RELATIONSHIP BETWEEN PROSTATOMEGALY, PROSTATIC MINERALIZATION, AND CYTOLOGIC DIAGNOSIS

Transcription

1 RELATIONSHIP BETWEEN PROSTATOMEGALY, PROSTATIC MINERALIZATION, AND CYTOLOGIC DIAGNOSIS CHRISTINA A. BRADBURY, JODI L. WESTROPP, RACHEL E. POLLARD Canine prostatic disease is commonly evaluated with abdominal ultrasound and radiographs. Mineralization of the prostate is often reported, but the clinical relevance of this finding is currently not known. The purpose of this study was to characterize the relationship between ultrasonographic and radiographic prostate mineralization and the final diagnosis. Medical records of 55 dogs with evidence of prostatomegaly or prostatic mineralization and a cytologic diagnosis were evaluated. Radiographs and ultrasound images were assessed for caudal retroperitoneal lymphadenopathy, vertebral lesions, or other signs of metastasis, and mineralization was assessed semiquantitatively. Twenty-two of 55 (40%) dogs had prostatic neoplasia. Regarding neoplasia, mineralization in neutered dogs had a positive predictive value (PPV) of 100%, a negative predictive value (NPV) of 50%, and a sensitivity and specificity of 84% and 100%, respectively. Mineralization in intact dogs had a PPV of 22%, an NPV of 96%, and a sensitivity and specificity of 67% and 77%, respectively. All neutered dogs with prostatomegaly but not prostatic neoplasia had bacterial prostatitis and were castrated within the previous 3 months. Intact dogs with prostatomegaly and mineralization but not neoplasia had paraprostatic cysts (n ¼ 3), benign prostatic hyperplasia (n ¼ 2) or prostatitis (n ¼ 2). Mineralization score was not indicative of neoplasia. In conclusion, neutered dogs with prostatic mineralization were very likely to have prostatic neoplasia. Intact dogs were unlikely to have prostatic neoplasia if no mineralization was found on radiographs or ultrasound. Veterinary Radiology & Ultrasound, Vol. 50, No. 2, 2009, pp Key words: prostatic mineralization, prostatomegaly, radiography, ultrasound. From the Department of Surgery and Radiological Sciences (Pollard), Department of Medicine and Epidemiology (Westropp), and the Veterinary Medical Teaching Hospital (Bradbury), School of Veterinary Medicine, University of California, Davis, CA. Presented in part at the American College of Veterinary Internal Medicine Annual Meeting, San Antonio, TX, June Address correspondence and reprint requests to Rachel E. Pollard, DVM, PhD, at the above address. repollard@ucdavis.edu Received July 30, 2008; accepted for publication September 25, doi: /j x Introduction ROSTATIC CARCINOMA IS a highly malignant neoplasm in Pthe dog with a prevalence of %. 1 Prostatic neoplasia must be distinguished from nonmalignant prostatic disease so that appropriate management can be initiated. Castrated dogs appear to be at risk for prostatic carcinoma, 2,3 with an odds ratio of compared with intact dogs. Castration after sexual maturity does not reduce the risk of developing prostatic carcinoma. 2 4 Adenocarcinoma is the most common tumor in intact dogs, while tumors of mixed morphology were more common in castrated dogs. 5 Of 11 castrated dogs with prostatic disease, eight had prostatic neoplasia and three had bacterial prostatitis. 6 However, two of the three castrated dogs with prostatitis had been neutered because of clinical signs associated with prostatic infection. This, in addition to a paucity of reports in the literature indicating otherwise, suggests it is rare for castrated dogs to develop prostatic disease other than neoplasia. At our institution, the presence of mineralization in the prostate identified with either abdominal radiographs or ultrasound is considered highly suspicious for prostatic neoplasia. Prostatic mineralization is considered even more suspicious for neoplasia if the dog is neutered. A previous study supports this postulation in that multifocal, irregularly shaped, parenchymal mineral opacities were only seen in dogs with prostatic carcinoma and one dog with chronic bacterial prostatitis. 7 Other reports describe the presence of mineralization associated with paraprostatic cysts but that the appearance of the cyst wall mineralization was characteristic and easily distinguishable from parenchymal mineralization. 8,9 The purpose of this retrospective study was to determine the underlying disease process in a large number of dogs with radiographic or ultrasonographic evidence of prostatic mineralization. This was compared with a population of dogs with an imaging diagnosis of prostatomegaly but no evidence of mineralization. We hypothesized that most sexually intact dogs with prostatic mineralization would have prostatic neoplasia or paraprostatic cysts. In the event of prostatic mineralization in dogs with nonneoplastic disease, we anticipated that mineralization would be less severe than that seen in dogs with neoplastic 167

2 168 BRADBURY,WESTROPP, AND POLLARD 2009 prostatic disease. Moreover, we anticipated that all neutered dogs with prostatic mineralization would have prostatic neoplasia. Materials and Methods Fig. 1. (A) A lateral radiograph of the caudal abdomen is shown with mineralization of the prostatic parenchyma visible (arrows). The degree of mineralization was graded as 2. (B) A sagittal ultrasound image of the prostate is shown with multiple hyperechoic foci representing mineralization throughout the parenchyma. The degree of mineralization was graded as 3. Dogs with radiographic or ultrasonographic evidence of prostatomegaly or prostatic mineralization imaged between March 1993 and April 2006 were identified. Dogs were included if a cytologic or histopathologic diagnosis had been made and if the radiographs or ultrasonographic images were available for review and there was evidence of prostatomegaly or prostatic mineralization. Dogs were excluded if the time between imaging and final diagnosis was greater than 1 month, or if there was questionable radiographic or ultrasonographic evidence of either prostatomegaly or prostatic mineralization. Records were reviewed by two of the authors (R.E.P. and C.M.B.) and information regarding signalment, radiographic and ultrasonagraphic findings, cytologic and histopathologic findings, and method of tissue sampling was recorded. When radiographs were available, the ratio of prostatic height to the pubic brim sacral promontory dimension as measured on the lateral abdominal radiograph was calculated. 7 Prostatomegaly was defined as a ratio greater than Radiographic prostatic mineralization was defined as the presence of mineral opacities within the prostate. 10 For ultrasound images, prostatomegaly was defined subjectively based on impressions noted by the clinician in the imaging report. Prostatic mineralization on abdominal ultrasonography was defined as areas of high reflectivity within the prostatic parenchyma that caused distal acoustic shadowing. 10 A semiquantitative scale of mineralization was established where a score of 0 indicated no mineralization, 1 indicated mild mineralization, 2 indicated moderate mineralization, and 3 indicated severe mineralization (Fig. 1). Radiographs and ultrasonographic images and reports were reviewed and evidence of caudal retroperitoneal lymphadenopathy, vertebral lesions, and pulmonary metastatic disease was recorded. Ultrasound-guided fine-needle aspiration (FNA) and cytologic evaluation are often used to determine the underlying etiology of prostatic disease. Although histopathologic diagnosis of prostatic disease remains the gold standard, there is agreement between cytologic and histopathologic diagnoses for prostatic disease. 11 Therefore, both types of sampling techniques were included in our study. All cytologic and histopathologic samples were prepared and evaluated in our institution. Samples for cytologic diagnosis were collected by ultrasound-guided FNA or traumatic catheterization. Histopathologic diagnosis was made from samples collected via biopsy or necropsy. When dogs had evidence of mineralization, they were divided into two groups based on neoplastic and nonneoplastic diagnoses. The age, prostatic mineralization score, and prostatic size were compared between neoplastic and nonneoplastic groups using an unpaired Student s t-test (Excel ). A P-value 0.05 was considered significant. The positive and negative predictive values (PPV, NPV), specificity, and sensitivity of evidence of mineralization in intact and in castrated dogs were calculated. Results Of 199 records reviewed, 55 dogs met the inclusion criteria. A variety of breeds were included with no breed be- Microsoft Corporation, Redmond, WA.

3 Vol.50,No. 2 PROSTATIC MINERALIZATION IN DOGS 169 Table 1. Cytologic Diagnosis for 55 Dogs with Prostatomegaly and/or Prostatic Mineralization Neoplastic Disease (n ¼ 22) Nonneoplastic Disease (n ¼ 33) Cytologic Diagnosis Number of Dogs (%) Cytologic Diagnosis Number of Dogs (%) Prostatic carcinoma 14 (64%) Prostatitis 22 (67%) Transitional cell carcinoma 5 (24%) Benign prostatic hyperplasia 7 (21%) Undifferentiated carcinoma 1 (4%) Paraprostatic cyst 3 (9%) Anal sac adenocarcinoma 1 (4%) Cystic hyperplasia 1 (3%) Lymphosarcoma 1 (4%) Dogs are divided into neoplastic and nonneoplastic categories. ing overrepresented. Thirty-four dogs (62%) were sexually intact while 21 (38%) were castrated. All dogs had either radiographs (n ¼ 17), ultrasound images (n ¼ 12), or both (n ¼ 26) available for review. Diagnoses were obtained in 21 dogs via prostatic aspirates, 19 by biopsy, two had traumatic catheterization, and in 13 dogs, a diagnosis was obtained at necropsy. Of the 55 dogs, 22 (40%) had prostatic neoplasia and 33 (60%) had nonneoplastic diseases. Table 1 defines the cytologic diagnosis in the 55 dogs. Nineteen of 22 (86%) dogs with neoplasia were castrated and three (14%) were intact. Eighteen of the 22 (82%) dogs with prostatic neoplasia had evidence of mineralization; of the four dogs without mineralization, two had prostatic carcinoma, one had transitional cell carcinoma, and one had anal sac adenocarcinoma. The mean ( standard deviation) age for dogs with neoplasia was 9.45 ( 1.97) years. Thirty-one of 33 (94%) dogs with nonneoplastic disease were intact. The two neutered dogs in the nonneoplastic group had been neutered within the previous 3 months due to clinical signs associated with prostatic disease and neither dog had evidence of mineralization. Seven of 33 (21%) dogs with nonneoplastic prostatic disease had evidence of mineralization; of those, two had benign prostatic hyperplasia, two had prostatitis, and three had paraprostatic cysts. The mean ( standard deviation) age for dogs without neoplasia was 8.34 ( 3.14) years. Age was not significantly different between neoplastic and nonneoplastic groups (P ¼ 0.07). Of the 43 dogs that had radiographs available for review, 14 (33%) had prostatic neoplasia and 29 (67%) had nonneoplastic prostatic disease. There was no significant difference in the ratio of prostatic height to the pubic brim sacral promontory dimension when comparing dogs with and without neoplasia [ vs (mean standard deviation); P ¼ 0.22]. Furthermore, when evaluating the mean prostatic mineralization score, no significant difference between dogs with prostatic neoplasia and those with nonneoplastic processes was found ( vs ; P ¼ 0.60). Additional radiographic lesions such as pulmonary metastatic disease (n ¼ 4), or periosteal reaction on the vertebra (n ¼ 3), ilium (n ¼ 1), or femurs (n ¼ 1), and/or enlarged caudal retroperitoneal lymph nodes (n ¼ 6) were seen in dogs with prostatic neoplasia. Only two dogs with nonneoplastic disease had retroperitoneal lymph node enlargement. Of the 38 dogs that had ultrasound images available for review, 18 (47%) had prostatic neoplasia and 20 (53%) had nonneoplastic prostatic disease. The mean mineralization score for dogs with prostatic neoplasia was significantly higher compared with dogs with nonneoplastic disease [ vs l; (mean standard deviation) P ¼ 0.02]. Assessment for additional lesions was compromised by a limited number of available images; however, six dogs with prostatic neoplasia had retroperitoneal lymph node enlargement, whereas only two dogs with nonneoplastic disease had lymphadenopathy. Of the 26 dogs that had both radiographs and ultrasound images available for review, the mineralization score was not significantly different for the two imaging modalities (P ¼ 0.74). Sixteen of 21 neutered dogs in this study had prostatic mineralization and all 16 (100%) had neoplasia. Nine of 34 intact dogs in this study had evidence of mineralization and two (22%) had neoplasia. The PPV and NPV, sensitivity, and specificity for the presence of mineralization in cases of neoplasia are shown for neutered and intact dogs in Table 2. Note that the PPV of prostatic mineralization in neutered dogs is 100%, meaning that the presence of mineralization indicates neoplasia. The high NPV indicates that intact dogs without prostatic mineralization are very unlikely to have neoplasia. Discussion Based on our results, castrated dogs with prostatic mineralization have a high likelihood of having prostatic neoplasia. Furthermore, intact dogs with prostatic mineralization may have diseases other than neoplasia including Table 2. Positive and Negative Predictive Values (PPV, NPV), Sensitivity, and Specificity Regarding the Presence of Mineralization and the Incidence of Neoplasia Sexual Status PPV (%) NPV (%) Sensitivity (%) Specificity (%) Intact Neutered Results for intact and neutered dogs are shown separately.

4 170 BRADBURY,WESTROPP, AND POLLARD 2009 paraprostatic cysts, benign prostatic hyperplasia, or prostatitis. Differentiation of these diseases is essential to provide veterinarians with the tools needed to manage each disorder properly. In people, prostatic disease is a frequent disorder in older men In 1990, surgery for benign prostatic hyperplasia was the most common surgical procedure performed in males. 15 Prostate cancer is now the most frequently diagnosed cancer in men, accounting for 33% of all diagnosed cancers in males in Prostatitis is the most common urologic diagnosis in men less than 50 years of age. 12 Although age and results from digital rectal exam provide some help in differentiating these disorders, further tests are needed for confirmation. These tests include measurement of serum prostate-specific antigen, transrectal ultrasound with needle biopsy, and magnetic resonance imaging. 14,17 19 The sonographic appearance of prostatic neoplasia and benign prostatic hyperplasia overlap 20 and punctate, brightly echogenic foci, presumed to be corpora amylacea deposits or areas of calcification, were often seen in both benign and malignant lesions in men. 21 Abdominal radiographs are not used as a test for human prostatic disease. On the contrary, abdominal radiographs are commonly used in dogs to evaluate the prostate gland. A radiographically normal prostate gland has been defined as a height less than 70% of the pubic brim sacral promontorydimensiononalateralview. 7 Moreover, if prostatic height exceeds 90% of the pubic brim sacral promontory dimension, this is suggestive of prostatic neoplasia, abscess, or paraprostatic cyst. 7 Although these guidelines did provide evidence for prostatic enlargement, no differentiation between neoplasia and other disease processes could be ascertained. Prostate gland mineralization can be identified using both radiography and ultrasonography. In prior work, dogs with prostatic mineralization were found to have prostatic neoplasia, but the association between the degree of mineralization and the specific disease was not characterized. 7,22 We suspected that dogs with severe mineralization would be more likely to have prostate cancer, and this was confirmed with regard to sonographic imaging. However, there was not an association between the degree of radiographic mineralization and diagnosis. One possible explanation is that ultrasound is more sensitive for detecting small amounts of mineralization, but in animals having both modalities the mineralization score was not significantly different between the two imaging modalities. Thus, the reason for this discrepancy is unknown. Also in prior work, ultrasonographic shadowing, consistent with mineralization or gas, did not improve differentiation between bacterial prostatitis and prostatic carcinoma; however, four out of five (80%) dogs with radiographic evidence of mineralization were ultimately diagnosed with prostatic neoplasia. 7 Our findings were similar in that 73% of dogs with radiographic evidence of mineralization, and 67% of dogs with ultrasonographic evidence of mineralization, had prostatic neoplasia. We found caudal retroperitoneal lymphadenopathy in both neoplastic and nonneoplastic conditions, as noted by others. 7 Additional radiographic findings such as pulmonary metastatic disease and periosteal reactions on the vertebra, ilium, or femur were only seen in dogs with prostatic neoplasia. In summary, neutered dogs with prostatic mineralization are likely to have prostatic neoplasia as the PPV and specificity in our population was 100%, whereas intact male dogs with prostatic mineralization may have either neoplastic or nonneoplastic prostatic disease. Intact dogs without radiographic or ultrasonographic evidence of prostate mineralization were unlikely to have prostatic neoplasia, having an NPV of 96%. The size of the prostate gland was not useful for differentiating between malignant and benign prostatic disease. Sonographic mineralization score, but not radiographic mineralization score, was significantly higher in dogs with malignant prostatic disease. Differentiation between neoplastic and nonneoplastic disease cannot be made in intact dogs based on concurrent prostatic mineralization and caudal retroperitoneal lymphadenopathy, but if pulmonary metastatic disease or periosteal reactions on vertebra, femur, or pelvic bones are seen, then neoplasia is highly likely. 1. Weaver AD. Fifteen cases of prostatic carcinoma in the dog. Vet Rec 1981;109: Obradovich J, Walshaw R, Goullaud E. The influence of castration on the development of prostatic carcinoma in the dog. 43 cases ( ). J Vet Intern Med 1987;1: Teske E, Naan EC, van Dijk EM, et al. Canine prostate carcinoma: epidemiological evidence of an increased risk in castrated dogs. Mol Cell Endocrinol 2002;197: Bell FW, Klausner JS, Hayden DW, et al. Clinical and pathologic features of prostatic adenocarcinoma in sexually intact and castrated dogs: 31 cases ( ). J Am Vet Med Assoc 1991;199: Cornell KK, Bostwick DG, Cooley DM, et al. Clinical and pathologic aspects of spontaneous canine prostate carcinoma: a retrospective analysis of 76 cases. Prostate 2000;45: REFERENCES 6. Krawiec DR, Heflin D. Study of prostatic disease in dogs: 177 cases ( ). J Am Vet Med Assoc 1992;200: Feeney DA, Johnston GR, Klausner JS, et al. Canine prostatic disease comparison of radiographic appearance with morphologic and microbiologic findings: 30 cases ( ). J Am Vet Med Assoc 1987;190: Head LL, Francis DA. Mineralized paraprostatic cyst as a potential contributing factor in the development of perineal hernias in a dog. J Am Vet Med Assoc 2002;221: , Zekas LJ, Forrest LJ, Swainson S, et al. Radiographic diagnosis: mineralized paraprostatic cyst in a dog. Vet Radiol Ultrasound 2004;45: Thrall D. Textbook of veterinary diagnostic imaging, 5th ed. Philadelphia, PA, USA: Saunders, 2007.

5 Vol.50,No. 2 PROSTATIC MINERALIZATION IN DOGS Powe JR, Canfield PJ, Martin PA. Evaluation of the cytologic diagnosis of canine prostatic disorders. Vet Clin Pathol 2004;33: Nickel JC. Prostatitis: myths and realities. Urology 1998;51: Sciarra F, Toscano V. Role of estrogens in human benign prostatic hyperplasia. Arch Androl 2000;44: Woolf SH. Screening for prostate cancer with prostate-specific antigen. An examination of the evidence. N Engl J Med 1995;333: Carter HB, Coffey DS. The prostate: an increasing medical problem. Prostate 1990;16: Jemal A, Siegel R, Ward E, et al. Cancer statistics, CA Cancer J Clin 2006;56: Albertsen PC, Hanley JA, Gleason DF, et al. Competing risk analysis of men aged 55 to 74 years at diagnosis managed conservatively for clinically localized prostate cancer. JAMA 1998;280: Kurhanewicz J, Swanson MG, Nelson SJ, et al. Combined magnetic resonance imaging and spectroscopic imaging approach to molecular imaging of prostate cancer. J Magn Reson Imaging 2002;16: Tong S, Downey DB, Cardinal HN, et al. A three-dimensional ultrasound prostate imaging system. Ultrasound Med Biol 1996;22: Lee F, Gray JM, McLeary RD, et al. Prostatic evaluation by transrectal sonography: criteria for diagnosis of early carcinoma. Radiology 1986;158: Burks DD, Drolshagen LF, Fleischer AC, et al. Transrectal sonography of benign and malignant prostatic lesions. AJR Am J Roentgenol 1986;146: Feeney DA, Johnston GR, Walter PA. Ultrasonography of the kidney and prostate gland. Has gray-scale ultrasonography replaced contrast radiography? Probl Vet Med 1991;3: