Learning curve and functional outcomes after laser enucleation of the prostate for benign prostate hyperplasia according to surgeon’s caseload

Laser enucleation of the prostate (LEP) has proven to be a safe and effective minimally invasive surgical treatment for bladder outlet obstruction (BOO) due to benign prostate hyperplasia (BPH) or prostate cancer (PCa) [1, 2]. Unlike transurethral resection of the prostate (TURP), it can be performed regardless of size and with less morbidity as compared to TUR-P and open simple prostatectomy [3, 4], even in patients requiring anticoagulation [5]. Although multiple enucleation techniques, using various energy sources, have been described over the past few years [6, 7], HoLEP remains the only size-independent option validated in randomized controlled trials [8, 9] and supported by level-1 evidence [10, 11]. As such HoLEP is recommended by European Association of Urology (EAU) and American Urology Association (AUA) guidelines [12, 13] and has become the gold standard for the surgical treatment of lower urinary tract symptoms due to BOO [11, 14, 15]. Despite its scientific status, diffusion into clinical practice is limited and classical TURP remains the predominant surgical BOO procedure by far [16]. An allegedly flat learning curve is regarded as main barrier against widespread clinical acceptance [17‐22]. However, more recently, a growing number of mentoring programs have been introduced, aiming to accelerate the learning curve [9, 22]. Especially one recent review [9] suggested that around 50 procedures might be sufficient to achieve a stable outcome, which could be reduced to about 25 in the case of structured training. We herein evaluated the learning curve of four novel LEP surgeons with regard to functional outcomes and complications. We postulated that applying a structured LEP mentoring program, stable functional outcomes and low complication rates can be achieved independent of the surgical caseload.

In our institutional database, we identified 677 eligible patients who underwent LEP at the University Hospital Frankfurt from November 2017 to January 2022. Of these, 84 (12%), 75 (11%), 82 (12%), 106 (16%), and 330 patients (49%) were operated by surgeons at A: < 25, B: 25–49, C: 50–99, D: 100–199, and E: ≥ 200 procedures.

We reported our institutional experience regarding safety and efficacy of LEP with regard to surgeon’s caseload. Our analyses yielded several important observations.

First, we reported a large series of nearly 700 patients, which is one of the biggest cohorts published concerning learning curve analyses [17‐19]. Moreover, our report does not represent a single surgeon´s experience but pooled results from four beginner surgeons, which make our results very robust and less susceptible to individual abilities and, as such, more generalizable.

Second, we found a highly significant and consistent decrease in OR time depending on surgeons’ caseload. This finding is plausible and consistent with virtually all previous reports that also confirmed a more time efficient procedure with growing experience. Brunckhorst et al. [12] reported a retrospective analysis of 253 consecutive cases of a single HoLEP surgeon. Consistent with our results, they found that enucleation speed was significantly increasing throughout the beginning of the learning curve, however, they found a plateau after 50–60 cases which is in contrast to our results, where we saw ongoing improvements in OR time until > 200 cases.

Third, we found excellent and stable functional outcomes independent of the surgeon’s caseload. This is in contrast to most of the other studies as highlighted in a systematic review by Kampantais et al. [9], which included 24 studies with a total of 5173 patients. They found ongoing improvements in functional outcomes until a plateau was reached after approximating 25–50 cases. This is in contrast to our findings, where we found stable functional outcomes for novices trained within our structured mentoring program independent of their caseload. However, the authors also concluded that a structured mentorship program would aid for faster progress, which might explain our better results.

Fourth, we evaluated major complication rates according to caseload and found no significant association between caseload and occurrence of major complications. Interestingly, the majority of publications report on functional outcomes or OR time. Only a minority of studies also reported on complications, when assessing the learning curve. This is worrisome as a low complication rate should be the utmost goal for every intervention, especially for a surgery, which is conducted for a benign disease. Rosenhammer et al. [25] found that ASA status and prostate cancer were independent predictors for grade ≥ 2 complications but not the caseload, which is consistent with our results, where we also found ASA status and prostate volume, but not caseload, as significant predictor of major complications. Moreover, Westhofen et al. [22] found the overall incidence of treatment-related adverse events was significantly higher without a training program. Conversely, Lerner et al. [19] showed a strong correlation between complication rates and learning curve in their study, especially when time between consecutive cases was more than 5 weeks. We did not evaluate the time in between the consecutive cases of the surgeons in our analysis, so we cannot compare these results. However, it appears logical especially without a structured mentoring program that already learned skills will diminish when time is too long in between two cases. Similarly, the institutional caseload seems important in this regard, as beside personal frequency of interventions, a high level of competence of the team in dealing with the intervention ensures high quality surgical outcomes [26].

Fifth, we applied Trifecta and Pentafecta criteria as defined by Robert et al. [21] to our results. These criteria of a perfect procedure consist of the combination of complete enucleation and morcellation within less than 90 min and without any conversion to standard TURP (Trifecta). In addition, information on complications or occurrence of stress urinary incontinence in 3 months after the operation are added for the Pentafecta criteria. Applying these criteria to our results, we achieved the Trifecta criteria mainly by reaching the expected OR time of less than 90 min after more than 50 cases, as we never used conversion to TUR-P. Similarly, the Pentafecta criteria were reached at 50 cases due to OR time because of consistently low rates of complications and urinary incontinence. This is comparable to the report by Peyronnet et al. [20] who met the criteria in four consecutive patients after the 40th procedure in the HoLEP group.

Taken together, we reported on a large series of nearly 700 patients with pooled results from four beginner surgeons, which make our results very robust and pertinent. We found a highly statistically significant improvement in OR time according to surgeon’s caseload. This is in accordance to virtually all other reports. However, we found stable functional outcomes and stable low complications rates, independent from the surgeon’s caseload. This is in contrast to many other studies that report an improvement of these variables according to the surgeon’s caseload. Apparently, a structured mentoring program over the first 50 cases, which was applied in our institution can help overcome the learning curve.

Several limitations apply to our study. Although the database is designed prospectively, the analysis of the learning curve is retrospective and thus selection biases may occur. Low number of cases in each caseload group limited depth of subgroup analyses. Furthermore, our prospective database may be influenced by a negative selection bias regarding the admittance of patients with a particularly high perioperative risk to our tertiary care university center. Nevertheless, we included all LEP patients since its implementation at our institution to allow most comprehensive analyses.

With a structured mentoring program, the safety and efficacy of LEP can be ensured even during the learning curve with excellent functional outcomes. Only the OR time decreases significantly with increasing experience of the surgeon.