Abstract
Objective
To compare recovery after laparoscopic hysterectomy (LH) and abdominal hysterectomy (AH).
Design
A parallel, open, non-inferiority, multicentre, randomised controlled, expertise-based surgery trial.
Setting
10 NHS (National Health Service) hospitals within the UK.
Participants
Women undergoing hysterectomy for a benign gynaecological condition.
Interventions
Consenting women of 18–55 years were randomised to LH or AH using a secure internet facility by a surgeon with self-declared expertise. Major complications were recorded by clinicians, and recovery was assessed by regular text messaging and postal questionnaires.
Primary and secondary outcome measures
Major surgical complications (Clavien-Dindo≥level 3) up to six completed weeks postsurgery, time to resumption of normal activities measured by the Patient-Reported Outcomes Measurement Information System Physical Function tool and quality of recovery at 24 hours (Quality of Recovery 15 score; 0–150).
Results
75 women were randomised before early curtailment of the trial; 32/39 (82%) and 30/36 (83%) women underwent LH and AH, respectively. Major complications occurred in 2/32 (6%) LH versus 4/30 (13%) AH groups. No difference in time to resumption of usual activities was found (median [IQR, n] 7.5 weeks (3.6–8.2, 25) LH vs 7.5 weeks (5.5–10.6, 26) AH groups or quality of recovery (mean [SD, n] 81.1 (13.4, 27) vs 72.3 (17.6, 22), respectively; adjusted mean difference 7.2, 95% CI −3.2 to 17.6).
Conclusions
No differences were found in complications or recovery between LH and AH. However, early cessation of the trial due to recruitment challenges limits clinical inferences. It is important that larger comparative trials are conducted now that LH, including robotics, is becoming adopted as standard practice.
Trial registration number
ISRCTN14566195, IRAS ID 287988.
Keywords: GYNAECOLOGY, Minimally invasive surgery, SURGERY
STRENGTHS AND LIMITATIONS OF THIS STUDY.
We used a novel, validated, personalised recovery tool (Patient-Reported Outcomes Measurement Information System Physical Function) disseminated at regular intervals via SMS messaging to gain meaningful, accurate information about recovery milestones and time to resumption of self-selected activities.
To minimise confounding arising from variation in surgeon proficiency, we included an expertise-based design, acquiring information regarding experience and caseload for individual surgeons. Randomisation was stratified to balance prognostic variables, and rates of follow-up for trial outcomes were high.
The main limitation is the relatively small sample of 75 patients. At the time of trial analysis, 17% of randomised patients had not had surgery due to prolonged waiting lists following the COVID-19 pandemic, medical contraindications or patients no longer wanting surgery.
Non-adherence to the allocated route of hysterectomy was noted in 10% of participants undergoing hysterectomy. Given the nature of the intervention, it was not possible to blind either the care providers, investigators or participants to their allocated group.
Introduction
Hysterectomies are increasingly performed by a laparoscopic route as the recovery times are perceived to be faster given the smaller incisions compared with conventional open surgery. Laparoscopic hysterectomy (LH) for benign indications was found in a Cochrane review of 28 trials (3431 women) to have significantly more urinary tract injuries compared with abdominal hysterectomy (AH),1 but the evidence was of low to moderate quality, and trials were criticised for the relative inexperience of surgeons in novel LH.2 No differences in the costs or outcomes between LH and AH were found apart from return to normal activities. However, these recovery data were from only three small trials using inconsistent and non-validated methods of measurement.
Despite the lack of clear evidence of overall benefit, the uptake of LH is increasing with greater familiarity and increased proficiency in the technique, aided by improved training and better surgical equipment.3 4 Patient’s values and preferences, especially around the perceived speed of recovery, may also be driving this trend.5
Thus, a large RCT (Randomised Controlled Trial) was designed to determine the effectiveness of contemporary LH compared with AH for a benign gynaecological condition, overcoming previous RCT limitations relating to expertise and validated recovery assessment. We assessed major complications and time to resumption of usual activities using a relevant, personalised, validated instrument (PROMIS-SF).6,8 We report the trial design and key outcomes until the trial was discontinued following implementation of the Department of Health and Social Care ‘Research Reset programme’ (https://www.nihr.ac.uk/researchers/managing-research-recovery.htm).9
Methods
The trial was closed prematurely on the grounds of futility due to inability to recruit at the anticipated rate and considered unlikely to be able to deliver the endpoints in the postpandemic environment.10
Study design
The detailed LAVA (LAparoscopic Versus Abdominal hysterectomy) trial protocol has been previously published.11 In short, a parallel, open, non-inferiority, multicentre, randomised controlled, expertise-based surgery trial was conducted. The non-inferiority design reflected the belief that LH is not substantially worse, in terms of major complications, and has superior recovery from the patient’s perspective. The trial aimed to recruit 3250 women requiring a hysterectomy for a benign gynaecological condition and who were suitable for either laparoscopic or open, abdominal techniques. Women were eligible for recruitment to the LAVA trial if they were aged between 18 and 55 years, able to give informed consent to participate and did not require concomitant gynaecological surgery for bladder/pelvic support or deep endometriosis requiring dissection of the para-rectal space.
Randomisation was provided by a secure online randomisation system at the Birmingham Clinical Trials Unit (available at http://www.trials.bham.ac.uk/lava). Participants were randomised online to either a laparoscopic or open, AH in a 1:1 ratio using a minimisation algorithm to ensure balance in the treatment allocation over the following variables: previous caesarean section (yes/no), BMI (Body Mass Index) (≤29.9, 30–34.9, ≥35 kg/m2), uterine size (≤12 weeks, >12 weeks), planned retention of cervix (yes/no) and recruiting centre.
Hysterectomy was undertaken by either a laparoscopic or an open abdominal route, by a surgeon who had self-declared as having expertise according to stated, minimum standards in LH, AH or both approaches to hysterectomy: 30 independent cases and 12 cases per year. The decision to remove or retain the cervix or remove and retain the ovaries was left to the discretion of the participant in consultation with her gynaecologist.
Outcomes
No core outcome set currently exists for hysterectomy, although one is under development and has been registered on the COMET initiative database (https://www.comet-initiative.org/Studies/Details/1845). Following trial closure, the outcomes we collected up until the final randomised participant, who had received surgery, reached 26 weeks postsurgery.
Primary outcome
The primary outcome was major complications up to six completed weeks postsurgery objectively ascribed in accordance with the validated and widely used Clavien-Dindo classification of surgical complications.12 They were defined as any of the following up to and including six full weeks’ postsurgery: (1) all Clavien-Dindo grade III–V complications, (2) Clavien-Dindo grade II complications of pulmonary embolus or blood transfusion, (3) haemorrhage≥1 L or (iv) major adverse anaesthetic event.
Key secondary outcome
The key secondary outcome was the time from surgery to resumption of usual activities using the validated, personalised PROMIS-SF (Patient-Reported Outcomes Measurement Information System Physical Function) questionnaire item bank v1.2.6 This was the first in-class use of this recovery tool in this patient population. In brief, participants were presented with 29 items covering relevant activities for our trial population selected from the entire 121 item bank.7 Every item contained five response categories. At baseline, participants were asked to select eight activities from this list of 29 that, in their view, would most reflect their day-to-day activities. In this way, participants created their personalised physical function short form. Full recovery was achieved once all eight personalised activities had returned to baseline score. Until all personalised activities returned to baseline value, participants were asked to complete this weekly for the first 12 weeks, then fortnightly from week 13 to week 26, after which follow-up was censored.
Other secondary outcomes
Inpatient assessments: postoperative pain measured using a Numerical Rating Scale (NRS); quality of recovery score taken from the Quality of Recovery 15 (QoR-15) questionnaire13 measured at 24 hours; time from operation to discharge in days where the day of operation will be counted as day 0 and each additional day estimates one more night in the hospital.
Up to 14 days after surgery: postoperative pain (NRS) and analgesia use.
Up to 6 weeks postsurgery: minor complications;11 representation and readmission to hospital; time away from normal activities
6 weeks postsurgery: quality of life (QoL) score using EuroQol-5D-5L index score and the visual analogue scale (VAS).14
12 weeks postsurgery: EuroQol-5D-5L index score and the VAS; time from surgery to work in days; work productivity and activity impairment scores using WPAI-GH (Work Productivity and Activity Impairment -General Health) questionnaire.15
12 months postsurgery: satisfaction with hysterectomy (5-point Likert scale ranging from very satisfied to very unsatisfied).
Complication data, as well as re-presentation and readmission information, were collected from the relevant case report forms completed by the local research team and participants on the day of surgery and postoperatively. Where necessary, the local principal investigators were contacted for further details obtained from the patient’s case notes.
Sample size
To enable 90% power to test the non-inferiority hypothesis at a one-sided 2.5% significance level (two-sided 5% level), assuming a 3% margin of non-inferiority and a major complication rate of 6% in the abdominal (control) group, 2634 participants were required. The estimate of 6% was taken from a similar previous comparative study.16
An extra consideration was the potential for clustering by surgeon due to the expertise-based design.17 18 The majority of surgeons were anticipated to deliver the type of surgery in which they possess expertise, but some would deliver both surgery types. Under the assumption that each of the 50 centres would use six surgeons (operating on approximately nine patients on average during the trial), along with an intracluster correlation estimate of 0.02, the sample size was increased by 16% to 3055.
A final inflation of 6% to account for loss to follow-up brought the final sample size total to 3250 participants.
Statistical methods
The trial was originally designed for the primary outcome to be analysed within a non-inferiority framework, but this plan was abandoned once the trial was stopped early given the low number of events and resulting impact on statistical power. A decision was made to present summary data only (frequencies and percentages) for binary outcomes given the low likelihood of being able to draw sensible inferences from these outcomes, but formal analysis was still planned for the time to event and continuous outcomes which were likely to be more sensitive to detect any differences between groups.
The key secondary outcome of time from surgery to resumption of normal activities was analysed using a Cox proportional hazard model, adjusting for the minimisation variables. Observations were censored at the last response time if no final date of recovery is available (up to a maximum of 26 weeks). A HR and 95% CI were produced to assess the superiority of LH compared with open abdominal surgery. A second analysis using restricted mean survival time19 was also undertaken. The area under the survival curve was restricted at the point where at least one group has no one left ‘at risk’ of recovering. Time from operation to discharge in days was also analysed in the same way.
For continuous responses measured at a single time point, mean differences and 95% CIs were estimated using linear regression models. For postoperative pain up to 14 days postsurgery, mean differences and 95% CIs were estimated through the use of a repeated measures mixed effect linear regression model. Parameters allowing for participant, treatment group, time and the minimisation variables were included.
Patient and public involvement
The LAVA trial benefitted from patient and public involvement throughout the trial life cycle: The development phase, including the choice of important outcomes, was informed by two face-to-face patient focus groups, a PPI (Patient and Public Involvement) survey of 945 women from the Hysterectomy Association and feedback from the Royal College of Obstetricians and Gynaecologists Women’s Voices and patient information department. Two PPI representatives were co-applicants and contributed to the trial design and the design and content of patient-facing documentation and information, as well as advising on strategies to enhance the visibility of the trial and optimise recruitment. The PPI co-applicants were invited to attend Trial Management Group and Trial Steering Committee meetings during the lifetime of the trial and are co-authors of this paper.
The trial was approved by the West Midlands-Edgbaston Research Ethics Committee (trial registration: ISRCTN14566195, IRAS ID 287988
Results
Patient characteristics and surgical details
Over 14 months prior to trial closure (September 2021–November 2022), 75 patients were randomised from nine of the 10 open sites. 62 (83%) patients underwent hysterectomy, and all provided the primary outcome, assessment of major complications. The assigned route of surgery was achieved in 31 (97%) and 25 (83%) in the LH and AH groups, respectively (figure 1). Of the five non-adherent patients randomised to AH, the route of surgery was changed to laparoscopic in four participants at their request and one because of surgeon preference (to vaginal hysterectomy). Lack of surgical equipment was cited as the reason for the non-adherent patient randomised to LH.
Figure 1. Consolidated Standards of Reporting Trials (CONSORT) flow diagram of participants through the LAparoscopic Versus Abdominal hysterectomy (LAVA) trial. FU, Follow Up.
The indication for surgery for the majority of participants in both groups was abnormal uterine bleeding followed by pelvic pain. The baseline characteristics of the participants are shown in table 1. Overall, 9% of women had a uterine size >12 weeks, 40% had undergone a caesarean section and the mean BMI was 30 (SD 5.2) kg/m2.
Table 1. Baseline characteristics of participants by group and overall in the LAparoscopic Versus Abdominal hysterectomy (LAVA) trial.
| Laparoscopic hysterectomy(n=39) | Open abdominal hysterectomy(n=36) | Overall(n=75) | ||
|---|---|---|---|---|
| Minimisation variables | ||||
| Previous caesarean section | Yes | 15 (38%) | 15 (42%) | 30 (40%) |
| No | 24 (62%) | 21 (58%) | 45 (60%) | |
| BMI (kg/m2) | ≤29.9 | 17 (44%) | 16 (44%) | 33 (44%) |
| 30–34.9 | 15 (38%) | 15 (42%) | 30 (40%) | |
| ≥35 | 7 (18%) | 5 (14%) | 12 (16%) | |
| Uterine size (weeks) | ≤12 | 35 (90%) | 33 (92%) | 68 (91%) |
| >12 | 4 (10%) | 3 (8%) | 7 (9%) | |
| Planned retention of cervix | Yes | 2 (5%) | 2 (6%) | 4 (5%) |
| No | 37 (95%) | 34 (94%) | 71 (95%) | |
| Other baseline variables | ||||
| Age (years) | Mean (SD, n) | 43.5 (5.5, 39) | 42.7 (6.3, 36) | 43.1 (5.9, 75) |
| Range | 32.0–55.0 | 29.0–54.0 | 29.0–55.0 | |
| Ethnicity | White | 35 (90%) | 32 (89%) | 67 (89%) |
| Black African | 0 (0%) | 1 (3%) | 1 (1%) | |
| Black Caribbean | 0 (0%) | 1 (3%) | 1 (1%) | |
| Asian | 2 (5%) | 1 (3%) | 3 (4%) | |
| Mixed other | 1 (3%) | 0 (0%) | 1 (1%) | |
| Do not know | 1 (3%) | 1 (3%) | 2 (3%) | |
| Parity | 0 | 1 (3%) | 1 (3%) | 2 (3%) |
| 1 | 8 (21%) | 6 (17%) | 14 (19%) | |
| 2 | 15 (38%) | 14 (39%) | 29 (39%) | |
| 3 | 8 (21%) | 8 (22%) | 16 (21%) | |
| >3 | 7 (18%) | 7 (19%) | 14 (19%) | |
| Planned conservation of ovaries | Yes | 18 (46%) | 23 (64%) | 41 (56%) |
| No | 21 (54%) | 13 (36%) | 34 (45%) | |
| Diabetes | Yes | 3 (8%) | 2 (6%) | 5 (7%) |
| No | 36 (92%) | 34 (94%) | 70 (93%) | |
| If yes to diabetes: | ||||
| Insulin dependent | Yes | 0 | 1 | 1 |
| No | 3 | 1 | 4 | |
| Indication for surgery according to main symptom | Abnormal vaginal bleeding | 22 (56%) | 24 (67%) | 46 (61%) |
| Pain of any type | 14 (36%) | 10 (28%) | 24 (32%) | |
| Pressure-related symptoms | 0 (0%) | 1 (3%) | 1 (1%) | |
| Other* | 3 (8%) | 1 (3%) | 4 (5%) | |
Other includes risk reducing surgery, fibroids and fibroids, pelvic pain and heavy menstrual bleeding.
Only one participant had a subtotal hysterectomy. Moderate or severe adhesions were present in 13 (41%) and 7 (23%) patients in the LH and AH groups, respectively. Deep endometriosis was encountered in one patient allocated to abdominal surgery. The median uterine weight was 131 g and 135 g in the LH and AH groups, respectively. The mean surgical time was comparable (90 min (SD 25.5) vs 91 min (SD 57.9)) as was estimated blood loss (EBL) (median 50 mL vs 100 mL) for LH compared with AH, respectively. Only one patient had an EBL>500 mL in the LH group compared with four in the AH group. No planned laparoscopic hysterectomies were converted to laparotomy. There was no difference in the relative experience of surgeons performing each LH or AH (see onlinesupplemental tables 1 2 for full operation and surgeon details, respectively).
Primary outcome: major complications
Major complications occurred in 2/32 (6.3%) participants in the LH group compared with 4/30 (13.3%) in the AH group. The corresponding rates excluding cross-over were 2/31 (6.5%) and 3/25 (12%), respectively. In the LH group, the complications were an intraoperative haemorrhage>1 L and a delayed thermal injury to a left ureter presenting 10 days later with a uretero-vaginal fistula requiring nephrostomy and eventual retrograde stenting 6 weeks later. In the AH group, there were three intraoperative haemorrhages>1 L and one postoperative pelvic haematoma requiring radiological/surgical intervention (table 2).
Table 2. Frequency of major surgical complications in the LAVA trial up to 6 weeks postsurgery by group.
| Major surgical complication* | Laparoscopic hysterectomy (n=32) | Open abdominal hysterectomy (n=30) | Overall (n=62) |
|---|---|---|---|
| Major postoperative complications | |||
|
1 (3%) | 3 (10%)† | 4 (6%) |
|
0 (0%) | 1 (3%) | 1 (2%) |
|
1 (3%) | 0 (0%) | 1 (2%) |
|
1 (3%) | 0 (0%) | 1 (2%) |
| Major anaesthetic complications | 0 (0%) | 0 (0%) | 0 (0%) |
| Any other complication not covered requiring surgical, endoscopic or radiological intervention‡ | 1 (3%) | 0 (0%) | 1 (2%) |
| Participant had any major complication?§ | |||
| Yes | 2 (6%) | 4 (13%) | 6 (10%) |
| No | 30 (94%) | 26 (87%) | 56 (90%) |
| Minor surgical complication* | |||
| Haemorrhage 500 mL to <1 L | 0 (0%) | 2 (7%) | 2 (3%) |
| Fever (presumed infection) requiring antibiotics | 2 (6%) | 2 (7%) | 4 (6%) |
| Pain uncontrolled by usual analgesic management | 1 (3%) | 1 (3%) | 2 (3%) |
| Urinary retention requiring recatheterisation | 1 (3%) | 3 (10%) | 4 (6%) |
| Catheterisation for longer than 72 hours | 0 (0%) | 3 (10%) | 3 (5%) |
| Pelvic haematoma not requiring radiological or surgical intervention | 1 (3%) | 0 (0%) | 1 (2%) |
| Wound infections/complications managed at the bedside or on the ward | 1 (3%) | 1 (3%) | 2 (3%) |
| Any other minor complication requiring pharmacological treatment with drugs other than antiemetics, antipyretics, analgesics, diuretics, electrolytes or physiotherapy¶ | 2 (6%) | 3 (10%) | 5 (8%) |
| Participant had any minor complication? | |||
| Yes | 5 (16%) | 9 (30%) | 14 (23%) |
| No | 27 (84%) | 21 (70%) | 48 (77%) |
Participants may experience multiple major surgical complications.
One of these participants had vaginal hysterectomy.
Other includes the following: presented to site with abdominal pain. Admitted for a pelvic ultrasound scan (USS). 2 days after admission still awaiting scan and started to complain of slight urinary incontinence. USS kidney–ureter–bladder (KUB); mild hydronephrosis at the left kidney. Diagnosed with a vesicovaginal fistula and resolved with a nephrostomy and ureteric stenting for 6 weeks.
Primary outcome measure.
Other includes constipation, catheter stuck and required USS-guided removal and bilateral multilocular cysts on pelvic ultrasound scan (USS) suggestive of haemorrhagic cysts.
Secondary outcomes: minor complications
Five minor surgical complications occurred within 6 weeks following surgery in the LH group and 11 in the AH group (table 2).
Secondary outcome: recovery and return to work
We found no evidence of a difference in personalised recovery (median weeks [IQR, n] 7.5 (3.6–8.2, 25) LH vs 7.5 (5.5–10.6, 26) AH groups; restricted mean survival time difference (weeks) −0.49, 95% CI −2.85 to 1.87). The HR for resumption of usual activities was 0.99 (95% CI 0.52 to 1.85) (figure 2).
Figure 2. Kaplan-Meier plot of time from surgery to resumption of normal activities by group.
Likewise, no evidence of a difference in quality of recovery at 24 hours postsurgery was observed between LH and AH (mean [SD, n] 81.1 (13.4, 27) vs 72.3 (17.6, 22), respectively; adjusted mean difference 7.2 (−3.2, 17.6). Median time to discharge was also comparable 1.0 night (IQR 1.0–1.0) LH versus 2.0 nights (IQR 2.0–3.0), restricted mean survival time difference −1.0 (95% CI −2.6 to 0.6) and HR 2.13 (1.19, 3.80). We found no evidence of difference in pain scores over the first 14 days following surgery (adjusted mean difference 0.28 (95% CI −0.34 to 0.89)) nor in the type of analgesia or median duration of use in the first 14 days (11 days; IQR 7–14 LH vs 13 days; IQR 9–14) (onlinesupplemental tables 4 5, respectively). Overall, 14 (23%) of participants represented to the hospital within six completed weeks of surgery: six (19%), median (IQR) attendance 1 (1-1) and eight (27%), median (IQR) attendance 2.0 (1-2) in the LH and AH groups, respectively. Three (5%) participants were readmitted to the hospital within six completed weeks: one (3%) in the LH group and two (7%) in the AH group (online supplemental table 6).
Of patients currently employed, 19 (86%) in the LH group versus 15 (79%) in the AH group had returned to work by 12 completed weeks postsurgery. The respective median (IQR) return times to work postsurgery were 50 days (42–70) versus 40 days (28–70). No differences in work participation were observed between LH and AH groups (online supplemental table 7).
Secondary outcome: quality of life and satisfaction
No significant differences were observed in generic health-related QoL posthysterectomy between LH and AH (online supplemental table 3). At 12 months after surgery, 17/17 (100%) of patients in the LH group were very satisfied or satisfied with their hysterectomy compared with 11/13 (84%) of patients in the AH group.
Discussion
Main findings
Six major complications occurred in participants in the LAVA trial, equating to a major complication rate of 6% for LH and 13% for AH. The numbers were too small to test the non-inferiority hypothesis for LH compared with AH. The four major complications in the AH group were all related to bleeding; intraoperative>1 L or postoperative pelvic haematoma. In the LH group, there was one intraoperative haemorrhage and one visceral injury; thermal damage of a ureter resulting in a fistula into the vagina. Only one patient (3%) in the LH group and two patients (7%) in the AH group were readmitted to the hospital within six completed weeks of surgery.
No clear differences in recovery between LH and AH were observed, but the small sample precludes any meaningful inferences. Recovery quality in hospital as well as pain scores and use of analgesia during the first 14 days were comparable. No differences in generic QoL were observed for LH and AH at 6 and 12 weeks after surgery. Median times to resumption of preoperatively selected activities of importance to individual patients were between 49 and 53 days, and the median time for returning to work was between 40 and 50 days, with no significant differences between LH and AH. 90% of participants were satisfied with the outcome of their hysterectomy at 1 year.
Strengths and limitations
We used a novel, validated, personalised recovery tool (PROMIS-SF)17,19 disseminated at regular intervals via SMS messaging to gain meaningful, accurate information about recovery milestones and time to resumption of self-selected activities. To minimise confounding arising from variation in surgeon proficiency, we included an expertise-based design,17 20 acquiring information regarding experience and caseload for individual surgeons. Randomisation was stratified to balance prognostic variables, and rates of follow-up for trial outcomes were high.
The main limitation is the relatively small sample of 75 patients. At the time of trial analysis, 17% of randomised patients had not had surgery due to prolonged waiting lists following the COVID-19 pandemic, medical contraindications or patients no longer wanting surgery. Non-adherence to the allocated route of hysterectomy was noted in 10% of participants undergoing hysterectomy. Given the nature of the intervention, it was not possible to blind either the care providers, investigators or participants to their allocated group.
Interpretation (in light of other evidence)
Systematic reviews and meta-analyses of previous RCTs have shown no overall benefit of LH over AH for benign indications.5 Indeed, the largest RCT to date showed a higher major complication rate, driven by urinary tract injuries, with LH.16 In our curtailed trial, one urinary tract injury occurred with LH; this visceral damage necessitates further invasive intervention, often takes months to resolve and may lead to issues with chronic pain or renal damage. In contrast, while bleeding complications were seen more commonly with AH, intraoperative bleeding can be resolved with blood and/or fluid replacement and postoperative pelvic haematoma with drainage and antibiotics, generally without prolonged morbidity or long-term symptoms.
Using a novel, personalised and validated recovery tool administered regularly by SMS, our trial found the time to full resumption of usual activities to be comparable between LH and AH. This finding is in contrast to earlier trials1 (six women, 618 patients) that suggested a quicker return to normal activities on average for women undergoing LH (22–25 days) compared with AH (37 days). In our trial, it took 7.5 weeks on average for full recovery regardless of the route of hysterectomy. Our recovery data are more valid than previous trials because a personalised recovery tool was used (ie, recovery endpoints that are important to individual women).13 14 These data will be useful in counselling women, regardless of the type of hysterectomy, about their recovery. In addition, there is a move to same-day discharge with LH,21 22 but our trial did not show a difference in time to discharge from hospital between LH and AH. It is possible that as enhanced recovery and same day discharge protocols after hysterectomy become more widespread, inpatient hospital stays may reduce more for LH over AH, or may see a comparable reduction in stays for both approaches.
Conclusion
The results of the LAVA trial suggest that patients can be counselled while there is a small, but significant risk of major complications, a problematic or protracted recovery is unusual following a hysterectomy for a benign condition being undertaken by either a laparoscopic or open route. Patients can be informed that most patients resume their normal activities at levels commensurate with their preoperative abilities within 8 weeks of their procedure and are satisfied with the result of surgery.
However, it is important that larger trials are conducted, now that LH is the default procedure in many centres throughout the NHS, to evaluate the relative safety and recovery benefits of LH and AH. It is important to justify the increasing equipment costs, including robotic technologies and concerns over environmental sustainability from greater use of disposable equipment in LH.23 24 Large, multicentre surgical trials that can answer relevant clinical questions are scarce because they are difficult to execute.25 26 The LAVA trial faced a lack of clinician equipoise,10 mostly favouring laparoscopic approaches to hysterectomy despite the absence of a clear evidence base to support this belief.1 Research is needed to identify interventions that can surmount the challenges posed by surgical trials, especially around the lack of surgeon equipoise, and then focus investment to provide innovative solutions and successfully deliver future surgical RCTs.
Supplementary material
Footnotes
Funding: Funding for the LAVA trial is provided by an award from the National Institute of Health Research Health Technology Assessment programme (Ref: NIHR128991).
Prepub: Prepublication history and additional supplemental material for this paper are available online. To view these files, please visit the journal online (https://doi.org/10.1136/bmjopen-2024-096265).
Provenance and peer review: Not commissioned; externally peer reviewed.
Patient consent for publication: Not applicable.
Ethics approval: This study involves human participants. The trial was approved by the West Midlands-Edgbaston Research Ethics Committee (REC reference: 21/WM/0019). Participants gave informed consent to participate in the study before taking part.
Patient and public involvement: Patients and/or the public were involved in the design, or conduct, or reporting, or dissemination plans of this research. Refer to the Methods section for further details.
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information.
References
- 1.Pickett CM, Seeratan DD, Mol BWJ, et al. Surgical approach to hysterectomy for benign gynaecological disease. Cochrane Database Syst Rev. 2023;8:CD003677. doi: 10.1002/14651858.CD003677.pub6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Donnez J, Squifflet J, Jadoul P, et al. Results of eVALuate study of hysterectomy techniques: high rate of complications needs explanation. BMJ. 2004;328:643. doi: 10.1136/bmj.328.7440.643. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Lee SH, Oh SR, Cho YJ, et al. Comparison of vaginal hysterectomy and laparoscopic hysterectomy: a systematic review and meta-analysis. BMC Womens Health. 2019;19:83. doi: 10.1186/s12905-019-0784-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Madhvani K, Curnow T, Carpenter T. Route of hysterectomy: a retrospective, cohort study in English NHS Hospitals from 2011 to 2017. BJOG. 2019;126:795–802. doi: 10.1111/1471-0528.15539. [DOI] [PubMed] [Google Scholar]
- 5.Poletti B, Stringer G, Furness K. Patient Experience Pre-Implementation of an Enhanced Recovery After Surgery Protocol: A Qualitative Investigation. J Multidiscip Healthc. 2024;17:1147–58. doi: 10.2147/JMDH.S453467. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Crins MHP, Terwee CB, Klausch T, et al. The Dutch-Flemish PROMIS Physical Function item bank exhibited strong psychometric properties in patients with chronic pain. J Clin Epidemiol. 2017;87:47–58. doi: 10.1016/j.jclinepi.2017.03.011. [DOI] [PubMed] [Google Scholar]
- 7.Terwee CB, Roorda LD, de Vet HCW, et al. Dutch-Flemish translation of 17 item banks from the patient-reported outcomes measurement information system (PROMIS) Qual Life Res. 2014;23:1733–41. doi: 10.1007/s11136-013-0611-6. [DOI] [PubMed] [Google Scholar]
- 8.van der Meij E, Anema JR, Leclercq WKG, et al. Personalised perioperative care by e-health after intermediate-grade abdominal surgery: a multicentre, single-blind, randomised, placebo-controlled trial. The Lancet. 2018;392:51–9. doi: 10.1016/S0140-6736(18)31113-9. [DOI] [PubMed] [Google Scholar]
- 9.Department of Health and Social Care, Northern Ireland Executive, Welsh Government, Scottish Government Saving and improving lives: the future of UK clinical research delivery. https://assets.publishing.service.gov.uk/media/60d20134d3bf7f4bcfe76629/The-future-of-UK-clinical-research-delivery.pdf Available.
- 10.Antoun L, Bevan S, Mahmud A, et al. Lessons learnt from the multi-centre LAparoscopic Versus Abdominal hysterectomy (LAVA) randomised controlled trial. Facts Views Vis Obgyn. 2024;16:35–45. doi: 10.52054/FVVO.16.1.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Antoun L, Middleton L, Smith P, et al. LAparoscopic Versus Abdominal hysterectomy (LAVA): protocol of a randomised controlled trial. BMJ Open. 2023;13:e070218. doi: 10.1136/bmjopen-2022-070218. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Clavien PA, Barkun J, de Oliveira ML, et al. The Clavien-Dindo classification of surgical complications: five-year experience. Ann Surg. 2009;250:187–96. doi: 10.1097/SLA.0b013e3181b13ca2. [DOI] [PubMed] [Google Scholar]
- 13.Stark PA, Myles PS, Burke JA. Development and psychometric evaluation of a postoperative quality of recovery score: the QoR-15. Anesthesiology. 2013;118:1332–40. doi: 10.1097/ALN.0b013e318289b84b. [DOI] [PubMed] [Google Scholar]
- 14.EuroQol - a new facility for the measurement of health-related quality of life. Health Policy. 1990;16:199–208. doi: 10.1016/0168-8510(90)90421-9. [DOI] [PubMed] [Google Scholar]
- 15.Reilly MC, Zbrozek AS, Dukes EM. The validity and reproducibility of a work productivity and activity impairment instrument. Pharmacoeconomics. 1993;4:353–65. doi: 10.2165/00019053-199304050-00006. [DOI] [PubMed] [Google Scholar]
- 16.Garry R, Fountain J, Mason S, et al. The eVALuate study: two parallel randomised trials, one comparing laparoscopic with abdominal hysterectomy, the other comparing laparoscopic with vaginal hysterectomy. BMJ. 2004;328:129. doi: 10.1136/bmj.37984.623889.F6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Cook JA, Elders A, Boachie C, et al. A systematic review of the use of an expertise-based randomised controlled trial design. Trials. 2015;16:241. doi: 10.1186/s13063-015-0739-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Roberts C. The implications of variation in outcome between health professionals for the design and analysis of randomized controlled trials. Stat Med. 1999;18:2605–15. doi: 10.1002/(sici)1097-0258(19991015)18:19<2605::aid-sim237>3.0.co;2-n. [DOI] [PubMed] [Google Scholar]
- 19.Royston P, Parmar MKB. The use of restricted mean survival time to estimate the treatment effect in randomized clinical trials when the proportional hazards assumption is in doubt. Stat Med. 2011;30:2409–21. doi: 10.1002/sim.4274. [DOI] [PubMed] [Google Scholar]
- 20.van Brummen HJ, Bruinse HW, van de Pol G, et al. Defecatory symptoms during and after the first pregnancy: prevalences and associated factors. Int Urogynecol J Pelvic Floor Dysfunct. 2006;17:224–30. doi: 10.1007/s00192-005-1351-0. [DOI] [PubMed] [Google Scholar]
- 21.Antoun L, Smith P, Afifi Y, et al. Short stay laparoscopic hysterectomy: An evaluation of feasibility and patient satisfaction. Facts Views Vis Obgyn. 2021;13:377–85. doi: 10.52054/FVVO.13.4.039. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.GIRFT National day surgery delivery pack. Version 1.0. 2020 https://www.gettingitrightfirsttime.co.uk/wp-content/uploads/2024/09/National-Day-Surgery-Delivery-Pack-V2.0-September-2024.pdf Available.
- 23.Siu J, Hill AG, MacCormick AD. Systematic review of reusable versus disposable laparoscopic instruments: costs and safety. ANZ J Surg. 2017;87:28–33. doi: 10.1111/ans.13856. [DOI] [PubMed] [Google Scholar]
- 24.Morrison JE, Jr, Jacobs VR. Replacement of expensive, disposable instruments with old-fashioned surgical techniques for improved cost-effectiveness in laparoscopic hysterectomy. JSLS. 2004;8:201–6. [PMC free article] [PubMed] [Google Scholar]
- 25.Pollock AV. Surgical evaluation at the crossroads. Br J Surg. 1993;80:964–6. doi: 10.1002/bjs.1800800807. [DOI] [PubMed] [Google Scholar]
- 26.Davies L, Beard D, Cook JA, et al. The challenge of equipoise in trials with a surgical and non-surgical comparison: a qualitative synthesis using meta-ethnography. Trials. 2021;22:678. doi: 10.1186/s13063-021-05403-5. [DOI] [PMC free article] [PubMed] [Google Scholar]