Hysterectomy is the most common gynecologic surgical procedure in the United States, with approximately 600,000 performed annually.1,2 Of those, an estimated 30% were approached vaginally. Vaginal hysterectomy long has been associated with less perioperative risk than hysterectomy using an abdominal approach.3–6 However, vaginal hysterectomy is increasingly performed in conjunction with pelvic reconstructive procedures, including anterior and posterior colporrhaphy.7 In fact, an estimated 11% of women undergo surgery for prolapse, incontinence, or both by the age of 80 years.8 Operating in the pelvis predisposes a woman to injury of the urinary tract, gastrointestinal tract, and major vessels and nerves because of their proximity to the female reproductive organs.9 Limited data quantify the safety of combined vaginal procedures.
Evaluation of perioperative morbidity for women undergoing reconstructive pelvic surgery has been compared with morbidity accompanying major gynecologic surgery, but, to our knowledge, no studies have compared complication rates in women undergoing vaginal hysterectomy with those having vaginal hysterectomy with reconstructive pelvic surgery. Therefore, our main objective was to estimate the perioperative and postoperative complication rates for women undergoing vaginal hysterectomy at a major tertiary care center who did and did not have concurrent reconstructive pelvic surgery. This is the initial step in assessing care and evaluating surgical outcomes as complication rates increasingly become a criterion valued by patients and payers alike.
MATERIALS AND METHODS
The study was approved by the Mayo Clinic institutional review board. Women aged 18 years or older who had a vaginal hysterectomy for a benign indication at Mayo Clinic from January 2004 through December 2005 were evaluated. Patients were included in the medical-record review if they underwent vaginal hysterectomy with or without salpingectomy, oophorectomy, pubovaginal sling, or reconstructive pelvic surgery. Any patient refusing to participate in research studies or who underwent any additional procedures, including laparoscopy or nongynecologic surgery, was excluded. Mayo Clinic’s extensive electronic medical records linkage system allows the complete ascertainment of patient data, including radiographic findings, laboratory results, patient communications, and scanned records from other institutions.
All women underwent vaginal hysterectomy with a Mayo-McCall culdoplasty, as described in the Mayo Clinic Manual of Pelvic Surgery,10 which resuspends the vaginal cuff to the uterosacral ligaments bilaterally. Cystoscopy was performed on most patients having reconstructive pelvic surgery, which included anterior colporrhaphy, posterior colporrhaphy, and combined anteroposterior colporrhaphy; there was no mesh or graft augmentation. Suprapubic catheters were placed in the majority of the women undergoing anterior colporrhaphy. Salpingectomy, oophorectomy, or pubovaginal sling surgery was performed concomitantly as indicated. All women received preoperative antibiotics. A resident, fellow, or both participated in all procedures.
This retrospective cohort study identified two patient groups: those who underwent vaginal hysterectomy and those who had vaginal hysterectomy with reconstructive pelvic surgery. Data on all patients were abstracted and collected in a database designed by the principal investigator (C.A.H.). The main outcome measures evaluated in this study were the following perioperative complications: unplanned intensive care unit (ICU) admission, any medical problem requiring intervention (eg, pulmonary edema requiring diuretics), and hospital readmission or reoperation within 9 weeks after operation. Any deviation from the surgical procedure due to an adverse event (eg, cystotomy) was coded as a complication. Additionally, the specific type of complication was collected and used for secondary analyses.
Comparisons between the two surgical groups were evaluated with the χ2 or Fisher exact test for nominal variables and the two-sample t-test for continuous variables. A multivariable logistic regression model was fit using stepwise and backward variable selection to identify a set of factors significantly different between the two groups. The factors considered in the modeling included the patient characteristics listed in Table 1 (with the exception of type of insurance and residency), the preoperative medical comorbid conditions listed in Table 2, and uterine weight, change in hemoglobin, and change in creatinine listed in Table 3. Additional logistic regression models were fit to assess the association between the presence of a complication (binary outcome) and the surgical group after adjusting for the factors identified in the previous modeling. We calculated that we would have 80% power to detect a 10% difference between women undergoing vaginal hysterectomy and those undergoing additional reconstructive pelvic surgery with the estimated 750 patients from January 2004 through December 2005.
Associations were summarized by calculating odds ratios (ORs) and corresponding 95% confidence intervals (CIs). Data were analyzed using SAS 9.0 (SAS Institute, Inc., Cary, NC). All calculated P values were two-sided; an α level less than .05 was considered statistically significant.
Of the 903 patients identified on the basis of the surgical index code for vaginal hysterectomy, the cases of 736 women were eligible for review. Table 1 summarizes the patient characteristics for those undergoing vaginal hysterectomy and those who had concurrent reconstructive pelvic surgery. Women undergoing vaginal hysterectomy with reconstructive pelvic surgery were older, were more likely to have public insurance, were more likely to be referred to Mayo Clinic from a distance (ie, selected on the basis of more severe comorbid conditions or disease status requiring management at a tertiary care medical center), and had a higher stage of prolapse. As such, women with additional reconstructive pelvic surgery were more likely to have prolapse as the indication for hysterectomy, whereas women with vaginal hysterectomy alone were more likely to have menorrhagia as the primary indication for hysterectomy.
A list of preoperative medical comorbid conditions is shown in Table 2. Women who underwent vaginal hysterectomy with reconstructive pelvic surgery were less often asthmatic and fewer were anemic (defined as hemoglobin less than 12.0 g/dL). They did, however, have a higher frequency of hypertension, cardiovascular disease, and congestive heart failure. They were also more likely to have taken corticosteroids within the preceding year.
Table 3 delineates the surgical procedures performed for women in both groups. Women undergoing concurrent reconstructive pelvic surgery were more likely to undergo salpingectomy, oophorectomy, pubovaginal sling procedures, cystoscopy, and placement of a suprapubic catheter. As expected, they also had longer procedure times, more intravenous fluid infusion, and a higher estimated blood loss; however, the frequency of blood transfusion did not differ between groups. Women undergoing vaginal hysterectomy alone had heavier uteri (more frequently the uterine weight exceeded 250 g), which corresponds to a higher billing code for increased difficulty. The length of hospitalization was longer for women undergoing vaginal hysterectomy with reconstructive pelvic surgery.
Seven patients who had vaginal hysterectomy with reconstructive pelvic surgery and 17 who had vaginal hysterectomy alone did not have complete records through the postoperative evaluation and were excluded from this assessment. The characteristics of those lost to follow-up did not differ from the characteristics of the remaining cohort, and excluding them did not affect the previously described analyses. Table 4 lists the complication rates, and Table 5 describes types of complications for each group. Women undergoing vaginal hysterectomy with reconstructive pelvic surgery were more likely overall to have a complication (147/336 [43.8%] compared with 77/376 [20.5%], OR 3.0, 95% CI 2.2–4.2, P<.001), defined as hospital readmission, reoperation, unplanned ICU admission, and medical problem requiring intervention. This association was significant (OR 3.0, 95% CI 1.5–6.2, P=.003) even after adjusting for factors significantly different between the two surgical groups (age, surgical indication, and change in hemoglobin). Women undergoing vaginal hysterectomy with reconstructive pelvic surgery were also more likely to have more medical problems requiring intervention (P<.001), the most common complication being urinary tract infection requiring antibiotic treatment, followed by pulmonary edema treated with diuretic therapy. In addition, the frequency of unplanned ICU admissions was higher among those with pelvic reconstruction (2.1% compared with 0%, P=.005). The indications for ICU admission included cardiac arrest (n=1), respiratory arrest (n=1), exacerbation of congestive heart failure (n=2), dysrhythmia (n=1), and pulmonary embolism (n=2).
Disregarding urinary tract infections, the overall complication rate was still higher among the women with concomitant reconstructive pelvic surgery (22.9% compared with 16.5%, OR 1.5, 95% CI 1.04–2.2, P=.03). Although there were significant differences in preoperative medical comorbid conditions between the two groups, the rates of hospital readmission and reoperation were similar for both groups during the index hospitalization and after discharge. Women undergoing vaginal hysterectomy alone underwent reoperation for ureteral obstruction (n=4), acute rectal pain (n=1), and acute vaginal bleeding (n=1). Women with additional reconstructive pelvic surgery underwent reoperation for ureteral obstruction (n=2), acute vaginal bleeding (n=2), intraabdominal hemorrhage, small bowel obstruction, and appendiceal epiploica necrosis (exploratory laparotomy, n=3). No mortality was observed within either group during the follow-up period.
In this study, women undergoing vaginal hysterectomy with concomitant reconstructive pelvic surgery were older and had a higher stage of prolapse, with prolapse as the primary indication for surgery. After adjusting for age, surgical indication, and change in hemoglobin, women in this subgroup were found to have more perioperative and postoperative complications. Specifically, morbidity was higher in those women undergoing concurrent reconstructive pelvic surgery with respect to unplanned ICU admission and medical problems requiring intervention. Urinary tract infection was the major contributor to increased morbidity within these patients; however, after disregarding urinary tract infection, more women undergoing additional reconstructive pelvic surgery had a complication than did women undergoing vaginal hysterectomy alone.
Certain complications were observed more frequently in women with reconstructive pelvic surgery. In particular, urinary tract infection constituted a large proportion of the postoperative complications. The frequency of urinary tract infection may be attributable to the routine use of suprapubic catheters in women undergoing anterior colporrhaphy, which are placed to reduce postoperative voiding dysfunction. Patients presenting with urinary complaints suggestive of infection were encouraged to obtain a urinalysis and urine culture before starting antibiotics. No patient was readmitted for sequelae from urinary tract infection. Furthermore, more patients with unplanned ICU admissions had preexisting cardiovascular disease. Factors associated with concurrent reconstructive surgery (ie, advanced age, longer operating room times, more intravenous fluid administration) could have exacerbated underlying medical conditions. It is also important to note that, at Mayo Clinic’s Methodist Hospital, where all gynecologic surgery is performed, the only cardiac monitoring available is in the ICU. Thus, any patient who required monitoring was transferred to the ICU.
Perioperative morbidity associated with gynecologic surgery has been evaluated by other researchers. Bai et al11 concludes that the postoperative complication rate among women undergoing abdominal hysterectomy or reconstructive vaginal surgery was no different (26.7% compared with 34.4%, respectively) despite the older age of patients undergoing reconstruction. Lambrou et al12 shows the overall perioperative complication rate for women undergoing reconstructive pelvic surgery to be 46%, 15% for readmission, and 4% for reoperation. The most important perioperative risk factors were the number of surgical procedures performed and intraoperative blood loss resulting in blood transfusions. Our study shows a similar perioperative complication rate of 44% in women undergoing vaginal hysterectomy with concurrent reconstructive pelvic surgery. However, almost half of these complications were urinary tract infections. Excluding these, the actual rate of complications in these women was 23%.
There are limitations of this study. By its nature, a retrospective review is limited in that not all outcome measures could be assessed and some important variables may have been excluded. In addition, the two groups were distinctly different; those having concurrent reconstructive pelvic surgery were older, had more comorbid conditions, and underwent more surgical procedures with longer operating room times and thus had a higher likelihood of unplanned ICU admissions and medical problems requiring intervention. Granted, many of the complications were minor (eg, urinary tract infection treated on an outpatient basis with oral antibiotics), but even minor medical problems may become severe in older patients. In addition, all of the operating surgeons are fellowship trained, and their surgical experience may not be generalizable to all gynecologic surgeons. The generalizability of this study may be limited because of the homogeneity of the two groups, with almost all patients being white. Finally, postoperative complications were determined through 9 weeks after surgery. Complications may have occurred after that point and not been recorded. This concern may be minimized because there was 96% data abstraction for the entire cohort, and any adverse event occurring more than 2 months after surgery may not be related immediately to the operation.
Despite these limitations, this study has many strengths. The main strength was the ability to quantify perioperative and postoperative morbidity as it applies to women undergoing vaginal hysterectomy with and without reconstructive pelvic surgery. Other studies have reviewed perioperative morbidity; however, they either included abdominal surgery as a comparison11,12 or explored the effect with respect to one group (eg, women who were elderly13–17 or obese18). Furthermore, the distribution of patients in both groups was relatively equal, and almost all patients were undergoing their first vaginal surgery. First procedures may minimize potentially higher rates of complications in the reconstructive pelvic surgery group due to prior operation, scarring, and distorted anatomy.
By quantifying perioperative and postoperative morbidity among women undergoing vaginal hysterectomy alone, direct comparisons can be made to those women undergoing additional vaginal reconstruction. We found an overall higher frequency of perioperative complications for all women undergoing vaginal hysterectomy than has been described previously in the literature. This higher frequency may be attributable to the relatively broad definitions for complications we applied as well as to the inclusion of minor adverse events. However, the definitions applied to identify morbid events have been used in other studies11,16–18 and capture serious and relevant complications.
Quantifying perioperative and postoperative morbidity in women undergoing vaginal hysterectomy with additional reconstructive pelvic surgery allows identification of higher-risk patients. Recognizing risk factors before surgery enables surgeons to plan appropriately in an attempt to reduce complications and positively affect the postoperative course. Additionally, quantifying morbidity allows for future prospective studies to evaluate the effect of risk reduction within these higher-risk groups.
1. Brill AI. Hysterectomy in the 21st century: different approaches, different challenges. Clin Obstet Gynecol 2006;49:722–35.
2. Whiteman MK, Hillis SD, Jamieson DJ, Morrow B, Podgornik MN, Brett KM, et al. Inpatient hysterectomy surveillance in the United States, 2000-2004. Am J Obstet Gynecol. 2008;198:34.e1–7.
3. Harris WJ. Early complications of abdominal and vaginal hysterectomy. Obstet Gynecol Surv 1995;50:795–805.
4. Dicker RC, Greenspan JR, Strauss LT, Cowart MR, Scally MJ, Peterson HB, 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.
5. Benassi L, Rossi T, Kaihura CT, Ricci L, Bedocchi L, Galanti B, et al. Abdominal or vaginal hysterectomy for enlarged uteri: a randomized clinical trial. Am J Obstet Gynecol 2002;187:1561–5.
6. Davies A, Hart R, Magos A, Hadad E, Morris R. Hysterectomy: surgical route and complications. Eur J Obstet Gynecol Reprod Biol 2002;104:148–51.
7. Luber KM, Boero S, Choe JY. The demographics of pelvic floor disorders: current observations and future projections. Am J Obstet Gynecol 2001;184:1496–501.
8. Olsen AL, Smith VJ, Bergstrom JO, Colling JC, Clark AL. Epidemiology of surgically managed pelvic organ prolapse and urinary incontinence. Obstet Gynecol 1997;89:501–6.
9. Khan F, Kenton K. Intraoperative injury in reconstructive pelvic surgery. J Pelvic Med Surg 2006;12:241–56.
10. Webb MJ. editor. Mayo Clinic manual of pelvic surgery. 2nd ed. Philadelphia (PA): Lippincott Williams & Wilkins; 2000. p. 210.
11. Bai SW, An JM, Shin JS, Park JH, Kim SK, Park KH. Comparison of perioperative complications between reconstructive pelvic surgery and general gynecologic surgery. Yonsei Med J 2004;45:281–6.
12. Lambrou NC, Buller JL, Thompson JR, Cundiff GW, Chou B, Montz FJ. Prevalence of perioperative complications among women undergoing reconstructive pelvic surgery. Am J Obstet Gynecol 2000;183:1355–8.
13. Toglia MR, Nolan TE. Morbidity and mortality rates of elective gynecologic surgery in the elderly woman. Am J Obstet Gynecol 2003;189:1584–7.
14. Mains LM, Magnus M, Finan M. Perioperative morbidity and mortality from major gynecologic surgery in the elderly woman. J Reprod Med 2007;52:677–84.
15. Parker DY, Burke JJ, Gallup DG. Gynecological surgery in octogenarians and nonagenarians. Am J Obstet Gynecol 2004;190:1401–3.
16. Sung VW, Weitzen S, Sokol ER, Rardin CR, Myers DL. Effect of patient age on increasing morbidity and mortality following urogynecologic surgery. Am J Obstet Gynecol 2006;194:1411–7.
17. Stepp KJ, Barber MD, Yoo EH, Whiteside JL, Paraiso MF, Walters MD. Incidence of perioperative complications of urogynecologic surgery in elderly women. Am J Obstet Gynecol 2005;192:1630–6.
© 2009 by The American College of Obstetricians and Gynecologists. Published by Wolters Kluwer Health, Inc. All rights reserved.
18. Chen CC, Collins SA, Rodgers AK, Paraiso MF, Walters MD, Barber MD. Perioperative complications in obese women vs normal-weight women who undergo vaginal surgery. Am J Obstet Gynecol. 2007;197:98.e1–8.