SEARCH

SEARCH BY CITATION

Keywords:

  • Laparoscopic supracervical hysterectomy;
  • patient satisfaction;
  • postoperative symptoms;
  • subtotal hysterectomy

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Contribution to authorship
  9. Details of ethics approval
  10. Research governance
  11. Funding
  12. Acknowledgements
  13. References

Objective  Evaluation of long-term outcomes following laparoscopic supracervical hysterectomy (LSH).

Design  Retrospective postal questionnaire.

Setting  Norwegian university teaching hospital.

Population  A total of 315 consecutive patients.

Methods  A questionnaire sent to all patients who underwent a LSH during 2004 and 2005.

Main outcome measures  Persistent vaginal bleeding and pelvic pain, patient acceptability of such symptoms and patient satisfaction following LSH.

Results  A total of 240 women (78%) completed the questionnaire. About 24% reported experiencing vaginal bleeding up to 3 years following their hysterectomy, although this was rated as minimal in 90% of cases, resulting in a mean bothersome score of 1.1 (SD 2.0) on a 10-point visual analogue scale (VAS). Women operated on by less experienced surgeons were more likely to report vaginal bleeding following surgery (P = 0.02). About 74% of women reported having menstrual pain prior to surgery, with a mean score of 6.8 (SD 2.1) (10-point VAS). Up to 3 years following surgery, 38% continued to experience menstrual pain, although this was significantly less intense with a mean score of 3.5 (SD 2.2) (P < 0.01). While all women reported a decrease in the amount of pain experienced following the hysterectomy, those having a hysterectomy because of endometriosis reported significantly higher levels of menstrual/cyclical pain after surgery compared with women who had a hysterectomy for other reasons (P < 0.01). Ninety per cent of women reported being satisfied with their surgery.

Conclusion  Although vaginal bleeding and pelvic pain are frequently observed following LSH, these symptoms are significantly reduced and patient satisfaction is high.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Contribution to authorship
  9. Details of ethics approval
  10. Research governance
  11. Funding
  12. Acknowledgements
  13. References

The benefits of laparoscopic hysterectomy compared with an open abdominal approach are well documented (less postoperative pain, better cosmetic results, fewer wound infections and earlier discharge from hospital).1 However, nearly 20 years after the first reports of different techniques of laparoscopic hysterectomy, abdominal hysterectomy is still the dominating surgical technique worldwide.2,3

Advocates of laparoscopic supracervical hysterectomy (LSH) suggest that the procedure is easier to perform, less invasive and carries a lower risk of ureteric injuries and infectious complications compared with total laparoscopic hysterectomy (TLH).4–6 Opponents of LSH, however, are concerned with the persistent risk of cervical stump symptoms such as vaginal bleeding and pelvic pain following the supracervical hysterectomy, causing patient distress and eventually repeated surgery.7 The incidence of cervical cancer after subtotal hysterectomy is so low in countries with routine cervical screening programmes, that there is no justification for removal of the cervix on this basis.8

There is no current evidence confirming the assertion that LSH carries a lower risk of complications compared with TLH.9 In the only prospective, randomised trial comparing complications and clinical outcome following LSH and TLH, no differences in complication rates between the two procedures were observed.10 In a prospective, randomised trial comparing the outcomes of LSH with hysteroscopic endometrial resection, LSH was found to be superior in terms of overall patient satisfaction, and the authors concluded that LSH may be considered the best surgical approach for the management of abnormal bleeding.11

The literature regarding clinical outcomes following LSH mainly consists of case series and retrospective reports. The occurrence of persistent vaginal bleeding following LSH is reported in the wide range of 0–25%.6,7,11–13 Furthermore, there have been discussions about whether the risk of pelvic pain is increased after LSH compared with TLH, and some authors recommend TLH in the presence of preoperative pelvic pain and/or endometriosis.14–16

At Ullevål University Hospital, LSH is the recommended procedure for patients with benign conditions requiring hysterectomy and with no history of previous cervical dysplasia. Consequently, 78% of the laparoscopic hysterectomies performed in 2005 were supracervical.17 The present study was performed to evaluate the occurrence of cervical stump symptoms (i.e. vaginal bleeding and pelvic pain), patient acceptability of cervical stump symptoms and overall patient satisfaction following LSH. Furthermore, factors that might influence the occurrence of cervical stump symptoms, such as preoperative pelvic pain and level of experience of the surgeon, were evaluated.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Contribution to authorship
  9. Details of ethics approval
  10. Research governance
  11. Funding
  12. Acknowledgements
  13. References

Following ethical approval, in January 2007, all women who were treated by LSH in the Department of Gynaecology, Ullevål University Hospital, Oslo, Norway during 2004 and 2005 were sent a questionnaire. Nonresponders were sent one reminder questionnaire 4 weeks following the original mail out. Consequently, women were contacted between 12 and 36 months after surgery.

The questionnaire was divided into two sections: the first section contained 14 questions about reasons for having the hysterectomy, menstrual pain and bleeding prior to the hysterectomy, knowledge and expectations surrounding continued bleeding, need for cervical screening after the hysterectomy and satisfaction about preoperative information. The second section of the questionnaire contained 14 questions about experiences of menstrual bleeding and pain following the hysterectomy and the impact this has on life activities, any treatments for bleeding and pain following the hysterectomy, any new symptoms related to the hysterectomy and overall satisfaction with the hysterectomy. Women were also invited to provide any additional information about their experiences in a free text section. Standard 10-point visual analogue scales (VAS) were used to measure pain intensity and extent to which pain and bleeding were bothersome. The remaining questions were either dichotomous ‘yes’ or ‘no’ responses, or they provided women with either four or five categories of responses to choose from. The majority of categories provided ordinal data.

The questionnaire was piloted on six women to determine whether women were able to comprehend the questions and whether there were any difficulties in completing it. Some minor changes were made to the wording of two questions in response to this pilot work. Since we were not measuring any complex constructs within the questionnaire, no further validity checks were necessary.

In addition, information about any subsequent related treatment as well as the surgeon’s experience was obtained from the medical records. Four senior consultants, each carrying out LSHs for more than 10 years, were identified as ‘experienced’ and the remaining surgeons were identified as ‘less experienced’.

LSH is the standard procedure in our department for women with a benign condition requiring hysterectomy and with no previous history of cervical dysplasia. The standardised operative technique for LSH at our department has been described previously.18 The amputation of the cervix was performed by a unipolar lap loop, followed by electrocoagulation by unipolar diathermy of the endocervix. Endometriosis was treated by bipolar electrocoagulation or excision, if detected during the procedure.

Data were analysed using SPSS for Windows (SPSS 14.0; SPSS, Inc., Chicago, IL, USA). Normally distributed continuous data from two groups were analysed using a two-sided independent samples Student’s t test and when paired, the paired samples t test. Categorical data were analysed using Pearson’s chi-square test. Forward stepwise logistic regression analysis was used to calculate the adjusted odds ratios (OR) for continued menstrual/cyclical pain, continued vaginal bleeding and patient satisfaction.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Contribution to authorship
  9. Details of ethics approval
  10. Research governance
  11. Funding
  12. Acknowledgements
  13. References

A total of 315 eligible women were identified as having had an LSH during 2004 and 2005 and were therefore sent a questionnaire. Seven women could not be contacted; six had moved to unknown addresses, and one had died from a nongynaecological condition. Two hundred and forty women returned the questionnaire, providing a response rate of 78%.

Of the 308 procedures, 153 (50%) were performed in 2004 and 155 (50%) in 2005. The response rates for 2004 and 2005 were 75 and 81%, respectively. There were no significant differences between responders and nonresponders in terms of age and incidence of repeated surgery.

Reasons for having the hysterectomy

Most women (70%) stated two or more reasons for having the hysterectomy, the dominating reasons being fibroids (68%) and/or heavy bleeding (67%). About 46% of women stated pain as a reason for hysterectomy and 16% stated endometriosis.

Menstrual bleeding

Thirteen women (5%) had reached the menopause before the hysterectomy, and two had medically induced amenorrhoea. Self-reported preoperative menstrual data were available for 220 women and the majority reported their menstruation to be very heavy (43%) or heavy (25%). The remaining women reported their preoperative bleeding as moderate (18%), normal (11%) and minimal (3%).

After the hysterectomy, 57 women (24%) reported continued menstrual bleeding, the majority of these women experiencing irregular bleeding patterns (44%). About 28% of women experiencing persistent bleeding reported regular bleeding, 12% bleeding in relation to sexual activity and the remaining 16% reported a combination of regular and irregular bleeding that was also related to sexual activity. The amount of continued menstrual bleeding was reported to be minimal in 90% of the women, and less than their normal preoperative periodic bleeding in the remaining 10%. The mean degree of bother caused by vaginal bleeding after the LSH, scored on a 10-point VAS, was 1.13 (SD 1.95).

Continued menstrual bleeding was less likely to be reported by women who had a hysterectomy because of pain (OR 0.41, 95% CI: 0.20–0.85), fibroids (OR 0.47, 95% CI: 0.23–0.93) and in older women (OR 0.89, 95% CI: 0.83–0.95) (Table 1). Continued menstrual bleeding following surgery was more likely to be reported in women with heavy periods prior to surgery (OR 4.07, 95% CI: 1.32–12.57) and in women treated by less experienced surgeons (OR 2.56, 95% CI: 1.24–5.27). About 50% of women stated that they had not been aware of the possibility that they might experience postoperative continued menstrual bleeding.

Table 1.  Adjusted risk estimates for experiencing persistent vaginal bleeding up to 3 years following LSH
 Number of women with vaginal bleeding (%)OR (95% CI)*
  • *

    Forward stepwise logistic regression analysis.

Pain as a reason for hysterectomyYes23 (21.5)0.41 (0.20–0.85)
No34 (26.2) 
Fibroids as a reason for hysterectomyYes33 (20.1)0.47 (0.23–0.93)
No24 (31.2) 
Heavy bleeding prior to hysterectomyYes41 (26.3)4.07 (1.32–12.57)
No16 (19.8) 
Experience of surgeonLess experienced20 (36.4)2.56 (1.24–5.27)
Experienced37 (20.3) 
AgeYears57 (23.8)0.89 (0.83–0.95)

Pain

A total of 178 women (74%) suffered from menstrual pain before the hysterectomy, with a mean score of 6.81 (SD 2.05) on a 10-point VAS. Eighty-nine women (38%) reported continued menstrual/cyclical pain up to 3 years following their hysterectomy, although their mean score of 3.53 (SD 2.16) was significantly less than before surgery (mean pain reduction 3.32, 95% CI: 2.72–3.92, P < 0.01). The intensity of pain prior to surgery varied according to the reasons leading to the hysterectomy (Table 2). Women having a hysterectomy for endometriosis or for pain, reported a mean preoperative pain score of 7.47 (SD 2.45) and 7.13 (SD 2.57), respectively.

Table 2.  Pain intensity prior to and after surgery by reasons for hysterectomy
Reason for hysterectomynMean pain score prior to surgery (SD)Mean pain score up to 3 years after surgery (SD)Mean difference in pain scores (SD)95% CI of differenceP value*
  • *

    Paired samples t test.

Fibroids1514.76 (3.42)1.08 (1.89)3.69 (3.34)3.15–4.22<0.001
Endometriosis377.47 (2.45)3.46 (2.91)4.01 (3.97)2.69–5.34<0.001
Heavy bleeding1515.64 (3.20)1.27 (2.03)4.38 (3.37)3.23–4.91<0.001
Pain1047.13 (2.57)1.85 (2.50)5.28 (3.54)4.59–5.97<0.001

While all women reported a significant decrease in the amount of pain experienced following hysterectomy, women having a hysterectomy because of endometriosis reported significantly higher levels of remaining menstrual/cyclical pain after surgery (mean pain score 3.46, SD 2.91), compared with women who did not report endometriosis as a reason for the hysterectomy (mean pain score = 0.94, SD 1.74) (mean difference: 2.43, 95% CI: 1.74–3.12, P < 0.001).

As the adjusted OR reveal (Table 3), increased intensity of preoperative pain resulted in a greater chance of experiencing pain postoperatively (OR 1.14, 95% CI: 1.04–1.25). Women reporting endometriosis as the reason for having the hysterectomy were also more likely to experience pain up to 3 years following surgery (OR 4.88, 95% CI: 2.10–11.34) as were women who had their surgery in 2004, when compared with women having surgery in 2005 (OR 1.89, 95% CI: 1.04–3.40).

Table 3.  Adjusted risk estimates for experiencing pain up to three years following LSH
 Number of women with pain (%)OR (95% CI)*
  • *

    Forward stepwise logistic regression analysis.

Pain prior to surgery10-point scale89 (37)1.14 (1.04–1.25)
Endometriosis as a reason for hysterectomyYes27 (73)4.88 (2.10–11.34)
No62 (31) 
Year of surgery200448 (31)1.89 (1.04–3.40)
200541 (26) 

Repeated surgery

In total, 22 women (7%) had further related surgery in the same department after the LSH, with the key procedures being adhesiolysis (n = 6) and extirpation of cervix uteri (n = 7) (Table 4). Of the seven women who went on to have their cervix removed (due to cervical stump symptoms), three had their original hysterectomy because of endometriosis. Of the women who reported experiencing continued menstrual bleeding after the hysterectomy, 7% underwent repeated surgery.

Table 4.  Surgical procedures performed after LSH
Surgical procedure after LSHNumber of patients (%)
Laparoscopic adhesiolysis6 (1.9)
Laparoscopic extirpation of cervix uteri7 (2.3)
Laparoscopic drainage of postoperative abscess1 (0.3)
Laparotomy with resection of bowel due to postoperative peritonitis1 (0.3)
Laparoscopic bilateral salpingo-oophorectomy and removal of cervix uteri (sarcoma uteri)1 (0.3)
Scar correction3 (1.0)
Umbilical hernia repair1 (0.3)
Tension-free vaginal tape procedure2 (0.6)

New symptoms following hysterectomy

Fifty-three (22%) women reported experiencing new symptoms following their hysterectomy, with the majority of symptoms appearing to relate to the menopause (vasomotoric symptoms, vaginal dryness, depression and gained weight) (Table 5).

Table 5.  New symptoms reported following the hysterectomy
New symptomsn (%)
Changes related to urination and/or defecation9 (3.8)
Vasomotor symptoms5 (2.1)
Vaginal discharge5 (2.1)
Vaginal dryness5 (2.1)
Pelvic pain5 (2.1)
Other (dyspareunia, depression, gained weight, reduced libido, cystitis, candida infection and fear of cervical cancer)24 (10.0)

Satisfaction with surgery

Almost all women reported being satisfied (20%) or very satisfied (70%) with the hysterectomy. Women who reported having a new symptom following their hysterectomy were less likely to report being very satisfied (OR 0.20, 95% CI: 0.10–0.38), as were women who had repeated surgery following their hysterectomy (OR 0.06, 95% CI: 0.01–0.53). Women who reported being satisfied with the preoperative information were more likely to report being very satisfied with the hysterectomy (OR 3.31, 95% CI: 1.78–6.15).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Contribution to authorship
  9. Details of ethics approval
  10. Research governance
  11. Funding
  12. Acknowledgements
  13. References

This is the first large study to report both the occurrence of long-term outcomes following LSH as well as the impact that these outcomes have on women’s experiences. The occurrence of continued menstrual bleeding in our study is high compared with the results of previous studies.6,7,11–13 This may partly be explained by different definitions of ‘vaginal bleeding’ in the different reports. We have included both regular and irregular bleeding as well as bleeding related to sexual activity in our study, whereas some previous studies have only reported the occurrence of vaginal bleeding on a regular monthly basis.6,12,13 When we only included regular bleeding in our analysis, the proportion decreased from 24 to 8%. Moreover, others included regular and irregular bleeding after the hysterectomy, but did not report bleeding related to sexual activity.7

Poor surgical technique resulting in uterine amputation above the level of the internal cervical ostium has been suggested as a cause of the high occurrence of vaginal bleeding following LSH.6 The results of our study, showing a significantly higher rate of postoperative vaginal bleeding in patients treated by less experienced surgeons, support this assertion. It is also possible that more meticulous destruction of any remaining endometrium by electrocoagulation of the cervical canal following the amputation is carried out by experienced surgeons. However, this remains to be demonstrated. Although we did not collect data relating to surgeon experience and complexity of hysterectomy, it is likely that the more experienced surgeons undertook the more complex hysterectomies and therefore, we may be underestimating the increased risk of continued menstrual bleeding that occurs with less experienced surgeons.

Although around one-quarter of women in our study reported continued vaginal bleeding, they rarely found the bleeding to be bothersome, and it did not affect their overall satisfaction with the hysterectomy, despite our finding that 50% of the women were not aware of the possibility of persistent bleeding. Since we also found that women who reported heavy bleeding prior to their hysterectomy were more likely to experience bleeding after surgery, it is likely that this residue bleeding represents a vast improvement from their preoperative state and therefore creates minimal bother. Indeed, only 7% of women with continued vaginal bleeding had repeated surgery. This proportion is modest compared with the results presented by Okaro et al., in which all women with cervical stump symptoms underwent repeated surgery.7

The proportion of women who reported suffering from preoperative menstrual pain in this study was relatively high, with almost half of the women reporting pain as a reason for hysterectomy. Although the postoperative pelvic pain was reduced after the procedure, a relatively large proportion of women, particularly those with endometriosis, reported continued pain. These results are similar to those previously reported by Ewen and Sutton.19 It has been suggested that LSH may be an inappropriate procedure in the presence of preoperative pelvic pain and/or endometriosis due to the risk of persistent pelvic pain, with some authors stating that endometriosis should be considered as a contraindication for LSH.14–16 Although the results of our study indicate a less favourable outcome following LSH in patients with endometriosis, these women did report a significant reduction in pain following their hysterectomy, and it is possible that they would not have had any greater reduction in pain if they underwent a total hysterectomy. To date, there is no evidence from randomised control trials to suggest that total hysterectomy in women with endometriosis results in greater pain reduction than subtotal hysterectomy. Furthermore, the effect of a subsequent removal of the cervix following an LSH due to pain does not appear to have been reported.

As cervical stump symptoms appear to be relatively common following LSH, it is important that women are informed preoperatively regarding the risk of persistent menstrual bleeding and/or pain. In our study, women who were not satisfied regarding information reported a significantly lower degree of total patient satisfaction, which illustrates the importance of proper and adequate preoperative information.

The strength of our study is the relatively large sample size, with an excellent response rate compared with previous studies reporting long-term outcomes following LSH.6,7,11–13 The major limitation of the study is that the data were collected retrospectively, and that we were not able to compare the results of LSH to those of TLH as would be possible with a controlled trial. While women may not have been able to recall their preoperative symptoms with great precision, we believe that they were able to present a reasonably accurate view of the whole experience from the pre- to postoperative time, and therefore the change in symptoms is likely to reflect improvements or deteriorations. While a controlled trial would provide useful comparative data, it is likely to be difficult to recruit women to take part in a trial as they may have personal wishes about keeping their cervix or not.

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Contribution to authorship
  9. Details of ethics approval
  10. Research governance
  11. Funding
  12. Acknowledgements
  13. References

This study demonstrates that while continued menstrual bleeding is relatively common following LSH, it is reported by women to be minimal and rarely bothersome. Consequently, patient satisfaction following the procedure is high and is not influenced by the presence of continued menstrual bleeding.

All women should be informed preoperatively regarding the risk of vaginal bleeding and/or pelvic pain. Women with endometriosis should be informed regarding the possibility of persistent pelvic pain and the increased risk of repeated surgery following the procedure. Furthermore, the surgical technique during LSH may be modified to reduce the occurrence of menstrual bleeding after the hysterectomy.

Contribution to authorship

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Contribution to authorship
  9. Details of ethics approval
  10. Research governance
  11. Funding
  12. Acknowledgements
  13. References

This study was submitted as M.L.’s dissertation for the MSc in Advanced Gynaecological Endoscopy, Postgraduate Medical School, University of Surrey, UK. All authors have contributed to the planning, performance and writing of the publication.

Details of ethics approval

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Contribution to authorship
  9. Details of ethics approval
  10. Research governance
  11. Funding
  12. Acknowledgements
  13. References

Approval from the Regional Committee for Medical Research Ethics in Eastern Norway (Regional etisk komite, Helse Øst) was granted for the study protocol on 26 June 2006 (reference number: 415-06177 1.2006.1976).

Research governance

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Contribution to authorship
  9. Details of ethics approval
  10. Research governance
  11. Funding
  12. Acknowledgements
  13. References

Permission from Ullevål Hospital’s Advisory Committee on the Protection of Patient Records (personvernansvarlig) was granted for on 15 May 2006 (reference number: 684).

Funding

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Contribution to authorship
  9. Details of ethics approval
  10. Research governance
  11. Funding
  12. Acknowledgements
  13. References

The Department of Gynaecology, Ullevål University Hospital covered expenses including printing and posting of the survey documents, as well as prepaid envelopes. No additional funding was required for the survey.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Contribution to authorship
  9. Details of ethics approval
  10. Research governance
  11. Funding
  12. Acknowledgements
  13. References

The authors thank Anne Birte Lømo for practical assistance during the performance of the survey.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Contribution to authorship
  9. Details of ethics approval
  10. Research governance
  11. Funding
  12. Acknowledgements
  13. References