Minimal renal pelvis dilation in cats diagnosed with benign ureteral obstruction by antegrade pyelography: a retrospective study of 82 cases (2012–2018)

Abstract

Objectives

The aim of the study was to describe renal pelvis (RP) and ureteral ultrasonographic measurements in a population of cats with confirmed benign ureteral obstruction (UO) by antegrade pyelography. The secondary objective was to further describe clinical findings associated with minimally dilated obstructed kidneys in an attempt to better understand its occurrence.

Methods

Retrospective case series of cats diagnosed with benign UO were confirmed by antegrade pyelography. Medical records were reviewed and signalment, diagnostic imaging results, serum creatinine (SCr) concentration and urine culture results were recorded. Each obstructed kidney was categorized into two groups: group 1 included all RP measurements ⩽4 mm and group 2 included all RP measurements >4 mm.

Results

A total of 82 cats with 114 obstructed ureters met the inclusion criteria. Fifty (61%) cats had a unilateral UO and 32 (39%) had a bilateral UO. Thirty (26%) kidneys were included in group 1 while 84 (74%) were included in group 2. Nine (8%) kidneys had an RP dilation ⩽2 mm. Median RP and ureteral diameters were 6.6 mm (range 1.1–37.0 mm) and 3.2 mm (range 0.0–11.0 mm), respectively. RP size correlated positively with ureteral diameter in the study population (P <0.0001), but not in group 1 when analyzed separately (P = 0.47). UO was secondary to stones in 80 (70%) ureters. Seventeen (21%) cats had a positive urine culture. At admission, 79 (96%) cats were azotemic with a median preoperative SCr concentration of 444 µmol/l (range 108–1326 μmol/l). The mean (95% confidence interval [CI]) preoperative SCr concentration was significantly higher in group 1 (762 µmol/l [498–1165 μmol/l]) than in group 2 (409 µmol/l [333–502 μmol/l]). RP size in the two groups correlated negatively with preoperative SCr concentration (P = 0.0002).

Conclusions and relevance

Feline UO may be associated with minimal RP dilation and the severity of RP and ureteral dilation can be highly variable. Absence of significant RP dilation does not rule out UO in cats.

Introduction

Since the first case series of cats with ureterolithiasis was published in 1998, feline ureteral obstruction (UO) has been increasingly diagnosed and treated.^1,2

Feline UO is a potentially life-threatening condition that may cause severe azotemia by decreasing renal blood flow and glomerular filtration rate secondary to increased pressure within the renal pelvis (RP) and ureter. Cats with UO often suffer from concurrent kidney disease. ³

UO in cats most commonly occurs secondary to obstructive ureterolithiasis, which are mainly composed of calcium oxalate.^4–7 Obstructive ureterolithiasis is now recognized as the most frequent cause of acute kidney injury in this species.^4,8 Other causes of UO include acquired ureteral stricture, ureteritis, pyelonephritis/pyonephrosis, ureteral neoplasia, blood clots, dried solidified blood calculi, congenital abnormalities (ureteral stenosis with or without ectopia, circumcaval ureter), surgical trauma and retroperitoneal fibrosis.^2,4,5,9–14

Ultrasonography is commonly used to assess cats with azotemia. Findings such as RP dilation and ureteral dilation proximal to the location of obstruction are usually seen with UO and the underlying cause may be identified. In other cases, such as with stricture or blood clot and dried solidified blood calculi, the cause of UO may be difficult to identify on ultrasound.^12,13 Diagnostic utility of ultrasonography may vary depending on the cause of the UO, with a sensitivity and specificity of 98% and 96%, respectively for detection of ureterolithiasis and 44% and 98%, respectively for detection of ureteral stricture. ⁷ Ultrasonography was considered inaccurate to diagnose feline UO in one study when using pyelography as the gold standard, as no significant differences in ultrasonographic signs were found between cats with an obstructed and non-obstructed ureter. ¹⁵ In the study by Lamb et al, only 71% of cats with a tentative diagnosis of UO, based on clinical data including ultrasound findings of the urinary tract, were shown to actually have a UO using pyelography.

The RP of cats and dogs with normal renal function may be too small to measure ultrasonographically, although RP widths reaching almost 4 mm have been observed in both species.^16,17 Diuresis or polyuria may result in RP dilation in cats and dogs.^16,18,19 The degree of RP dilation varies among cats with normal renal function and those values may overlap with those observed in cats with renal disease such as chronic kidney disease (CKD), pyelonephritis or UO, thereby complicating its interpretation.^16,17 In two previous studies, RP dilation of >13 mm was invariably attributed to UO, but the majority of the obstructed kidneys had an RP dilation of <13 mm and there was overlap with other renal diseases.^16,17

Renal diverticula and ureters in healthy dogs and cats are barely distended by urine flow and therefore not normally visible during ultrasound examination. ²⁰ The luminal diameter of a non-obstructed distal feline ureter has been reported to be approximately 0.4 mm; however, how this measurement was obtained is difficult to verify. ²¹ The ureters of 10 healthy control cats were evaluated ultrasonographically and showed no measurable ureteral dilation. ¹⁷ As with RP dilation, the degree of ureteral dilation may also overlap among cats with UO and other renal diseases such as pyelonephritis.^16,17

The authors have observed several cases of UO in cats with minimal RP dilation and distended ureters. Previous studies in cats with RP dilation as small as 2 mm, and studies in humans, also reported this observation.^2,16,22,23 Although the presence of RP dilation and dilated pelvic diverticula is an important ultrasonographic sign of UO, it may fail to occur or occur later in the disease process. Absence of significant RP dilation in feline UO may lead to a misdiagnosis and have important negative repercussions on the patient if left untreated. The degree of RP dilation secondary to UO may reflect the degree and duration of outflow obstruction, urine output and renal tissue compliance against increased pressure within the kidney.

The main objective of this study was to describe RP and ureteral ultrasonographic measurements in a population of cats with confirmed benign UO by antegrade pyelography. The secondary objective was to further describe clinical findings associated with minimally dilated obstructed kidneys in an attempt to better understand its occurrence.

Materials and methods

Case selection

Medical records of cats treated by subcutaneous ureteral bypass (SUB) device placement for partial or complete unilateral or bilateral benign UO at the Centre Hospitalier Universitaire Vétérinaire of the University of Montreal, from January 2012 to December 2018 were reviewed.

Cats were included in the study if they had an initial diagnosis of benign UO on abdominal ultrasound based on the concurrent findings of RP dilation and ureteral dilation proximal to the location of obstruction and subsequently confirmed via antegrade pyelography. A fluoroscopic-guided antegrade pyelogram was performed intraoperatively by one of the authors (MD or CV) before each SUB device placement to confirm the UO. After access to the RP was achieved with an 18 G over-the-needle catheter, a urine sample was submitted for aerobic bacterial culture and antimicrobial susceptibility testing. Fluoroscopic-guided antegrade pyelography was performed by injecting sterile saline (0.9% NaCl) mixed with iohexol (in equal proportions) into the RP so as to achieve RP distension and ureteral filling. If a complete or partial UO was confirmed, a SUB device was placed in the obstructed kidney. A complete UO was defined as RP and ureteral dilation up to an abrupt termination of the contrast medium in the ureter associated with a complete lack of contrast medium distally. A partial UO was defined as RP and ureteral dilation up to an abrupt decrease in ureteral diameter associated with a delayed transit of contrast medium into the urinary bladder.

To meet the inclusion criteria, an abdominal ultrasound examination had to be performed by a board-certified radiologist prior to antegrade pyelography. Ultrasound images were retrospectively reviewed by a board-certified radiologist when specific measurements of the RP in transverse plane and/or ureter were not reported on the ultrasound report. Presence of pelvic diverticula dilation, defined by visible pelvic diverticula, were also reported on reviewed ultrasound images. For patients that underwent multiple abdominal ultrasound examinations prior to surgery, the images acquired immediately prior to surgery were selected for the study. When several measurements of the RP in transverse plane were recorded during the same ultrasound examination, mean RP size was calculated.

Each obstructed kidney was divided into two groups: group 1 included small RP measurements of ⩽4 mm and group 2 included larger RP measurements of >4 mm. This cut-off was selected by the authors based on RP width variations that can be seen in cats with normal renal function in order to better describe a population of cats with benign UO associated with minimally dilated RP and to simplify comparison with other variables. ¹⁵

Cats were excluded from the study if malignant UO was diagnosed, if no ultrasound examination was performed by a board-certified radiologist prior to antegrade pyelography or if ultrasound measurements of RP and ureteral diameters were not available at the time of the study.

Medical records review

For each cat enrolled in the study, data extracted from the medical record included signalment, body weight, side of UO, urinary tract ultrasonographic findings including RP size in transverse plane and ureteral diameter of the obstructed kidney, suspected cause of UO, presence of dilated pelvic diverticula, serum creatinine (SCr) concentration upon arrival and at discharge date, urine culture results and method of urine collection (intraoperative pyelocentesis and/or preoperative cystocentesis) and if antibiotics were given <24 h or >24 h before urine collection were all recorded.

SCr concentration at presentation to the hospital was measured by a blood gas analyzer with a limit of detection of 1326 µmol/l. A cat was considered azotemic if SCr concentration was ⩾140 µmol/l as recommended by the International Renal Interest Society. ²⁴

Statistical analysis

In cases of bilateral UO, kidneys and associated ureters were individually analyzed.

SCr concentration for cats with bilateral UO presenting both a small RP measurement ⩽4 mm and a larger RP measurement >4 mm was not taken into account during comparison of this parameter between both groups, given the simultaneous affiliation with the two groups. Only cats with unilateral UO were considered when evaluating the relationship between RP size or ureteral diameter and SCr concentration. SCr concentration of cats euthanized before discharge were not taken into account when evaluating SCr concentration at discharge.

Data for RP size, ureteral diameter and SCr concentration were log10-transformed to normalize the distributions. An Anderson–Darling normality test was used. Variation in RP size or ureteral diameter in relation to several independent variables including group, patient’s age, cause of UO, pelvic diverticula dilation and urine culture result obtained by pyelocentesis were examined with a linear mixed model with animal identification as a random effect adjusting for unequal variances when needed. A linear mixed model was also used to examine the relationship between RP size and ureteral diameter.

A Pearson’s correlation test was used to evaluate the relationship between RP size or ureteral diameter and SCr concentration of unilaterally obstructed kidneys at presentation and at discharge.

The Cochran–Mantel–Haenszel χ² test was used to assess the difference of distribution of identified causes of UO and concurrent pelvic diverticula dilation between groups 1 and 2.

All analyses were performed with statistical software (SAS, version 9.3), and values of P <0.05 were considered statistically significant.

Results

Cats

A total of 82 cats were enrolled in the study, including 32 (39%) cats with bilateral UO and 50 (61%) cats with unilateral UO: 29 (58%) were left sided and 21 (42%) were right sided. Five cats were excluded from the study; one had a malignant UO at the bladder trigone, two were operated on after hours and ultrasound was performed by an internist and two had nephroliths completely filling the RP, preventing appropriate measurements. SUBs were placed in all cats.

The population consisted of 44 (54%) spayed female cats and 38 (46%) castrated male cats with a median age of 9.0 years (range 2.5–17.2 years) and body weight of 4.06 kg (range 2.15–9.25 kg). There were 66 domestic shorthair cats, six Siamese, two Abyssinian and one each of the following: Himalayan, Cornish Rex, Bengal, Balinese, Burmese, Persian, Tonkinese and Birman.

In the final study analysis, there was a total of 114 obstructed kidneys and ureters. Thirty (26%) kidneys were included in group 1 and presented with a small RP measurement ⩽4 mm and 84 (74%) were included in group 2 and presented with a larger RP measurement >4 mm. A total of nine (8%) obstructed kidneys had an RP measurement ⩽2 mm.

Ultrasound findings

The median RP and ureteral diameters in the study population were 6.6 mm (range 1.1–37.0 mm) and 3.2 mm (range 0.0–11.0 mm), respectively. The median ureteral diameter in group 1 and 2 were 2.5 mm (range 1.0–11.0 mm) and 3.8 mm (range 0.0–11.0 mm), respectively. When all obstructed kidneys and ureters were considered, a statistically significant positive relationship was identified between RP and ureteral diameters (R ²  = 58.9%, P <0.0001). However, no significant difference was found between these two variables in group 1 when analyzed separately (P = 0.47). Information on renal diverticula (dilated or not) was reported in 67% and 71% of obstructed kidneys in groups 1 and 2, respectively. Among them, pelvic diverticula dilation was present in 65% and 92% of obstructed kidneys in groups 1 and 2, respectively. Mean RP size (95% confidence interval [CI]) with and without concurrent pelvic diverticula dilation were 8.3 mm (6.6–10.3 mm) and 3.5 mm (2.6–4.8 mm), respectively. RP size was significantly larger if concurrent pelvic diverticula dilation was present (P <0.0001). Ureteral diameter was not associated with pelvic diverticula dilation (P = 0.27). There was no significant association between RP size (P = 0.65) or ureteral diameter (P = 0.37) and the patient’s age.

Of the 114 obstructed ureters, 80 (70%) were associated with ureterolithiasis based on ultrasound examination, which included 18 (60%) ureters in group 1 and 62 (74%) ureters in group 2. The prevalence of UO caused by ureterolithiasis was not statistically different between the two groups (P = 0.56). For the remaining 34 obstructed ureters, a specific etiology could not be identified on ultrasound. A ureteral stricture or ureteritis was suspected in 31 obstructed ureters. No etiology was identified in the remaining three obstructed ureters as their entire course could not be followed during the ultrasound examination. Mean RP size (95% CI) associated and not associated with UO caused by ureterolithiasis were 7.6 mm (6.3–9.2 mm) and 5.2 mm (3.9–6.9 mm), respectively. RP size was significantly larger in kidneys with UO caused by ureterolithiasis compared with those without ureterolithiasis (P = 0.023). Ureteral diameter was not associated with the occurrence of ureterolithiasis (P = 0.68).

Creatinine values

The median SCr concentration of all cats with a UO was 444 µmol/l (range 108–1326 μmol/l) preoperatively and 184 µmol/l (range 85–780 μmol/l) the day of discharge. The mean SCr concentration at discharge (95% CI) did not differ between group 1 (210 µmol/l [156–282 μmol/l]) and group 2 (197 µmol/l [173–226 μmol/l]) (P = 0.71). Five cats did not survive to discharge and were euthanized. Among these, four had a bilateral UO and one had a unilateral UO. Ten cats with bilateral UO (31%) were excluded from SCr concentration statistical analysis between the two groups because they were presenting both a small and a larger RP. In group 1, the median preoperative SCr concentration was 923 µmol/l (range 141–1326 μmol/l) and the mean preoperative SCr concentration (95% CI) did not differ between cats with bilateral (602 µmol/l [245–1326 μmol/l]) and unilateral UO (932 µmol/l [699–1243 μmol/l]) (P = 0.33). In group 2, the median preoperative SCr concentration was 361 µmol/l (range 108–1326 μmol/l) and the mean preoperative SCr concentration (95% CI) was higher in bilateral UO (607 µmol/l [414–892 μmol/l]) than in unilateral UO (349 µmol/l [277–440 μmol/l]) (P = 0.017). The mean (95% CI) preoperative SCr concentration was significantly higher in group 1 (762 µmol/l [498–1165 μmol/l]) than in group 2 (409 µmol/l [333–502 μmol/l]) (P = 0.01). RP size in the two groups correlated negatively with preoperative SCr concentration (Pearson’s r = −0.52, P = 0.0002) but not with SCr concentration at discharge (Pearson’s r = −0.13, P = 0.38). There was no significant correlation between ureteral diameter and preoperative SCr concentration (Pearson’s r = −0.21, P = 0.16) or SCr concentration at discharge (Pearson’s r = −0.10, P = 0.50).

From the 81 cats for which an SCr concentration was recorded, 78 (96%) were azotemic at admission and 64 (79%) were azotemic on the day of discharge. The three cats without azotemia at admission were included in group 2. Two had a unilateral UO and one had a bilateral UO.

Urine culture and sensitivity

Urine culture results from preoperative cystocentesis and intraoperative pyelocentesis were available for 48 (59%) and 81 (99%) cats, respectively. A total of 42 (51%) cats received antibiotics preoperatively: <24 h prior to surgery in 12 (29%) cats and >24 h prior to surgery in 30 (71%) cats. Antibiotics used include beta-lactamases alone, fluoroquinolones alone or given together with beta-lactamases in 12 (28%), 20 (48%) and 10 (24%) cats, respectively. Among available pyelocentesis and cystocentesis urine samples, seven (9%) and 13 (27%) cultured positive, respectively. Seventeen of 82 (21%) cats had a positive urine culture. Escherichia coli, Staphylococcus species and Enterococcus faecalis were isolated in 12 (70%), three (18%) and two (12%) cultures, respectively.

Results from both pyelocentesis and cystocentesis urine cultures were available in 47 (57%) cats and were in agreement in 37 (79%) cats (34 negative cultures and three positive cultures) and in disagreement in 10 (21%) cats. Nine urine cultures obtained by cystocentesis preoperatively were positive while intraoperative urine cultures obtained by pyelocentesis were negative. Among these patients, eight received antibiotics between the time of cystocentesis and pyelocentesis and one did not receive any antibiotic therapy. One urine culture obtained by pyelocentesis intraoperatively was positive (E coli) while preoperative urine culture obtained by cystocentesis was negative.

The mean size (95% CI) of the RP was significantly larger in obstructed kidneys with positive urine culture obtained by pyelocentesis (13.9 mm [7.8–24.9 mm]) compared with those with negative urine culture (6.4 mm [5.5–7.5 mm]) (P = 0.013). Ureteral diameter was not associated with urine culture results (P = 0.22).

Discussion

This study revealed a large group of cats with confirmed benign UO associated with minimal RP dilation. Twenty-six percent of kidneys had an RP diameter <4 mm and 8% of kidneys had an RP diameter <2 mm. Considering these results, UO, regardless of its severity, may be associated with minimal RP dilation that may not differ from unobstructed cats or cats with other renal-related diseases.

Ultrasonographic measurement of the RP alone may fail to correctly identify obstructed kidneys in cats and therefore a UO cannot be excluded based on the absence of significant RP dilation. Previous studies have reported this finding.^2,17 Ultrasonographic ureteral dilation may help better assess UO in cats; however, absence of dilation of the ureter or a minimally dilated ureter may also be associated with UO, especially if the obstruction is at the ureteropelvic junction. The presence of ureteritis and/or fibrosis could also limit its distension. In cats with suspected UO, RP and ureteral diameters should be evaluated by ultrasound and a search for the cause of obstruction should be undertaken. In cases with equivocal ultrasound changes compatible with UO, antegrade pyelography may be considered to confirm or exclude an obstruction. Antegrade pyelography is considered the most accurate test to diagnose and identify the site of UO.^9,25 In a retrospective study of 49 cats with a tentative diagnosis of UO based on ultrasonography undergoing pyelography, only 71% of the 65 kidneys subjected to pyelography were shown to have UO. ¹⁵ In another recent study of 33 kidneys from cats and dogs that underwent antegrade pyelography, 42% had a complete UO, 27% had a partial UO and 21% had functional UO. ²⁶

RP and ureteral diameters in cats with UO in this study were slightly lower than those reported previously. Previous studies reported a median maximal RP width of 11.5 mm (range 2.0–29.0 mm, n = 79), 9.2 mm (range 2.0–35.5 mm, n = 186), 9.0 mm (range 3.0–30.0 mm, n = 46), 6.8 mm (range 1.2–39.1 mm, n = 17) and 10.4 mm (range 3.1–19.3 mm, n = 11) and maximal median ureteral diameter of 3.5 mm (range 1.0–11.0 mm, n= 79), 3.3 mm (range 0.0–10.0 mm, n = 172) and 3.7 mm (range 0.0–14.0 mm, n = 11) in cats with UO.^2,15–17,22 The lower RP and ureteral diameters in our study may be explained by a tendency to identify and treat UO more aggressively at our institution or reflect a different population of cats with UO.

As reported previously, most of the UOs in our feline population were caused by ureterolithiasis. Other causes of UO were difficult to confirm on ultrasound examination. Considering the poor sensitivity of ultrasound in identifying ureteral stricture, many cats were suspected to have a stricture. ⁷ RP size was significantly larger in kidneys with UO caused by ureterolithiasis compared with those without ureterolithiasis. Other causes of UO less likely identified on ultrasound, such as ureteral stricture, ureteritis, pyelonephritis, blood clots or dried solidified blood calculi, may therefore cause a lesser degree of obstruction. The degree of UO and ability to identify strictures on pyelography in comparison to ultrasound could not be determined in this retrospective study. Cats with a positive urine culture obtained by pyelocentesis had significantly larger RP diameters suggesting that pyelonephritis may have contributed to the UO; however, these cats all had ureterolithiasis identified during ultrasonography.

Despite the small RP, SUBs were successfully placed in all patients (both nephrostomy and cystostomy catheters). In patients with a small RP preventing formation of the nephrostomy catheter’s loop, nephrostomy tubes were placed extending into the ureter and not looped.

In previous studies, bacteriuria was identified in 25–33% of cases of feline UO.^2,22 These studies were based on the microscopic examination of urine sediment or urine culture results by either cystocentesis or pyelocentesis. The lower prevalence of positive urine cultures in the present study (21%) may be explained by our diagnostic criteria being based solely on urine culture from urine samples obtained by pyelocentesis and cystocentesis and the high prevalence of preoperative antibiotic therapy, often prescribed by the primary veterinarian. The prevalence of urinary tract infection in the present study could be underestimated given that preoperative urine culture results were not available in 41% of cats; 51% of cats received a preoperative antibiotic and eight cats had a positive urine culture by cystocentesis and a negative urine culture by pyelocentesis after receiving antibiotics. Bacterial isolates in this study are similar to others previously described with E coli being the most frequently isolated followed by Gram-positive cocci. Urinary tract infection seems to be rather uncommon in cats with UO compared with dogs in which over half of patients had a positive urine culture.^27,28 Although 79% of urine culture results by cystocentesis appeared to be consistent with those by pyelocentesis, this result may be underestimated from the high prevalence of antibiotics prescribed preoperatively. Urine cultures obtained by cystocentesis were similar to urine cultures obtained by pyelocentesis. Nevertheless, the authors recommend a urine culture obtained by pyelocentesis when performing an antegrade pyelography or during intraoperative SUB device placement in feline UO. One cat in this study had a positive intraoperative urine culture obtained by pyelocentesis while preoperative urine culture obtained by cystocentesis was negative. The preoperative urine sample obtained by cystocentesis was taken 6 days prior to surgery; therefore, a bacterial infection could have developed during this time.

Many factors may affect the degree of RP dilation in feline UO. The degree of hydronephrosis secondary to UO may reflect the degree and duration of outflow obstruction, urine output and renal tissue compliance against increased pressure within the collecting system. Cats with partial UO may have a small RP initially if the condition is acute or if they are volume depleted during ultrasound examination. RP dilation may be enhanced with administration of high rates of fluids or diuretics.^16,18,19,29 It is now well known that cats with CKD have a certain degree of interstitial inflammation and fibrosis and tubular atrophy that are more severe with advanced disease and positively correlated with the severity of azotemia.^30,31 Renal parenchymal and/or capsular fibrosis associated with CKD could limit RP dilation and/or renomegaly in an obstructed kidney owing to decreased compliance of fibrotic tissue despite increased pressure within the renal urine collection system. ³² However, to the authors’ knowledge, this theory is not proven in veterinary medicine and further prospective studies are needed to explore this hypothesis. Performing renal biopsies at the time of surgery or renal histology following death may aid in determining the underlying process preventing RP dilation. However, as most patients are severely azotemic at the time of surgery, this is not routinely pursued.

CKD is a frequent finding in cats with UO. Many of these cats have ultrasonographic evidence of CKD at admission.^4,5,15,33 Of the 46 kidneys with proven UO following pyelography in a study of azotemic cats, 39% had ultrasonographic signs compatible with pre-existing CKD including irregular kidney shape, cortical scars, reduced kidney size and parenchymal calcification. ¹⁵ In another study of 47 cats with ureterolithiasis, 94% of the affected cats had ultrasonographic evidence of chronic changes associated with kidney disease. ³³ Of the 117 cats that underwent ureteral surgery or ureteral stent placement in one study, 73% had evidence of chronic renal changes during ultrasound examination. ⁵ In a study of 41 cats with UO treated with a ureteral stent or a SUB device, 23 cats (56%) had a known history of CKD of at least 6 months. ³

A positive association between urolithiasis and feline CKD has been reported. ³⁴ This study showed a significantly higher prevalence of CKD among cats with upper urinary tract urolithiasis (33/59 [56%]) than cats without urolithiasis (20/67 [30%]). However, as no causality between urolithiases and CKD was assessed during the study, the authors conclude that it remains unclear whether urolithiasis is a predictive factor for, or a consequence of, CKD. ³⁴ Similar associations have also been identified in human medicine where the presence of urolithiasis, nephrolithiasis in particular, has been associated with a two-fold higher risk of development of CKD independent of other known CKD risk factors.^35–37 The relationship between CKD and urolithiasis in veterinary medicine remains unclear and requires further investigations. Although the exact mechanism of urolithiasis-associated kidney damage is yet to be elucidated, several pathways have been proposed, including intermittent upper urinary tract obstruction and infection, renal parenchymal mineral deposition, parenchymal inflammation and fibrosis and renal injury from urolithiasis treatments. ³⁵ UO in rodents is considered the most widely used model to study the mechanisms of tubulointerstitial fibrosis due to its rapid development together with tubular atrophy. ³⁸ Alternatively, CKD could play a role in urolithiasis formation through abnormal handling of minerals and metabolites. ³⁴

This study, along with previous retrospective studies, has shown that the vast majority of cats with benign UO are azotemic at admission regardless of whether the obstruction is unilateral or bilateral.^2,22 Another interesting finding is that cats in group 1 had a significantly higher SCr concentration at admission than cats in group 2 and that RP size correlated negatively with preoperative SCr concentration. This finding could be the result of more advanced CKD in group 1 with greater fibrosis and less tissue compliance in these patients. However, more controlled investigations are needed to explore this theory. Several parameters not controlled in this study could also have influenced RP and SCr concentration measurements. There was, however, no significant correlation between RP size and SCr concentration at discharge. Explanations for this finding could be attributed to the small number of cats evaluated and the overall improved azotemia following resolution of UO, resulting in less SCr concentration variation between cats.

There was an absence of correlation between the diameter of the RP and ureter in group 1 when analysed separately compared with all kidneys. This finding is well illustrated in Figure 1, where minimal RP dilation is associated with severe hydroureter secondary to obstructive ureterolithiasis. This could be explained by the presence of renal parenchymal and capsular fibrosis with variable fibrosis and inflammation of the ureter, thus limiting the expansion of the RP and/or the ureter. Ureteral fibrosis secondary to UO has been shown in rats after iatrogenic ureteral ligation. ³⁹ A similar process could occur in cats, therefore preventing ureteral dilation over time. While RP size was generally larger with concurrent pelvic diverticula dilation, kidneys in group 1 often had dilated pelvic diverticula. Limited RP expansion owing to fibrosis with sustained pressure could result in dilated pelvic diverticula.

Figure 1

Ultrasound images of the left kidney and ureter (a,b) and lateral radiograph of the abdomen (c) of a cat with benign ureteral obstruction prior antegrade pyelography. Images (a,b) show minimal renal pelvis dilation (1.2 mm) and a proximal ureteral dilation of 1.7 mm. As the ureter continues caudally, there is an increase in its diameter up to 7.5 mm in its caudal portion up to a large ureterolith (measuring 7.7 mm in length). Image (c) shows large ureteroliths in the caudal ureter (dashed arrow) and several smaller ureteroliths throughout its course to the kidney (arrows) where nephroliths are also visible (arrowheads)

This study has a number of limitations, many of which are related to its retrospective nature. Lack of standardization of ultrasound examinations and inter-radiologist variation could have affected RP and ureteral measurements. Lack of prior information regarding the degree of UO (complete or partial) is another important limitation of the present study. UO in the population of this study were all confirmed by one of the authors (MD or CV) prior to SUB device placement based on the criteria outlined in the ‘Materials and methods’ section. Data regarding pelvic diverticula were only available in approximatively 70% of ultrasound reports in both groups, which could have contributed in overestimating the number of dilated pelvic diverticula, especially in the first group where absence of pelvic diverticula dilation may not have been mentioned. Information regarding patient hydration and prior medical history, the amount of fluids administered prior to ultrasound examination, function of the contralateral kidney with unilateral UO and the duration of obstruction were unable to be taken into account due to the retrospective nature of this study and may have influenced RP or SCr concentration measurements and evaluation. The smaller number of cats in group 1 may be another limitation in this study. A larger population of cats may have yielded a different median SCr concentration and therefore influenced comparison between unilateral and bilateral UO in group 1 and between both groups. Finally, the analyzer used to measure SCr concentration had a maximum detection limit of 1326 µmol/l, which limited the analysis of this parameter. Given these uncontrolled variables, results regarding RP diameter and SCr concentration should be assessed cautiously.

Conclusions

The results of our study show that feline UO may be associated with minimal RP dilation and the severity of RP and ureteral dilation can be highly variable. Absence of significant RP dilation does not rule out UO in cats and attention to ureteral diameter and further examinations, such as antegrade pyelography, may be considered to confirm a UO in cases with equivocal ultrasonographic changes. Early recognition and decompression are crucial in these patients in order to preserve renal function as much as possible. Further controlled prospective studies are needed to evaluate the pathophysiology behind minimally RP dilation in feline UO.