Out of the 452 patients underwent prostate biopsy (Table 1), 230 (50.9%) cases were diagnosed with prostate cancer, the overall positive rate of biopsy cores was 24.6% (1416 of 5759 cores). Average age was 70.1 years, average PSA level was 22.0 ng/mL, average prostate volume was 57.0 mL, and average PSA density was 0.51 ng/mL2. There was no significant difference in age, PSA level, prostate volume, and PSA density between two groups. The overall detection rate in TR group appeared to be higher than in TP group in our study (56.1% vs 44.4%, OR: 1.612, 95% CI: 1.111–2.340, p = 0.012). To minimize confounding factors of PSA level and PSA density in two groups, CMH test was performed, showing that the cancer detection rate in TR group was still higher than in TP group both when adjusting with PSA level (OR: 1.529, 95% CI: 1.036–2.255, p = 0.032), or when adjusting with PSA density (OR: 1.514, 95% CI: 1.030–2.226, p = 0.035). Similarly, the positive rate of biopsy cores was higher in the TR group than in the TP group (27.5% vs 21.0%, OR: 1.429, 95% CI: 1.264–1.616, p < 0.001).
Table 1
Clinical characteristics of patients underwent transrectal or transperineal prostate biopsy.
Variables
|
All patients
(n = 452)
|
Transrectal biopsy
(n = 245)
|
Transperineal biopsy
(n = 207)
|
p value
|
Age at biopsy (years)
|
|
|
|
|
Average/Median
|
70.1 ± 8.4/70
|
70.0 ± 8.7/70
|
70.3 ± 8.1/70
|
0.769
|
Range
|
45–88
|
45–88
|
51–88
|
PSA (ng/mL)
|
|
|
|
|
Average/Median
|
22.0 ± 20.6/13.8
|
23.2 ± 21.8/14.1
|
20.6 ± 19.0/13.7
|
0.634
|
Range
|
0.3–99.2
|
0.3–99.2
|
0.8–97.6
|
Prostate volume (mL)
|
|
|
|
|
Average/Median
|
57.0 ± 36.0/49
|
54.4 ± 31.6/46
|
60.0 ± 40.3/52
|
0.151
|
Range
|
10–334
|
11–204
|
10–334
|
PSA density (ng/mL2)
|
|
|
|
|
Average/Median
|
0.51 ± 0.59/0.27
|
0.55 ± 0.64/0.29
|
0.47 ± 0.51/0.26
|
0.181
|
Range
|
0.01–3.63
|
0.01–3.63
|
0.02–2.40
|
Prostate cancer diagnosis rate (n, %)
|
230/452
(50.9%)
|
138/245
(56.1%)
|
92/207
(44.4%)
|
0.012
|
Positive rate of
biopsy cores (n, %)
|
1416/5759
(24.6%)
|
876/3185
(27.5%)
|
540/2574
(21.0%)
|
< 0.001
|
Furthermore, to compare the prostate cancer detection rate of two biopsy approaches in patients with different clinical characteristics, subgroup analysis was performed according to serum PSA level, prostate volume, PSA density and PI-RADS (prostate imaging reporting and data system) score. As shown in Table 2, In subgroup analysis stratified by PSA level, there were no significant differences in cancer detection rate between the TR and TP groups for each subgroup. When stratified by prostate volume, the TR group have a higher cancer detection rate than the TP group among the patients with 30–80 mL prostate volume (56.3% vs 40.5%, OR: 1.892, 95% CI: 1.165–3.073, p = 0.010), as well as in the subgroup with > 80 mL prostate volume (36.6% vs 16.3%, OR: 2.967, 95% CI: 1.060–8.305, p = 0.034). However, there was no significant difference between two biopsy approaches in cancer detection rate among the patients with ≤ 30 mL prostate volume. Similarly, cancer detection rates did not significantly differ between TR and TP group according to different PSA density. In addition, PI-RADS score can be assessed in 235 patients (130/245 in TR group, 105/207 in TP group) by MRI, when stratified by PI-RADS score (≤ 3, > 3), TR group had a higher cancer detection rate than TP group (83.9% vs 65.7%, OR: 2.734, 95% CI: 1.255–5.957, p = 0.010) in patients with PI-RADS score of > 3, however, did not significantly differ with TP group on cancer detection rate in patients with PI-RADS score of ≤ 3.
Table 2
Comparison of prostate cancer detection rate according to PSA level, prostate volume and PSA density between transrectal and transperineal prostate biopsy.
Variables
|
All patients
(n, %)
|
Transrectal biopsy
(n, %)
|
Transperineal biopsy
(n, %)
|
p value
|
PSA (ng/mL)
|
|
|
|
|
≤ 10
|
63/152 (41.4)
|
40/87 (46.0)
|
23/65 (35.4)
|
0.190
|
10–20
|
56/140 (40.0)
|
29/63 (46.0)
|
27/77 (35.1)
|
0.188
|
20–100
|
111/160 (69.4)
|
69/95 (72.6)
|
42/65 (64.6)
|
0.280
|
Prostate volume (mL)
|
|
|
|
|
≤ 30
|
74/96 (77.1)
|
38/53 (71.7)
|
36/43 (83.7)
|
0.163
|
30–80
|
134/272 (49.3)
|
85/151 (56.3)
|
49/121 (40.5)
|
0.010
|
> 80
|
22/84 (26.2)
|
15/41 (36.6)
|
7/43 (16.3)
|
0.034
|
PSA density (ng/mL2)
|
|
|
|
|
≤ 0.15
|
23/92 (25.0)
|
14/42 (33.3)
|
9/50 (18.0)
|
0.091
|
> 0.15
|
207/360 (57.5)
|
124/203(61.1)
|
83/157 (52.8)
|
0.118
|
PI-RADS score
|
|
|
|
|
≤ 3
|
27/87 (31.0)
|
14/49 (28.6)
|
13/38 (34.2)
|
0.573
|
> 3
|
112/148 (75.7)
|
68/81 (83.9)
|
44/67 (65.7)
|
0.010
|
Table 3
Comparison of positive rate of biopsy cores according to T stage and ISUP grade between transrectal and transperineal prostate biopsy.
Variables
|
All patients
(n, %)
|
Transrectal biopsy
(n, %)
|
Transperineal biopsy
(n, %)
|
p value
|
T stage
|
|
|
|
|
T1 + T2
|
381/1160 (32.8)
|
231/754 (30.6)
|
150/406 (36.9)
|
0.029
|
T3 + T4
|
725/1065 (68.1)
|
433/598 (72.4)
|
292/467 (62.5)
|
0.001
|
ISUP grade
|
|
|
|
|
1
|
79/542 (14.6)
|
49/351 (14.0)
|
30/191 (15.7)
|
0.582
|
2–3
|
369/853 (42.3)
|
192/455 (42.2)
|
177/398 (44.5)
|
0.504
|
4–5
|
955/1520 (62.8)
|
622/975 (63.8)
|
333/545 (61.1)
|
0.297
|
Among 230 patients diagnosed with prostate cancer by biopsy, T stages were assessable in 175 patients (104/138 in TR group, 71/92 in TP group). We further analyzed the difference in the positive rates of biopsy cores between the two biopsy approaches in patients with different T stages. As shown in Table 3, the TR group have a higher positive rates of biopsy cores among the patients with T3 and T4 stages (72.4% vs 62.5%, OR: 1.573, 95% CI: 1.213–2.038, p = 0.001), while TP group have a higher rate among the patients with T1 and T2 stages (30.6% vs 36.9%, OR:0.754, 95% CI: 0.585–0.972, p = 0.029). Additionally, except for 1 patient in TR group diagnosed with neuroendocrine carcinoma, ISUP (international society for urological pathology) grades were obtained according to the highest pathology score core for each individual patient. When stratified by ISUP grade, there were no significant differences in positive rates of biopsy cores between the TR and TP groups for each subgroup.