Trial registration: Nederlands Trial Register: Trial number NTR 1795. Acronym: Ligasure (http://www.trialregister.nl/trialreg/admin/rctview.asp?TC=1795).
Electrosurgical bipolar vessel sealing versus conventional clamping and suturing for vaginal hysterectomy: a randomised controlled trial
Article first published online: 24 AUG 2012
© 2012 The Authors BJOG An International Journal of Obstetrics and Gynaecology © 2012 RCOG
BJOG: An International Journal of Obstetrics & Gynaecology
Volume 119, Issue 12, pages 1473–1482, November 2012
How to Cite
Lakeman, M., The, S., Schellart, R., Dietz, V., ter Haar, J., Thurkow, A., Scholten, P., Dijkgraaf, M. and Roovers, J. (2012), Electrosurgical bipolar vessel sealing versus conventional clamping and suturing for vaginal hysterectomy: a randomised controlled trial. BJOG: An International Journal of Obstetrics & Gynaecology, 119: 1473–1482. doi: 10.1111/j.1471-0528.2012.03484.x
- Issue published online: 11 OCT 2012
- Article first published online: 24 AUG 2012
- Accepted 17 July 2012. Published Online 24 August 2012.
- electrosurgical bipolar vessel sealing;
- pain score;
- vaginal hysterectomy
Please cite this paper as: Lakeman M, The S, Schellart R, Dietz V, ter Haar J, Thurkow A, Scholten P, Dijkgraaf M, Roovers J. Electrosurgical bipolar vessel sealing versus conventional clamping and suturing for vaginal hysterectomy: a randomised controlled trial. BJOG 2012;119:1473–1482.
Objective To compare the effects of electrical bipolar vessel sealing and conventional suturing on postoperative pain, recovery, costs and micturition symptoms in women undergoing vaginal hysterectomy.
Design Randomised controlled trial.
Setting Eight teaching hospitals in the Netherlands.
Population One hundred women scheduled to undergo vaginal hysterectomy for benign conditions excluding pelvic organ prolapse.
Methods Women were randomised to vessel sealing or conventional surgery. The quality of life related to pelvic floor function was assessed using validated questionnaires before surgery and 6 months after surgery. Pain scores and recovery were assessed using a diary, including daily visual analogue scale scores, starting from the day before surgery until 6 weeks after surgery.
Main outcome measures Visual analogue scale pain scores, surgery time, blood loss, complications, quality of life related to pelvic floor function and costs.
Results The evening after surgery, women in the vessel-sealing group reported significantly less pain (5.7 versus 4.5 on a scale of 0–10, P = 0.03), but after that pain scores were similar. Operation duration was shorter for vessel sealing (60 versus 71 minutes, P = 0.05). Blood loss and hospital stay did not differ. We observed no major difference in costs between the two interventions (2903 versus 3102 €, P = 0.26). Changes in micturition and defecation symptoms were not affected by the surgical technique used.
Conclusion Using vessel sealing during vaginal hysterectomy resulted in less pain on the first postoperative day, shorter operating time, similar morbidity and similar pelvic floor function. No major differences in costs were found between the two interventions.
Hysterectomy is one of the most commonly performed gynaecological procedures. The commonest indications for hysterectomy are all benign conditions such as abnormal uterine blood loss, dysmenorrhoea, abdominal pain or mechanical obstruction caused by an enlarged uterus.1,2 The main goal of removal of the uterus is to improve quality of life. Therefore possible side-effects of hysterectomy, which negatively influence the quality of life, have to be carefully evaluated and if possible limited.
If technically feasible, a vaginal approach is preferred based on the reported benefits of shorter hospitalisation, fewer complications and lower costs.2,3 One of the main limitations of the vaginal route is that it offers relatively limited space for surgical access to vascular pedicles.4 To create adequate visualisation, traction is applied to the tissue. This might cause not only increased postoperative pain, but also nerve damage, possibly explaining the increased rate of postoperative micturition symptoms found after vaginal hysterectomy.5,6
Electrosurgical bipolar vessel sealing systems have been developed to seal large tissue bundles and blood vessels up to 7 mm in diameter. By using this technique only one clamp has to be inserted through the vagina to secure the vessels and cut the tissue, instead of one clamp and one scissors. This might shorten the operation duration as the result of a limitation in surgical steps, but might also reduce postoperative pain because less traction is needed to pull the uterus downward, which theoretically reduces postoperative pain. Furthermore, using a single clamp, as when using a vessel-sealing device, allows the surrounding tissues to be cut closer to the uterus, which is important in preventing innervational damage, because the more lateral aspects of the uterine supportive ligaments are known to be more densely innervated.7 By limiting the traction applied to the tissue and the innervational damage, the development of micturition symptoms after hysterectomy might be prevented.
A few previous randomised controlled trials have proved this technique to be safe and efficacious, and reported possible advantages of using vessel sealing such as less postoperative pain, shorter operation time, less blood loss and shorter hospital stay.4,7–10 However the effect of these findings on postoperative recovery, micturition symptoms and on the costs for the procedure has never been examined before.
The first aim of this study is therefore to compare the effect of vessel sealing and conventional clamping on postoperative pain, recovery and micturition symptoms and second, to evaluate the costs of using vessel sealing during vaginal hysterectomy.
A multicentre randomised controlled trial comparing bipolar vessel sealing with conventional suturing in vaginal hysterectomy was performed in eight teaching hospitals in the Netherlands. Women scheduled to undergo vaginal hysterectomy for benign indications, were included between April 2009 and March 2011. Exclusion criteria were: symptomatic prolapse and concomitant procedures. In all hospitals the medical ethics committee approved this study. After the women had signed the informed consent they were allocated to one of the treatment groups by computerised randomisation. Randomisation was performed using block randomisation with a block size of ten and 1 : 1 allocation of treatment. Assignment of the treatment was only known by the operating physician. Both the women and the medical team responsible for the postoperative care were blinded to the allocated intervention. Concealment of allocation was assured by not providing information to the woman during the first postoperative year about which surgical technique was used. Also, the nurses and physicians involved in the postoperative care were not informed about treatment allocation.
Abdominal hysterectomy was performed if the size and descent of the uterus did not allow vaginal hysterectomy. Gynaecologists and residents performing the hysterectomies in this study had none or limited experience with the use of vessel-sealing techniques during vaginal hysterectomy. They were all given extensive training on its working mechanism and use before the start of the study.
Vaginal hysterectomy was performed according to a standardised protocol. A circumferential vaginal incision was made around the anterior portion of the cervix below the line of the bladder. The pouch of Douglas was opened posteriorly and a retractor was postioned. The bladder was then dissected from the vagina anteriorly. The sacrouterine ligaments were clamped transected and sutured using Vicryl No. 0 sutures. These ligatures were left long to suspend the vault later. In the case of conventional technique, the other pedicles were clamped, transected and sutured using Vicryl No. 0 sutures. If case the woman was assigned to the vessel-sealing technique the pedicles were clamped, transected and sealed using the Ligasure™ Impact (LF4200) in combination with the ForceTriad™ energy platform (Covidien; Tyco Healthcare, Valleylab, CO, USA). The vault was closed similarly in both groups using Vicryl No. 1 sutures. Double suturing was not routinely performed and no effort was taken to prevent back-bleeding from the uterus.
The amount of blood loss was the sum of the volume collected by a suction device during the procedure and the estimated volume of the total number of gauzes used during the procedure. This amount was estimated by the operation assistant.
Pain medication was administered according to the hospital protocol. This included morphine on indication and nonsteroidal anti-inflammatory agents in combination with paracetamol until 3 days after surgery. At home women could take pain medication on indication. The amount of pain medication used was reported in the pain diary.
Primary outcome was postoperative pain, measured by the visual analogue scale (VAS) during the first week after surgery. Secondary outcomes were duration of surgery, amount of blood loss, complications during surgery or hospital stay, late complications related to surgery, time to resume normal daily activity, quality of life and costs.
All women were asked to complete the following validated, disease-specific, quality-of-life questionnaires before surgery and 6 months after surgery:
- 1The Urogenital Distress Inventory (UDI) assesses the presence and experienced bother of pelvic floor symptoms.11,12 The UDI scores range from zero to 100, with zero representing women without any symptoms and 100 representing women who encounter all possible symptoms and experience these all as maximally bothersome. In the Dutch validated version, symptoms measured with the UDI are divided into five domains: discomfort/pain, urinary incontinence, overactive bladder, genital prolapse and obstructive micturition.
- 2The Defecation Distress Inventory (DDI) assesses the presence and experienced bother of defecation symptoms.12,13 The questions are divided into five domains: constipation, obstructed defecation, pain, incontinence for faeces and incontinence for flatulence. The DDI scores range from zero to 100, with zero representing women without any symptoms and 100 representing women who encounter all possible symptoms and experience these all as maximally bothersome.
- 3The Incontinence Impact Questionnaire (IIQ) measures the impact of micturition and defecation symptoms on different aspects of quality of life.11,12 The questions are divided into five domains: mobility, physical functioning, social functioning, emotional functioning and embarrassment. The IIQ scores range from zero to 100, with zero representing women who experience no bother related to micturition and defecation symptoms and 100 representing women who experience maximal bother.
- 4The RAND-36 is a questionnaire that was developed in the USA in the late 1980s as part of the Medical Outcomes Study, a longitudinal investigation of the self-reported health status of women with a range of chronic conditions.14 It consists of 36 items organised into nine scales (general health perceptions, health change, physical functioning, role limitations caused by physical problems, role limitations caused by emotional problems, social functioning, bodily pain, vitality, and mental health). Each scale yields a score in the range from zero (worst possible score) to 100 (best possible score).
Women were asked to keep a diary starting from the day before surgery until 6 weeks after surgery. They were instructed to answer questionnaires in the evening and to do this at a fixed time. The diary consisted of several questionnaires:
- 1A VAS ranging from 0 to 100 to assess the experienced pain was filled in daily during the first week after surgery and 2 and 6 weeks thereafter by all women.
- 2A self-developed questionnaire consisting of questions to evaluate the presence and experienced bother of bodily pain, having a bladder catheter, having food restrictions, being impaired in self-care, having intravenous infusion, undergoing investigations (e.g. radiograph, blood samples), and having impaired mobility after hysterectomy. Each of these questions inquired whether any type of discomfort was present, and quantified the degree of discomfort experienced from each particular symptom. This was recorded on a prespecified categorical response scale: not at all, slightly, moderately, or very much. Patients answered this questionnaire daily during the first week after surgery.
Unit costing and costs
The costs associated with vessel sealing versus conventional clamping and suturing for vaginal hysterectomy have been described from the perspective of the healthcare provider. The time horizon was 6 weeks following randomisation.
Costs of the index operation and admission, re-interventions and readmissions, blood transfusions, outpatient hospital care and emergency room, and out-of-hospital care by the general practitioner, physiotherapist and company doctor have been calculated as the product sum of recorded healthcare use and the corresponding unit costs. In-hospital healthcare use was recorded by clinical report form, whereas outpatient hospital and out-of-hospital use of resources were gathered with a patient questionnaire at 6 weeks following randomisation. In the absence of a reason to expect the type of surgery to influence the time needed to prepare for the operation or the time of recovery and in holding bay, pre-calculation unit costs of the index procedures (Vessel sealing versus Conventional approaches) as well as vaginal or abdominal reoperations following bleeding incidents were derived from expert-opinion-based estimates of the input of personnel (times of operator, anaesthesiologist, pharmacist, nursing), material (disposables), and apparatus (based on depreciation and production) as well as the overhead involved. Only the time in the operating theatre was gathered empirically from recordings of the operation duration for each woman and subsequently averaged over women by type of surgery.
Unit costs of blood transfusions (packed cells) and outpatient hospital and out-of-hospital healthcare components were derived from the 2010 Dutch costing guideline for healthcare research.15 After price-indexing, all costs were expressed in Euros calculated for the base year 2010. All unit costs are presented in Table 1.
|Unit||Unit costs (€)|
|Inpatient hospital care|
|Index admission and readmission, general ward||Day||463||Dutch costing guideline 2010|
|Vessel sealing||Session||1328||Consisting of the costs of staff and time in theatre (€616), pharmacy (€68), vessel-sealing device (€445), depreciation of force triad (€2), and other materials including depreciation (€197). Except for the vessel-sealing device and the use of the force triad, these costs include the costs of overhead and housing|
|Conventional surgery||Session||953||Consisting of the costs of staff and time in theatre (€688), pharmacy (€68), and other materials including depreciation (€197). These costs include the costs of overhead and housing|
|Reoperation vaginally||Session||886||Consisting of the costs of staff (€622), pharmacy (€68), and other materials including depreciation (€197). These costs include the costs of overhead and housing|
|Reoperation abdominally||Session||1263||Consisting of the costs of staff (€979), pharmacy (€140), and other materials including depreciation (€144). These costs include the costs of overhead and housing|
|Outpatient hospital and out-of-hospital care|
|General practitioner||Visit||28||Dutch costing guideline 2010|
|Outpatient hospital||Visit||73||Dutch costing guideline 2010|
|Paramedic||Visit||36||Dutch costing guideline 2010|
|Emergency room||Visit||153||Dutch costing guideline 2010|
|Company doctor||Visit||28||Dutch costing guideline 2010|
Mean postoperative pain during the first week after surgery as measured by the VAS was taken as the primary outcome. Based on a previous study a difference in postoperative pain of 15% (pain score of 40 versus 34) was considered to be clinically relevant.16 With a power of 80%, a two-sided significance level of 0.05 and a standard deviation of 10, a total of 90 women (45 in each group) was needed using a Student’s t-test for unpaired data. Anticipating a 10% drop out, we aimed to include 100 women (50 in each group).
Analysis was performed according to the intention-to-treat principle. Surgical outcomes were compared using unpaired t-tests for continuous variables and Pearson’s chi-square test for dichotomous variables. The VAS scores were compared using analysis of variance repeated measurement analysis in which the daily VAS score was treated as the within-subject repeated measure and the surgical technique used as a between-subject categorical factor. The domain scores of the UDI, DDI and IIQ were calculated. To account for preoperative differences in UDI and DDI domain scores we decided to calculate P values using repeated measurement. Survival analysis by log-rank test was used to compare the difference in time to resume normal daily activities between both interventions. A P value below 0.05 indicated statistical significance.
Because of skewed distributions, costs differences were assessed by calculating 95% confidence intervals for the mean differences after correction for bias and using accelerated non-parametric bootstrapping, drawing 1000 samples of the same size as the original sample separately for each group and with replacement.17 Statistical analyses were performed using software (SPSS 18.0; SPSS, Inc., Chicago, IL, USA).
Figure 1 shows the Consort flow diagram of this study. From 1 June 2009 to 1 May 2011, 100 women participated in this study. At 6 months after surgery only 75% of the women returned the questionnaire.
In Table 2 the participants’ characteristics are shown. At baseline, women receiving conventional surgery were older by 4 years on average. This difference was mainly caused by three women aged over 70 years who were all randomised in the conventional group. The indication for hysterectomy was nonsignificantly more often dysmenorrhoea in the vessel-sealing group but this did not lead to the more regular use of pain medication preoperatively. We analysed if there were any imbalances in treatment assignment between the eight participating hospitals, none were found (data not shown). As most of the procedures were performed by trainees we also analysed if the experience of the surgeon differed between both treatment arms, no difference was observed.
|Conventional surgery||Vessel sealing|
|Number of women||51||49|
|Age (years)*||48.1 (10.0)||44.0 (6.3)|
|Parity (median, range)||2.0 (1–5)||2.0 (0–6)|
|Body mass index (kg/m2)||27.9 (5.0)||26.5 (5.4)|
|Previous abdominal surgery†||23 (51.1%)||27 (57.4%)|
|Comorbidity†||23 (52.3%)||16 (34.0%)|
|Chronic obstructive pulmonary disease||6||7|
|Main indication for hysterectomy|
|Abnormal uterine blood loss||37||33|
|Postmenopausal blood loss||2||0|
|Preoperative use of pain medication|
|Yes||15 (29%)||20 (41%)|
|No||21 (41%)||17 (35%)|
|Missing||15 (29%)||12 (24%)|
Table 3 shows the operative parameters for both techniques. The operating time was shorter in the vessel-sealing group (59.7 versus 71.3 minutes P = 0.05), whereas the amount of blood loss and duration of hospital stay did not differ. Two women, one in each group, were converted to abdominal procedure during the initial procedure because vaginal removal of the uterus was not possible. Three women needed to return to the theatre because of bleeding complications: two women from the conventional group, of whom one underwent an abdominal re-operation during admission and the other was re-admitted to undergo a vaginal re-operation. One woman from the vessel-sealing group was re-admitted for a vaginal re-operation. Two more women were re-admitted to the hospital; one for observation of bleeding, which appeared to be in the normal range, and one woman was admitted because of pain complaints which were self-limiting and thought to be ovulation pain.
|Conventional surgery (n = 51)||Vessel sealing (n = 49)||P value|
|Duration of surgery (minutes)||71.3 (27.0)||59.7 (28.3)||0.05|
|Blood loss (ml)||247.7 (206.4)||231.4 (263.4)||0.74|
|Hospital stay (days)||4.0 (1.5)||3.7 (0.9)||0.20|
|Complications during surgery|
|Bladder perforation||1 (2%)||1 (2%)||1.0|
|Blood loss >500 ml||3 (6%)||3 (6%)||1.0|
|Complications after surgery|
|Haematoma||1 (2%)||0 (0%)|
|Blood transfusion||2 (4%)||2 (4%)||1.0|
|Re-operation||2 (4%)||1 (2%)||0.58|
|Re-admission to the hospital||2 (4%)||3 (6%)||0.61|
In Figure 2 the VAS scores assessing postoperative pain are shown. Using repeated measurement analyses we found no significant effect of the surgical technique on postoperative pain during the first week after surgery (P = 0.71). We found that only the evening after surgery, postoperative pain scores differed for the two performed surgical techniques (P = 0.02). Starting from the first day after surgery, pain scores were comparable, as was the amount of administered pain medication (data not shown). Days till return to normal daily activity, as indicated by the woman, were also comparable (vessel-sealing group 27.2 (SE 2.4) days, conventional group 28.6 (SE 2.0) days, P = 0.66). We performed a subanalysis by excluding the three women aged over 70 to make both groups comparable by age. After this adjustment, no significant effect was found of surgical technique on pain scores during the first postoperative week, and on postoperative recovery. We also added treatment centre to the repeated measurement analyses to exclude an interaction between used surgical technique and centre of treatment, no interaction was found (P = 0.39).
In Table 4 the preoperative and postoperative UDI and DDI domain scores are shown. To account for preoperative differences in UDI and DDI domain scores repeated measurement analyses were performed. No statistically significant differences between interventions were observed.
|Before surgery||6 months after surgery||P value*|
|Conventional surgery||Vessel sealing||Conventional surgery||Vessel sealing|
|Overactive bladder||24.3 (4)||13.5 (3)||12.8 (4)||13.5 (4)||0.77|
|Incontinence||26.1 (4)||14.5 (3)||19.2 (3)||9.7 (3)||0.58|
|Obstructive micturition||10.6 (3)||12.1 (3)||7.7 (2)||9.3 (3)||0.71|
|Pain||31.8 (4)||35.1 (4)||14.0 (3)||16.7 (4)||0.55|
|Prolapse||7.6 (3)||6.0 (2)||0.0 (0)||1.0 (1)||0.42|
|Constipation||12.3 (3)||10.7 (3)||10.2 (4)||14.7 (3)||0.20|
|Obstructive defecation||7.2 (3)||7.4 (2)||6.0 (3)||6.9 (3)||0.82|
|Pain||7.6 (3)||7.8 (3)||6.6 (3)||6.6 (3)||0.73|
|Faecal incontinence||0.8 (1)||2.2 (1)||0.4 (0)||2.3 (2)||0.66|
|Flatulence incontinence||16.7 (4)||15.2 (4)||11.1 (3)||15.7 (5)||0.06|
The costs related to both interventions are shown in Table 5. The inpatient hospital costs were nonsignificantly higher in the vessel-sealing group (2903 versus 3102 €, mean cost difference 198 €, P = 0.26). It is mainly explained by the extra costs of the Ligasure device and force triad (i.e. 447 €). However, the decreased staff costs as a result of the shorter operation duration when performing vessel sealing (conventional surgery 688 € versus vessel sealing 616 €) and the lower costs of hospital stay after vessel sealing (conventional surgery 1852 € versus vessel sealing 1713 €) partly compensate for the extra costs of the vessel-sealing device.
|Conventional surgery||Vessel sealing||Vessel sealing minus Conventional surgery||P value|
|n||Mean costs (95% CI)||n||Mean costs (95% CI)||Mean costs (95% CI)|
|Inpatient hospital care||51||2903 (2651–3225)||49||3102 (2958–3250)||198 (−139 to 473)||0.257|
|Outpatient hospital and out-of-hospital care||35||57 (37–79)||37||115 (71–167)||57 (10–113)||0.037|
|Total||51||2943 (2692–3264)||49||3188 (3040–3341)||246 (−121 to 545)||0.179|
The costs of outpatient hospital and out-of-hospital care (i.e. care by the general practitioner, physiotherapist and company doctor) were significantly higher in the vessel-sealing group compared with conventional surgery. This is mainly explained by four women, who were all randomised into the vessel-sealing group, and had an extensive number of outpatient hospital visits: one woman because of constipation, one because of pain complaints, one because of urinary tract infection and one because of palpitations. The groups did not differ significantly by the total costs.
We performed a randomised controlled trial to evaluate the effects of vessel sealing on postoperative pain, surgery-related parameters, pelvic floor symptoms and costs. We found that postoperative pain did not appear to differ after the first night after surgery. The use of vessel sealing reduced the time of the procedure. The quality of life related to pelvic floor function was similar following both techniques. No statistically significant differences in costs emerged.
Before further interpreting these data some issues need to be addressed.
Despite proper computerised randomisation the conventional group was older than the vessel-sealing group. This difference was mainly caused by three women aged over 70 who were all randomised in the conventional group. When analysing pain scores and recovery without these three women, still no major differences occurred.
This study was performed in eight different hospitals; one could argue that because of the numerous different surgeons, differences in operative statistics could be masked. However, because this is also the case in daily practice, we think that it only increases the amount to which these results can be generalised.
When considering the indication for hysterectomy, nonsignificantly more women were operated for dysmenorrhoea in the vessel-sealing group. This might influence the perception of postoperative pain, which was our primary outcome. However, as most of the time this was a combination of dysmenorrhoea and menorrhagia, and no significant differences in the preoperative regular use of pain medication were found, we do not think this influenced our findings.
Furthermore, our loss to follow up was larger than expected. At 6 months after surgery we sent the questionnaire to the participant’s home address, nonresponders received a second questionnaire 1 month after the first one and 2 weeks after sending the questionnaire nonresponders were contacted by phone. Despite these efforts, only 75% of the women returned the questionnaire. The moderate response rate can probably be explained by the fact that this was a quite ‘easy’ study for the woman to participate in, this might have reduced the commitment of the woman to participate in long-term followup. Furthermore the completion of the pain diary was time consuming, which might have caused women to stop completing it. The nonresponders were equally distributed over the different centres. While conducting the trial, differential loss to follow up occurred, 31% in the conventional group and 18% in the vessel-sealing group. To evaluate to what extent this differential loss to follow-up could have affected our results we studied whether women who did not respond to the pain diary or 6 months follow-up questionnaire differed by baseline (age, body mass index, parity) and operative characteristics (duration of surgery, blood loss, hospital stay, complications) from women who did respond. No clear patterns emerged.
To further evaluate the potential impact of our missing data on the reported results we studied three extreme scenarios: all loss to follow-up receives the worst pain score observed in that study group (scenario A) or the best pain score observed (scenario B), or only the women in the Ligasure group receive the worst pain score observed, while the women in the conventional group receive the best pain score observed.
The pain results (i.e. significantly increased pain in the conventional group on the evening following surgery, no difference during the remainder of the first week) were robust under scenario A and partly under scenario B. Under scenario C the conventional group experienced less pain during follow up. Hence, the robustness of our study results specifically failed in one scenario. Considering however that we did not find any differences in baseline and operative characteristics between women with and without missing data, scenario C may not be highly likely. We conclude that the differential loss to follow up among the studied groups in theory could affect the study results, but is actually unlikely to do so given the similarity of responders and non-responders to patient diaries and questionnaires on baseline and operative characteristics.
One of the strengths of this trial is that women as well as the medical staff (other than the operating surgeon) were blinded for the used technique. As a result, bias in the counselling on experienced pain, based on the women’s or the nurses’ preferences, was prevented. Another strong characteristic is that we used the same validated questionnaire before and after surgery to evaluate micturition and defecation symptoms.
Our main outcome was postoperative pain. In line with two previous studies reporting postoperative pain be reduced shortly after vaginal hysterectomy when using vessel sealing,8,10 we also found decreased postoperative pain the evening after surgery in the vessel-sealing group. However, we also found that this difference in postoperative pain disappeared during the first day after surgery. This is probably explained by the overall low pain scores found in this study, after the first postoperative day. We based our power analyses on results we found in a previous study among women scheduled for abdominal hysterectomy.16 We expected a difference in postoperative pain of 15% (pain score of 40 versus 34). From the first day after surgery pain scores were lower than these values. Why pain scores are so much higher after abdominal hysterectomy is probably because of the added pain of the abdominal incision, which adds to the pain of hysterectomy itself.3
In contrast with the low pain scores, the mean hospital stay might seem quite long in our study in comparison to previous studies.8,10,18 However in comparison to other studies among women scheduled for vaginal hysterectomy this hospital stay still falls within the normal range.3 The large spread in duration of hospital stay is probably explained by local cultural factors, and by surgeon, as well as participant, expectations.19 In line with a recent meta-analysis,19 hospital stay in our study population was somewhat shorter among women in the vessel-sealing group; however, this difference was not statistically significant.
In concordance with our findings, previous studies comparing vessel sealing with conventional clamping, all reported reduced operation time.4,7,9,10,18 The most plausible explanation for this finding is that energy-based vessel-sealing devices allow for rapid vessel sealing, coagulation and transection of the pedicle in one handheld tool.20 The clinical relevance of an 11.6-minute difference in operating time is a point of discussion, especially as this 15% shortening of operation duration did not seem to induce a quicker recovery or shorter hospital admission. However every minute shorter the operation lasts, does reduce the costs of the procedure. The 11-minute reduction in operating time already saves 72 €. This partly compensated for the extra costs of the device.
Some previous studies also found a decrease in blood loss using vessel sealing.9,10,18 We could not confirm such an advantage of the vessel-sealing technique.7,9,10,18. The studies that found an advantage in blood loss were mostly small studies, with serious methodological limitations.19 In these studies, the estimated blood loss was lower using both techniques than the values found in our study. This is probably explained by the exclusion of more complex hysterectomies in these studies. Another larger study that also included more complex hysterectomies did find an estimated blood loss in the same range as our measurements.7
We expected to find the vessel-sealing technique to be more expensive than the conventional technique, but the difference in costs remained limited to a nonsignificant 198 €. The extra costs of the Ligasure device and force triad of 447 € were compensated for in part by shorter operation duration (72 € saved) and the lower costs of hospital stay (139 € saved).
Before the study it was hypothesised that vessel sealing might reduce the prevalence and experienced bother of micturition and defecation symptoms after surgery through a reduction in damage to the autonomic innervation.21 During the first 6 months after surgery such a difference was not observed. To some extent, this can be explained by the rarity of postoperative pelvic floor disorders in both study groups, which is expressed by the low UDI and DDI domain scores at 6 months after surgery. The decrease in UDI and DDI domain scores at baseline may be explained by elimination of bothersome symptoms like menorrhagia or dysmenorrhoea, which have been shown to decrease women’s disease-specific quality of life.22–24 Another explanation may be that follow up was too short to observe a significant difference. Other previous studies with a short follow up were also not able to demonstrate an increase in urinary symptoms after hysterectomy,25–28 whereas studies with long-term follow up did find increased urinary incontinence and voiding difficulties.6,29–31
Surgery seems to be faster using vessel sealing. However, the shorter operation duration does not completely compensate for the costs of the Ligasure clamp. Patients may benefit from the use of a vessel-sealing technique because the postoperative pain during the first night after surgery is less.
Disclosure of interests
MMEL, TS, VD, JFH and MGWD have nothing to disclose. RS initiated a study with AMS. AT is a tutor for laparoscopic surgery for Johnson & Johnson, a faculty member of laparoscopic training course for fellows sponsored by Covidien, a member of the Olympus Surgical Advisory Group for Gynaecology, and part of the international faculty for training in LESS single port laparoscopy. JPWRR is an investigator who initiated studies with AMS and Pohl Boskamp.
Contribution to authorship
JPRWR and MMEL are responsible for the conception and design. Acquisition of data was performed by MMEL, ST, RS, VD, AH, AT, PS and JPWRR. Analysis and interpretation of data were performed by MMEL, MGWD and JPWRR and the manuscript was drafted by MMEL. Critical revision of the manuscript for intellectual content was carried out by ST, RS, VD, JFH, AT, PS, MGWD and JPWRR. MMEL and MGWD were responsible for the statistical analysis.
Details of ethics approval
The study received approval from the medical ethical committee of the Academic Medical Centre of Amsterdam on 3 March 2009. The study also received approval from the local medical ethical committees from the other six teaching hospitals in which the study was conducted.
This study was funded by an unrestricted research grant from Covidien. The study group was responsible for the study design, data collection and writing of the article. Covidien did not have access to the data or influence on the protocol, statistical analysis or manuscript writing.
- 3Surgical approach to hysterectomy for benign gynaecological disease. Cochrane Database Syst Rev 2006;2:CD003677., , , , , .
- 11Short forms to assess life quality and symptom distress for urinary incontinence in women: the Incontinence Impact Questionnaire and the Urogenital Distress Inventory. Continence Program for Women Research Group. Neurourol Urodyn 1995;14:131–9., , , , .
- 15Manual for Costing: Methods and standard costs for economic evaluations in health care. [In Dutch]. Amstelveen: Dutch Health Insurance Executive Board, 2010., , .