Hysterectomy is the most common non-pregnancy-related gynaecologic surgical procedure performed in the USA, with one in three women having a hysterectomy by the age of 60 1. Most hysterectomies in the UK are abdominal (70–90%), with only 10–30% performed vaginally and less than 5% laparoscopically 2. The 2005 Cochrane review of surgical approaches to hysterectomy for benign gynaecological diseases concluded that, where possible, vaginal hysterectomy (VH) should be performed in preference to abdominal hysterectomy (AH) 3. When compared with AH, VH is associated with fewer unspecified infections/febrile episodes, shorter hospital stay and quicker return to normal activities. There was no evidence of benefits for laparoscopic hysterectomy (LH) compared with VH, whereas the operating time was increased in LH 3. Although VH is considered to be the quickest and most cost-effective manoeuvre for removal of the uterus, it is used in only 23% of the hysterectomies performed in the USA. Traditionally, many surgeons have limited indications for VH, these indications not including narrow vagina, bulky uterus, nonprolapsed uterus, pelvic adhesion, scarred uterus, obesity and nulliparity 4. Currently, 20% or more of deliveries are carried out by caesarean section in most countries. With this trend likely to flourish, it is conceivable that the surgeon of tomorrow will increasingly have to encounter the problem of hysterectomy in patients with one or multiple previous caesarean sections. Sheth and Allahbadia 5 confirmed that the chief concern in this group of patients is the risk of injury to the bladder and how to get access to the anterior cul-de-sac. The current study assessed the safety and efficacy of the vaginal route for hysterectomy in comparison with the abdominal route in patients who had undergone previous pelvic surgery, which represents a significant challenge for the majority of gynaecologists.
Patients and methods
The study was carried out in Ain Shams University Maternity Hospital over a period of 20 months from December 2010 to August 2012. Ethical approval was granted from the local scientific research committee on August 2010. Women waiting to undergo a hysterectomy procedure for benign disease were approached, recruited and provided their written consent to participate after receiving information about the study. By then, each enrolled participant was assigned the next available number in the concealed sequence in a computer-generated randomization plan, and this done by a physician who played no role in patients’ enrolment. Fifty-four consecutive cases that underwent hysterectomy for benign disease underwent a preoperative assessment and were assigned randomly to one of the following two groups: group 1 (the case group), which included 27 patients who had undergone VH (n=27), and group 2 (the control group), which included 27 patients who had undergone AH (n=27). In both groups, all patients had a history of previous abdominopelvic gynaecologic or obstetric operation. The types of operations were as follows: ovarian cystectomy, salpingo-oophorectomy, salpingectomy, myomectomy, caesarean section and appendectomy. Inclusion criteria for both groups were as follows: first-degree or second-degree uterine descent, uterine size 16 or less weeks, benign pathology, ample vaginal canal, wide and deep posterior and lateral vaginal fornices, and subpubic angle more than 90°. Patients with the cervix flushed with the vagina, narrow vagina, uterine size more than 16 weeks, absent uterine descent, no adequate uterine mobility, pelvic mass, pelvic malignancy or previous successful repair of vesicovaginal fistula were excluded from the study. Participants were subjected to a complete preoperative assessment by a full assessment of history, complete examination and examination under anaesthesia to confirm the previous findings and to assess uterine size, mobility, descent, cul-de-sac and vaginal adequacy. The decision was then made randomly on whether to proceed vaginally or abdominally after fulfilment of the inclusion criteria was established. A total of 64 patients were examined; 10 patients were excluded after examination under anaesthesia for reasons such as narrow vagina, uterine size more than 16 weeks, no uterine mobility and absent uterine descent, and all underwent AH but were excluded from the study. The remaining 54 patients fulfilled the inclusion criteria and were included in the study and were divided into two equal groups: 27 patients underwent VH and 27 patients underwent AH. Participants received general, spinal or epidural anaesthesia. Heaney’s technique was used in the VH group with or without debulking techniques such as myomectomy, morcellation, bisection, polypectomy or coring. It was common to encounter adhesions between the bladder and the cervix in some patients, and this made the surgery technically more difficult. Sharp dissection was attempted initially, keeping as close to the cervix as possible, to minimize the chances of bladder injury. In some patients, in whom dense adhesions prevented access to the uterovesical fold of the peritoneum, the lateral window technique or utilization of the uterocervico-broad ligament space technique was very useful as shown by Sheth 6, wherein the dense adhesions in the midline, the lateral space between the leaves of the broad ligament adjoining the uterocervical border near the isthmic notch (the uterocervico-broad ligament space) was usually free of adhesions.
The primary outcome measures were the operative time and blood loss, whereas the secondary outcome measures were hospital stay and intraoperative and postoperative complications. Each patient in the study was tested for the following endpoints: the total operative time was calculated and it was defined as the time from initial mucosal incision to closure of the vaginal cuff with satisfactory haemostasis (in group 1, VH group), and time from skin incision to skin closure (in group 2, AH group). Operative blood loss was estimated by weighing the swabs (by the anaesthetist and auxiliary theatre staff). Concomitant procedures included separate calculation of operative time and blood loss. The postoperative haemoglobin and haematocrit were measured for all participants at 24 h after the procedure. Hospital stay was identified for each case (<48 h or >48 h) and, in cases of delayed discharge, the reason was identified and recorded. Complications were reported, for example, the need for blood transfusion, urinary bladder injury or conversion to laparotomy.
Sample size and statistical analysis
Twenty-seven cases in each group would allow a difference of 28% points in the rate of complications, assuming that the rate is 51 in the first group with a power of 90% using a 5% significance level [rate of success (can be determined from published studies)=28% difference] (confidence interval=95%, accuracy of ±28 percentage point, significant level 5%, power=90%). Required sample size was 27 in each group.
Data were presented as mean and SD for numeric parametric data and numbers and % for categorical data. Variation between two groups was compared using Student’s t-test for numeric parametric data, Fischer’s exact and χ2-tests for categorical data. P value of less than 0.05 was considered statistically significant. All statistical calculations were carried out using statistical package for the social science (SPSS, version 15; SPSS Inc., Chicago, Illinois, USA) for Microsoft Windows.
Fifty-four consecutive cases over a period of 20 months who underwent hysterectomy for benign disease were finally analysed. Twenty-seven cases underwent VH and 27 cases underwent AH. The VH group included 27 women ranging in age from 41 to 62 years, with a median age of 48 years; their parity ranged from 2 to 6, with a median value of 4. The AH group included 27 women, ranging in age from 43 to 61 years, with a median age of 50 years; their parity ranged from 1 to 5, with a median value of 3. For the VH group, the most common surgery was a caesarean section (13 cases), representing 48.1% of VH cases. For the AH group, the most common surgery was a caesarean section (15 cases), representing 55.6% of AH cases. The most common indication for hysterectomy in the VH group was dysfunctional uterine bleeding (nine cases, representing 33.3% of the cases). The most common indication for hysterectomy in the AH group was fibroid uterus (14 cases), representing 51.9% of cases. The observed differences in age, parity, type of previous abdominopelvic surgeries and indications of hysterectomy between the two groups were not statistically significant. Participants in the VH group had a shorter operating time compared with patients who underwent AH (73.7±21 vs. 107.8±20 min, respectively, P<0.001). The difference in the measured blood loss intraoperatively was significantly higher in the AH group compared with the VH group. There was no statistically significant difference between group 1 and group 2 in preoperative haemoglobin percentages (11.7±1.3 vs.11.6±1.2, respectively, P>0.05); however, a significant lower haemoglobin percentage was observed on the first day after surgery in patients of group 2 compared with those of group 1 (9.8±1.3 vs. 10.8±1.3, respectively, P<0.05), with a smaller decrease in the haemoglobin level in group 1 compared with group 2 (0.86 vs. 1.8 g, P<0.001). No significant difference was found between the VH and the AH group in intraoperative complications. The most frequent minor postoperative complication was febrile morbidity and was observed most often in the AH group (25.9%). Other complications included blood transfusion, haematoma formation, urinary tract infection, paralytic ileus and surgical site infection, which occurred most frequently in the AH group; however, the difference was not statistically significant. There was a statistically significant difference between the VH and AH groups in the postoperative course, with less need for analgesics and a shorter hospital stay in the VH group, indicating a smooth postoperative period and rapid recovery and return to normal life.
The current study examined whether the transition from the abdominal to the vaginal route of hysterectomy in patients with previous pelvic surgery is safe and easy. Postoperative mobilization was earlier in the VH group of patients. The study proved that postoperative patient comfort was very much increased with VH. This was proved by studies carried out by many investigators 7–9. Sawkar 10 measured the pain scoring on postoperative day 3 in the visual analogue scale; the mean pain score in the VH group was 2.88 cm whereas the mean pain score in the AH group was 6.48 cm. The difference in the pain rating scoring between the two groups was found to be statistically significant, with a P value of less than 0.05, in agreement with the results of our study 10. Silva-Filho etal. 11 confirmed a better postoperative quality of life in women who underwent VH compared with those who underwent AH. The vaginal approach of hysterectomy was documented to be associated with less postoperative pain and rapid return to social life compared with AH 12,13. Some authors suggest that the vaginal route is less invasive and should be the gynaecological surgeon’s first choice 14. VH yields superior results in terms of inflammatory response compared with AH 15. On reviewing the literature, minimal blood loss is particularly desirable in patients undergoing hysterectomy for benign conditions as they often have anaemia and iron deficiency. Doucette et al.16 reported the same findings; they found that blood loss and the decrease in haemoglobin were less in the study VH group than in the AH group (5.7% compared with 6.5% for AH, P=0.009). In contrast to our study, the frequency of intraoperative haemorrhages was significantly higher in patients who underwent VH with a history of caesarean sections than those without a history of caesarean sections in one study 17. On reviewing the literature, a shortening of the operative time was found in patients who underwent VH with previous pelvic surgery compared with patients who had not undergone a previous pelvic surgery 18. Kovac and Robert 19 preferred VH when the pathological lesion was confined to the uterus with an average weight not more than 280 g and when a postoperative febrile complication is expected. Our data clearly do not support Sheth and Malpani’s 20 recommendation that VH should be avoided in women with a history of more than two previous caesarean deliveries (although we had only two patients, which are not sufficient for a comparison). However, their results were in agreement with ours in terms of the fact that the vaginal route is the route of choice for performing a hysterectomy in patients with a previous caesarean section 20. Nevertheless, in a more recent study, other investigators performed VH in 312 patients with a history of two or more caesarean sections using a four-step method to safeguard the bladder and access the vesicouterine peritoneum, and reported that VH was performed in 311 patients, and the abdominal route was used in only one patient because of haemorrhage. Only one patient had bladder trauma, which was promptly repaired 21. Taylor et al.8 reported the same results as those of our study that length of hospital stay increased significantly with total AH than VH (mean 3.9 vs. 2.6 days, P<0.001).
Our results are also in agreement with the results of Doucette et al.16 in terms of a shorter hospital stay for the VH study group (2.1 days) than the AH group (2.7 days, P<0.001). Matteson et al.22 carried out a study to evaluate the factors associated with increased hospital charges for hysterectomy, with a specific focus on differences on the basis of surgical approach. They found that blood loss greater than 1000 ml and operative time of 105 min or more were strongly associated with higher charges for hysterectomy. Other factors associated with higher charges included postoperative fever, increasing length of hospitalization, the use of prophylactic antibiotics and the AH surgical approach compared with VH 22. In a recent debate, investigators suggested that gynaecologists who include vaginal surgery in their armamentarium are better equipped to serve their patients. Recently, there is a debate whether there is enough trained residents on vaginal hysterectomy in the era of using laparoscopic hysterectomy. It is now mandatory for all gynaecologists to re-evaluate their teaching programme 23,24.
There are several limitations to the analysis reported in this study. First, although this study attempted to control for all patient and surgical variables that could potentially increase the difficulty of the hysterectomy, two important variables were overlooked during data collection: patient weight and uterine weight. Both extreme patient weight and large uterine size are factors that limit visibility and, therefore, complicate dissection and advancement of the bladder flap. The data on uterine size could potentially be added to the analysis in the future after revisiting the medical records, but accurate data on patient weight are unlikely to be available. The second limitation of the study was the fact that only incidental cystotomies recognized intraoperatively were included in the final analysis. Harkki-Siren et al.25 reported that only 58% of bladder injuries in a retrospective review of 62 379 hysterectomies were diagnosed intraoperatively. This figure indicated over 40% to be diagnosed up to 30 days postoperatively, many as fistulas 25. Next, the data on the number of previous vaginal deliveries were incomplete and, therefore, not included in the final analysis. Boukerrou and colleagues concluded that a history of at least one vaginal delivery was not significant when looking at complications at the time of VH in patients with a previous caesarean section. Unger et al.26, however, reported that the overall complication rate, when considering incidental cystotomy, blood loss and operative time, was reduced for patients who had undergone at least one previous vaginal delivery (3.2 vs. 17.6%, P=0.004) 26.
On the basis of our study, VH appears to be the preferred hysterectomy technique for women with previous pelvic surgery, and every gynaecological surgeon should be familiar with this procedure. VH had a shorter operating time, less blood loss, fewer febrile episodes or unspecified infections, shorter duration of hospital stay and more rapid return to normal activities. Previous pelvic operation did not make VH technically more difficult; thus, this study did not find any significant benefits in performing AH over VH in patients with previous pelvic surgery, and major complications were almost the same; thus, VH should be performed in preference to AH where possible.
Conflicts of interest
There are no conflicts of interest.
1. Farquhar CM, Steiner CA. Hysterectomy rates in the United States, 1990–1997. Obstet Gynecol 2002; 99:229–234.
2. Maresh MJ, Metcalfe MA, McPherson K, Maresh MJ, Metcalfe MA, McPherson K, et al.. The VALUE national hysterectomy study: description of the patient and their surgery. Br J Obstet Gynecol 2002; 109:302–312.
3. Johnson N, Barlow D, Lethaby A, Johnson N, Barlow D, Lethaby A, et al.. Surgical approach to hysterectomy for benign gynaecological disease. Cochrane Database Syst Rev 2006; 2:CD003677.
4. Kives SL, Levy BS, Levine RL. Laparoscopic-assisted vaginal hysterectomy
: American Association of Gynecologic Laparoscopists’ 2000 membership survey. J Am Assoc Gynecol Laparosc 2003; 10:135–138.
5. Sheth SS, Allahbadia GNSheth SS, Studd JWW. Hysterectomy after previous abdominopelvic surgery. Vaginal hysterectomy
2002; 11London:Martin Dunitz Ltd;127–137.
6. Sheth SSStud J. Vaginal hysterectomy
. Progress in obstetrics and gynaecology 1993; 10London:Martin Dunitz Ltd;317–340.
7. Garg PK, Deka D, Malhotra N. Non-descent vaginal hysterectomy
for benign condition. A better proposition than abdominal hysterectomy. J Obs Gynaecol Today 2002; 6:345–346.
8. Taylor SM, Romero AA, Kammerer-Doak DN, Qualls C, Rogers RG. Abdominal hysterectomy for the enlarged myomatous uterus compared with vaginal hysterectomy
with morcellation. Am J Obstet Gynecol 1996; 162:994–998.
9. Dewan R, Agarwal S, Manisha, Minocha B, Sen Soumendra K. Non-descent vaginal hysterectomy
– an experience. J Obstet Gynecol Ind 2004; 54:376–378.
10. Sawkar H. Vaginal hysterectomy versus totalabdominal hysterectomy in non descent uterus: a comparative study
2009.Karnataka, India:Department of Obstetrics and Gynecology, Adichunchanagiri Institute of Medical Sciences, Rajiv Ghandi University of Health Science.
11. Silva-Filho AL, Werneck RA, de Magalhães RS, Belo AV, Triginelli SA. Abdominal vs vaginal hysterectomy
: a comparative study of the postoperative quality of life and satisfaction. Arch Gynecol Obstet 2006; 274:21–24.
12. Benassi L, Rossi T, Kaihura CT. Abdominal or vaginal hysterectomy
for enlarged uteri: a randomized clinical trial. Am J Obstet Gynecol 2002; 187:1561–1565.
13. Gimbel H, Zobbe V, Andersen BM, Filtenborg T, Gluud C, Tabor A. Randomised controlled trial of total compared with subtotal hysterectomy with one-year follow up results. Br J Obstet Gynaecol 2003; 110:1088–1098.
14. Sheth SS. The scope of vaginal hysterectomy
. Eur J Obstet Gynaecol Reprod Biol 2004; 115:224–230.
15. Ribeiro SC, Ribeiro RM, Santos NC, Pinotti JA. A randomized study of total abdominal, vaginal and laparoscopic hysterectomy. Int J Gynaecol Obstet 2003; 83:37–43.
16. Doucette RC, Sharp HT, Alder SC. Challenging generally accepted contraindication to vaginal hysterectomy
. Am J Obstet Gynecol 2011; 184:1386–1389.
17. Boukerrou M, Lambaudie E, Collinet P, Crépin G, Cosson M. A history of caesareans is a risk factor in vaginal hysterectomies. Acta Obstet Gynecol Scand 2003; 82:1135–1139.
18. Robert J, Pedros S. Vaginal hysterectomy
following pelvic operation. Obstet Gynecol 1967; 30:394–398.
19. Kovac S, Robert F. Hysterectomy outcomes in patients with similar indications. Obstet Gynecol 2000; 95:787–793.
20. Sheth SS, Malpani AN. Vaginal hysterectomy
following previous caesarean section. Int J Gynaecol Obstet 1995; 50:165–169.
21. Sheth SS. Vaginal hysterectomy
in women with a history of 2 or more cesarean deliveries. Int J Gynaecol Obstet 2013; 122:70–74.
22. Matteson KA, Peipert JF, Hirway P, Cotter K, DiLuigi AJ, Jamshidi RM. Factors associated with increased charges for hysterectomy. Obstet Gynecol 2006; 107:1057–1063.
23. Sheth SS, Paghdiwalla KP, Hajari AR. Vaginal route: a gynaecological route for much more than hysterectomy. Best Pract Res Clin Obstet Gynaecol 2011; 25:115–132.
24. Goolab BD. Vaginal hysterectomy
and relative merits over abdominal and laparoscopically assisted hysterectomy. Best Pract Res Clin Obstet Gynaecol 2013; 27:393–413.
25. Harkki-Siren P, Sjoberg J, Tiitinen A. Urinary tract injuries after hysterectomy. Obstet Gynecol 1998; 92:113–118.
26. Unger JB, Meehs GR. Vaginal hysterectomy
in women with history of previous cesarean delivery. Am J Obstet Gynecol 1998; 179:298–299.