Parts of this chapter are reproduced from Hashim et al.2 This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) license, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: http://creativecommons.org/licenses/by/4.0/. The text below includes minor additions and formatting changes to the original text.
Summary and interpretation of main findings
The UNBLOCS trial compared two technologies for treating BPO in the UK NHS: ThuVARP and the gold-standard TURP.
Patients were randomised between the two arms at the point of surgery. The two arms were generally well balanced at baseline for sociodemographic and clinical characteristics, with the groups also equally divided between those presenting with urinary retention and those with bothersome LUTS secondary to BPO. Participants had a mean age of 70 years. The proportion of participants receiving their randomised treatment was much higher in the TURP arm (98%) than in the ThuVARP arm (75%), with eight participants requiring a conversion to TURP mid-procedure.
Primary outcomes
The primary aim of the trial was to determine whether ThuVARP and TURP are equivalent in terms of the co-primary outcomes of patient-reported IPSS and Qmax at 12 months after surgery.
The two procedures appear to be equivalent for symptom relief measured by the IPSS, and therefore this blinded trial has shown that patients’ perceptions’ of urinary tract symptoms are equivalent when the actual treatment is unknown. Both procedures achieved a much improved IPSS from that reported at baseline, dropping from a mean score of around 21 for participants with LUTS (baseline data could not be collected for retention participants) to a mean score of around 6 at 12 months post surgery for all participants. This is as expected from BPO surgery, demonstrating that both procedures improve subjective symptom scores.
Maximum urine flow rate improved after both procedures, from a mean of < 10 ml per second at baseline for participants with LUTS (baseline data could not be collected for retention participants) to a mean of > 20 ml per second at 12-month follow-up for all participants. ThuVARP gives a lower maximum urinary flow rate at 12 months than TURP, deeming the treatments non-equivalent in Qmax. Changing the test to superiority provided evidence to suggest that TURP is superior to ThuVARP with respect to Qmax.
The result for Qmax is based on the assumption that a difference of > 4 ml per second has clinical significance. As discussed in Chapter 2, Equivalence margin, no minimal clinically important difference in Qmax is accepted in the literature. A figure of 2 ml per second has been quoted previously; however, following discussion between clinicians, in this trial the higher value of 4 ml per second was chosen. Therefore, the finding of non-equivalence and superiority is in the context of uncertainty about what is considered as a clinically significant difference for Qmax, and superiority was demonstrated for TURP against a relatively high value of clinical significance. However, the improvement in Qmax from baseline seen in both arms would be considered a clinically successful outcome from BPO surgery.
Secondary outcomes
The ThuVARP procedure did not lead to fewer complications, as had been expected, with results suggesting that the procedures were similar in this respect. Both procedures were equally safe perioperatively, with a risk of no more than 4% for any individual complication, and no incidence of TUR syndrome (dilutional hyponatraemia due to fluid absorption). In the 12 months post surgery, a similar incidence of clot retention (≈5%) and urethral stricture (≈3%) was observed, as well as of infections including urinary tract infection (≈35%), sepsis (≈1%) and pyrexia of unknown origin (≈3%). Overall, ≈20% of participants in both arms suffered a serious adverse event.
Of particular note when considering complications is the level of clinically significant bleeding, which was experienced equally between the two arms; around 5% of participants experienced bleeding to the point that their haemoglobin had to be measured, and around 2% of participants required a blood transfusion. One of the advantages quoted in the literature of laser procedures over standard treatment is a reduction in the risk of bleeding and transfusion rates; however, no such difference was observed in this trial. A further advantage of laser procedures that has previously been reported is a reduction in the length of hospital stay. However, in this trial the length of stay was 48 hours in both arms, which is the national average, and certainly shorter than that in the Chinese trials of the thulium laser published in the literature.13,14
Other surgical outcomes, such as the length of indwelling catheterisation time and change in levels of haemoglobin and serum sodium pre- and postoperatively, were also very similar between the arms. Serum sodium did not reduce significantly after surgery, suggesting that serum sodium checks may not be required postoperatively, as have previously been conducted. Both groups had a median indwelling catheterisation time postoperatively of approximately 2 days.
In terms of urinary measures on the whole, taking into account IPSS, International Consultation on Incontinence Questionnaire, Qmax and post-void residual, all results were in favour of the TURP procedure on visual appraisal, albeit by a small amount and without any statistical significance. Patient-reported outcome measures in this trial looked at urinary symptoms, quality of life, sexual symptoms and satisfaction with surgery, and no significant difference was found between the arms in these. However, we cannot rule out a clinically important difference in patient-reported nocturia, with indication of an increased incidence at 12 months post surgery in the ThuVARP arm. This result is difficult to rationalise, although it may be a heat-induced effect; however, the other storage symptoms do not reflect this. It is also possible that charring from the laser could cause irritation and urinary frequency/nocturia. It should be noted that satisfaction with surgery and quality of life after surgery were rated highly in both arms, with no significant difference between the arms.
Subgroup analyses suggest a greater improvement in Qmax after receiving TURP than after receiving ThuVARP among patients diagnosed with LUTS, with no difference between the arms indicated in participants diagnosed with urinary retention. In addition, there is a suggestion that men aged < 70 years may benefit more from TURP than from ThuVARP in terms of Qmax. Further exploration of these potential effects would be of interest for future research.
Of note is the finding that only 75% of participants randomised to ThuVARP received their complete randomised treatment, as opposed to 98% of participants who underwent TURP procedures. A large proportion of changes in treatment were because of equipment failure in the ThuVARP arm, with 18 participants being changed to TURP straight away or converting mid-procedure. The size of the prostate also resulted in nine conversions to TURP. The proportion of participants receiving conversions, compared with those receiving ThuVARP, was relatively balanced across the centres, with between 6% and 25% of participants in each centre receiving a conversion from ThuVARP to TURP. The high rate of conversions may have been due to several factors. As surgeons had generally completed a mean of only seven cases before proceeding to trial cases, there may have been an element of ‘teething problems’ with the laser, and a continued learning curve past the assessment of competence. However, this is unlikely to be the cause, as, when breaking the recruitment period per surgeon into halves, the conversion rate in the first half was 11%, whereas it increased to 28% in the second half. Potentially, as surgeons became more experienced with the laser, their judgement of when procedures needed to be converted may have altered. Equally, some surgeons expressed a reluctance to operate on large prostates with the laser early in the trial, with some exclusions of patients with larger prostates from the trial. The increased number of conversions could therefore also have reflected the increasing confidence of surgeons to operate on larger prostates with the laser. Future research into the comparative effectiveness of ThuVARP and TURP in large prostates would also be of interest.
An interesting finding, although one not prespecified, was that the treatments differed in pathology diagnostic detection. Following the procedure, 193 participants in each arm had a prostate histology, among whom 13% in the TURP arm were diagnosed with prostate cancer, compared with only 5% in the ThuVARP arm. When looking at the histology samples, pathologists were usually alerted to laser TURP cases by the fact there was charring of the tissue. On reviewing the weight of resected tissue, the amount of tissue available for histology was noted to be also significantly reduced in those who had laser procedures. This is due to the vaporesection conducted by the thulium laser, which vaporises a proportion of the tissue as well as resecting. The difference in histological finding may therefore be due to both a reduction in tissue and the tissue being damaged by charring from the laser. The significance of this finding is that prostate cancer diagnoses may be missed as a result of the limited histology available, although TURP is not part of the diagnostic pathway for prostate cancer. The results of this study may therefore have implications that the National Cancer review55 may wish to consider. Further work into this area could include categorising the prostate cancer findings by stage and grade from pathological reports to further establish the clinical relevance of the findings.
The trial was conducted across four teaching hospitals and three district general hospitals in the UK, and included a variety of hospitals to increase the generalisability of the results. The majority of recruitment occurred in the teaching university affiliated hospitals, which suggests that research is easier to conduct in these hospitals because there are more resources available. However, this does not mean that smaller hospitals cannot participate in research, as demonstrated by successful recruitment across all sites.
Economic evaluation
The results of the economic evaluation indicate that, from an NHS secondary care perspective and an NHS perspective, participants in the ThuVARP arm had slightly higher costs and fewer QALYs than those in the TURP arm, but these differences were consistent with chance. The probability that ThuVARP is the cost-effective option compared with TURP at a threshold of £20,000 per QALY is only 24%, which increases slightly to 25% when all NHS costs are taken into account.
The expected lower costs in the ThuVARP arm resulting from the fact that this can be carried out as a day case procedure did not materialise. This is to some extent an artefact of the trial, in that participants had to be listed as a day-case or an inpatient stay prior to admission. According to Hospital Episode Statistics data from 2013/14,56 the mean length of stay for TURP procedures was 2.9 days, as opposed to 1.8 days for laser procedures already in use in the NHS. However, Hospital Episode Statistics data for 2016/1757 report a reduction in the mean length of stay for TURP to 2 days, equal to that for laser procedures. Indeed, a shift was made in two of our study sites to conduct TURP procedures as day case during the trial. No difference in actual length of stay was observed between the TURP and ThuVARP arms during the trial. The effect on costs of being able to carry out ThuVARP as a day-case procedure in the other five centres was examined; this showed that ThuVARP was cheaper than TURP in this scenario (£3909 vs. £4245), which led to a positive INMB of £99 (95% CI –£547 to £745). This implies that if TURP did not become a day-case procedure, and if ThuVARP was always a day-case procedure, then the two operations would not differ in cost-effectiveness.
It is also worth noting that all hospitals will have a TURP machine. Therefore, although an annual equivalent cost (including machine maintenance) was calculated for both procedures, there would be an upfront cost to purchase the laser machine that would not in reality be mirrored by a need to purchase a TURP generator, until the generator required replacing. The cost of purchasing a generator is approximately 10% of the purchase cost of a laser machine and therefore trusts may not be inclined to fund this upfront higher laser cost. The majority of the equipment cost for the TURP procedure is in the consumables used, contrary to the laser cost, the majority of which results from the reusable equipment, including the laser machine. There is a high rate of conversions from ThuVARP to TURP, which during this trial meant that both sets of equipment were required for 24% of procedures. The effect of this in relation to the overall costs in a sensitivity analysis showed that the need to have TURP capital and reusable equipment always available increased the cost in the laser arm by only £9.
The finding that the ThuVARP procedure took on average 21 minutes longer than the TURP procedure is important to consider. This is the case both from a patient perspective, in terms of increased time under anaesthetic, and from the perspective of service delivery. It is possible that a continued learning curve of the ThuVARP procedure past assessment of competence had an impact on this finding; however, there was no evidence of any learning curve effect. If this difference is maintained, then it is possible that an additional small case could be added on to an operating list if TURPs are performed, which could have implications for reducing waiting lists.
Qualitative study
The qualitative study investigated the participants’ experiences of both procedures in more depth and found that participants reported satisfactory outcomes in the main, regardless of procedure. Similar journeys were reported following both procedures. Participants described a period of uncertainty about whether or not their expectations for outcome would be met, which preceded a largely satisfactory outcome. The period of uncertainty was characterised by concerns regarding pain, bleeding and residual urinary symptoms. Some negative influence on recovery of the participants’ mood was reported, but this had largely resolved by the time of interview. Participants reported satisfaction overall with the outcome but highlighted that understanding and being appropriately informed about what to expect during recovery were essential to helping them cope with the recovery period.
Slower recovery associated with previous catheter use was intimated, which was perceived to be because their bladder had had poorer function before the operation. Supported by conservative strategies, such as pelvic floor exercises and medication where necessary, bladder function was perceived to have improved or to be improving.
Most participants were delighted that their quality of life had been restored and for some this was quicker than they had expected. A sentiment of returning to a more ‘youthful’ form was discussed. This exploration highlighted that perceptions of male LUTS as ‘an old man’s problem’ remain entrenched. Engagement with this group (the study population) to educate about these perceptions may be indicated in order to encourage men to seek treatment that could benefit their lives.
Active enquiry about sexual matters was made. Participants who had experienced issues with their sex lives either as a result of the surgery or that predated the intervention reported that they had been rarely asked about this aspect and would like the opportunity to disclose this in order to receive appropriate signposting.
With regard to future practice, accurate information provision is key to patients’ satisfaction and expectations about the outcome. Participants achieved their expected outcome regardless of procedure, with no evidence of preference for one type of surgery. Exploration of de novo symptoms, such as those relating to sexual matters, is welcome so that patients can receive appropriate guidance rather than feeling that they simply have to ‘put up’ with their situation.
Strengths and limitations
The main strength of this study is the successful recruitment to target, despite the challenges of consenting patients to a trial in which randomisation happened at the point of surgery, and particularly at a time when NHS trusts were experiencing significant pressures, with delays to elective surgery as a result of operating theatre availability and hospital bed shortages, particularly during the winter months. In addition, the number of patients who withdrew from the trial was small, and follow-up rates were exceptional.
In terms of blinding, this seems to have been successful as participants were not able to guess which procedure they had received. This reduced the bias in subjective improvement of symptoms and quality of life, and we would recommend this method of randomising at the point of patient anaesthesia for blinding in future surgical trials where logistically possible.
An additional strength of this study was the inclusion of catheterised participants, who are logistically a difficult population to include in trials of BPO surgery. The inclusion of these participants meant that adjustment for baseline for the primary analysis was not possible; however, the trial team believe that this is outweighed by the increased generalisability.
Limitations lay with the reporting of complications, with different qualities of reporting across staff and sites. Interpretation of reporting criteria may have varied across sites and there is always an element of clinical judgement, for example in defining what is a normal expectation for surgery and what is a complication. Bleeding is a good example of where the level that constitutes a complication is subjective, as some degree of bleeding would be expected during surgery. Therefore, the team focused on reporting bleeding to the extent that haemoglobin levels needed to be checked for a possible blood transfusion, and whether or not a blood transfusion was conducted. The team analysed the complications according to the main tick boxes provided in the CRFs. The ‘other’ section was completed to varying degrees and therefore the decision was taken to abandon these, while considering any serious adverse events which were reported separately. However, it should be noted that the data on serious adverse events were collected for the purposes of safety reporting during the trial and not as outcome data. Therefore, categorisation of expectedness and relatedness has been determined by the principal investigators at the sites and not independently verified, and levels of reporting also significantly varied across sites.
A further limitation lay in the measurement of prostate size, which was obtained by surgeons carrying out digital rectal examination in the operating theatre before surgery but once the patient was anaesthetised. We are aware that digital rectal examination is not an accurate measure of prostate size,58 and that a transrectal ultrasound would have provided more accurate data. However, this trial has taken the pragmatic approach of following routine care in the NHS, which usually involves only a digital rectal examination before surgery. It should be noted that, in general, this did not affect clinical outcome, with only one participant sent for emobilisation because he had a very large prostate. In general, rectal examination can reveal a very large prostate, which would have deemed such patients as unsuitable for TURP and also may indicate prostate cancer.
An additional limitation of randomising patients at the point of surgery was that the patients had to be listed as either day case or inpatient before randomisation. Therefore, no differential could be made between the two operations prior to randomisation. At two trial sites, a move had already been made to conduct TURP procedures as day case, and therefore the majority of both procedures were listed as day case. The other five trial sites listed all procedures as inpatient procedures. This trial was therefore not set up so that sites could make a judgement on whether they wished to conduct either procedure as day case or inpatient, but rather decisions had to be made that applied to both operations. In some respects this is a limitation, as sites that would not have conducted TURP as a day-case procedure may have chosen to conduct ThuVARP as a day-case procedure. However, in other respects it is a strength, as sites could not apply any bias in whether they perceived one or the other operation to be conducted as day case. Ultimately, length of stay, after the patient has been listed as day case or inpatient, is determined by the patient’s health status and therefore we believe that this trial presents a fair comparison of the two procedures.
A statistical limitation was that the data collected at 3 months could not be made use of in the imputation model, as originally planned, owing to modelling issues (collinearity and lack of convergence).
Overall evidence and generalisability
The results of this trial will undoubtedly play a part in future decision-making for doctors and patients involved with treatment for LUTS. NICE in 2010 suggested that if offering surgery for managing voiding LUTS presumed to be secondary to BPO, to only consider laser vaporisation techniques, bipolar transurethral vaporisation of the prostate (TUVP) or monopolar or bipolar transurethral vaporisation resection of the prostate as part of a RCT that compares these techniques with TURP. The UNBLOCS trial has undertaken this recommendation for the ThuVARP technique.
The trial has shown that the ThuVARP and TURP procedures are equally effective in improving IPSS, and although TURP is superior in improving maximum urine flow rate, both procedures improved urinary flow to what is considered a clinically successful level. The two procedures also demonstrated very similar results across virtually all other clinical and patient-reported outcomes. The results of the economic evaluation indicate differences in costs and QALYs between the two arms, which could be a result of chance, with the total adjusted mean secondary care cost for the ThuVARP arm only £9 more than the TURP arm, and a QALY difference of 0.01 between the arms in favour of TURP. This marginal cost difference supports the acceptability of both techniques within the NHS. It should be noted, however, that ThuVARP took on average 21 minutes longer than TURP, which has implications for patients, in terms of an increased time under anaesthetic, and for service delivery within the NHS. In addition, a higher proportion of ThuVARP cases required conversion to TURP, which may indicate that both sets of equipment would need to be available for conducting ThuVARP. However, the data overall suggest that both procedures are viable in the UK NHS from a clinical and cost perspective.
The finding of a smaller improvement in Qmax levels after ThuVARP than after TURP is consistent with a recent systematic review and meta-analysis59 comparing ThuVARP with TURP, as is a longer operating time for ThuVARP. The anticipated benefits of the laser of reduced catheterisation time, length of hospital stay and bleeding were not observed in this trial, contrary to the recent evidence review. However, the review is based on the limited evidence available, it includes non-RCTs due to the lack of RCTs, and the majority of its studies were conducted in Asia across different health systems.59
The National Institute for Health and Care Excellence also recommends monopolar transurethral vaporisation of the prostate or HoLEP at centres specialising in the technique with mentorship arrangements in place due to a long learning curve, reducing their generalisability. The short period of training that surgeons undertake to achieve competence in ThuVARP demonstrates that the two techniques are similar in terms of the operative skills required, and therefore its comparative ease of generalisability.
The trial also supports NICE’S recommendation of offering monopolar and bipolar transurethral resection surgery of the prostate for managing voiding LUTS presumed secondary to BPO. There is no strong clinical finding suggesting that ThuVARP is not an effective procedure for BPO, although TURP may have marginal benefits. Findings from the UNBLOCS trial provide evidence that could be used to update the current NICE guidelines on the efficacy of new technologies in the NHS.
The trial provides evidence about the risks of both procedures that can be used to update the literature, allowing patients to be more informed at the point of consent about the risks of these procedures, especially with regard to side effects. In addition, the result that serum sodium levels did not reduce significantly after surgery indicates that the routine review of urea and electrolytes and full blood counts post TURP may not be necessary. This would have cost-saving implications for the NHS, as traditional teaching suggests that these blood tests should be carried out after TURP. The trial will help to define a pathway for patients undergoing endoscopic resections of the prostate.
Recommendations for research
- Longer-term follow-up of trial participants would demonstrate whether or not there is any sustained difference between the arms in reoperation rates over time. This would provide a more comprehensive perspective on whether or not the lower Qmax achieved by ThuVARP had any relevance in terms of requiring further future interventions for LUTS especially at an earlier time span than TURP.
- Of interest would be future research into the comparative effectiveness of ThuVARP and TURP for patients with large prostates of > 80 ml that are operable by specialists, which may extend the scope of the operation further than current EAU guidelines.
- Comparison of the enucleation procedures using thulium and holmium lasers, as ThuVARP is easier to learn and adopt compared with laser enucleation techniques.
Implications for health care
The trial provides evidence for the clinical effectiveness and cost-effectiveness of laser vaporesection of the prostate compared with the gold-standard TURP in men with moderate to severe bothersome LUTS, including urinary retention, who are considering surgery for BPO. The potential advantages of reduced blood loss, shorter hospital stay and earlier return to normal activities have been thought to make laser prostate resection techniques attractive to both patients and health-care providers, although there is uncertainty about the extent of symptom improvement, and about improvement in quality of life in both the short and the long term. The UNBLOCS RCT addressed these questions for the ThuVARP technique.
Both operations were shown to be effective for treating BPO. Although they were equivalent in terms of IPSS, their equivalence could not be demonstrated in terms of Qmax, for which, despite a relatively small difference in resulting flow rate, TURP was shown to be superior. The results generally suggest little difference between this procedure and the standard TURP procedure from a clinical and an overall cost point of view, although marginal benefits of TURP were indicated in terms of reduced operative time, increased available histology for incidental cancer detection and a higher rate of patients undergoing their randomised procedure. Both operations achieved a high level of patient satisfaction with surgery, and improved quality of life, which is supported by the qualitative element of this study. Overall, the results suggest that it may be appropriate that new treatment alternatives continue to be compared with TURP.
This study also provides the added benefit of up-to-date information on the patient-reported and clinical outcomes associated with standard TURP, which can be used to inform clinical decisions and patient expectations, the importance of which was highlighted in the qualitative element of this study.