:99–103.
12. Nichols DH. Vaginal hysterectomy. In: Nichols DH, ed. Gynecologic and obstetric surgery. Philadelphia, Pennsylvania: Mosby, 1993:297–333.