In our study, ICLs were most commonly diagnosed in older dogs (mean age of 10.6 years). This finding is in agreement with previously published literature, with RCs most commonly identified in dogs 8-14 years of age [1, 3, 5–7], with reports of affected dogs less than 8 years of age being sparse . In humans, the incidence of simple RCs (defined as those with good acoustic enhancement, absence of echoes within the lesion, and sharply marginated smooth walls)  increases with age, ranging from 0.22 to 0.55% in children and up to 36% in people over 80 years of age . No breed predisposition was identified in our study with mixed breed dogs most commonly affected. Renal cysts have been described in a variety of breeds in the literature including American Staffordshire terrier, German shepherd, Yorkshire terrier, mixed breed dogs, and miniature pincher [1, 3, 5–7]. Males were overrepresented in our study, representing 15 of 18 (83.3%) affected dogs. In the veterinary literature, an equal distribution of male [1, 3, 7] and female [5, 6, 42] dogs has been reported. In humans with simple RCs, a male-to-female ratio of 2.8:1 in adults and 1.6:1 in children has been identified [35, 43]. The male predisposition identified in our study should be interpreted with caution due to the relatively low number of dogs included in the study. Interestingly, no sex predilection has been identified for other renal cystic diseases such as PKD and PNPs in dogs and cats [9, 10, 14, 16, 44–47].
In 6 of 18 dogs in our study, ICLs were an incidental finding, with 5 dogs having no clinical signs and 1 dog having clinical signs related to acute pancreatitis. In 3 previously published canine cases, the RC was an incidental finding during either a routine evaluation following previous mast cell tumor excision , further investigation for multiple mammary masses , or a regular health check . In humans, RCs are mostly asymptomatic and incidentally identified . Of the dogs that were symptomatic in our study, the most common presenting signs were decreased appetite and lethargy, which is similar to what has been previously described in veterinary literature [2, 5, 6]. In humans, <10% of RCs become symptomatic, with symptoms including flank pain, flank mass, hypertension, hematuria, and fever secondary to infection [34, 36]. In 1 multi-institutional retrospective study involving children with symptomatic RCs, the most common reported symptom was abdominal pain .
Abnormalities during physical examination were commonly identified in our study, with a palpable abdominal mass, tense abdomen, and abdominal distention most frequent. Pain on abdominal palpation was a common finding in a previous retrospective case series involving dogs and cats but was only identified in 2 dogs in our study . The identification of an enlarged and irregular kidney on abdominal palpation was a consistent finding in all 6 previously reported cats with RCs [2, 37]. While the presence of a palpable abdominal mass was a relatively common finding (5 of 18) in the dogs of our study, this appears uncommon based on previously published canine RC cases, reported in only 1 of 11 dogs [1–7].
Systemic hypertension has been described as a common finding in dogs and cats with RCs . Systolic blood pressure measurements had values >140 mmHg were identified in 7 dogs in which blood pressure was measured in our study. Similarly, in a previous retrospective study, systemic hypertension was identified in all 5 dogs with RCs (≥160 mmHg) . In humans, it has been shown that the presence of simple RCs is associated with increased prevalence of systemic hypertension with older male patients being more commonly affected [48–50]. Furthermore, human patients with multiple RCs, large RCs or peripherally located RCs have a higher likelihood of being hypertensive [48–50].
In humans, a CT based classification system known as the Bosniak classification is routinely used for categorizing renal cystic lesions according to their likelihood of malignancy . This classification includes 5 categories, namely Bosniak categories (BC) I, II, IIF, III and IV . Studies correlating histopathology of cystic lesions and their classification according to Bosniak system identified that 0% of BC I lesions, 0.09% of BC II lesions, approximately 10% of BC IIF, approximately 50% of BC III lesions, and 90% of BC IV lesions were malignant [51–54]. In humans this classification is used to guide the most appropriate treatment, with BC I and II lesions typically ignored, BC IIF lesions monitored, and BC III and IV excised unless substantial comorbidities or limited life expectancy would favor observation instead . In our study, the choice of treatment was based on surgeon preference. In line with the study inclusion and exclusion criteria, malignancy was not suspected in any of the cases herein on preoperative imaging or cytologic analysis and therefore did not influence the surgeon’s decision regarding the most appropriate treatment. A classification system similar to the Bosniak is lacking in veterinary medicine and may be useful to guide the most appropriate treatment in animals with ICLs. This is highlighted by the fact that 6 of 13 dogs with histopathology in our study were diagnosed with neoplasia, 4 of which were malignant.
Based on the results of our study, PCD was the least successful treatment, with 4 of 5 dogs undergoing additional treatment (repeat PCD or different) because of ICL recurrence. In our study, we included dogs in the PCD category only when ICL drainage was intended to be a definitive treatment and not just for the purpose of obtaining a sample for cytologic analysis. This technique has only been described in 1 other previous case, with the RC found to have regained almost 100% its original size after 1 month . In humans, PCD of simple RCs is associated with a high recurrence rate of up to 80% and therefore is considered an ineffective treatment option [34, 36, 55, 56]. This high recurrence rate is related to continued fluid production associated with the presence of fluid-secreting epithelium lining the cyst wall .
Unlike PCD alone, sclerotherapy is aimed at destroying the secretory epithelium of RCs and is the most commonly described treatment for RCs in dogs and cats [2, 3, 6, 7, 55]. In people, no consensus has been reached regarding the most successful sclerotherapy protocol with regard to volume of sclerosing agent, duration of treatment, number of treatments, however, ethanol is the most commonly used sclerosing agent [34, 36]. Similarly, ethanol was used in all cases that underwent sclerotherapy in our study. Two different sclerotherapy protocols have been described for the management of RCs in veterinary literature and were also used in 4 of 7 dogs in our study. One protocol involved 2 injections on a single occasion of 95% ethanol followed by drainage 3 minutes after each injection . The second protocol involved a single injection of 95% ethanol left in situ for 20 minutes during which time the position of the patient was changed every 5 minutes to ensure uniform distribution of the sclerosing agent . A modification of the latter protocol was used in 3 of 7 dogs of the present study where the sclerosing agent was left intracavitary for 40 minutes in an effort to increase the chances of destruction of the secretory epithelium. In humans, prolonged contact of the secretory epithelium with the sclerosing agent has been shown to result in a lower rate of cyst recurrence [55, 57].
The overall rate of intra-procedural/operative complications in our study was low, with only a minor intraprocedural complication having occurred during sclerotherapy (identification of mild perinephric leakage of the sclerosing agent), which was not associated with patient morbidity and did not require intervention. Leakage of sclerosing agent is a rare complication in people that usually does not result in serious problems . It is also possible that a certain degree of leakage is to be expected and that the occurrence of this complication in dogs is underreported. In a previous report , mild abdominal hemorrhage requiring no intervention was described as an intraoperative complication during sclerotherapy in 1 cat and 1 dog but was not identified in our study. No intraprocedural or postprocedural complications have been reported for canine RCs treated with PCD, ureteronephrectomy or cyst fenestration/omentalization [1, 4–6]. The highest rates of post-procedural/operative complications occurred after PCD and sclerotherapy in our study and included the requirement for revision treatment because of ICL recurrence in most cases. The rate of postoperative complications after deroofing and ureteronephrectomy was low and none required further intervention.
In our study, 3 of 11 ICLs treated with sclerotherapy were found to have involuted at last follow-up imaging (either after sclerotherapy or until the date of revision treatment), 3 were <50% their original size and 5 were ≥50% their original size. Of the 8 ICLs that did not involute, only 2 underwent revision treatment, both of which were ≥50% their original size at follow-up imaging. Of the remaining 6 ICLs that did not involute, 3 were <50% their original size and 3 were ≥50% their original size. These findings, however, should be interpreted with caution because whether these 8 ICLs would have eventually involuted with longer follow-up imaging is unknown. In humans, complete cyst regression can take in excess of 12 months, and therefore, persistence of the cyst on follow-up abdominal ultrasound prior to this time may not signify treatment failure . In the early follow-up period, cyst refilling following sclerotherapy may be secondary to reactive or inflammatory fluid collections, which may eventually disappear . A similar phenomenon has been described in 2 dogs in which the RC partially refilled following sclerotherapy before complete resolution at 1 and 8 months, respectively [3, 6].
Based on the inclusion/exclusion criteria of our study, we did not include ICLs that were suspected to be neoplastic on initial imaging performed. Interestingly, 6 of 13 ICLs with histopathology available were diagnosed with neoplasia, with 4 of these 6 being consistent with malignancy. Primary renal neoplasms are rare in dogs and cats, representing less <2% of all tumors in these species, with renal cell carcinoma being the most common malignant tumor [59, 60]. Cystadenocarcinoma was found in 2 of 4 dogs with malignant neoplasia in our study. This neoplasm has been most frequently described in German shepherd dogs and infrequently in other breeds, is usually bilateral in presentation, and is associated with concurrent nodular dermatofibrosis [61, 62]. Both affected dogs in our study were toy breeds (bichon frisé and miniature pinscher), neither had skin nodules and only 1 had bilateral ICLs. All 6 dogs with neoplastic ICLs in our study had preoperative/intraoperative cytologic analysis of cystic fluid performed and no evidence of neoplastic cells were identified in any dog. In humans with renal cystic masses, cytology is not commonly performed due to its low sensitivity (50%) in detecting malignancy . In people, CT and ultrasound of ICLs is preferred over cytology for the management of ICLs . Based on the results of our study, cytology of aspirates obtained from ICLs cannot be recommended as a method to exclude neoplasia in dogs. Based on the poor sensitivity of cytology for detection of neoplastic cells and the fact that 4 of 13 dogs with histopathology were diagnosed with malignancy, there may be reason for concern with performing renal sparing techniques such as sclerotherapy and deroofing. Conversely, the potential for progression in size of contralateral ICLs with progressive loss of renal function should be carefully considered prior to performing ureteronephrectomy on the most severely affected side in dogs with bilateral lesions. An interesting observation from our study is that dogs with histologically confirmed RCs tend to have >1 ICL in the ipsilateral or contralateral kidney compared with dogs with histologically confirmed neoplasia. This observation requires further investigation and potentially could be used to guide the most appropriate treatment in dogs with ICLs. The identification of renal epithelial cells on cytology is commonly reported in cases of canine RCs but were identified in only a single case in our study [1, 3, 7]. No bacteria were seen cytologically in any dogs in our study, which is similar to what has been reported previously [1, 3, 7]. Furthermore, culture of cystic fluid yielded no bacterial growth in our study, which is in agreement with the majority of previously reported cases in the literature [1–3, 6, 7].
In our study, 3 of 6 ICLs fully resolved at 45, 51 and 61 days, respectively, following deroofing. All 3 were confirmed as histologically benign RCs. In 1 additional dog diagnosed with RC, the lesion was found to have partially resolved at 44 days after deroofing. Whether the RC of this dog would have further reduced in size with longer follow-up is unknown. Deroofing has been described in 2 dogs with RCs in the literature, both of which had RC resolution at 4.5 and 7 months, respectively . One of these cases was treated with laparoscopy similar to 2 dogs in our study. In humans, laparoscopic deroofing is considered the gold standard for management of symptomatic RCs, especially large cysts in young patients and in cases of failed sclerotherapy or PCD . The overall recurrence rate following laparoscopic deroofing in people is 19% and studies have shown that when combined with omentalization the recurrence rate decreases to 0% [34, 64, 65]. The effect of omentalization in ICL resolution rate cannot be extrapolated from the results presented herein due to the low number of cases. A lower rate of RC recurrence has been reported following laparoscopic deroofing in humans compared with single session sclerotherapy, which is in agreement with the results of our study [66–68]. Whether omentalization in addition to deroofing offers an advantage over deroofing alone with drainage into the abdominal cavity where the omentum naturally resides is unknown. It is important to note that in cases with malignant ICLs, deroofing and omentalization may be inappropriate as it may increase the chances of intraperitoneal tumor spread.
Ureteronephrectomy was most commonly performed as an index treatment in our study, with only 1 dog undergoing this procedure as a revision. This is most likely attributable to individual surgeon preference and even though none of the ICLs included in this study demonstrated radiologic evidence of malignancy, it is possible that a renal sparing technique was considered inappropriate by the attending clinician. This procedure has also been described in 2 dogs with RCs in the veterinary literature but is usually reserved for unilateral renal tumors without evidence of metastasis or contralateral renal impairment [4, 5, 69]. Of those that underwent ureteronephrectomy, 3 were diagnosed with RCs on histopathology, and 4 with neoplasia, 3 malignant and 1 benign. As discussed previously, a concern with performing ureteronephrectomy as index treatment in cases with concurrent smaller ICLs in the contralateral kidney is the potential for progression in size of such ICLs and progressive loss of renal function.
We recognize several important limitations in our study. Due to its retrospective nature, the accuracy of recorded data relies on the completeness of the medical records. The number of cases included is larger than what has been reported previously but remains small. This is a reflection of the infrequent nature of ICLs in dogs. Although a multi-institutional study offered the possibility of gathering a larger number of dogs, such a study design brings with it differences in case management, surgical technique, and follow-up. The study inclusion period extended over 15 years and therefore differences in surgical technique will exist over such a period of time. Due to the non-prospective nature of the study, the selected treatment was not randomized but rather based on individual surgeon preference. The timing of follow-up examinations and imaging studies was not standardized in our study and therefore whether certain ICLs would have eventually resolved with greater follow-up time following a particular treatment cannot be ascertained. Histopathology was not available for 5 dogs in our study. Comparison of the rate of requirement for revision treatment between individual techniques is challenging due to the overall low number of cases within individual treatment categories and the inclusion of neoplastic lesions within certain categories.