NICE commissioned an external assessment centre (EAC) to review the evidence submitted by the company. This section summarises that review. Full details of all the evidence are in the project documents on the NICE website.
Relevant evidence comes from 1 systematic review and 1 English language translation of an uncontrolled case series
3.1 In the original UroLift medical technologies guidance, the EAC considered:
one systematic review summarising 9 studies (reporting outcomes for 452 to 680 people, depending on the outcome)
one uncontrolled case series (reporting outcomes from 20 people).
The EAC identified no further evidence. The studies relevant to the decision problem in the scope were:
nine studies in the systematic review including 2 papers on a randomised controlled trial (RCT; the LIFT study; McVary et al. 2014; Roehrborn et al. 2013) and 7 uncontrolled before‑and‑after studies (Cantwell et al. 2014; Chin et al. 2012; Delongchamps et al. 2012; McNicholas et al. 2013; Shore et al. 2014; Woo et al. 2011 and 2012)
one English language translation of an uncontrolled case series (Abad et al. 2013).
For full details of the clinical evidence, see section 3 of the assessment report in supporting documentation.
3.2 In the original guidance, there was no published evidence directly comparing the UroLift System with the comparator technologies highlighted in the scope. So, the EAC did an evidence synthesis of the outcomes in the UroLift studies. It compared them with those reported with transurethral resection of the prostate (TURP) and holmium laser enucleation of the prostate (HoLEP) in a systematic review (Li et al. 2014). During the consultation period, initial results from the BPH6 RCT comparing UroLift with TURP became available.
3.3 The EAC's evidence synthesis showed that both TURP and HoLEP were associated with greater improvements in International Prostate Symptom Score (IPSS) than UroLift at all time points. Overall changes within a 2‑year period ranged from -17.34 to -19.7 with TURP and -17.68 to -20.88 with HoLEP, compared with -9.22 to -11.82 with UroLift. Maximum urinary flow (Qmax) and post-void residual improvements were also greater with TURP and HoLEP.
3.4 The EAC's evidence synthesis reported that the IPSS quality-of-life score improved by 2.22 to 2.48 points for people having UroLift treatment. However, this was less than the improvement after TURP (2.99 to 3.18 points) and HoLEP (2.64 to 3.24 points). An increase of 1 to 3 points is generally considered to represent a minimum important change.
3.5 The EAC's evidence synthesis showed that sexual function is not negatively affected after using UroLift. In fact, small, statistically significant improvements (0.3 to 0.4 points, based on combined sexual health scores reported in the meta-analysis) were reported. Changes in sexual function were poorly reported in the TURP and HoLEP studies, which made it difficult to accurately assess the effect of these technologies. Expert advice was that deterioration in sexual function was well described and seen in practice in some people having TURP or HoLEP.
3.7 In 7 studies there were statistically significant improvements in symptom severity (IPSS score) and in 4 studies there were improvements in Benign Prostatic Hyperplasia Impact Index (BPHII) score up to 5 years after the UroLift procedure. These studies were Roehrborn et al. 2015; Bozkurt et al. 2016; Rukstalis et al. 2016; Bardoli et al. 2017; Roehrborn et al. 2017; Sievert et al. 2018; Eure et al. 2019 and Rubio et al. 2019; Rukstalis et al. 2018.
3.8 Compared with TURP, people having UroLift reported smaller improvements in IPSS scores up to 12 months after the procedure (Sonksen et al. 2015; Gratzke et al. 2016). Compared with Rezum, people having UroLift reported greater improvements in IPSS scores at 30 days after the procedure (Tutrone and Schiff, 2020).
3.9 Qmax improved up to 5 years after UroLift treatment in most studies (Roehrborn et al. 2015; Bozkurt et al. 2016; Rukstalis et al. 2016; Roehrborn et al. 2017; Sievert et al. 2018; Rubio et al. 2019; Rukstalis et al. 2018). However, in Eure et al. (2019) Qmax decreased up to 6 months after the procedure and no significant difference in Qmax was reported by Bardoli et al. (2017).
3.10 In 4 studies there was a statistically significant improvement (up to 12 months) in post-urination residual volume (Bozkurt et al. 2016; Rukstalis et al. 2016; Bardoli et al. 2017; Sievert et al. 2018). In Gratzke et al. (2016) Incontinence Severity Index scores remained unchanged up to 2 years after UroLift treatment.
3.11 TURP produced greater improvements in Qmax and post-urination residual volume up to 24 months after the procedure compared with UroLift (Sonksen et al. 2015; Gratzke et al. 2016).
3.12 In most studies, the UroLift procedure did not result in statistically significant changes in erectile dysfunction. This was assessed using the International Index of Erectile Function and the Sexual Health Inventory for Men (SHIM) questionnaires (Bozkurt et al. 2016; Rukstalis et al. 2016; Rubio et al. 2019). However, in people with obstructive median lobes, there were improvements in both measures up to 12 months after the procedure (Rukstalis et al. 2018). The amount of change in SHIM scores did not differ much between UroLift and TURP (Sonksen et al. 2015; Gratzke et al. 2016) but was statistically significantly better with UroLift than Rezum (Tutrone and Schiff, 2020).
3.13 In 5 studies, Male Sexual Health Questionnaire for Ejaculatory Dysfunction (MSHQ-EjD) scores after UroLift and other treatments were reported. In 2 of these, there were improvements over time after UroLift (Roehrborn et al. 2015; Rukstalis et al. 2018). In 2 other studies improvements over time with UroLift compared with TURP were not statistically significant (Sonksen et al. 2015; Gratze et al. 2016). In 1 study the difference in scores between people who had UroLift or Rezum was not statistically significant at 30 days follow up (Tutrone and Schiff, 2020).
3.14 After TURP 74% of people needed catheterisation for more than 24 hours compared with 45% after UroLift (Sonksen et al. 2015). After UroLift 57% of people needed post-procedure catheterisation compared with 87% after Rezum (Tutrone and Schiff, 2020). Catheterisation time after UroLift was statistically significantly shorter than with Rezum (1.2 days compared with 4.5 days; Tutrone and Schiff, 2020).
3.15 Eleven studies measured quality of life, with 8 showing a statistically significant improvement up to 5 years after UroLift treatment. Quality-of-life scores for people having UroLift were statistically significantly better than for people having Rezum (Tutrone and Schiff, 2020). In Sonksen et al. 2015 and Gratzke et al. 2016 there were no statistically significant differences between quality-of-life scores after TURP and UroLift at up to 12 and 24 months, respectively.
3.16 In 1 study (Sonksen et al. 2015) hospitalisation times were reduced for UroLift (time to discharge 1.0 days) compared with TURP (1.9 days).
3.17 One small study (Rukstalis et al. 2018) including 45 people described the clinical effectiveness of using UroLift in people with an obstructive median lobe. UroLift reduced BPHII and IPSS scores of symptom severity and improved sexual function (MSHQ-EjD score), quality-of-life measures and urological outcomes (Qmax values). The changes were statistically significant.
3.18 All 6 NICE shared learning case studies suggested that UroLift was beneficial when used in the NHS, resulting in improved IPSS and quality-of-life scores, reduced surgery times and reduced hospital stay. In 1 case study, the use of either general or local anaesthetic was compared, and no statistically significant differences were reported in IPSS, quality-of-life and pain scores after the procedure (NHS Fife, 2020).
The company's updated cost model is based on the original model but Rezum is a comparator and median lobe treatment is included
3.19 The company updated the original economic model to include Rezum as a comparator and median lobe treatment. Clinical parameters for UroLift were based on the LIFT study, using 5‑year post-procedure data (Roehrborn et al. 2017). The original guidance was based on clinical parameters from the same trial at 1 and 2 years after the procedure (Roehrborn et al. 2013 and 2014). For full details of the cost evidence, see section 4 of the assessment report update.
3.20 The EAC updated some of the model's parameters, including the cost of incontinence to cover the 5‑year time horizon, the consumables costs for TURP procedures and the NHS reference costs.
3.21 The overall cost of UroLift was reduced by £200 per surgery because of adjustments in the number of devices implanted and the duration of surgery. The number of implants per surgery was reduced from 4 to 3.5 and theatre time was decreased from 30 minutes to 14 minutes based on submitted audit data. These data were collected from NHS trusts over the past 3 years for 552 people who had treatment. The findings were supported by local audits carried out in NHS trusts and described in NICE shared learning case studies (NHS Fife 2020; Natarajan 2020; Dhanasekaran 2020b; Royal Devon and Exeter NHS Trust 2020; Norfolk and Norwich NHS Trust 2019).
3.22 Changing the follow up for UroLift surgery from a face-to-face consultation to a telephone consultation reduced the cost by £72.33 per consultation. This was based on an EAC cost of £37.00 for 20 minutes of band 6 nurse time.
3.23 In the model update the costs of bipolar TURP and monopolar TURP increased compared with the original guidance. This was because of an increase in consumable costs for bipolar TURP, and to a lesser extent for monopolar TURP. The cost of managing incontinence was also applied to the whole population who have treatment instead of only when treatment has failed.
The revised EAC base-case analysis shows that UroLift is cost saving when compared with all comparators
3.24 The EAC's revised base-case analysis showed that when UroLift is done as an outpatient procedure, UroLift is cost saving, per person, by:
£1,006 compared with bipolar TURP
£1,267 compared with monopolar TURP and
£1,255 compared with HoLEP.
When UroLift is done as a day-case procedure, it is cost saving, per person, by:
£96 compared with Rezum
£981 compared with bipolar TURP
£1,242 compared with monopolar TURP and
£1,230 compared with HoLEP.
The EAC concluded that UroLift is cost saving compared with monopolar TURP, bipolar TURP and HoLEP in the base case and in the company's and EAC's scenarios.
3.25 The UroLift economic model was compared with the model used in NICE's medical technologies guidance on Rezum. The committee concluded that there were too many uncertainties to draw firm conclusions about the costs of using Rezum compared with the costs of using UroLift. However, the Rezum base-case model results showed that Rezum was cost saving when compared with UroLift. The key parameters that were changed in the UroLift model were theatre time, length of stay and type of consultation after UroLift. If length of hospital stay were the same for Rezum and UroLift, Rezum would be cost saving compared with UroLift. However, the EAC's sensitivity analysis concluded that UroLift was only cost saving compared with Rezum if theatre time for the procedure was less than 16.7 minutes.