Ceylon Medical Journal
Trans rectal ultra sound guided prostate biopsies: a single centre
experience in Sri Lanka
M D S Lokuhetty1, H D Wijesinghe2, D T Abeysuriya3, U C Samarasinghe4 and N D Perera5
(Index words: carcinoma prostate, prostate biopsy, Sri Lanka, serum PSA)
Background Trans rectal ultrasound guided prostate
biopsy (TRUS) was introduced to Sri Lanka in 2002.
Objectives 1. To study clinicopathological features of
males subjected to TRUS biopsy 2. To compare
estimation of tumour burden by two methods in
carcinoma prostate (CaP).
Methods 749 symptomatic males subjected to TRUS
biopsy over 64 months at a single centre. Information
was retrieved from case records. Tumour burden in CaP
was calculated as: 1. Calculated tumour burden (CTB) –
total percentage tumour in each core/total number of
cores 2. Percentage positive biopsy cores (PPBC) –
number of positive cores / total number of cores X 100.
SPSS 15.0, student's t test and Spearman’s rank
correlation coefficients were used for statistical analysis.
Results 35.2% had CaP, microacinar in type. 34.88% were
poorly differentiated. CaP was frequent among older
patients (P<0.00001). The prostate volume in CaP was
significantly lower than in the benign group (P<0.05).
Prostate specific antigen (PSA) level was significantly
higher in CaP (P<0.00001). A 99.6% sensitivity and 4.7%
specificity was observed at PSA of 4ng/ml for detecting
CaP. Specificity was 98% at 25.5ng/ml, with a sensitivity
of 44.4%. CTB and PPBC had similar correlations with
biochemical/histological parameters of CaP and were
strongly correlated (0.786).
Interpretation Males with CaP were older, had higher
PSA levels and smaller prostates. A cut off level of PSA
>4ng/ml could be used for directing symptomatic patients
for TRUS biopsy to detect CaP, keeping in mind that
specificity is 98% only at 25.5ng/ml. Both CTB and PPBC
could be used to calculate tumour burden in TRUS with
Carcinoma prostate (CaP) accounted for 4.9% (301
cases) of all cancers affecting Sri Lankan men in 2005 .
Trans rectal ultrasound guided core biopsy of the prostate
(TRUS biopsy), is considered a safe and accurate
diagnostic test for CaP . This study aims to document
our experience with 749 TRUS biopsies in symptomatic
men, since introduction of the technique to Sri Lanka in
1,2,3Department of Pathology, Faculty of Medicine, University of Colombo, 4Nawaloka Hospital, Colombo and 5National
Hospital of Sri Lanka.
Correspondence: HDW, e-mail: <email@example.com>. Received 25 July and revised version accepted 11
December 2008. Competing interests: none declared.
Vol. 54, No. 1, March 2009
1. To study clinicopathological features of
symptomatic men subjected to TRUS biopsy of
prostate, with a view to formulate future diagnostic
strategies for CaP.
2. To compare estimation of tumour burden by
calculated tumor burden (CTB) and percentage
positive biopsy core (PPBC) methods in diagnosed
Materials and methods
749 TRUS biopsies performed over a period of 64
months at a single tertiary care hospital in the private sector
were included in the study. All these men presented with
lower urinary tract symptoms with or without abnormal
digital rectal examination (DRE) findings. DRE and
ultrasound examination of the prostate followed by
biopsies, were performed by a single radiologist using an
18 G needle. A minimum of 6 tissue cores were obtained
from right and left sides of the base, middle and the apical
regions of the prostate. Additional cores were obtained
from nodules, if present.
The age and serum PSA (obtained prior to DRE and
ultrasonography) were retrieved from case records. The
prostate volume was obtained from documented
ultrasonography findings. Prostate specific antigen
density (PSAD) was calculated by dividing the serum PSA
in ng/ml by the volume of prostate in milliliters. Hematoxylin
and eosin stained slides were examined by a single
pathologist and categorised as follows 1. Benign with/
without inflammation 2. CaP (acinar/ductal) and atypical
glands suspicious, but not diagnostic of carcinoma. CaP
were graded and scored based on the Gleason scoring
The tumour burden was calculated as:
= Total of percentage tumor in
Total number of tissue cores
biopsy cores (PPBC)
= Number of positive tissue
cores × 100
Total number of tissue cores
Presence of high-grade prostatic intra epithelial
neoplasia (PIN) and perineural invasion were documented.
SPSS 15.0 software was used for analysis. Student’s t test
was used to compare means and Spearman’s rank
correlation coefficients were used to assess correlations.
264 (35.2%) were diagnosed to have CaP on TRUS
biopsy. 454 (60.6%) were diagnosed as benign. Small acini
that were suspicious but not diagnostic of carcinoma were
seen in 31 cases (4.2%). The age of the study population
ranged from 33-90 years with a mean age of 69.3 years
(SD=8.2 years). The mean age of men with CaP was 71.3
years (SD=7.8 years). The mean age of those diagnosed
as benign was 68.1 years (SD=8.2 years). Older men
had malignant disease more often than younger men
(P<0.00001) (Figure 1).
Figure 1. Distribution of patients with benign and
malignant diagnosis according to age
Serum PSA was measured in 716 men (Table 1). The
mean serum PSA was 51.2 ng/ml (SD=80.0 ng/ml) in men
with diagnosed CaP. It was 20.1 ng/ml (SD=31.1 ng/ml) in
the benign group (Table 2). This difference was
statistically significant (P< 0.00001). When the receiver
operating curve (ROC curve) was applied, a sensitivity of
99.6% and a specificity of 4.7% were seen at a serum PSA
level of 4 ng/ml in detecting CaP. A specificity of 98% was
observed only at serum PSA of 25.5 ng/ml, however with
a low sensitivity (44.4%). Application of recommended
age specific PSA cut off values  showed no change in
sensitivity and no significant increase in specificity as
compared to a cut off serum PSA of 4 ng/ml. The mean
Serum PSA Density (PSAD) for the CaP and the benign
groups were 1.3 ng/ml2 (SD=1.8 ng/ml2) and 0.56 ng/ml2
(SD=2.1 ng/ml2) respectively. This difference was not
statistically significant (p>0.05). The mean prostate volume
was 41.6ml (SD=26.8 ml) in the CaP group. The non-cancer
group had a mean prostate volume of 52.3 ml (SD=32.8
ml). This difference was statistically significant (P<0.001).
CaP was small acinar in type in 99.2% (262/264). An
additional large duct component was present in two. The
combined Gleason score of CaP group is shown in table 3.
Majority (60/172-34.88%) had poorly differentiated
carcinomas with a Gleason score of 8 to 10. There was a
trend for increasing levels of serum PSA and PSAD with
higher Gleason scores (Table 3). High grade prostatic
intraepithelial neoplasia (PIN) was seen in a man,
associated with a moderately to poorly differentiated
adenocarcinoma of small acinar type. Perineural invasion
was present in 19 (7.1% - 19/264) men with CaP. There was
Age of benign patientsAge of cancer patients
Ceylon Medical Journal
no association between perineural invasion and the
Gleason's score (P>0.05).
Spearman’s rank correlation coefficients for
associations of PPBC and CTB with the biochemical and
Table 1. Serum PSA levels in relation to histological diagnosis
<4 ng/ml 4.1-10 ng/ml>10.1 ng/mlTotal number
Benign21 137275 433
Malignant1 55 196 252
Benign with suspicious foci1 1218 31
Total23 204489 716
histological parameters are shown in Table 4. None of the
differences of correlations between CTB and PPBC
were statistically significant (Table 4). However, a strong
correlation of 0.780 was noted between PPBC and CTB.
Table 2. Means of PSA, PSAD and prostate volume in relation to histological diagnosis
PSAD prostate volume
Benign 20.17 (SD = 31.15) 0.56 (SD = 2.12)52.36 (SD = 32.87)
Malignant51.22 (SD = 80.01)1.31 (SD = 1.87) 44.54 (SD = 55.97)
Benign with suspicious foci27.75 (SD = 64.28)0.5 (SD = 1.02) 56.04 (SD = 29.98)
Table 3. Distribution of the combined Gleason’s score in patients with CaP and its
association with serum PSA and PSAD
Differentiation Number of
2-5 Well 2 1.1624.171.06
6 Moderately58 33.7224.58 0.76
7Moderately to poorly 52 30.23 55.991.29
8-10 Poorly60 34.8888.992.14
Table 4. Spearman’s rank correlation coefficients for association of PPBC and
CTB with biochemical and histological parameters of CaP
PPBC Calculated tumor
Difference between the
PSA level0.398 (p<0.01)0.245 (p<0.01) Not significant
PSA density0.435 (p<0.01)0.329 (p<0.01)Not significant
Combined Gleason’s score 0.400 (p<0.01)0.295 (p<0.01)Not significant
Perineural invasion-0.176 (p<0.01)-0.175 (p<0.05)Not significant
Vol. 54, No. 1, March 2009
The prevalence of CaP in this TRUS guided biopsy
series is 35.2%. Brawer et al reported a prevalence of 30.5%
for CaP in 187 TRUS biopsies of men who had PSA levels
greater than 4.0 ng/ml . The prevalence observed in this
study is slightly higher.
Mean PSA level of men with CaP was significantly
higher than those with benign disease. The mean prostate
volume in patients with CaP was significantly lower than
in the benign group; however the difference between the
mean PSAD values of the two groups was not statistically
significant. These observations can be explained by
current knowledge that malignant prostate epithelial cells
are capable of producing more PSA per gram of tissue
than normal or hyperplastic cells . It also confirms that
PSAD is not reliable for exclusion of a man with high PSA
A sensitivity of 99.6% for CaP detection was obtained
at a cut off serum PSA level of 4 ng/ml. Therefore it is
desirable that all symptomatic men with serum PSA levels
above 4 ng/ml are directed towards further investigations
including TRUS biopsy. However, it should be borne in
mind that the specificity for CaP is only 4.7% at this level
of serum PSA. As the specificity for CaP increases with
rising serum PSA (36.8% at PSA 10.05 ng/ml and 98% at
PSA of 25.5 ng/ml) a raised level of PSA especially above
25.5 ng/ml makes a symptomatic man more likely to have
CaP. It is documented that most men with slightly elevated
PSA levels (4-10 ng/ml) do not have carcinoma . Our
observations were similar.
The pathology of CaP in Sri Lanka is hitherto not
documented. 99.2% of the CaP in this series was of small
acinar type. The majority were poorly differentiated with a
combined Gleason score of 8-10. High grade PIN was
detected in one case in association with CaP. None had
high grade PIN in the absence of carcinoma. Documented
incidence for high grade PIN in TRUS is 4.6% . However,
in these countries serum PSA is used as a screening tool
for CaP in asymptomatic men . Therefore it is likely that
men with pre-invasive stage of disease (high grade PIN)
are detected more often than in our setting.
Recent studies show that both the number of cores
positive for cancer and the total percentage of core length
involved by cancer independently predict extra capsular
extension and positive surgical margins . This
information obtained by TRUS biopsies play a critical role
in patient management. Having calculated the burden of
tumour in the study sample by two different methods,
PPBC showed a higher correlation with the PSA level, PSA
density and combined Gleason's score when compared
with CTB. The difference in correlations by the two
methods however were not significant. In view of the strong
correlation observed between them, it is possible to use
both PPBC and CTB to calculate tumour burden in TRUS
biopsies. However, calculation of CTB is time consuming
and laborious, where as PPBC is more user-friendly. PPBC
was shown to be related to the preoperative serum PSA
level, Gleason score, clinical stage, extra-prostatic disease,
seminal vesicle involvement and to have a relationship
with biological outcome after radical prostatectomy .
Similarly PPBC was related to the serum PSA level and the
Gleason score in this study. Although radical prosta-
tectomy is still not widely performed in Sri Lanka, it is
offered more often in therapeutic protocols of localized
CaP. Therefore, the results of this study on tumour burden
on TRUS biopsies will be useful when CaP diagnosed on
TRUS biopsy are directed to curative surgery.
We gratefully acknowledge the contributions made
by Prof Rajitha Wickremasinghe of the University of
Kelaniya, who guided us in the statistical analysis of data
and all surgical colleagues who referred symptomatic men
for TRUS biopsy.
1. National Cancer Control Programme. Cancer Registry:
Cancer Incidence in Sri Lanka 2005.
2. Lippman HR, Ghiatas AA, Sarosdy MF. Systematic
transrectal ultrasound guided prostate biopsy after negative
digitally directed prostate biopsy. Journal of Urology 1992;
3. Bostwick DG. Grading prostate cancer. American Journal
of Clinical Pathology 1994; 102: S38-56.
4. Watson E, Jenkins L, Bukach C, Austoker J. The PSA test
and prostate cancer: information for primary care. NHS
Cancer Screening Programmes, Sheffield: 2002.
5. Brawer MK, Chetner MP, Beatie J. Screening for prostate
carcinoma with prostate specific antigen. Journal of Urology
1992; 147: 841-5.
6. Stamey TA. Diagnosis of prostate cancer: a personal view.
Journal of Urology 1992; 147: 830-2.
7. Magi-Galluzi C, Epstein JI. Threshold for diagnosing prostate
cancer over time. Human Pathology 2003; 34: 1116-8.
8. Egan AJM, Bostwick DG. Prediction of extraprostatic
extension of prostate cancer based on needle biopsy findings:
perineural invasion lacks significance on multivariate analysis.
American Journal of Surgical Pathology 1997; 21: 1496-500.
9. Smith RA, Cokkinides V, Eyre HJ. American Cancer Society
Guidelines for the Early Detection of Cancer, 2005. CA A
Cancer Journal for Clinicians 2005; 55: 31-44.
10. Zhou M, Epstein JI. The reporting of prostate cancer on
needle biopsy; prognostic and therapeutic implications and
the utility of diagnostic markers. Pathology 2003; 35: 472-9.
11. Antunes AA, Srougi M, Dall’Oglio MF, Crippa A,
Campagnari JC, et al. The percentage of positive biopsy
cores as a predictor of disease recurrence in patients with
prostate cancer treated with radical prostatectomy. British
Journal of Urology International 2005; 96: 1258-63.