The Diagnosis Of Bladder Outlet Obstruction In Male LUTS

Part I Development of a clinical nomogram for predicting bladder outlet obstructionBACKGROUND Male lower urinary tract symptom (LUTS) can be divided into storage (irritative) symptoms, voiding (obstructive) symptoms, and post-voiding symptoms. As to the aged male, the most common cause is benign prostatic hyperplasia (BPH) and its subsequent bladder outlet obstruction (BOO). Benign prostatic hyperplasia (BPH) is one of the most common diseases in the male, and the morbidity of BPH increases with age in the male. Histological benign prostatic hyperplasia may initially occur from the age of 40. Then probably 50% of men suffer from LUTS/BPH at their age of 60. At the age of 80, approximately 83% of the male would get in trouble with LUTS/BPH. Mostly, BOO is the first critical factor leading to the moderate-to-severe LUTS. BOO does not only badly depress the patient’s quality of life, but also lead to the dysfunction of the detrusor, the hydrops of the upper urinary tract and the renal function deterioration in the end.Currently, the urodynamic s pressure-flow study (PFS) is recommended as golden standard to diagnose BOO by the International Continence Society (ICS). However, the urodynamic study is rather limited in clinical use. Firstly, the urodynamic device is relatively expensive and the matched medical consumable materials may be costly. So, a large amount of the regional hospitals and basic hospitals in town can’t afford, and lots of patients in straitened circumstances also can’t pay for the examination charge. Secondly, how to perform, interpret, and report pressure-flow examinations are far from perfection due to the complexity and speciality of the pressure-flow study. If the operation and the report were not normative, there would be terribly variation and error which may lead to misdiagnose. Finally, urodynamic study is an invasive measurement, which is associated with a considerable rate of complications and morbidity including discomfort and pain, urinary retention, gross hematuria, urinary tract infection and fever. The blindly and indiscriminate use of the urodynamic study will do harm to the patients seriously.Therefore, how to diagnose and classify BOO is absolutely important, and the diagnostic results may affect the decision-making and change the following clinic therapy. Because the patients with BOO usually need to undergo the surgery to relieve the obstruction. However, how to detect and evaluate the patients who really deserve to carry out the urodynamic study? Are there any ordinary and noninvasive methods of diagnosing bladder outlet obstruction in male LUTS? Although numerous researches have been conducted home and abroad, there is still a clinical problem expecting more studies.OBJECTIVE To evaluate and assess the efficacy and validity of the most common and noninvasive parameters in daily clinical practice for detecting bladder outlet obstruction compared with the golden standard pressure-flow study. Find out the best parameters to develop a clinical nomogram for predicting the probability of the bladder outlet obstruction in male LUTS in our population. With the purpose of saving the expenditure, simplifying the procedure, and providing useful reference in clinical practice.MATERIALS AND METHORDS Retrospectively analyze the outpatients and inpatients of male LUTS from November 2003 to November 2015 in Guangzhou First People’s Hospital. Inclusion criteria:i. Male LUTS, aged>40 yr. ii. PSA<4ng/ml; if PSA> 4ng/ml, a pathological diagnosis must be needed to exclude the prostate cancer, iii. Complete the pressure-flow study. Exclusion criteria:i. Neurogenic baldder. ii. Prostate cancer, iii. Bladder cancer, iv. Patients with bladder diverticulum and bladder stone, v. Patients with urethrostenosis. vi. Patients with acute urinary tract infection, vii. Patients with urinary tuberculosis, viii. Patients with pelvic fracture and urethral injury, ix. History of pelvic surgery and pelvic radiotherapy, x. Drug therapy in 2 weeks, including al-Adrenoceptor antagonists, muscarinic receptor antagonists, 5a-reductase inhibitor and other medicine which may affect the function of bladder.PSA was analyzed by the Abbott Axsym (Abbott Corporation) PSA assay before June 2006 and the Roche Elecsys 2010 PSA assay (Roche Group) after June 2006. Prostate volume (PV) and transition zone volume (TZV) were determined by ultrasonography measurement and was calculated using the mathematical formula, PV (or TZV) (ml)= 0.52 x anterior-posterior diameter (cm)×transverse diameter (cm) x superior-inferior diameter (cm). The Uroflowmetry was taken and the pressure-flow study (PFS) was conducted by the professional urologist. The operation was abided by the Good Urodynamic Practice according to the International Continence Society (ICS).Collecting the data, including the urodynamic parameters, BOO parameters, and other clinical parameters. Urodynamic parameters:Qmax, Qave, Voiding volume, Pdet.Qmax, PVR, RF, Abrams/Griffiths number, BOOI.BOO parameters were divided into BOO and none-BOO, according to the ICS nomogram and Schaefer’s nomogram respectively. Other clinical parameters:prostate volume, transitional zone volume, transitional zone index, prostate specific antigen. The clinical parameters were analyzed with Spearman rank correlation. Receiver operating characteristic (ROC) curve was used to evaluate the efficiency of each clinical parameter for detecting bladder outlet obstruction. Univariable and multivariable analyses were used to test the effect of each clinical parameter for detecting bladder outlet obstruction. Then, find out the best independent predictors on the diagnosis of bladder outlet obstruction, and develop the clinical nomogram for predicting the probability of the bladder outlet obstruction. Statistic analyses were performed using SPSS (Version 21). All statistical tests were two-sided with P<0.05 considered statistically significant. The nomogram was conducted with R statistical software (Version 3.2.1).RESULT A total of 1866 patients were retrieved, of which 267 patients were unable to void, or the PFS curve were unsatisfactory, and were excluded. The data from1599 patients were analyzed. Of the 1599 patients, the median age was 71.0 (64.0/76.0), the median PV was 45.0(28.0/65.2) ml, the median TZI was 23.1(13.7/38.1) ml, the median PSA was 2.8(1.1/6.5) ng/ml. Of the 1599 patients, 1241 patients were diagnosed according to the ICS nomogram,55.9%(694/1241) were BOO and 44.1%(547/1241) were none-BOO.1319 patients were diagnosed according to the Schaefer’s nomogram,60.3%(796/1319) were BOO (Schaefer Class>3) and 39.7%(523/1319) were none-BOO (Schaefer Class<3).The most common used clinical parameters included PV, TZV, PSA, Qmax, and PVR. Analyzed with Spearman rank correlation, the clinical parameters show significant correlation.1) Significant negative correlation was found between Qmax and PV, TZV, PSA, PVR.2) Significant positive correlation was found between PV and PSA, PVR, especially between PV and PSA, and the correlation coefficient was up to 0.690.3) Significant positive correlation was found between TZV and PSA, PVR, especially between TZV and PSA, and the correlation coefficient was up to 0.678.4) None significant correlation was found between PSA and PVR.Considering the ICS-nomogram as the golden standard and using the parameter PV to detect bladder outlet obstruction (BOO), we found that, the area under the ROC curve (AUC) for PV to detect BOO was 0.803, while the AUCs for TZV, TZI were 0.807,0.698 respectively. The best cut-off of PV to detect BOO was 37.72ml, which had a sensitivity of 81.7%, a specificity of 69.1%, and a positive predictive value of 78.3%. The best cut-off of TZV to detect BOO was 19.35ml, which had a sensitivity of 78.6%%, a specificity of 71.4%, and a positive predictive value of 80.1%. When detecting BOO according to the Schaefer’s nomogram, we found the AUC for PV was 0.806, while the AUCs for TZV, TZI were 0.814,0.713 respectively. The results according to the Schaefer’s nomogram coincide well with the results according to the ICS nomogram.Considering the ICS-nomogram as the golden standard and using the parameter PSA to detect BOO, we found that, the AUC for PSA to detect BOO was 0.775. When cut-off was 2.5ng/ml, the sensitivity was 72.0% the specificity was 70.3%, and the positive predictive value was 74.9%.The best cut-off of PSA to detect BOO was 3.61ng/ml, which had a sensitivity of 60.9%, a specificity of 82.2%, and a positive predictive value of 80.8%. When detecting BOO according to the Schaefer’s nomogram, we found the AUC for PV was 0.773. The best cut-off was 2.44ng/ml, which had a sensitivity of 72.2%, a specificity of 70.7%, and a positive predictive value of 77.6%. The results according to the Schaefer’s nomogram coincide well with the results according to the ICS nomogram.Considering the ICS-nomogram as the golden standard and using the parameter PVR and RF to detect BOO, we found that, the AUC for PVR and RF was 0.641 and 0.679, respectively. When detecting BOO according to the Schaefer’s nomogram, we found the AUC for PVR and RF was 0.642 and 0.684, respectively. The results according to the Schaefer’s nomogram coincide well with the results according to the ICS nomogram. Results indicate that PVR and RF were not superior to PV, TZV, and PSA.Considering the ICS-nomogram as the golden standard and using the parameter Qmax to detect none BOO, we found that the AUC for Qmax was 0.740. The best cut-off of Qmax was 11.85ml/s, which had a sensitivity of 69.9%, a specificity of 64.8%, and a positive predictive value of 72.48%. When detecting none BOO according to the Schaefer’s nomogram, we found the AUC for Qmax was 0.721. The results according to the Schaefer’s nomogram coincide well with the results according to the ICS nomogram.Based on the univariable analyses testing the effect of Qmax, PV, TZV, PSA, and PVR on the diagnosis of BOO, we conduct the multivariable analyses, and select Qmax, PV, PSA as the best clinical parameters to predict BOO. The Logistic regression equation:Log(p)=0.3324-0.2018*Qmax+0.0266*PV+1.1351*PSA. We finally developed a nomogram model which could predict the probability of the bladder outlet obstruction. According the internal validation of the nomogram, we get a concordance index of 0.854. Results indicated that, this nomogram showed high accuracy (85.4%) for the detection of BOO in male LUTS.CONCLUSION Compared with the golden standard pressure-flow study, the most common and noninvasive parameters in male LUTS such as Qmax, PV, TZV, TZI, PSA would be of a certain practical value, and TZV, PV, PSA, Qmax may be better in detecting bladder outlet obstruction (BOO) with the AUC of 0.807,0.803, 0.775,0.740, respectively. According to clinical practice, we developed a nomogram model with parameters of Qmax, PV, and PSA, which presented a higher accuracy (85.4%) in detecting BOO. The nomogram would help predicting BOO simply, noninvasively, individually, accurately, and providing more valuable reference and guidance in clinical decision.Part II Construction of a clinical nomogram in order to classify bladder outlet obstruction and detrusor underactivityBACKGROUND In our clinical urodynamic practice, we found that numerous of male patients with LUTS may suffer from detrusor underactivity (DU), while lots of them may have potential bladder outlet obstruction (BOO) simultaneously. However, we couldn’t evaluate the bladder contractility and distinguish the detrusor underactivity from the ICS-Nomogram. Therefore, we may misdiagnose some of the patients of potential bladder outlet obstruction. How to identify BOO and/or DU? Currently, there is no consensus on the standardisation of current concepts, and the diagnostic criteria and thresholds for DU have not yet been established or unified.OBJECTIVE To construct a bladder outlet obstruction-bladder contractility nomogram for the purpose of classifying bladder outlet obstruction, detrusor underactivity, and bladder outlet obstruction together with detrusor underactivity. This nomogram may provide valuable reference and guidance in both our clinical urodynamics and clinical practice.MATERIALS AND METHORDS Retrospectively analyze the outpatients and inpatients of male LUTS from November 2003 to November 2015 in Guangzhou First People’s Hospital. The inclusion criteria and exclusion criteria were the same as part 1, and the operation of the pressure-flow study and the quality control were also abided by the Good Urodynamic Practice according to the International Continence Society (ICS). The age, maximum flow rate (Qmax), average flow rate (Qave), detrusor pressure at maximum flow rate (Pdet.Qmax), post-voiding residual (PVR), bladder capacity (BC), voiding efficiency (VE), bladder outlet obstruction index (BOOI), bladder contractility index (BCI), maximum Watts factor (Wmax) were analyzed. After eliminating those of BOOI< 0, we finally got 1278 patients into analysis, and an iterative procedure was implemented using the computer program Matlab (The Mathworks). Based on the available Wmax-BOOI data points, the best fitting (second order polynomial) curve was defined. The iterative algorithm and linear interpolation was applied to determine the specific percentile values (< 10th, 25th,50th,75th, and> 90th). In this way, all the patient were subdivided into six subgroups, and One-way ANOVA was used for the comparison of one parameter over the groups, and independent samples t-test was used to compare parameters between the two percentile groups. A P-value<0.05 was considered statistical significant. All statistical analyses were done with SPSS, version 21.RESULT In this study, patients without detrusor overactivity had significantly different bladder capacity (300.4 vs.155.9ml), voiding efficiency (71.3% vs.84.4%), bladder contractility index (89.1 vs.107.5), maximum Watts factor (8.7 vs.13.0 W/m2), and detrusor pressure at maximum flow rate (48.1 vs.73.5 cmH2O), compared with patients with detrusor overactivity. (p<0.001)As the grade of BOO rose up, the Qmax decreased from 12.6(8.0-15.6) ml/s to 5.2(3.4-7.5) ml/s, the PVR increased from 10.0(0.0-45) ml to 80.0(25.0-175.0) ml, the VE decreased from 96.1%(67.8-110.1%) to 55.0%(34.1-84.5%), the BCI increased from 66.5(47.4-84.0) to 148.3(134.5-166.6), the Wmax increased from 5.6(4.0-8.4) W/m2 to 18.0(15.4-20.9)W/m2 and the PdetQmax increased from 20.8(16.8-27.1) cmH2O to 129.1(114.5-144.7) cmH2O.Comparison of the six percentile groups revealed that, Wmax, Qmax, Qave, PdetQmax, VE, and BCI significantly decreased with decreasing percentile group (P < 0.001), whereas PVR and bladder capacity significantly increased (P< 0.001). In contrast, BOOI ((P=0.669) remained stable over the different percentile groups. The comparison of values between the two lowest percentile groups didn’t show statistically significant differences for the majority of investigated parameters, while significant differences between patients of<25th percentile groups and those of the 25th-50th percentile groups were seen for Wmax (7.8 vs.10.2 W/m2, P< 0.001), Qmax(7.6 vs.9.4ml/s, P<0.001), Qave (3.6 vs.4.4ml/s, P<0.001), PVR (86.4 vs. 61.4ml, P<0.001), Bladder Capacity (275.9 vs.241.9ml, P<0.001), VE (40.3%vs. 75.4%, P<0.001), BCI (79.4 vs.95.7, P<0.001). Patients in the<25th percentile groups were significantly different compared to patients in higher percentiles with regard to PVR, voiding efficiency, and (cystometric) bladder capacity, which may indicate DU in clinical practice. Therefore, we propose the diagnosis of DU for patients below the 25th percentile in the nomogram.CONCLUSION Based on our clinical urodynamic dataset, we construct a bladder outlet obstruction-bladder contractility nomogram, which may be useful to classify bladder outlet obstruction, detrusor underactivity, and bladder outlet obstruction together with detrusor underactivity in mlae LUTS. This nomogram may provide valuable reference and guidance in both our clinical urodynamics and clinical practice, but need more clinic studies to confirm its accuracy and practicability.