1Archivio Italiano di Urologia e Andrologia 2020; 92, 1 ORIGINAL PAPER Free prostate-specific antigen outperforms total prostate-specific antigen as a predictor of prostate volume in patients without prostate cancer Sinan Avci, Efe Onen, Volkan Caglayan, Metin Kilic, Murat Sambel, Sedat Oner University of Health Sciences, Bursa Yuksek Ihtisas Training and Research Hospital, Department of Urology, Bursa, Turkey. Objective: In the management of benign prostatic hyperplasia (BPH), urology guide- lines recommend medical or surgical treatments according to different prostate volumes (PV). The aim of this study was to analyze the relationships between PV and age, total and free prostate specific antigen (tPSA, fPSA) and fPSA/tPSA ratio in patients without histologically proven prostate cancer. Materials and methods: A retrospective analysis was made of the data of 1334 patients who underwent transrectal ultra- sound (TRUS)-guided prostate biopsy between January 2016 and October 2018. A total of 438 patients with available data for age, tPSA and fPSA levels and PV calculated by TRUS were enrolled in the study. Patients with chronic prostatitis pathology in addition to BPH were also noted and evaluated as a separate group. Results: There were significant correlations between PV and age, tPSA, fPSA, fPSA/tPSA ratio (r = 0.210, r = 0.338, r = 0.548, r = 0.363 respectively). In multivariate linear regression analysis, fPSA was found to be the only predictor for PV (p < 0.001) when compared to age (p = 0.097), tPSA (p = 0.979) and fPSA/tPSA ratio (p = 0.425). In patients with chronic prostatitis pathology there were significant correla- tions between PV and age, tPSA, fPSA, fPSA/tPSA ratio (r = 0.279, r = 0.379, r = 0.592, r = 0.359, respectively). The multivariate linear regression analysis showed a signifi- cant correlation only between PV and tPSA and fPSA/tPSA ratio but not with fPSA and age (p = 0.008, p = 0.015, p = 0.430, p = 0.484, respectively). In men with only BPH pathology there were significant correlations between PV and age, tPSA, fPSA, fPSA/tPSA ratio (r = 0.223, r = 0.385, r = 0.520, r = 0.287, respectively) In multivariate linear regression model the significant correlation was shown only between PV and fPSA (p < 0.001). Conclusions: Although tPSA was significantly correlated with PV in patients without prostate cancer, the correlation between fPSA and PV was much stronger. However, it should be kept in mind that the efficacy of fPSA may be limited in patients with clinically unknown prostatic inflammation. KEY WORDS: Benign prostatic hyperplasia; Chronic prostatitis; Free prostate-specific antigen; Free prostate-specific antigen/total prostate-spesific antigen ratio; Prediction; Prostate-specific anti- gen; Prostate volume. Submitted 18 July 2019; Accepted 1 September 2019 Summary No conflict of interest declared. DOI: 10.4081/aiua.2020.1.1 INTRODUCTION Lower urinary tract symptoms (LUTS) have traditionally been related to bladder outlet obstruction, which is often caused by benign prostatic enlargement resulting from the histological condition of benign prostatic hyperplasia (BPH) (1, 2). Prostate volume (PV) predicts symptom progression and the risk of complications such as uri- nary retention (3). PV may also be determinative in the decision for BPH treatment. For example, in urology guidelines the use of 5 alpha reductase inhibitors in prostate volumes > 40 cc is recommended for medical treatment. Surgical treatment options also vary accord- ing to the prostate volume in guidelines (4). Therefore, it is important to know the PV correctly for the progno- sis and treatment of the disease. Digital-rectal examination (DRE) is the simplest way to assess PV, but the correlation to PV is poor. Underestimation of PV by DRE increases with increasing transrectal ultrasound (TRUS) volume, particularly where the volume is > 30 mL (5). TRUS is more accurate in determining PV than DRE and transabdominal ultra- sound so it is the standard method recommended for measurement of PV (6, 7) However, it is an expensive, time-consuming, and uncomfortable modality for the ini- tial evaluation of men with LUTS. Therefore, another cheaper and simply applicable method, other than DRE is needed to predict the PV correctly in our daily practice. The prediction of PV can be based on total and free prostate specific antigen (PSA). Both PSA forms predict the TRUS prostate volume (± 20%) in > 90% of cases (8, 9). The aim of this study was to analyze the relationships between total PSA, free PSA, age and prostate volume in patients with histologically proven BPH. Furthermore, the implication of the use of free PSA as a proxy marker to estimate PV was analyzed. MATERIALS AND METHODS The data of 1334 men who underwent transrectal ultra- sound (TRUS) guided prostate biopsy between January 2016 and October 2018 were analyzed retrospectively. Approval for the study was granted by the Local Ethics Committee (reg. no: 2011-KAEK-25 2018/11-03). Patients with pathological results of cancer, prostatic intraepithelial neoplasia (PIN) or atypical small acinar proliferation (ASAP), aged < 40 years, with PSA levels > 30 ng/dl, with a history of 5alpha-reductase inhibitor therapy, phytotherapy or any invasive therapy for BPH were excluded. Patients who had a cystoscopy, colonoscopy, TRUS, prostate biopsy, acute prostatitis, Sinan_Stesura Seveso 01/04/20 18:51 Pagina 1 Archivio Italiano di Urologia e Andrologia 2020; 92, 1 Sinan Avci, Efe Onen, Volkan Caglayan, Metin Kilic, Murat Sambel, Sedat Oner 2 urinary tract infection and urinary retention during the previous month were also omitted. Those with a subse- quent positive prostate biopsy were also excluded, and those with negative prostate biopsies were included. The pathology reports with chronic prostatitis in addition to BPH were also noted. The PV of the patients were calculated by measuring three dimensions of the prostate with TRUS, and using the ellipsoid formula (PV=height*width*length*0.52). For prostate enlargement, a volume of 40 ml was con- sidered as the cut-off value. Serum PSA levels were meas- ured using the chemiluminescent microparticle immunoas- say (CMIA) method prior to any prostate manipulation, including DRE, TRUS and biopsy. A total of 438 patients who met the inclusion criteria, with the available data of age, total-free PSA levels and PV calculated by TRUS were enrolled in the study. Patients were stratified by age into three groups: < 60 years, 60-70 years and > 70 years. Patients with PSA lev- els < 10ng/dl and PSA levels between 10 ng/dl and 30ng/dl were also evaluated as two separate groups. Data obtained in the study were analysed using SPSS ver- sion 15.0 software (SPSS, Inc., Chicago, IL, USA). Correlation and linear regression analyses were per- formed to evaluate the relationships between age, total PSA, free PSA and PV. Receiver operating characteristics (ROC) curves were constructed to evaluate the ability of free PSA to predict PV for the entire cohort and each sub- group. A value of p < 0.05 was accepted as statistically significant. RESULTS A total of 896 patients with any exclusion criteria or with incomplete data were excluded from the study. The remaining 438 patients had a mean age of 64.82 ± 7.18 years, median total PSA value of 6.82 ng/dl (min- max = 2.75-29.85), median free PSA value of 1.68 (min- max = 0.24-11.25), median free PSA/total PSA ratio of 0.235 (min-max = 0.02-0.82) and median PV of 74 cc (min-max = 40-422). The baseline characteristics of the entire cohort and each subgroup are shown in Table 1. Statistically significant correlations were determined between PV and age, total PSA, free PSA, free PSA/total PSA ratio when the entire cohort was analyzed (p < 0.001 and r = 0.210, p < 0.001 and r = 0.338, p < 0.001 and r = 0.548, p < 0.001 and r = 0.363 respectively) (Table 2). Free-PSA was found to be the only predictor for PV (p < 0.001) in the multivariate linear regression model when compared to age (p = 0.097), total PSA (p=0.979) and free PSA/total PSA ratio (p = 0.425) (Table 3). Patients with chronic prostatitis pathology and patients with pathology reported as BPH only were evaluated sep- arately. In the chronic prostatitis group there were signif- icant correlations between PV and age (p = 0.011 and r = 0.279), total PSA (p < 0.001 and r = 0.379), and free PSA (p < 0.001 and r = 0.592), fPSA/tPSA ratio (p < 0.001 and r = 0.359) (Table 2). In the multivariate linear regres- sion model, a significant correlation was shown only between PV and tPSA, fPSA/tPSA (p = 0.008, p = 0.015 respectively) (Table 3). In the BPH only group, there were significant correlations between PV and age (p < 0.001 and r = 0.223), total PSA (p < 0.001 and r = 0.385), free PSA (p < 0.001 and r = 0.520) and fPSA/tPSA ratio (p < 0.001 and r = 0.287) (Table 2). In the multivariate linear regression model, a significant correlation was shown only between PV and fPSA (p < 0.001) (Table 3). There were correlations between free PSA and PV in all three age groups (< 60 years p < 0.001 and r = 0.546, 60- 70 years p < 0.001 and r = 0.506, > 70 years p < 0.001 and r = 0.483) (Table 4). There were correlations between free PSA and PV when the cohort was separated according to total PSA value as below or above 10 ng/dl Table 1. Characteristics of the patient population. Number Age (years) Total PSA (ng/dl) Free PSA (ng/dl) fPSA/tPSA ratio Prostate volume (cc) of patients (mean ± SD) (median, min-max) (median, min-max) (median, min-max) (median, min-max) Age groups < 60 121 56.05 ± 3.66 5.59 1.20 0.208 61 (28%) (2.75-29.85) (0.24-7.74) (0.03-0.69) (40-290) 60-70 224 65.43 ± 2.72 7.19 1.70 0.24 80 (51%) (2.9-28.62) (0.36-9.56) (0.02-0.82) (40-422) > 70 93 74.72 ± 2.94 9.05 2.21 0.26 83 (21%) (2.75-28.63) (0.50-1.25) (0.07-0.55) (40-297) Pathology results Chronic 114 65.69 ± 6.19 6.37 1.71 0.252 81 prostatitis group (26%) (2.9-24.09) (0.48-9.56) (0.09-0.69) (40-297) Only BPH group 324 64.39 ± 7.09 6.86 1.805 0.248 79.5 (74%) (2.75-29.85) (0.24-0.26) (0.03-0.82) (40-422) tPSA levels < 10 ng/dl 325 63.93 ± 7.09 5.995 1.43 0.239 68.50 (74%) (2.75-9.91) (0.24-5.28) (0.03-0.82) (40-234) > 10 ng/dl 113 67.34 ± 6.91 13.84 3.05 0.230 95.0 (26%) (10.04-29.85) (0.36-1.25) (0.02-0.56) (40-422) Total cohort 438 64.82 ± 7.19 6.82 1.69 0.235 74 (100%) (2.75-29.85) (0.24-1.25) (0.02-0.82) (40-422) tPSA: Total prostate-specific antigen; fPSA: Free prostate-specific antigen; BPH: Benign prostatic hyperplasia; SD: Standard deviation; min: Minimum; max: Maximum. Sinan_Stesura Seveso 01/04/20 18:51 Pagina 2 (PSA < 10 ng/dl p < 0.001 and r = 0.494, PSA > 10 ng/dl p < 0.001 and r = 0.512) (Table 4). The cut-off level for free PSA was determined as 1.285 ng/dl for the prediction of prostate volume > 40 cc (Table 5). The cut-off levels are shown in Table 5 for the other subgroups in which free PSA was significant in predict- ing PV. The receiver operating characteristic (ROC) curves of each group for fPSA in the prediction of prostate vol- ume < 40cc or > 40cc are shown in Figure 1. DISCUSSION In the management of benign prostatic hyperplasia (BPH), urology guidelines recommend medical or surgical treat- ments according to different prostate volumes (PV). Therefore, accurate determination of PV is crucial for the choice of treatment and for the prediction of treatment outcomes such as the probability of urinary retention and the need for surgery (10-13). Nevertheless, BPH is a progressive disease and that progression is related to pro- static enlargement (14-16). Transrectal ultrasound (TRUS) is more accurate in deter- mining PV than transabdominal ultrasound so it is the ref- erence method used to measure PV (6, 7). However, it is an expensive, time-consuming and uncomfort- able method, and the equipment is not available in most primary set- tings. Moreover, in the initial eval- uation of BPH patients, neither TRUS nor transabdominal ultra- sound is recommended according to the urology guidelines. Digital- rectal examination (DRE) is simple to perform and useful for estimat- ing the PV but it may assess small prostates as larger, and large ones as smaller. Therefore, there is a need for a reliable, practical and cheap method as an alternative to ultrasonography and DRE in daily practice (17). Several independent investigators have verified the log-linear rela- tionship between serum total PSA and PV in different populations and races with similar results (18). Hochberg et al. (19) and Coban et al. (20) found the corre- lation coefficient between total PSA and PV to be 0.39 and 0.41, respectively. Similarly, in the pres- ent study, a significant correlation was determined between total PSA and PV (r = 0.33) but this result was not confirmed in the multivariate analysis (Table 3, p: 0.979). In agreement with the current study multivariate analy- sis findings, some investigators have proposed that the variability in the relationship between total PSA and PV can preclude the accurate prediction of the prostate volume using the total PSA alone for an individual patient (9, 21). 3Archivio Italiano di Urologia e Andrologia 2020; 92, 1 Free PSA as useful tool to predict prostate volume Table 3. Multivariate analysis of factors effecting the PV. Unstandardized Standardized 95% CI coefficients coefficients t p ß SE ß Lower bound Upper bound Total cohort Age 0.450 0.265 0.075 1.696 -0.71 0.972 0.091 tPSA 0.023 0.890 0.003 0.026 -1.726 1.772 0.979 fPSA 11.640 3.178 0.427 3.663 5.394 17.886 < 0.001 fPSA/tPSA ratio 25.800 31.322 0.067 0.824 -35.763 87.363 0.411 Chronic prostatitis group Age 0.476 0.677 0.068 0.703 -0.873 1.825 0.484 tPSA 7.312 2.706 0.671 2.702 1.924 12,699 0.008 fPSA -6.615 8.341 -0.224 -0.793 -23.220 9.991 0.430 fPSA/tPSA ratio 169.948 68.279 0.456 2.489 34.015 305.880 0.015 Only BPH group Age 0.494 0.385 0.076 1.283 -0.265 1.252 0.201 tPSA -1.437 1.387 -0.158 -1.036 -4.169 1.295 0.301 fPSA 17.079 4.879 0.615 3.500 7.465 26.692 < 0.001 fPSA/tPSA ratio -27.425 46.072 -0.067 -0.595 -118.201 63.351 0.552 PV: Prostate volume; tPSA: Total prostate-specific antigen; fPSA: Free prostate-specific antigen; SE: Standart error; CI: Confidence interval. Table 5. Receiver operating characteristic (ROC) curves for free PSA to predict whether prostate volume is > 40cc or < 40cc. AUC SE p 95% CI Sensitivity (%) Specificity (%) Cutoff levelT Total cohort 0.780 0.036 < 0.001 0.709-0.851 72.8 73.5 1.285 Only BPH group 0.749 0.054 < 0.001 0.643-0.855 66.5 76.0 1.495 Age < 60 years 0.782 0.059 < 0.001 0.667-0.897 77.0 71.4 0.875 Age 60-70 years 0.738 0.067 0.001 0.606-0.869 71.8 72.2 1.365 Age > 70 years 0.854 0.074 < 0.001 0.708-0.999 90.4 80.0 1.325 PSA < 10 ng/dl 0.753 0.044 < 0.001 0.666-0.840 75.5 71.8 1.105 PSA > 10 ng/dl 0.892 0.054 < 0.001 0.787-0.998 89.3 80.0 1.660 AVAUC: Area under curve; SE: Standart error; CI: Confidence interval. Table 4. Correlations between free PSA and PV in different age groups and total PSA levels. Age tPSA levels < 60 years 60-70 years > 70 years < 10 ng/dl > 10 ng/dl n = 121 (28%) n = 224 (51%) n = 93 (21%) n = 325 (74%) n = 113 (26%)  Correlation Correlation Correlation Correlation Correlation coefficient p coefficient p coefficient p coefficient p coefficient p PV-fPSA 0.546 < 0.001 0.506 < 0.001 0.487 < 0.001 0.494 < 0.001 0.473 < 0.001 PV: Prostate volume; tPSA: Total prostate-specific antigen; fPSA: Free prostate-specific antigen. Table 2. Correlations between prostate volume and age, total PSA, free PSA, free PSA/total PSA ratio. Total cohort Chronic prostatitis group Only BPH group Correlation p Correlation p Correlation p coefficient coefficient coefficient PV - Age 0.210 < 0.001 0.279 0.011 0.223 < 0.001 PV - tPSA 0.338 <0.001 0.379 < 0.001 0.385 < 0.001 PV - fPSA 0.548 < 0.001 0.592 < 0.001 0.520 < 0.001 PV-fPSA/tPSA 0.363 < 0.001 0.359 < 0.001 0.287 < 0.001 PV, prostate volume; tPSA, total prostate-specific antigen; fPSA, free prostate-specific antigen. Sinan_Stesura Seveso 01/04/20 18:51 Pagina 3 Archivio Italiano di Urologia e Andrologia 2020; 92, 1 Sinan Avci, Efe Onen, Volkan Caglayan, Metin Kilic, Murat Sambel, Sedat Oner 4 Although there have been numerous studies investigat- ing the correlation of total PSA and PV in patients with BPH, the relationship between free PSA and PV has received little attention. To the best of our knowledge, there have only been seven studies that have examining the relationship between free PSA and PV (8, 9, 20, 22- 25). All of these studies showed that free PSA was supe- rior to total PSA for predicting PV, and this result was also demonstrated in the multivariate analyses in five of the aforementioned studies (8, 9, 20, 24, 25). In the cur- rent study, both total PSA and free PSA were correlated with PV (r = 0.33, r = 0.54, respectively) but multivari- ate analysis showed a significant relationship only between free PSA and PV (p < 0.001). The superiority of the free PSA to total PSA found in the current study is consistent with findings in literature (8, 9, 20, 22-25). There were significant correlations between total PSA and PV in the multivariate analysis of 3 of the 5 previ- ously mentioned studies (8, 20, 24), whereas in the other two studies, a significant relationship was found only between free PSA and PV, as was the case in this current study (9, 25). Prostatic inflammation appears to play a role in BPH pathogenesis and progression (26, 27) but in the afore- mentioned studies (8, 9, 20, 22, 24, 25), other than Mao et al. (23), who reported that patients with pathological results of only BPH were included in the study, there is no information about the presence of inflammatory con- ditions such as chronic prostatitis in the pathology results of patients. According to the pathology results of the current study, patients were divided into two groups as only BPH and BPH with chronic prostatitis. Significant correlations were found between free PSA and PV in both groups. However, in multivariate analysis, the BPH only group showed a significant correlation between free PSA and PV, similar to the entire cohort, but that correlation was not found in BPH with chronic prostatitis group, whereas there was a significant correlation between total PSA and PV. It can be hypothesized that free PSA may be less influenced by prostatic inflammation in which serum total PSA elevation may occur as a result of dis- ruption of the normal prostatic architecture, in other words, the greater increase in free PSA may result from a larger benign prostate tissue and prostatic inflammation contributes a greater increase to total PSA than free PSA. In the above-mentioned studies (8, 20, 24) where a sig- nificant correlation was determined between total PSA and PV in multivariate analysis, this result could be attributed to possible prostatic inflammation. However, as prostatic inflammation is a pathological diagnosis, it may not be known before biopsy, especially in patients without symptoms associated with such a pathological condition. According to the current study results, the correlation between free PSA and prostate volume was comparatively decreased in the case of inflammation in the prostate. Therefore, the recommendations of previ- ous studies and the current one for the use of free PSA to predict PV may be relatively limited in men with clini- cally unknown prostatic inflammation. In this respect, the importance of defining this point, which has not been explored in previous studies, should be emphasized. The relationship between free PSA and PV according to different age ranges and total PSA values was also exam- ined in this study. The patients were stratified into three age groups of < 60 years, 60-70 years and > 70 years. Figure 1. ROC curves of each group for free PSA to predict whether prostate volume is > 40 or < 40 cc. The areas under curve and p values of each group are shown in Table 5. Sinan_Stesura Seveso 01/04/20 18:51 Pagina 4 In all three age groups a significant correlation was deter- mined between free PSA and PV and the highest degree of correlation was found in the group aged < 60 years (r = 0.54, r =0.50, r = 0.48 respectively). In three previ- ous studies where patients were similarly classified according to age, a correlation between free PSA and PV was shown in all age groups (22-24). However, the age groups with the highest correlations were different from the current study. In two of the studies (23, 24) the high- est correlations were seen in the group aged 60-70 years, while in the other study (22) it was the group of patients > 70 years. In some studies (20, 22, 23) patients with total PSA values > 10 ng/dl have been excluded to reduce the possibility of including patients with prostate cancer. Nevertheless, in some studies (8, 24) there is no infor- mation about the upper limit of total PSA while in anoth- er study (9), patients with total PSA > 10 ng/dl were included. In addition, no study has evaluated the corre- lation between free PSA and PV in patients with total PSA values > 10 ng/dl. In the current study, the correlation was analysed between free PSA and PV in total PSA-strat- ified cohorts as total PSA above or below 10ng/dl. The data obtained showed that the value of the correlation coefficient was slightly greater for the total PSA < 10 ng/dl cohort than for the total PSA > 10 ng/dl cohort, suggesting that free PSA may correlate better to PV when the possibility of patients with prostate cancer decreases (r = 0.494, r = 0.473, respectively). However, from another perspective, because of the small difference in the correlation values between the groups, it can be said that in patients with total PSA > 10 ng/dl, free PSA can be used safely for the prediction of PV. In the current study, the diagnostic performance of free PSA as a proxy for PV was evaluated using ROC curves for each group, and free PSA was determined to be sig- nificant for PV with AUC values ranging from 0.73 to 0.89 for all subgroups (Table 5). In these analyses, the PV threshold was 40cc, which is of great importance as guidelines have suggested not prescribing 5α-reductase inhibitors to patients with a prostate volume < 40cc (4). In the ROC curves of the current study, when the cutoff value of free PSA was taken as 1.28 ng/dl to predict prostate volume > 40cc, sensitivity and specificity were determined as 72.8% and 73.5%, respectively and the AUC was 0.78 for the entire cohort. This result of AUC as 0.78 for free PSA to predict whether PV was > 40cc or < 40cc was slightly better than the values reported in previous studies (AUCs for references 8, 20, 22-24 were 0.72, 0.75, 0.71, 0.75, 0.75, respectively). There continues to be value in the use of free/total PSA for the stratification of the risk of prostate cancer and to decide on a biopsy for patients with 4-10 ng/mL total PSA and negative DRE. A previous study reported that prostate cancer was detected by biopsy in 56% of men with free/total PSA < 0.10, but in only 8% with free/total PSA > 0.25 ng /mL (28). Those studies indicate that the probability of BPH increases as free PSA levels increase. Therefore, free PSA is more closely related to BPH and this link is parallel to the current study results. In the multivariate analysis results, the patients with chronic prostatitis pathology showed a significant relationship between free/total PSA and PV. This finding can be con- sidered to be related to inflammation-induced total PSA increase, as discussed above. When the results were evaluated of the relationship between age and PV, a significant correlation (r = 0.21) was determined, similar to other studies in the literature (8, 9, 20, 22, 23) but that correlation was not seen in the multivariate analysis (p: 0.091). This was consistent with the study of Morote et al. (9) whereas the opposite was reported in studies by Kayikci et al. (7) and Coban et al. (20) (p: < 0.01 and < 0.01, respectively). The present study is one of a limited number of trials sug- gesting that free PSA is a strong predictor for PV and that it is better than total PSA. Initially, the efficacy of free PSA at predicting PV in patients with inflammatory pathology reports in addition to BPH and total PSA levels >10ng/dl were reported. These conditions were then evaluated as separate groups to eliminate any bias. In addition, sepa- rate cut-off levels for free PSA in the prediction of PV were established for different subgroups of patients. This study had some limitations, primarily the retrospective nature of the study, the probability of occult cancers that could not be detected by biopsy and the criteria used for sub- ject recruitment on the basis of the indications for prostate biopsy rather than a clinical diagnosis of BPH. Nevertheless, the data suggest that because of the ability to obtain more accurate estimates of the PV without the help of more expensive, invasive diagnostic evaluations, free PSA could provide a more reasonable contribution in the proper management of patients with BPH. CONCLUSIONS Although total PSA was significantly correlated with PV, this correlation was not shown in multivariate analyses unlike free PSA. The superiority of free PSA may be used to estimate PV with easily obtained serum tests and could be a useful tool for therapeutic decison-making and longitudinal follow-up in patients with BPH. However, in patients with prostatic inflammation, con- sidering that there is a significant relationship between free PSA and PV only in univariate analysis, it should be kept in mind that the efficacy of free PSA may be limited in this group of patients. REFERENCES 1. Abrams P, Cardozo L, Fall M, et al. The standardisation of ter- minology of lower urinary tract function: report from the Standardisation Sub-committee of the International Continence Society. Neurourol Urodyn 2002; 21:167-178. 2. Kupelian V, Wei JT, O’Leary MP, et al. Prevalence of lower uri- nary tract symptoms and effect on quality of life in a racially and ethnically diverse random sample: the Boston Area Community Health (BACH) Survey. Arch Intern Med. 2006; 166:2381-2387. 3. Wilkinson AG, Wild SR. Is pre-operative imaging of the urinary tract worthwhile in the assessment of prostatism?. Br J Urol. 1992; 70:53-57. 4. Gravas S, Cornu JN, Drake MJ, et al. Guidelines on the Management of Non-neurogenic Male LUTS. Uroweb 2018. Available from:http://uroweb.org/guideline/treatment-of-non-neuro- genic-male-luts/. 5Archivio Italiano di Urologia e Andrologia 2020; 92, 1 Free PSA as useful tool to predict prostate volume Sinan_Stesura Seveso 01/04/20 18:51 Pagina 5 Archivio Italiano di Urologia e Andrologia 2020; 92, 1 Sinan Avci, Efe Onen, Volkan Caglayan, Metin Kilic, Murat Sambel, Sedat Oner 6 5. Roehrborn CG. Accurate determination of prostate size via digi- tal rectal examination and transrectal ultrasound. Urology. 1998; 51:19-22. 6. Terris MK, Stamey TA. Determinetion of prostate volume by tran- srectal ultrasound. J Urol. 1991; 145:984-987. 7. Trop-Pedersen S, Juul N, Jakobsen H. Transrectal prostatic ultra- sonography: equipment, normal findings, benign hyperplasia and cancer. Scand J Urol Nephrol. Suppl 1988; 107:19-25. 8. Kayikci A, Cam K, Kacagan C, Tekin A, Ankarali H. Free prostate-specific antigen is a better tool than total prostate-specific antigen at predicting prostate volume in patients with lower urinary tract symptoms. Urology. 2012; 80:1088-1092. 9. Morote J, Encabo G, Lopez M, de Torres IM. Prediction of Prostate Volume Based on Total and Free Serum Prostate-Specific Antigen: Is It Reliable?. Eur Urol. 2000; 38:91-95. 10. Jacobsen SJ, Jacobsen DJ, Girman CJ, et al. Natural history of prostatism: risk factors for acute urinary retention. J Urol. 1997; 158:481-487. 11- Jacobsen SJ, Jacobsen DJ, Girman CJ, et al. Treatment for benign prostatic hyperplasia among community dwelling men: the Olmsted County study of urinary symptoms and health status. J Urol. 1999; 162:1301-1306. 12. Bosch JL, Bohnen AM, Groneveld FP. Validity of digital rectal examination and serum prostate specific antigen in the estimation of prostate volume in community-based men aged50 to 78 years: the Krimpen study. Eur Urol. 2004; 46:753-759. 13. Marberger MJ, Andersen JT, Nickel JC. Prostate volume and serum prostate-specific antigen as predictors of acute urinary reten- tion: combined experience from three large multinational placebo- controlled trials. Eur Urol. 2000; 38:563-568. 14. Kaplan SA, Roehrborn CG, McConnell JD, et al. Long-term treatment with finasteride improves results in a clinically significant reduction in total prostate volume compared to placebo over the full range of baseline prostate sizes in man enrolled in the MTOPS trial. J Urol. 2008; 180:1030-2. Discussion 1032-1033. 15. Kaplan SA, Lee JY, Meehan AG, et al. Long-term treatment with finasteride improves clinical progression of benign prostatic hyper- plasia in men with an enlarged versus a smaller prostate: data from the MTOPS trial. J Urol. 2011; 185:1369-1373. 16. Emberton M, Andriole GL, de la Rosette J et al. Benign prostat- ic hyperplasia: a progressive disease of aging men1. Urology 2003;61:267-273. 17. Roehrborn CG, Girman CJ, Rhodes T, et al. Correlation between prostate size estimated by digital rectal examination and measured by transrectal ultrasound. Urology. 1997; 49:548-557. 18. Trivedi MR, Choudhary BA. Digital rectal examination, tran- srectal ultrasound and prostate specific antigen as a triple assess- ment diagnostic tool for benign enlargement of prostate. Natl J Med Res. 2015; 5:244-248. 19. Hochberg DA, Armenakas NA, Fracchia JA. Relationship of prostate-specific antigen and prostate volume in patients withbiopsy proven benign prostatic hyperplasia. Prostate. 2000; 45:315-319. 20. Coban S, Doluoglu OG, Keles I, et al. Age and total and free prostate-specific antigen levels for predicting prostate volume in patients with benign prostatic hyperplasia. The Aging Male. 2016; 19:124-127. 21. Roehrborn CG, Boyle P, Gould AL, Waldstreicter J. Serum prostate-specific antigen as a predictor of prostate volume in men with benign prostatic hyperplasia. Urology. 1999; 53:581-589. 22. Choi H, Park JY, Shim JS, et al. Free prostate-specific antigen provides more precise data on benign prostate volume than total prostate-specific antigen in Korean population. Int Neurourol J. 2013; 17:73-77. 23. Mao Q, Zheng X, Jia X, et al. Relationships between total/free prostate-specific antigen and prostate volume in Chinese men with biopsy-proven benign prostatic hyperplasia. Int Urol Nephrol. 2009; 41:761-766. 24. Masuda H, Kawakami S, Sakura M, et al. Performance of free PSA better than total PSA for estimation of prostate volume in eld- erly men without prostate cancer (abstract). Eur Urol Suppl. 2011; 10:32. 25. Canto EI, Singh H, Shariat SF, et al. Serum BPSA outperforms both total PSA and free PSA as a predictor of prostatic enlargement in men without prostate cancer. Urology. 2004; 63:905-910. 26. Ficarra V, Rossanese M, Zazzara M, et al. The role of inflam- mation in lower urinary tract symptoms (LUTS) due to benign pro- static hyperplasia (BPH) and its potential impact on medical thera- py. Curr Urol Rep. 2014; 15:463. 27. He Q, Wang Z, Liu G, et al. Metabolic syndrome, inflammation and lower urinary tract symptoms: possible translational links. Prostate Cancer and Prostatic Dis. 2016; 19:7. 28. Catalona WJ, Partin AW, Slawin KM, et al. Use of the percent- age of free prostate-specific antigen to enhance differentiation of prostate cancer from benign prostatic disease: a prospective multi- center clinical trial. JAMA 1998; 279:1542-1547. Correspondence Sinan Avci, MD (Corresponding Author) sinavci@yahoo.com Efe Onen, MD efe17@yahoo.com Volkan Caglayan, MD volkantuysuz@hotmail.com Metin Kilic, MD kilicmetin@hotmail.com Murat Sambel, MD muratsambel@hotmail.com Sedat Oner, Associate Professor sedatoner@yahoo.com University of Health Sciences, Bursa Yuksek Ihtisas Training and Research Hospital, Department of Urology, Bursa, Turkey Sinan_Stesura Seveso 01/04/20 18:51 Pagina 6