British Journal of Radiology (2006) 79, 801-803
© 2006 British Institute of Radiology
doi: 10.1259/bjr/50628431
Multiple hyperechoic testicular lesions are a common finding on ultrasound in Cowden disease and represent lipomatosis of the testis
J Woodhouse, MRCS
1 and
M M Ferguson, FDSRCPS
2
1 Department of Urology, Christchurch Hospital, Christchurch, 2 Department of Stomatology, University of Otago, Dunedin, New Zealand
Correspondence: Professor Martin M Ferguson, Department of Stomatology & Health Sciences, University of Otago, P.O. Box 647 Dunedin, New Zealand.
 |
Abstract
|
|---|
Cowden disease (CD) is a genetic disease associated with multiple hamartomas and malignant neoplasms. During investigations for possible subnormal fertility, a series of eight males with CD underwent ultrasound scanning of their testes. Our findings detail the seven adult patients that were found to have multiple and bilateral testicular lesions on ultrasound. These lesions in Cowden's patients represent a newly described testicular pathology – lipomatosis of the testis. Here we detail the radiological findings. Ultrasound findings showed multiple (estimated 40+) discreet lesions randomly scattered thoughout the parenchyma of each affected testis. The lesions themselves had heterogeneous echo texture, but all being hyperechoic and with a variable size from 1 mm to 6 mm. One patient underwent MR examination of the lesions in which a high T1 signal was seen, but no useful characteristic features were identifiable.
|
Introduction
|
|---|
Cowden disease (CD; also known as Multiple Hamartoma Syndrome), is marked by increased cellular proliferation of ectodermal, mesodermal and endodermal tissues [1, 2]. These neoplasms can be benign or malignant. In particular, cancer of the breast, thyroid, endometrium and skin has been associated with CD. Diagnosis is based on operational criteria laid out by the "International Cowdens syndrome consortium" and supported by genetic testing. Mucocutaneous lesions comprise the predominant feature and are a principal way of suspecting the diagnosis. Phenotypic expression is variable and thus some cases probably remains undiagnosed. Despite it being a dominantly inheritable condition, the prevalence of the disease remains relatively low with more recent estimates at 1 per 250 000 [3].
A singular lesion within the parenchyma of a testis is not an uncommon finding with ultrasound scanning, but when found they are regarded as highly suspicious of malignancy. Multiple lesions of the testes are infrequent and usually represent microlithiasis. Multiple testicular lesions in the context of CD has been previously reported [4], but only in the context of a single patient. No evidence was provided to establish the histopathology in this case. Our findings are suggestive that these lesions are a very frequent occurrence in CD. This ultrasound finding is now known to represent a new histopathological entity called lipomatosis of the testis [5]. Here we detail the radiological aspects of this new condition.
|
Methods
|
|---|
All male patients with CD known to the authors were involved in this study, which was initially directed as a study into fertility issues and CD and subject to local ethics approval. The study was prospective and there were no exclusions made. All of the eight males had clinical features of CD and the diagnosis was confirmed by identification of the mutation on chromosome 10q.23. Ages ranged from 16 years to 58 years (mean = 38 years). The patients were interviewed and examined with attention initially on issues surrounding possible low fertility. Structural assessment of the testes in all eight was performed with ultrasound. Scanning was performed with a range of machines in four separate institutions from both the private and public sectors. Scanners included Philips HDI 5000 using high frequency linear probes. Exact scanner models were unable to be determined for all scans performed. All sonographers held the diploma of Australian Society of Ultrasound in Medicine and Biology. All scans were reviewed retrospectively by consultant radiologists at Christchurch Public Hospital. In some cases, some scans were assessed in real time by those consultants. Following identification of anatomical testicular lesions, all patients had serum tumour marker assays performed (alpha fetoprotein, beta human chorionic gonadotropin, and lactate dehydrogenase) and were offered interval ultrasound. Two patients had had previous testicular ultrasound for comparison. Overall, five of the eight had repeated scans with intervals ranging from 5 months to 89 months for comparison. One patient underwent MRI (General Electric Sigma Horizon LX 1.5 Tesla, Wisconsin, USA) with T1, T1 fat saturated, T1 fat saturated post-gadolinium enhancement (repetition time (TR) 600 ms, echo time (TE) 14 ms, field of view (FOV) 10 cm, slice thickness 3 mm, gap 0.5 mm, phase 224, frequency 384, electronic train length 3; 3 inch surface coil) and T2 sequences (TR 3500 ms, TE 96 ms, FOV 10 cm, slice thickness 3 mm, gap 0.5 mm, phase 192, frequency 256, electronic train length 12; 3 inch surface coil), attempting to further characterize these lesions. Four of eight subjects elected to undergo open testicular biopsy under general anaesthetic due to lack of any definitive pathological correlation.
|
Results
|
|---|
Of the cohort, all were normal on urogenital physical examinations with none of the lesions described being palpable. One patient previously had a testicular lump observed by ultrasound and discharged as benign, and this was still palpable and non tender. Another patient reported generalized low grade testicular tenderness. All were of normal sexual development.
All patients had at least one ultrasound and five had repeat interval scans. Ultrasound showed that all had testes of normal size and contour, but the seven adults had multiple (estimates of 40+ per testes) hyperechogenic foci of approximately 1–6 mm size within the parenchyma of both testicles (Figure 1
). Most lesions were small and non-shadowing. Only in one testis were acoustic shadows seen (Figure 2). The lesions were randomly scattered throughout the testis in all cases. The lesions did not demonstrate increased vascularity on Doppler imaging. There were no other focal parenchymal abnormalities seen. Five of seven had interval repeat ultrasound for comparison (range 5–89 months). Within the limits of ultrasound analysis, all lesions were reported as remaining stable. However, owing to the number of lesions in each testis it was not possible to categorically determine this.
View larger version (166K):
[in this window]
[in a new window]
|
Figure 1. Longitudinal ultrasound view of the right testis demonstrates well defined multiple non-shadowing small round lesions that are hyperechoic. The lesions are of various sizes and are non-confluent. No hypoechoic regions or calcifications are seen.
|
|
View larger version (166K):
[in this window]
[in a new window]
|
Figure 2. Longitudinal ultrasound of testis in one case demonstrated that some of the lesions displayed acoustic shadowing. The lesions here are similarly hyperechoic. No hypoechoic regions or calcifications are seen.
|
|
Two patients had undergone previous testicular ultrasound performed for other reasons available for retrospective comparisons. One patient, 3 years previously, had four serial scans for a suspicious scrotal lump. These had shown a calcified nodule on the surface of the testicle, but with poorly defined regions of echogenicity within the parenchyma. The original interpretation was that these regions had represented microlithiasis. Films were reviewed in light of our findings and the opinion was that these lesions were not microlithiasis. Another patient had had two scans carried out 7 years previously. Identical small echogenic non-shadowing foci were present at that point. With both patients, as initial scans had been for other reasons, it was impossible to comment with certainty as to the stability of the lesions; although it did seem that these lesions had not progressed significantly.
In an effort to further characterize the lesions non-invasively, one patient underwent MRI of the scrotum. MRI showed multiple (greater than 15) tiny focal high-signal areas within both testes, probably corresponding to the lesions seen on the ultrasound images, although these were very poorly visualized in all sequences used. Lesions were best seen on the T1 series (Figure 3). Insufficient detail was gained to be diagnostic.
View larger version (79K):
[in this window]
[in a new window]
|
Figure 3. SagittalT1 weighted spin echo sequence (repetition time (TR) 600 ms, echo time (TE) 14 ms, 3 mm slice thickness, field of view (FOV) 10 cm) of the left testis. The scan demonstrates tiny hyperintense areas (white arrows) within the testis parenchyma consistent with fat.
|
|
|
Discussion
|
|---|
CD is a rare disease and as such our cohort of eight patients represents a relatively large population of CD patients. The identification of the above described multiple bilateral testicular lesions on ultrasound and MR was initially documented by Lindsay [4] in a single CD patient, and has not been documented outside the context of CD. In this single case, the lesions were assumed to be hamartoma based on the lesions having a fat component on MRI, but no histopathological proof was offered for this diagnosis. Our series of CD patients were all impalpable incidentalomas found while investigating subfertility, probably explaining why they have not been extensively documented previously. Our group has established that these lesions represent lipomatosis of the testis in the four patients who underwent biopsy. Biopsy microscopic findings showed "interstitial lipomatosis consisting of nests of adipocytes within the testicular interstitium. No intratubular calcification or intratubular germ cell neoplasia was seen". There appeared to be no detrimental effect on spermatogenesis, nor on testicular biochemical function [5].
The diagnosis of malignancy was considered before diagnostic biopsy was performed. Serial scanning using ultrasound could not absolutely qualify the stability of the lesions as lesions were too numerous to enable the identification of individual lesions for comparative assessment on repeated scans. MRI unfortunately proved inadequate to be useful diagnostically. Hence biopsy was a diagnostic option that was offered to all these patients and testicular biopsies were performed on some of these patients (but not on all, once it was established that the lesions seen on ultrasound were not tumours). The appearances seen in our series represent benign testicular lipomatosis (hamartomatous growths). Ultrasound findings of primary testicular malignancies tend to be hypoechoic and homogeneous (seminoma) or are well defined with cystic elements and calcifications (teratoma). Lymphoma, leukaemia, and metastatic disease is uncommon and has variable appearance on ultrasound from focal hypoechoic lesions to a diffusely altered echogenicity [6, 7] – not similar to the findings of testicular lipomatosis seen in our cohort, nor the lesions described by Lindsay [4].
Benign testicular lesions are uncommon and are difficult to distinguish from malignancy on ultrasound. Because of this, they often result in orchidectomy. The high signal on MR T1 imaging supported a fatty component to the lesions and does lend itself to a diagnosis of hamartomas (as suggested by Lindsay) or lipomas as is the diagnosis. Testicular lipomas are reported, although they are very rare and have only been described only as singular lesions [8, 9].
Microlithiasis is the most commonly documented testicular pathology occurring as a multiple entity, and consists of multiple randomly scattered punctuate lesions in the testicular parenchyma (with variable acoustic shadowing) of approximately 1–3 mm in size representing calcified deposits in the lumen of seminiferous tubules. As they represent calcifications, they have a very high echogenicity and are thus distinctly different to the lesions we find in the Cowden's testes.
Testicular lipomatosis is a novel entity. We have detailed how the ultrasound appearances differ distinctively from other pathologies. These lesions are not described outside the context of CD and thus are apparently pathognomic for CD and could be used as a major diagnostic criterion for CD in adult males. A finding on ultrasound of these lesions should thus alert the radiologist to the diagnosis of CD. In our series, 7 of 8 of males with CD had these lesions. Phenotypic expression in CD often is not manifest until aged 20–30 years [10] and it remains to be seen if the eighth male – currently with no testicular lesions and aged 16 years – will eventually develop these lesions. The long term behaviour of these lesions is unknown to date.
|
Acknowledgments
|
|---|
The Authors would like to acknowledge the kind help of Hamish Fraser, Andrew Lang, Diane Leighton and Tim Buckingham for their help and advice regarding this work. We would also like to acknowledge the staff of Christchurch Public Hospital Radiology department and the Southern Cross Radiology department where the majority of scans were performed.
Received for publication October 5, 2005.
Revision received March 22, 2006.
Accepted for publication March 30, 2006.
|
References
|
|---|
- Lloyd KM, Dennis M. Cowden's disease. A possible new symptom complex with multiple system involvement. Ann Intern Med 1963;58:136–42.[Abstract/Free Full Text]
- Weary PE, Gorlin RJ, Gentry WC Jr, Comer JE, Greer KE. Multiple hamartoma syndrome (Cowden's disease). Arch Dermatol 1972;106:682–90.[Abstract/Free Full Text]
- Nelen MR, et al. Novel PTEN mutations in patients with Cowden disease: absence of clear genotype-phenotype correlations. Eur J Hum Genet 1999;7:267–73.[CrossRef][Medline]
- Lindsay C, Boardman L, Farrell M. Testicular hamartomas in Cowden disease. J Clin Ultrasound 2003;31:481–3.[CrossRef][Medline]
- Woodhouse JB, Delahunt B, English SF, Fraser HH, Ferguson MM. Testicular lipomatosis in Cowden's syndrome. Mod Pathol 2005;18:1151–6.[CrossRef][Medline]
- Hamm B. Differential diagnosis of scrotal masses by ultrasound. Eur Radiol 1997;7:668–79.[Medline]
- Howlett DC, Marchbank ND, Sallomi DF. Pictorial review. Ultrasound of the testis. Clin Radiol 2000;55:595–601.[CrossRef][Medline]
- Harper M, Arya M, Peters JL, Buckingham S, Freeman A, O'Donoghue EP. Intratesticular lipoma. Scand J Urol Nephrol 2002;36:223–4.[CrossRef][Medline]
- Honore LH. Fatty metaplasia in a postpubertal undescended testis: a case report. J Urol 1979;122:841–2.[Medline]
- Nelen MR, et al. Localization of the gene for Cowden disease to chromosome 10q22–23. Nat Genet 1996;13:114–6.[CrossRef][Medline]
This article has been cited by other articles:
|
|
|
|
 
R. N. Walker, T. J. Murphy, and M. L. Wilkerson
Testicular Hamartomas in a Patient With Bannayan-Riley-Ruvalcaba Syndrome
J. Ultrasound Med.,
August 1, 2008;
27(8):
1245 - 1248.
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
