Received: 13 June 2019 | Revised: 6 March 2020 | Accepted: 7 April 2020

DOI: 10.1111/andr.12797

ORIGINAL ARTICLE

Diagnostic accuracy of physical examination compared with

color Doppler ultrasound in the determination of varicocele
diagnosis and grading: Impact of urologists’ experience

Marcello S. Cocuzza1 | Bruno C. Tiseo1,2 | Victor Srougi1 | Guilherme J. A. Wood1 |

Joao P. G. F. Cardoso1 | Sandro C. Esteves3,4 | Miguel Srougi1

1
Departamento de Urologia, Faculdade de
Medicina, Hospital das Clinicas HCFMUSP, Abstract
Universidade de Sao Paulo, Sao Paulo, Brazil Background: Treatment of palpable varicocele in infertile men with abnormal semen
2
Hospital Israelita Albert Einstein, São Paulo,
parameters is widely accepted, and physical examination (PE) remains a cornerstone
Brazil
3
ANDROFERT, Andrology and Human
for recommending varicocele repair. However, identification of clinical varicocele
Reproduction Clinic, Campinas, Brazil during PE can be challenging for both urology residents and consultants.
4
Department of Surgery (Division Objective: To compare the diagnostic accuracy of PE to color Doppler ultrasonography
of Urology), University of Campinas
(UNICAMP), Campinas, Brazil (CDU) for the diagnosis of varicocele in experienced and non-experienced examiners.
Materials and methods: Diagnostic accuracy study involving 78 patients attending a
Correspondence
Marcello S. Cocuzza, Departmento de university-based infertility unit. Patients underwent scrotal PE by both experienced
Urologia, Faculdade de Medicina, Hospital (over 10 years experience in male infertility) and non-experienced urologists (senior
das Clinicas HCFMUSP, Universidade de Sao
Paulo, Rua. Dr. Ovidio Pires de Campos, 225; residents), and were subjected to CDU. varicocele diagnosis and varicocele grading
7º andar - Sala 710F, Sao Paulo 05403-000, were compared between examiner groups and to CDU. Accuracy measures were
Brazil.
Email: mcocuzza@uol.com.br evaluated, and interobserver agreement was estimated using unweighted kappa sta-
tistics. A subgroup analysis for normal and high body mass index (BMI) was also per-
formed for the same variables.
Results: Accuracy of PE for varicocele diagnosis was 63.5% with a positive predictive
value (PPV) of 75.5%. The specificity and PPV of PE were higher among experienced
than non-experienced urologists (82.0% CI: 74.27-88.26 and 81.1% CI: 74.39-86.44%
vs 67.2% CI: 58.33-75.22 and 70.6% CI: 64.52-76.08, respectively). Agreements on
varicocele diagnosis (k: 0.625 vs 0.517) and grading (k: 0.548 vs 0.418) by PE were
higher among experienced than non-experienced urologists. Differences between
eutrophic and overweight/obese patients were also suggested.
Discussion and conclusions: PE performed by infertility specialists identify patients
with varicocele more precisely than non-specialists. However, PE alone has subopti-
mal accuracy for varicocele diagnosis. Our results indicate that PE should be followed
by CDU to decrease the number of false positives and increase the diagnostic accu-
racy of varicocele diagnosis.

KEYWORDS

Doppler ultrasonography, male infertility, testis, varicocele

© 2020 American Society of Andrology and European Academy of Andrology

1160 | 
wileyonlinelibrary.com/journal/andr Andrology. 2020;8:1160–1166.
 |

20472927, 2020, 5, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/andr.12797 by Nat Prov Indonesia, Wiley Online Library on [17/04/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
COCUZZA et al. 1161

1 | I NTRO D U C TI O N testicular pain evaluation were asked to enroll in this diagnostic ac-
curacy study. Inclusion criteria were (a) history of primary infertility
Varicocele repair is widely used for male infertility treatment.1,2 The with at least one-year duration; (b) age between 25 and 45 years;
European Association of Urology (EAU)1 recommends that diagno- and (c) no previous history of scrotal surgery, cryptorchidism, or tes-
sis of varicocele should be initially made by physical examination ticular torsion. Exclusion criteria included hydrocele, scrotal masses
(PE) and confirmed by color Doppler ultrasound (CDU). This con- and genital or urinary anomalies detected by either PE or ultrasound.
trasts to the American Urological Association (AUA) recommenda- Patients with incomplete or missing data, and patients who were un-
tions, which states that PE alone is enough to diagnose varicocele.3 able to be submitted to CDU evaluation were excluded from the
Guidelines agree in treatment for patients with a palpable varico- analysis. A total of 78 consecutive patients were enrolled. Data col-
cele when associated with infertility and add two other criteria for lection was planned before physical and ultrasound examinations.
varicocele treatment: (a) normal female partner fertility or a poten- Recruitment period ranged from June 2012 to August 2013. The
tially treatable cause of infertility, and no concerns regarding time study complied with the standards for the reporting of diagnostic
to conception; and (b) abnormal semen parameters. When these accuracy (STARD statement). Institutional review board approval
conditions are met, fair evidence indicates a beneficial effect in in- was obtained for conducting the investigation.
creasing chances of natural and assisted conception.1,4 Up to this day, there are no published data that tried to compare
Clinical examination with the patient standing in a warm room is diagnosis accuracy of PE from infertility specialists and non-special-
currently the standard method for varicocele diagnosis, with a sensi- ists. In order to calculate sample size, we estimated that specialists
5,6
tivity and specificity of approximately 70%. Although venography would correctly identify varicocele in 60% of the testicular units, while
is considered the gold-standard method for diagnosing blood reflux, the diagnosis accuracy of non-experienced urologists would be 10%
it is rarely used except when conducted in conjunction with thera- lower, of 50%. When comparing one sample proportions, in a 1-sided
peutic vein occlusion. CDU has been considered the best diagnostic fashion, we estimated that the sample size needed to achieve statis-
tool and is frequently applied due to its non-invasiveness nature and tical significance would be of 153 testicular units for each examiner.9
an accuracy of approximately 90% when compared with venogra-
phy.5 One or more spermatic veins greater than 2.5-3.0 mm in di-
ameter (at rest and with Valsalva maneuver) on CDU examination 2.2 | Physical examination
7
usually correlates with varicocele on PE.
Notably, none of the guidelines aforementioned recommend All patients were examined for varicocele presence by two experi-
varicocelectomy in patients with subclinical varicocele, identified enced urologists (>10-year working experience as infertility specialist
by CDU alone.1,3 Subclinical varicocele treatment is not associated in a university-based hospital) and two non-experienced urologists
with significant benefits concerning semen parameters improve- (senior residents attending the urology program), blind to previous
ment or pregnancy rates.8 The latest Cochrane review on varicocele diagnosis. All doctors consecutively examined each patient in the
treatment effect in subfertile men describes minimal increase in the same room, in private, and an assistant nurse recorded the data.
odds ratio (OR) for natural pregnancy (OR 1.47 95% CI 1.05 to 2.05) Patients were examined in standing position, and the room tempera-
when patients with both subclinical and clinical varicocele have been ture was kept between 24 and 28 degrees Celsius. The sequence of
treated compared to observation. By contrast, treatment of only men medical examination was determined by sorting color cards. Each
with palpable varicocele resulted in an unequivocal increase in the examiner was blind to previous results. varicocele was classified
likelihood of pregnancy (OR = 2.39 95% CI 1.56 to 3.66), provided and graded based on the modified criteria of Dubin and Amelar10
4
the other aforesaid criteria were met. These factors highlight PE as (a) absent (no palpable varicocele), (b) grade 1 (palpable with aid
importance as a major determinant for recommending varicocele of Valsalva maneuver), (c) grade 2 (palpable without Valsalva), or (d)
treatment. However, PE might be inconclusive or difficult to perform grade 3 (visible). Body mass index (BMI) was also obtained in all men
in cases of low-grade varicocele, previous scrotal surgery, obesity, by kg/m2 formula.
concomitant hydrocele, or scrotal tenderness/hypersensitivity.
In this study, we assess diagnostic accuracy of PE performed by
experienced and non-experienced urologists in terms of varicocele 2.3 | Color Doppler ultrasound examination
identification and grading.
Our study relied on the results of CDU to determine the diagnostic
accuracy of PE for varicocele diagnosis. The ultrasound study of the
2 | M ATE R I A L A N D M E TH O DS varicocele should be performed with high-frequency linear probes
and with devices able to evaluate blood flow.11 After completing
2.1 | Study population the PEs, patients were conducted to CDU. CDU was performed in
a different warm room to mimic same conditions of PE. All patients
Patients attending our university-based male infertility outpatient were examined in standing and supine positions by a single expe-
clinic for initial infertility evaluation, varicocele assessment, or rienced radiologist with over twenty-year experience in urological
 |

20472927, 2020, 5, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/andr.12797 by Nat Prov Indonesia, Wiley Online Library on [17/04/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
1162 COCUZZA et al.

ultrasound. Varicocele was diagnosed by CDU with the following Interobserver agreements for varicocele diagnosis by PE were
criteria: (a) at least two venous channels with a diameter greater calculated using unweighted kappa statistics. Kappa provides a nu-
than 3 mm and (b) flow reversal with or without Valsalva maneuver merical rating (κ correlation coefficient) standardized to lie on a −1 to
for 1 second or more.12 Otherwise, patients were deemed as having 1 scale, where 1 is perfect agreement, and 0 is exactly what would
no signs of varicocele on CDU examination.13 An assistant nurse be expected by chance. Negative values would indicate agreement
recorded the data, and both the nurse and the radiologist were less than what could be expected by chance. Kappa value of > 0.8
blinded to previous PE results obtained by the urologists. All pa- was considered as almost perfect agreement, between 0.6 and 0.8
tients underwent physical and CDU examinations on the same day, as substantial agreement, between 0.4 and 0.6 as moderate agree-
at the same institution, and under similar examining conditions. ment, between 0.2 and 0.4 as fair agreement, while < 0.2 as poor
No adverse effects from performing either the PE or the CDU agreement.14
were reported. All analyses were performed with SPSS® package for Windows®
version 22.0 (SPSS Inc.). All statistical tests were two-tailed, and
P < .05 was considered statistically significant.
2.4 | Statistical analysis

Continuous variables are described by mean and standard devia- 3 | R E S U LT S

tion (SD). Categorical variables are presented as proportions and
percentages of the total. Sensitivity, specificity, positive (PPV), 3.1 | Demographics
and negative (NPV) predictive values of PE—both overall and by
examiners’ categories—were determined using the results of CDU Our patient population comprised mostly of non-obese men with-
as the gold standard for assessing venous reflux in the spermatic out obvious scrotal abnormalities. Median patient was 37 years
vein. (IQR: 28-43). Median BMI was 26.7 kg/m2 (95% IQR: 25.7-27.7),
with 41% of eutrophic patients, 36% of overweight, and 23% of
TA B L E 1 Distribution of varicocele diagnosis according to grade obese patients.
on physical examination A distribution of severity of the target condition—varicocele
grade—according to the 4 study observers is shown in Table 1.
No.
Observer Varicocele Grade 1 Grade 2 Grade 3

1 83 (53.8%) 31 (20.1%) 34 (22%) 6 (3.8%)

3.2 | Association of physical examination and CDU
2 82 (53.2%) 45 (29.2%) 19 (12.3%) 8 (5.1%)
for varicocele diagnosis
3a 84 (54.5%) 34 (22%) 29 (18.8%) 7 (4.5%)
4a 102 (66.2%) 23 (14.9%) 23 (14.9%) 6 (3.8%)
Varicocele was detected in 47 patients (60.3%) and 61 patients
Note: n = 154. (78.2%) by PEs performed by male infertility specialists and resi-
a
Experienced observers. dents, respectively.

TA B L E 2 Accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of physical examination
performed by experienced and non-experienced observers, for detection of varicocele

Varicocele Accuracy Sensitivity Specificity PPV NPV

N (%) (95% CI) (95% CI) (95% CI) (95% CI) (95% CI)

Ultrasounda 90 -- -- -- -- --
(58.4%)
Physical examination
Overallb 265 63.47% 55.56% 74.61% 75.47% 54.42%
(43.01%) (59.53-67.29) (50.26-60.76) (68.82-79.82) (70.98-79.47) (51.03-57.76)
Experienced urologistsc 122 66.23% 55% 82.03% 81.15% 56.45%
(39.61%) (60.65-71.50) (47.42-62.41) (74.27-88.26) (74.39-86.44) (51.97-60.83)
Non-experienced 143 60.71% 56.11% 67.19% 70.63% 52.12%
urologistsc (46.43%) (55.02-66.20) (48.53-63.48) (58.33-75.22) (64.52-76.08) (47.01-57.19)

Note: Color Doppler ultrasound was considered the gold-standard method for assessing vein reflux in spermatic veins and its results were used for
calculating the measures of accuracy.
a
Each spermatic cord was considered as a unit (in total, 154 units were examined).
b
Each spermatic cord was considered as a unit and were examined by four examiners (in total, 616 units were examined).
c
Each spermatic cord was considered as a unit and examined by two examiners (in total, 308 units were examined).
 |

20472927, 2020, 5, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/andr.12797 by Nat Prov Indonesia, Wiley Online Library on [17/04/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
COCUZZA et al. 1163

F I G U R E 1 Receiver operating
characteristics curves showing the
performance of experienced (blue
line) and non-experienced (green line)
urologists to diagnose varicocele by
physical examination in comparison
to color Doppler ultrasound. The area
under the curve (AUC) was higher for
experienced urologists (AUC: 0.685;
95% CI 0.625-0.745) than inexperienced
urologists (AUC: 0.616; 95% CI: 0.553-
0.680)

TA B L E 3 Agreement between experienced and non- 3.3 | Association of varicocele diagnosis and
experienced urologists on varicocele diagnosis and grading using varicocele grades by different observer categories on
unweighted Kappa statistics physical examination
Varicocele diagnosis κ Varicocele grading
(95% CI) κ (95% CI) The disagreement rates between experienced and non-experienced
Experienced 0.6253 (0.5037-0.7470) 0.5489 observers concerning the presence (any grade) or absence of vari-
urologists (0.4372-0.6607) cocele during PE were 18.2% and 24%, respectively (Table 3). This
Residents 0.5170 (0.3815-0.6526) 0.4181 resulted in a substantial agreement between experienced observers
(0.3024-0.5339) (κ correlation coefficient: 0.625; 95% CI 0.504-0.747; P < .001) and
Experienced 0.5188 (0.4235-0.6141) 0.4479 a moderate agreement between residents (κ correlation coefficient:
urologists vs. (0.3645-0.5313) 0.517; 95% CI 0.3815-0.653; P < .001).
residents
Disagreement between experienced observers regarding vari-
Note: κ (Kappa coefficient) - Kappa value of > 0.8 is considered cocele grade was observed in 26% of the spermatic cords examined
as almost perfect agreement; between 0.6 and 0.8 as substantial
(κ correlation coefficient: 0.549 (95% CI 0.437-0.661; P < .001). var-
agreement; between 0.4 and 0.6 as moderate agreement; between 0.2
and 0.4 as fair agreement; while < 0.2 as slight agreement.
icocele grade disagreement was higher between residents (36.4%
of the spermatid cords examined), with kappa measure of agree-
ment of 0.418 (95% CI 0.324-0.534; P < .001) (Table 3).
In the CDU evaluation, varicocele was found in 56 patients
(71.8%) and 90 testicular units (58.4%). Unilateral and bilateral
varicocele were found in 21 (26.9%) and 35 (44.9%) patients, re- 3.4 | Subgroup analysis
spectively. Two patients had only one testicle due to previous or-
chiectomy. Therefore, 154 units, comprised of spermatic cord and Considering potential limitations and difficulties of PE on overweight
testis, were examined. and obese patients, patients were divided in two subgroups accord-
Overall, accuracy of PE for detecting varicocele was 63.5%, al- ing to eutrophic (BMI < 24.9) and overweight/obese (BMI > 24.9).
beit higher among experienced (66.2% CI: 60.5-71.5) than non-ex- Accuracy, sensitivity, specificity, PPV, and NPV for both experi-
perienced counterparts (60.7% CI:55.02-66.2) (Table 2). Specificity enced and non-experienced urologists also suggest difference be-
(82.0% CI: 74.27-88.26% vs 67.2% CI: 58.33-75.22) and PPV (81.1 tween these groups, although still higher for experienced urologists
CI: 74.39-86.44% vs 70.6% 64.52-76.08) were also higher among ex- (Table 4).
perienced observers than non-experienced observers, respectively. The disagreement rates between experienced and non-ex-
Sensitivity and NPV were similar between groups. The ROC curve perienced observers concerning the presence (any grade) or ab-
with both experienced and non-experienced observers is provided in sence of varicocele during PE in eutrophic patients were 19%
Figure 1. and 23.8%. For overweight/obese, the disagreement rates were
 |

20472927, 2020, 5, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/andr.12797 by Nat Prov Indonesia, Wiley Online Library on [17/04/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
1164 COCUZZA et al.

TA B L E 4 Subgroups analysis: accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of
physical examinationa: (A) Performed by experienced observers, for detection of varicocele in eutrophic and overweight/obese patients. (B)
performed by non-experienced observers, for detection of varicocele in eutrophic and overweight/obese patients

Accuracy (95% CI) Sensitivity (95% CI) Specificity (95% CI) PPV (95% CI) NPV (95% CI)

(A) Experienced urologists

Eutrophic 71.43% 67.5% 78.26% 84.38% 58.06%
(62.70-79.12) (56.11-77.55) (63.64-89.05) (75.35-90.51) (49.37-66.29)
Obese 62.64% 45.00% 84.15% 77.59% 55.65%
(55.17-69.68) (35.03-55.27) (74.42-91.28) (66.77-85.64) (50.65-60.53)
(B) Non-experienced urologists
Eutrophic 62.70% 62.5% 63.04% 74.63% 49.15%
(53.64-71.15) (50.96-73.08) (47.55-76.79) (66.04-81.65) (40.30-58.06)
Obese 59.34% 51.00% 69.51% 67.11% 53.77%
(51.83-66.55) (40.80-61.14) (58.36-79.20) (58.27-74.88) (47.63-59.80)
a
Color Doppler ultrasound was considered the gold-standard method for assessing vein reflux in spermatic veins and its results were used for
calculating the measures of accuracy.

TA B L E 5 Subgroup analysis: Agreement between experienced and non-experienced urologists on varicocele diagnosis and grading in
eutrophic and overweight/obese patients using unweighted Kappa statistics

Varicocele diagnosis Varicocele diagnosis Varicocele grading Varicocele grading

eutrophic κ (95% CI) overweight/obese κ (95% CI) eutrophic κ (95% CI) overweight/obese κ (95% CI)

Experienced urologists 0.6323 (0.4582-0.8064) 0.5971 (0.4219-0.7723) 0.7030 (0.5684-0.8379) 0.5842 (0.4386-0.7299)
Residents 0.5220 (0.3111-0.7329) 0.5030 (0.3229-0.6830) 0.5758 (0.4074-0.7441) 0.5144 (0.3662-0.6625)

Note: κ (Kappa coefficient) - Kappa value of > 0.8 is considered as almost perfect agreement; between 0.6 and 0.8 as substantial agreement; between
0.4 and 0.6 as moderate agreement; between 0.2 and 0.4 as fair agreement; while < 0.2 as slight agreement.

17.6% and 24.2%, respectively. Experienced observers still hold varicocele grade than varicocele diagnosis among observes.
higher κ correlation coefficient than non-experienced for diagno- Notably, a perfect agreement (kappa coefficient > 0.8) concern-
sis and grading in both eutrophic and obese/overweight patients ing varicocele diagnosis and grading was not achieved even among
(Table 5). experienced observers. In a previous study, a 26% disagreement
rate was reported concerning varicocele diagnosis by PE between
two observers, but the impact of the physician's experience was
4 | D I S CU S S I O N A N D CO N C LU S I O N S not assessed.11 By contrast, our results suggest that clinician's ex-
perience plays a role in varicocele diagnosis by PE, as more mis-
Our results indicate that experienced observers (male infertility diagnosis was made by urology residents than by male infertility
specialists) identify varicocele more precisely than non-experienced specialists.
observers (senior urology residents). Experienced observers are par- Our study relied on the results of CDU to determine the di-
ticularly accurate in identifying spermatic cords without varicocele agnostic accuracy of PE for varicocele diagnosis. The ultrasound
in which no venous reflux was seen on CDU (true negatives). The study of the varicocele should be performed with high-frequency
PPVs indicated that non-experienced observers tend to overesti- linear probes and with devices able to evaluate blood flow.15 For
mate frequency of varicocele. However, the overall accuracy of PE the correct detection of blood flow, CDU must be calibrated to
to diagnose the presence of venous reflux to internal spermatic veins detect a slow flow (7.5 kHz). The evaluation should be performed
was suboptimal. Approximately 1 in every 4 varicocele detected by in the supine and then the upright position, with and without a
PE had no confirmatory evidence by CDU overall. In our patient Valsalva maneuver, in order to obtain a complete evaluation of the
population, 18% and 33% of patients examined by experienced and blood flow in the seminal cord veins.16 Blood vessels are first stud-
non-experienced would receive recommendation for varicocele re- ied in a gray scale and then with the color Doppler and the pulse
pair based on the presence of a varicocele on PE despite the absence Doppler. We used the criteria of Chiou et al to confirm the pres-
of confirmatory reflux by CDU, thus potentially resulting in unneces- ence of reflux into the internal spermatic veins.12 These authors
sary procedures. observed a sensitivity of 93% and a specificity of 85% when the
Experienced urologists in our study tend to show better CDU was compared to PE. In their study evaluating 64 patients, all
agreement than non-experienced ones concerning varicocele di- moderate to large varicocele found on PE were positive by CDU
agnosis and grading. Disagreement rates were higher concerning diagnosis using the Chiou et al scoring system, but the same group
 |

20472927, 2020, 5, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/andr.12797 by Nat Prov Indonesia, Wiley Online Library on [17/04/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
COCUZZA et al. 1165

had only a 68% positive rate by traditional CDU diagnostic criteria examination. Thus, it is imperative to find an accurate method of eval-
13
relying solely on vein diameter. uation of varicocele to avoid unnecessary procedures.
Varicocele seems to be more commonly found in men with nor-
mal BMI than overweight and obese counterparts.17,18 However, C O N FL I C T O F I N T E R E S T
it is possible that the lower frequency of varicocele detection in The authors declare no conflicts of interest.
overweight/obese men is related to PE difficulties in such patients.
Subgroup analysis suggests that even among experienced urolo- AU T H O R S ’ C O N T R I B U T I O N S
gists, accuracy, sensitivity, specificity, and agreement could differ Cocuzza was responsible for study supervision; Cocuzza and Victor
between eutrophic and overweight/obese. Therefore, our results Srougi were responsible for study conception and design; Wood and
from mostly non-obese population might be conservative as the Tiseo were responsible for acquisition of data; Wood was respon-
presence of obesity might make the diagnosis of varicocele by PE sible for statistical analysis and data interpretation, Cardoso and
more equivocal. Esteves were responsible for manuscript drafting, Miguel Srougi and
The EAU guidelines on male infertility recommend that varico- Esteves were responsible for critical revision of the manuscript.
cele diagnosed by clinical examination be confirmed by ultrasound
investigation and color Duplex analysis.1 By contrast, both the ORCID
AUA and the American Society for Reproduction Medicine (ASRM) Marcello S. Cocuzza https://orcid.org/0000-0002-5465-4304
Practice Committees recommend PE alone for varicocele diagnosis. Bruno C. Tiseo https://orcid.org/0000-0003-4497-2971
According to these guidelines, CDU examination should not be used Victor Srougi https://orcid.org/0000-0001-8346-3833
for routine screening and detection of varicocele in patients with- Guilherme J. A. Wood https://orcid.org/0000-0003-3712-4843
out a palpable scrotal abnormality. 2 Our results support the EAU Joao P. G. F. Cardoso https://orcid.org/0000-0003-0066-4619
recommendations as we observed that PE alone overestimated the Sandro C. Esteves https://orcid.org/0000-0002-1313-9680
frequency of varicocele. Use of PE followed by confirmatory CDU
might provide a more accurate identification of patients eligible for REFERENCES
varicocele repair, thus potentially increasing treatment efficiency. 1. Jungwirth A, Giwercman A, Tournaye H, et al. European Association
Nevertheless, the efficacy of repairing varicocele detected solely by of Urology guidelines on Male Infertility: the 2012 update. Eur Urol.
2012;62:324-332.
PE or a combination of PE and CDU concerning semen quality and
2. Practice Committee of the American Society for Reproductive M,
pregnancy outcomes deserves further investigation. Society for Male R, Urology. Report on varicocele and infertility: a
Proper identification of candidates for treatment is crucial as the committee opinion. Fertil Steril. 2014;102:1556-1560.
existing evidence does not support varicocele repair to treat subclin- 3. Male Infertility Best. Practice Policy Committee of the American
Urological A, Practice Committee of the American Society for
ical varicocele.19-21 Although mild improvements in seminal parame-
Reproductive M. Report on optimal evaluation of the infertile male.
ters might occur after subclinical varicocele repair, pregnancy rates Fertil Steril. 2006;86:S202-S209.
do not significantly improve in treated patients.8,22 4. Kroese AC, de Lange NM, Collins J, Evers JL. Surgery or emboliza-
Our study has limitations concerning the number of enrolled tion for varicoceles in subfertile men. Cochrane Database Syst Rev.
2012;10:CD000479.
patients and doctors performing PE. These limitations are inher-
5. Trum JW, Gubler FM, Laan R, van der Veen F. The value of palpation,
ent to the public university-based male infertility unit in which the varicoscreen contact thermography and colour Doppler ultrasound
study was conducted. Also, the incidence of varicocele in our study in the diagnosis of varicocele. Hum Reprod. 1996;11:1232-1235.
group was higher than that reported in the literature among infer- 6. Gat Y, Bachar GN, Zukerman Z, Belenky A, Gorenish M. Physical ex-
tility patients, since we included patients with known varicocele amination may miss the diagnosis of bilateral varicocele: a compar-
ative study of 4 diagnostic modalities. J Urol. 2004;172:1414-1417.
diagnosis and patients referred for testicular pain evaluation. This
7. Pilatz A, Altinkilic B, Kohler E, Marconi M, Weidner W. Color
patient selection was intentional, in order to obtain a larger amount Doppler ultrasound imaging in varicoceles: is the venous diameter
of varicocele patients in the studied population. We used CDU as sufficient for predicting clinical and subclinical varicocele? World J
the gold-standard measure for varicocele diagnosis. Although con- Urol. 2011;29:645-650.
8. Unal D, Yeni E, Verit A, Karatas OF. Clomiphene citrate versus var-
sidered the most sensitive test, venography invasiveness limits its
icocelectomy in treatment of subclinical varicocele: a prospective
clinical practice. Nevertheless, we applied strict criteria concerning randomized study. Int J Urol. 2001;8:227-230.
the methods for conducting PE and had all CDU examinations car- 9. Chow S-C, Wang H, Shao J. Sample Size Calculations on Clinical Trials,
ried out by a single senior experienced radiologist. 2nd edn. London: Chapman & Hall; 2008.
10. Dubin L, Amelar RD. Varicocele size and results of varicoce-
In conclusion, although PE is the standard diagnostic method for
lectomy in selected subfertile men with varicocele. Fertil Steril.
clinical varicocele, physical examination findings often rely on some 1970;21:606-609.
degree of subjective interpretation by observers. Our data showed 11. Hargreave TB, Liakatas J. Physical examination for varicocele. Br J
high disagreement during varicocele diagnosis and grading by physi- Urol. 1991;67:328.
12. Chiou RK, Anderson JC, Wobig RK, et al. Color Doppler ultrasound
cal examination performed by male infertility specialists, and an even
criteria to diagnose varicoceles: correlation of a new scoring system
higher disagreement between residents. Varicocele repair outcomes with physical examination. Urology. 1997;50:953-956.
may be significantly affected by the lack of objectivity in physical
 |

20472927, 2020, 5, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/andr.12797 by Nat Prov Indonesia, Wiley Online Library on [17/04/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
1166 COCUZZA et al.

13. Schiff JD, Li PS, Goldstein M. Correlation of ultrasound-measured 21. Marsman JW, Schats R. The subclinical varicocele debate. Hum
venous size and reversal of flow with Valsalva with improvement Reprod. 1994;9:1-8.
in semen-analysis parameters after varicocelectomy. Fertil Steril. 22. Yamamoto M, Hibi H, Hirata Y, Miyake K, Ishigaki T. Effect of vari-
2006;86:250-252. cocelectomy on sperm parameters and pregnancy rate in patients
14. Viera AJ, Garrett JM. Understanding interobserver agreement: the with subclinical varicocele: a randomized prospective controlled
kappa statistic. Fam Med. 2005;37:360-363. study. J Urol. 1996;155:1636-1638.
15. Hamada A, Esteves SC, Agarwal A. Varicocele Classification. In:
Alaa Hamada, Sandro Esteves, Ashok Agarwal, eds. Varicocele and
S U P P O R T I N G I N FO R M AT I O N
Male Infertility Current Concepts, Controversies and Consensus. 1st
ed: Springer International Publishing; 2016.p. 37-43 Additional supporting information may be found online in the
16. Liguori G, Trombetta C, Garaffa G, et al. Color Doppler ultrasound Supporting Information section.
investigation of varicocele. World J Urol. 2004;22:378-381.
17. Tanrikut C, Goldstein M. Varicocele: size does matter. J Urol.
2006;176:1912-1913.
How to cite this article: Cocuzza MS., Tiseo BC., Srougi V,
18. Handel LN, Shetty R, Sigman M. The relationship between varico-
celes and obesity. J Urol. 2006;176:2138-2140. et al. Diagnostic accuracy of physical examination compared
19. Jarow JP, Ogle SR, Eskew LA. Seminal improvement follow- with color Doppler ultrasound in the determination of
ing repair of ultrasound detected subclinical varicoceles. J Urol. varicocele diagnosis and grading: Impact of urologists’
1996;155:1287-1290.
experience. Andrology. 2020;8:1160–1166. https://doi.
20. Dhabuwala CB, Hamid S, Moghissi KS. Clinical versus subclinical
varicocele: improvement in fertility after varicocelectomy. Fertil
org/10.1111/andr.12797
Steril. 1992;57:854-857.