Article Text
Statistics from Altmetric.com
Cruikshank SH, Kovac SR. Randomized comparison of three surgical methods used at the time of vaginal hysterectomy to prevent posterior enterocele. Am J Obstet Gynecol. 1999 Apr;180:859-65.
Question
In women who are having vaginal hyster- ectomy, which vaginal cuff suspen- sion technique provides the best sur-gical prophylaxis against enterocele formation?
Design
Randomised (concealed), unblinded, controlled trial with 3-year follow-up.
Setting
Wright State University, Dayton, Ohio, United States.
Patients
100 women who were 24 to 77 years of age (mean age 40 y, mean parity 2 deliveries) and were having total vaginal hysterectomy. The exclusion criterion was a primary diagnosis of symptomatic pelvic organ prolapse involving the posterior vaginal segment. Follow-up at 3 years was 98%.
Intervention
Women were allocated to 1 of 3 surgical methods for closing the posterior supe- rior compartment of the vagina: a Moschcowitz-type purse-string closure, in which uterosacral and cardinal ligaments were sutured, the peritoneum of the anterior rectal wall was reefed, the cul-de-sac of Douglas was obliterated, and supporting structures were drawn to the midline (n = 33); a modified McCall culdeplasty, in which the uterosacral and cardinal ligaments were sewn, the posterior cul-de-sac was closed, the supportive structures were approximated in the midline, and the suture was externalised so that the posterior superior vaginal apex was elevated (n = 33); and closure of the peritoneum only (n = 34).
Main outcome measure
Postoperative enterocele.
Main results
At 3 years, women who had the McCall-type closure had fewer posterior en-teroceles than did women who had the Moschcowitz-type or peritoneal closure (P = 0.004) (Table).
Conclusion
3 years after vaginal hysterectomy, the McCall cuff suspension technique was more effective than the Moschcowitz or peritoneal purse-string closure for preventing asymptomatic enterocele formation.
Source of funding: Not stated.
For correspondence: Dr. S.H. Cruikshank, School of Medicine, University of South Carolina, Columbia, SC 29208, USA. FAX 803-641-3494.
Incidence of enteroceles with McCall-type closure (McC), Moschcowitz-type closure (Mos), and peritoneal closure (Per) at 3 years after vaginal hysterectomy*
Comparison Enterocele Event rates RRR (95% CI) NNT (CI)
stage
McC vs Per 1 6% vs 24% 74% (3 to 94) 6 (3 to 283)
2 0% vs 15% 100% (26 to 100) 7 (4 to 29)
McC vs Mos 1 6% vs 12% 48% (-125 to 88) Not significant
2 0% vs 18% 100% (39 to 100) 6 (3 to 16)
Mos vs Per 1 12% vs 24% 50% (-41 to 83) Not significant
RRI (CI) NNH
2 18% vs 15% 20% (-57 to 241) Not significant
*Abbreviations defined in Glossary; RRR, RRI, NNT, NNH, and CI calculated from data in article.
Commentary
A growing postmenopausal population has increased the need for rigorously designed studies of interventions for symptomatic pelvic relaxation. Cruikshank and Kovac did a randomised study of surgical techniques intended to prevent enterocele formation in women having vaginal hysterectomy. This is the first controlled trial to examine the effect of vaginal cuff suspension techniques for the prophylaxis of enteroceles.
The same surgeons, all highly skilled in vaginal surgery, did all of the procedures, thus ensuring that variation in training and experience would not confound the results of the intervention. The 3-year follow-up period, with assessment of patient outcomes by the same surgeons who used an objective, inter nationally recognised standard for the description of pelvic structural pathologic findings, is notable.
The structural pelvic dynamics of patients are likely to vary depending on the diagnoses that lead to hysterectomy (e.g., leiomyomata, incontinence, prolapse). This variation is partially controlled for by excluding women with a primary diagnosis of "symptomatic pelvic organ prolapse involving the posterior vaginal segment." Similarly, coincident surgical procedures (e.g., sacrospinous fixation, posterior repair) done at hysterectomy may also affect the likelihood of enterocele formation. The distribution of these variables by treatment allocation is not described. The authors are careful to point out that only 8 patients reported a vaginal bulge at follow-up, and none has required reparative surgery. The treatment allocation of these sympto-matic patients is not reported.
The application of the rigours of the ran-domised controlled study design to any sur-gical technique is fraught with pitfalls. The authors have avoided the most common and problematic ones. They show the superiority of the McCall-type culdeplasty for the prevention of asymptomatic posterior enteroceles. The generalisability of their findings to surgeons of lesser skill remains to be seen.
Francisco A. Garcia, MD
University of Arizona
Tucson, Arizona, USA