Hypothesis / aims of study
Benign prostatic hyperplasia (BPH) is one of the most common causes of lower urinary tract symptoms (LUTS) in men, and current international guidelines recommend a stepwise approach to the treatment of BPH/LUTS (1). Surgery remains the gold-standard in severe or refractory LUTS, and transurethral resection of the prostate (TURP) is the procedure of choice for the majority of men with BPH/LUTS.
Although considered a safe procedure, sexual side effects after TURP are still an important issue. In recent years, new approaches to TURP have been developed, with the objective of reducing morbidity while maintaining the benefits of the procedure. In that regard, ejaculation preserving techniques are a promising development, with ejaculation preserving TURP (epTURP) showing similar symptomatic and functional outcomes to the classic technique.
To our knowledge, no diagnostic feature has been firmly established as a predictor for TURP outcomes. However, the results of this procedure are heterogenous, with studies reporting a failure in symptomatic relief in around 12% of patients (2), raising the possibility that such predictors exist, at least for some patients. These may however be statistically concealed in the published studies, due to the analysis of the studied populations as a whole, irrespective of important factors such as Bladder Outlet Obstruction (BOO) severity. Therefore, the aim of this study is to analyze whether pre-operative BOO severity may affect a possible influence of prostate resected weight in TURP outcomes. Moreover, since there is a lack of literature reporting which patients are good candidates for less extensive prostate resections, our second goal is to analyze whether BOO severity may distinguish between adequate and inadequate candidates for these techniques.
Study design, materials and methods
We conducted a retrospective analysis of patients submitted to TURP in a university hospital between February 2011 and November 2015. Exclusion criteria were previous LUTS surgery, prostate cancer, urethral stricture or voided volumes <125mL in uroflowmetry. Pre-operative data was retrieved, including clinical history, comorbid conditions, medications, uroflowmetry and prostate volume (determined by transrectal ultrasound). Post-operatively, weight of the resected dry specimen and post-operative uroflowmetry values were considered.
Patients were stratified in two groups according to pre-operative Qmax, following the Siroky-Liverpool nomograms, which define a cut-off value of 10mL/s as a very strong predictor of BOO (≤2 standard deviations of the mean for a voided volume ≥125mL).
Results
A total of 185 patients were included, with a mean age of 58.5 (±7.2) years and a mean pre- and post-operative Qmax of 8.8 ± 3.6 and 14.9 ± 7.2 mL/s, respectively. The mean change in Qmax after surgery was 6.2 ± 7.1 mL/s. Other demographic and clinical characteristics are displayed in Table 1.
In the whole sample analysis, no statistically significant correlations were found between absolute resected prostate weight or percentage of resected weight and post-operative Qmax (r2=0.019, p=0.063 and r2=0.019, p=0.064, respectively). Similarly, the pre/post-operative difference in Qmax showed no correlation with the resection weight (r2=0.006, p=0.290) or the percentage of resected prostate weight (r2=0.006, p=0.283).
When stratifying patients into two groups according to pre-operative Qmax <10mL/s (n=121) and ≥10mL/s (n=64), no relevant differences in demographic or clinical characteristics were found between the groups (Table 2).
In the group of patients with pre-operative Qmax<10mL/s, post-operative Qmax was correlated with absolute resected prostate weight (r2=0.038, p=0.032), as well as with percentage of resected prostate weight (r2=0.051, p=0.013). In these patients, the difference in pre/post-operative Qmax was also strongly associated with absolute resected prostate weight (r2=0.036, p=0.037) and percentage of resected prostate weight (r2=0.040, p=0.029).
Neither of the above-mentioned correlations were established in the group of patients with pre-operative Qmax≥10mL/s. Absolute resected prostate weight and percentage of resected prostate weight were not associated with post-operative Qmax (r2=-0.001, p=0.796 and r2=-0.009, p=0.458, respectively), nor with peri-operative change in Qmax (r2=0.009, p=0.463 and r2=-0.018, p=0.294, respectively).
Interpretation of results
In agreement with previous studies, our analysis didn’t show an association between the extension of adenoma resection and post-operative outcomes in the whole sample analysis. Similarly, other reports explored the influence of pre-operative prostate volume in this correlation, yet no differences were noted (3).
Even though bladder outlet obstruction may only be diagnosed via urodynamic studies, Siroky-Liverpool uroflowmetry nomograms predict this condition with great efficacy using bladder volume and maximum flow rate. As stated by the authors, a Qmax<10/mL is a strong predictor of a clinically relevant BOO for bladder volumes of 125mL or higher. This cut-off value was thus used to dichotomize between patients considered as severely obstructed (Qmax<10mL/s) and patients with mild to moderate voiding dysfunction (Qmax≥10mL/s).
Our analysis suggests that in patients with pre-operative Qmax<10mL/s, a more thorough resection of the prostate is associated with better surgical outcomes. This association was not present in patients with pre-operative Qmax≥10mL/s, although surgery was beneficial in both groups. As suggested by other authors, prostate initial volume could play a role in this relationship, since a larger amount of tissue may need to be resected in order to treat BOO in larger rather than in smaller prostates. However, no such influence seems to exist, as the ratio between resected weight/prostate volume is similarly correlated with post-operative Qmax only in the group of patients with pre-operative Qmax<10mL/s. These results suggest that severely obstructed patients may profit from a complete adenoma resection. Conversely, men with higher maximum flow rates may be good candidates for techniques with less morbidity, such as epTURP. If further studies confirm our results, surgeons should be encouraged to adapt their TURP technique to the patients’ pre-operative clinical details and expectations concerning surgical side effects, in a patient-tailored way.
The present study has several limitations. First, the retrospective design may be a source of bias. Second, we did not consider symptom scales such as IPSS in our analysis, mainly due to a high level of missing data. Even though previous studies proved a high correlation between maximum flow rate and IPSS, LUTS grading and change after surgery would have been of great value in the analysis. Another source of relevant information would be urodynamic studies, which in our center are not routinely performed to all men with BPH/LUTS. Furthermore, in our study, the resected prostate weight was measured in the Pathology laboratory after fixation with formaldehyde, using precision scales. This fixation method results in a considerable reduction in specimen weight, and therefore this parameter, while valid for analysis within our studied group, is not directly comparable to previous studies.