Postobstructive diuresis is defined as polyuria after relief of a urinary tract obstruction1,2 and can occur following relief of obstruction from intraluminal, extraluminal, and neurogenic causes at any location along the urinary tract.3 Ureteral obstruction in cats may be caused by uroliths (including solidified blood calculi), tumors, strictures, blood clots, and iatrogenic ligation.4–7 In veterinary medicine, newer techniques for the treatment of urinary tract obstructions have become available, increasing the number of affected animals that may be treated successfully. As a result, better understanding is needed for how to provide optimal perioperative management for animals with an obstructed urinary tract as well as for managing POD in treated animals.
Postobstructive diuresis is clinically important because it can lead to dehydration and profound electrolyte and water distribution disturbances if not managed appropriately.8 The incidence and clinical outcome of POD in veterinary species are not well described; however, in a study9 on 28 cats with urethral obstruction, POD was defined as a urine output > 2 mL/kg/h (0.91 mL/lb/h), and 20 of those 28 (71%) cats developed POD. In addition to showing that cats were more likely to develop POD if they were acidemic at initial evaluation, investigators found that diuresis began ≤ 6 hours after relief of obstruction and continued for up to 4 days in some cats.9 Further, the same definition of POD was used in a study10 that showed that 50 of 57 (88%) cats developed POD ≤ 48 hours following relief of urethral obstruction.
A description of POD after renal decompressive surgery in companion animals is lacking. The primary objectives of the study presented here were to describe POD in cats undergoing surgical placement of ureteral stents or SUBs for ureteral obstruction and to identify factors associated with duration and maximum severity of POD in cats.
Materials and Methods
Case selection criteria
The medical records of cats with unilateral or bilateral ureteral obstruction diagnosed between August 2010 and December 2014 at the UCD-VMTH were evaluated. Cats were included if they underwent renal decompressive surgery with placement of a ureteral stenta or an SUBb; their ureteral obstructions were secondary to urinary stones or precursors, neoplasia, or strictures; and they were confirmed to have developed POD.
Procedures
Data collected from medical records included signalment, body weight, BCS (on a scale of 1 to 9, where 1 = cachectic, 4 to 5 = ideal, and 9 = grossly obese), hydration status (as assessed by skin turgor, mucous membrane moisture, presence of chemosis, serous ocular or nasal discharge, and precorneal tear film quality), body temperature (rectal or aural), heart rate, respiratory rate, reason for assessment, results of preoperative serum biochemical and acid-base analyses performed ≤ 24 hours before surgery, and results of urinalysis and bacterial culture of urine samples. All serum biochemical analyses were performed at the UCD-VMTH Clinical Diagnostic Laboratory, and results of serum biochemical and acid-base analyses performed ≤ 24 hours before surgery were chosen as baseline values.
All cats underwent abdominal ultrasonography performed by a board-certified radiologist. The presence of CKD (eg, exemplified by irregular kidney borders, decreased kidney size, cortical cysts and infarcts, or decreased corticomedullary distinction consistent with fibrosis) and unilateral or bilateral obstruction (eg, exemplified by pyelectasis, hydronephrosis, ureteral dilation, or evidence of obstructive ureteral calculi) were determined by ultrasonography. Positive-contrast antegrade pyelography was performed to confirm ureteral obstructions.
All renal decompressive surgeries were performed by, or under the direction of, a single surgeon (WTNC) and with fluoroscopic guidance. Data regarding the type of surgery performed (stent vs SUB and unilateral vs bilateral); duration of anesthesia; fluid therapy before, during, and after anesthesia; administration of medications known to affect urine output (eg, furosemide, mannitol, dexmedetomidine, angiotensin-converting enzyme inhibitors, and dextrose); and urine output were recorded. Cats were grouped by whether they underwent unilateral or bilateral surgery, then were further subgrouped according to whether a stent or SUB was placed (ie, unilateral stent, bilateral stent, unilateral SUB, or bilateral SUB groups).
For cats in which intraperitoneal drains were placed, the quantities of fluid collected through the drains were recorded, as were the results of biochemical analyses performed on the fluids collected. In addition, serum and abdominal fluid concentrations of creatine and potassium were compared to determine the presence of a uroabdomen. Results of postoperative serum biochemical and acid-base analyses as well as hospitalization outcomes (eg, hours in ICU, duration of hospitalization, resolution of azotemia, and survival to discharge) were also recorded.
Maximum severity of POD was determined for each cat as its greatest urinary output (mL/kg/h) on the basis of urine volume collected in a closed urinary collection system over a 6-hour period. To maintain consistency with existing veterinary literature, duration of POD was defined as the length of time that urine output was > 2 mL/kg/h while a urinary catheter was in place. For the purposes of the study and because of the inherent difficulty of accurately quantifying urine output in the absence of a urinary catheter, POD was considered to have ended at the time of urinary catheter removal.
Statistical analysis
Continuous variables (eg, body weight, body temperature, heart rate, respiratory rate, and clinical laboratory findings) were assessed for normality with the Shapiro-Wilk test. Variables that did not have a normal distribution were reported as median and IQR, whereas normally distributed variables were reported as mean ± SD. The main outcomes of interest were maximum severity of POD (defined as the highest recorded urine output [mL/kg/h]) and duration of POD (defined as hours with urine output > 2 mL/kg/h). Independent t tests or Wilcoxon rank sum tests were used to compare duration and maximum severity of POD between cats with and without anorexia, lethargy, vomiting, anuria, signs of abdominal pain, polydipsia, pollakiuria, hematuria, polyuria, or stranguria and between cats with and without previous diagnoses of CKD or AKI. In addition, the Wilcoxon rank sum test was used to compare duration and maximum severity of POD between cats that did and did not have resolution of azotemia while hospitalized and between cats that did or did not survive to hospital discharge. The duration and maximum severity of POD were also compared across predefined surgery groups (unilateral vs bilateral) with independent t tests and across predefined surgery subgroups (unilateral stent vs bilateral stent vs unilateral SUB vs bilateral SUB) with the Kruskal-Wallis test. Categorical variables (eg, type of surgery, medications received, and survival to hospital discharge) were reported as percentages of the total. The Fisher exact test (when the expected count in any cell was < 5) or χ2 test was used to compare whether the likelihood of survival to hospital discharge differed between surgery groups (unilateral vs bilateral) and subgroups (stent vs SUB). Spearman rank-order correlation analyses were performed to assess for linear relationships between continuously distributed variables (including age, body weight, duration of signs, preoperative PCV and urine specific gravity, duration in the ICU, duration hospitalized after surgery, and results of preoperative biochemical and blood gas analyses) and POD duration and maximum severity. Univariate analyses were performed; for all comparisons, values of P < 0.05 were considered significant, and all tests were 2 sided. For cats that had multiple surgeries for multiple episodes of ureteral obstruction, only data from the first episode of obstruction treated at the UCD-VMTH were included in the statistical analysis. All analyses were performed with available software.c
Results
Animals
A search of the UCD-VMTH medical records identified 37 cats with ureteral obstruction treated between August 2010 and December 2014 that met the case inclusion criteria. Twenty-five of the 37 (68%) cats were castrated males, and 12 (32%) were spayed females. Cats were recorded as domestic shorthair (n = 25/37 [68%]), domestic medium-hair (5/37 [14%]), and domestic longhair (4/37 [11%]) cats and 1 (3%) each of a Siamese, Sphinx, and Ocicat. Median age was 118 months (IQR, 93 to 145 months; range, 5 to 204 months). Age did not correlate significantly with duration (ρ = 0.009, P = 0.959) or maximum severity (ρ = −0.281, P = 0.092) of POD.
Patient history, physical examination, and clinical signs
Before evaluation at the UCD-VMTH, the median duration for which owners perceived signs of illness in their cats was 9 days (IQR, 3 to 30 days); however, duration of signs did not correlate significantly with duration (ρ = 0.196, P = 0.245) or maximum severity (ρ = 0.119, P = 0.484) of POD. In addition, fluid therapy (IV or SC, alone or in combination) had been administered to 28 of the 37 (76%) cats at home or at a referring veterinary hospital, and hydration status on initial examination at the UCD-VMTH was available for 36 cats. Five of the 36 (14%) cats (all 5 of which had been hospitalized previously) were noted to have been overhydrated, 20 (56%) cats (8 of which had been hospitalized previously) were adequately hydrated, and 11 (30%) cats (5 of which had been hospitalized previously) were dehydrated. Hydration status was not significantly (P = 0.384 and P = 0.086, respectively) associated with duration or maximum severity of POD.
Cats in which CKD had been diagnosed previously (n = 21/37 [57%]) had significantly (P = 0.040) less maximum severity of POD than did cats in which CKD had not been diagnosed previously (16 [43%]); however, a previous diagnosis of CKD was not significantly (P = 0.683) associated with duration of POD (Table 1). Neither the duration nor the maximum severity of POD differed significantly (P = 0.627 and P = 0.944, respectively) for cats in which AKI had been diagnosed previously (n = 9/37 [24%]), compared with cats in which AKI had not been diagnosed previously (28 [76%]).
Results of independent t test analysis to identify associations between previous diagnosis of CKD or AKI and subsequent duration and maximum severity of POD in 37 client-owned cats surgically treated with ureteral stent or SUB placement for ureteral obstruction.
| POD maximum severity* | POD duration | ||||
|---|---|---|---|---|---|
| Diagnosis | No. (%) of cats | Mean ± SD (mL/kg/h) | P value | Mean ± SD (h) | P value |
| CKD previously diagnosed | 0.040 | 0.683 | |||
| Yes | 21 (57) | 7.4 ± 4.3 | 55.4 ± 24.4 | ||
| No | 16 (43) | 11.2 ± 7.1 | 58.9 ± 26.3 | ||
| AKI previously diagnosed | 0.944 | 0.627 | |||
| Yes | 9 (24) | 8.7 ± 5.2 | 53.3 ± 23.3 | ||
| No | 28 (76) | 9.1 ± 6.3 | 58.1 ± 25.8 | ||
Values of P ≤ 0.05 were considered significant.
Greatest urinary output.
Body condition scores were noted in the medical records for 36 cats, of which 6 (17%) cats had BCSs of 2 or 3, 14 (39%) cats had BCSs of 4 or 5, and 16 (44%) cats had BCSs of 6 to 9. Median BCS was 5 (range, 2 to 9). Median body weight for all 37 cats was 4.8 kg (10.6 lb; IQR, 4.0 to 5.6 kg [8.8 to 12.3 lb]; range, 1.3 to 10.5 kg [2.9 to 23.1 lb]), and body weight correlated negatively and significantly (ρ = −0.369, P = 0.025) with maximum severity of POD in that cats with lower body weight had greater severity of POD. Body temperatures were within reference range (37.5° to 39.2°C [99.5° to 102.5°F]) for 26 of the 34 (76%) cats for which the parameter was recorded; however, 7 (21%) cats were hypothermic (< 37.5°C [99.5°F]), and 1 (3%) cat was hyperthermic (39.8°C [103.7°F]). Heart rates were within reference range (160 to 220 beats/min) for 32 of the 37 (86%) cats; however, 3 (8%) cats were bradycardic (< 160 beats/min), and 2 (5%) cats were tachycardic (> 220 beats/min). Additionally, 13 of the 37 (35%) cats were tachypneic (respiratory rate > 40 breaths/min; reference range, 16 to 40 breaths/min).
Of the clinical signs reported (Table 2), the presence of vomiting (n = 25/37 [68%]) or lethargy (24/37 [65%]) was significantly (P = 0.032 and P = 0.006, respectively) associated with greater maximum severity of POD. Anorexia (n = 27/37 [73%]), signs of abdominal pain (7/37 [19%]), pollakiuria (5/37 [14%]), polydipsia (4/37 [11%]), gross hematuria (3/37 [8%]), polyuria (1/37 [3%]), and stranguria [1/37 [3%]) were also noted; however, these signs were not associated with duration or maximum severity of POD. Cats that were anuric before surgery (n = 8/37 [22%]) had significantly (P = 0.018) longer durations of POD than did cats that were not anuric before surgery (29/37 [78%]); however, the maximum severity of POD was not significantly (P = 0.438) different between these 2 groupings. Of the 8 cats with anuria, 4 cats had bilateral ureteral obstructions, and 4 cats had unilateral ureteral obstructions.
Results of analyses with the independent t test or the Wilcoxon rank sum test to identify preoperative clinical signs associated with duration and maximum severity of POD for the 37 cats in Table 1.
| Clinical signs | No. (%) of cats | POD maximum severity P value | POD duration P value |
|---|---|---|---|
| Anorexia | 27 (73) | 0.141 | 0.438 |
| Vomiting | 25 (68) | 0.032 | 0.215 |
| Lethargy | 24 (65) | 0.006 | 0.588 |
| Anuria | 8 (22) | 0.438 | 0.018 |
| Signs of abdominal pain | 7 (19) | 0.121 | 0.300 |
| Pollakiuria | 5 (14) | 0.657 | 0.329 |
| Polydipsia | 4 (11) | 0.883 | 0.205 |
| Hematuria | 3 (8) | 0.824 | 0.939 |
| Polyuria | 1 (3) | — | — |
| Stranguria | 1 (3) | — | — |
Values of P ≤ 0.05 were considered significant.
— = Not calculated (too few animals for statistical analysis).
Preoperative clinicopathologic findings
Serum biochemical analyses results—Median serum creatinine concentration for cats before surgery was 6.1 mg/dL (IQR, 3.1 to 15.1 mg/dL; reference range, 1.1 to 2.2 mg/dL), and 34 of the 37 (92%) cats had serum creatinine concentrations > 2.2 mg/dL. Serum creatinine concentration correlated positively and significantly (ρ = 0.553, P < 0.001; and ρ = 0.430, P = 0.005; respectively) with duration and maximum severity of POD. Median BUN concentration before surgery was 84 mg/dL (IQR, 40 to 204 mg/dL; reference range, 18 to 33 mg/dL), and 31 of the 37 (84%) cats had a BUN concentration > 33 mg/dL. Concentration of BUN correlated positively and significantly (ρ = 0.546, P < 0.001; and ρ = 0.800, P = 0.004; respectively) with duration and maximum severity of POD. Preoperatively, 34 of the 37 (92%) cats were azotemic (creatinine > 2.2 mg/dL or BUN > 33 mg/dL). Median serum phosphorus concentration before surgery was 6.4 mg/dL (IQR, 4.6 to 12.0 mg/dL; reference range, 2.2 to 6.3 mg/dL). Twenty-two of the 37 (59%) cats had hyperphosphatemia, and 1 (3%) cat had hypophosphatemia. Serum phosphorus concentration correlated positively and significantly (ρ = 0.596, P = < 0.001; and ρ = 0.502, P < 0.001; respectively) with duration and maximum severity of POD. Mean ± SD serum sodium concentration before surgery was 151.0 ± 4.4 mmol/L (reference range, 151 to 158 mmol/L). Twelve of the 37 (32%) cats had hyponatremia, and 1 (3%) cat had hypernatremia. Serum sodium concentration correlated negatively and significantly (ρ = −0.334, P = 0.038) with maximum severity of POD. Median serum potassium concentration before surgery was 4.5 mmol/L (IQR, 4.0 to 5.6 mmol/L; reference range, 3.6 to 4.9 mmol/L). Sixteen of the 37 (43%) cats had hyperkalemia, and 2 (5%) cats had hypokalemia. Serum potassium concentration correlated positively and significantly (ρ = 0.438, P = 0.005; and ρ = 0.379, P = 0.017; respectively) with duration and maximum severity of POD. Mean ± SD serum albumin concentration before surgery was 3.0 ± 0.6 g/dL (reference range, 2.2 to 4.6 g/dL). Three of the 37 (8%) cats had hypoalbuminemia, whereas the albumin concentrations for the remaining 34 (92%) cats were within reference range. Serum albumin concentration before surgery was not predictive of POD duration or maximum severity.
Blood gas analyses—In 34 of the 37 (92%) cats, blood gas analyses were performed immediately before or after induction of anesthesia. Eighteen of the 34 (53%) samples collected were venous, and 16 (47%) samples were arterial. Mean ± SD blood pH was 7.3 ± 0.08 (reference range, 7.35 to 7.45), and mean ± SD standard base excess was −7.5 ± 7.5 mEq/L (reference range, −5.0 to 0 mEq/L). Standard base excess was < −5.0 mEq/L for 24 of the 34 (71%) cats and within the reference range for the remaining 10 (29%) cats tested. Mean ± SD serum bicarbonate concentration before surgery was 18.4 ± 3.5 mEq/L (reference range, 18 to 23 mEq/L). Serum bicarbonate concentration was < 18 mEq/L in 14 of the 34 (41%) cats, within reference range for 18 (53%) cats, and > 23 mEq/L in 2 (6%) cats. Preoperative serum pH, standard base excess, and bicarbonate concentration were not associated with duration or maximum severity of POD.
Urinalysis and bacterial culture of urine—Median urine specific gravity before surgery was 1.012 (IQR, 1.010 to 1.015; reference range, 1.035 to 1.060; n = 32). Urine specific gravity had a negative correlation (ρ = −0.467, P = 0.004) with duration of POD, but a positive correlation (ρ = 0.378, P = 0.021) with maximum severity of POD. Median urine pH was 6 (range, 5 to 8; reference range, 5 to 7.5; n = 30), and 28 of the 30 (93%) cats for which results were available had a urine pH within the reference range. The remaining 2 cats had a urine pH of 8. Bacterial culture was performed on urine samples obtained by cystocentesis (n = 30) or through a urinary catheter (4) from 34 of the 37 (92%) cats; however, only 3 of the 34 (9%) samples yielded bacterial growth. Escherichia coli was identified in all 3 samples, and a coinfection with Enterococcus faecalis was identified in 1 sample.
Surgical placement of stents or SUBs
During surgery, all cats received urethral catheters, 16 (43%) cats received intra-abdominal drains at the surgeon's discretion, 20 (54%) cats underwent unilateral ureteral stent placement, 7 (19%) cats underwent bilateral ureteral stent placement, 6 (16%) cats underwent unilateral SUB placement, and 4 (11%) cats underwent bilateral SUB placement. Neither the laterality (unilateral [n = 26] vs bilateral [11]) nor type (stent placement [27] vs SUB placement [10]) of surgeries was meaningfully associated with duration or maximum severity of POD, hours in the ICU, duration hospitalized after surgery, or survival to hospital discharge.
Median duration from initial evaluation at the UCD-VMTH to renal decompressive surgery was 42 hours (IQR, 24 to 70 hours; range, 5 to 201 hours). During this time, cats were managed medically to allow for spontaneous renal decompression or to stabilize cats for surgery. Median duration of anesthesia was 6.0 hours (IQR, 5.0 to 7.0 hours; range, 4.0 to 8.8 hours) and included time needed for induction of anesthesia, patient preparation, surgery, and obtainment of postoperative radiographs for all cats. Additionally, 34 of the 37 (92%) cats had placement of an esophageal feeding tube during the same anesthetic episode, which was included in the duration of anesthesia.
During the anesthetic procedures, 32 of the 37 (86%) cats received IV fluids (median fluid rate, 3.9 mL/kg/h [1.8 mL/lb/h]; IQR, 1.7 to 6.5 mL/kg/h [0.8 to 3.0 mL/lb/h]; range, 0 to 19.2 mL/kg/h [0 to 8.7 mL/lb/h]), and 2 of the 32 (6%) cats received < 1 mL/kg/h (0.45 mL/lb/h) of fluid therapy support during the anesthetic procedure. The most common type of IV fluid administered was lactated Ringer solution (n = 26/32 [81%]), followed by sterile saline (0.9% NaCl) solution (7/32 [22%]), 5% dextrose in water (2/32 [6%]), a buffered crystalloid solutiond (1/32 [3%]), and sterile water (1/32 [3%]). Five of the 32 (16%) cats received > 1 type of IV fluid during the anesthetic procedure.
Perioperative medications
Medications known to affect urine output that were administered to cats during the perioperative period (spanned from 12 hours before surgery to the time when collection of urine output data ceased after surgery) included mannitol (n = 17/37 [46%]), dexmedetomidine (15/37 [41%]), dextrose (13/37 [35%]), and furosemide (3/37 [8%]). No cats received an angiotensin-converting enzyme inhibitor during the perioperative period. The administration of these drugs was not meaningfully associated with duration or maximum severity of POD.
Postoperative urine output
Postoperative care was similar for all cats. Urinary catheters were maintained at the discretion of the attending clinician, provided that the catheters were tolerated by the cats. Because of risks associated with maintaining indwelling urinary catheters, catheters were removed as soon as it was reasonable to monitor hydration status by other means (eg, body weight measurements, estimated urine output, and serial physical examinations). Median durations of POD, ICU care, and hospitalization after surgery were 60 hours (IQR, 36 to 72 hours; range, 18 to 108 hours), 65 hours (IQR, 42 to 92 hours; range, 0 to 201 hours), and 6 days (IQR, 5 to 8 days; range, 2 to 15 days), respectively. Median maximum urine output was 7.6 mL/kg/h (3.5 mL/lb/h; IQR, 4.9 to 11.4 mL/kg/h [2.2 to 5.2 mL/lb/h]; range, 2.6 to 31.9 mL/kg/h [1.2 to 14.5 mL/lb/h]; reference range, 1 to 2 mL/kg/h). Median duration from end of surgery to greatest urine output was 12 hours (IQR, 6 to 24 hours; range, 6 to 60 hours), and median duration that urinary catheters were in place was 60 hours (IQR, 42 to 88 hours; range, 21 to 164 hours). At the point of urinary catheter removal, 4 of the 37 (11%) cats had urine output within reference range, whereas the remaining 33 (89%) cats still had high urine output (IQR, 3.1 to 5.4 mL/kg/h [1.4 to 2.5 mL/lb/h]; range, 2.5 to 33.1 mL/kg/h [1.1 to 15.1 mL/lb/h]). For the 16 cats that received an intra-abdominal drain, median duration that the drain was in place was 85 hours (IQR, 63 to 98 hours; range, 35 to 211 hours), and 1 cat was euthanized with the drain still in place. Samples of abdominal fluid collected through intra-abdominal drains were analyzed for 7 of the 16 cats, and samples from 4 cats contained creatinine or potassium concentrations consistent with uroabdomen.11 For each cat with documented uroabdomen, the volume of abdominal fluid removed through an intra-abdominal drain was included in the total urine output.
Duration of POD correlated positively and significantly (ρ = 0.684, P < 0.001; and ρ = 0.397, P = 0.012; respectively) with hours in the ICU and days hospitalized after surgery, but was not associated significantly (P = 0.504 and P = 0.094, respectively) with survival to hospital discharge or resolution of azotemia while hospitalized. Maximum severity of POD was not associated significantly (P = 0.525, P = 0.174, P = 0.504, and P = 0.558, respectively) with hours in the ICU, days hospitalized after surgery, survival to hospital discharge, or resolution of azotemia while hospitalized.
Hospitalization and survival to discharge
Thirty-four of the 37 (92%) cats survived to hospital discharge, and for these cats, the median durations for ICU care and for hospitalization after surgery were 63 hours (IQR, 42 to 91 hours; range, 0 to 201 hours) and 6 days (IQR, 6 to 8 days; range, 2 to 15 days), respectively. Median BUN concentration immediately before hospital discharge of the 34 surviving cats was 30 mg/dL (IQR, 23 to 47 mg/dL; range, 13 to 92 mg/dL), with 20 of the 34 (59%) cats having had BUN concentrations within reference range immediately before discharge. Median serum creatinine concentration immediately before hospital discharge was 2.2 mg/dL (IQR, 1.7 to 2.9 mg/dL; range, 0.6 to 7.9 mg/dL), with serum creatinine concentration < 2.2 mg/dL in 18 of the 34 (53%) cats. Azotemia had resolved during hospitalization for 15 of the 34 (44%) cats that had azotemia before surgery, 17 of the 34 (50%) surviving cats were still azotemic immediately before discharge, and all 3 cats that died or were euthanized were azotemic at last evaluation. Of the 21 cats in which CKD had been diagnosed previously, 7 (33%) had resolution of azotemia while hospitalized, whereas 12 (57%) did not, and 2 (10%) died or were euthanized while in the hospital.
Absolute changes in clinicopathologic results
When the absolute changes in clinicopathologic findings (findings ≤ 24 hours before surgery vs 1 day after surgery) were assessed, the maximum severity of POD had significant positive correlations with absolute changes in BUN, creatinine, and potassium concentrations, but significant negative correlations with sodium and bicarbonate concentrations as well as pH and standard base excess (Table 3). Additionally, duration of POD had significant positive correlations with absolute changes in BUN, creatinine, and potassium concentrations. None of the absolute changes in clinicopathologic findings from before surgery to 1 day after surgery were associated with survival to hospital discharge, duration in the ICU, or resolution of azotemia.
Results of Spearman ρ coefficient analysis to assess potential correlations between duration and maximum severity of POD and absolute changes in serum analytes from findings obtained before surgery versus findings obtained both 1 day after surgery and immediately before hospital discharge for cats in the previous tables.
| Serum analytes before surgery vs 1 day after surgery | Serum analytes before surgery vs immediately before hospital discharge | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| POD maximum severity | POD duration | POD maximum severity | POD duration | |||||||||
| Analyte | No. of cats | Absolute change: median (IQR) or mean ± SD | Spearman ρ | P value | Spearman ρ | P value | No. of cats | Absolute change: median (IQR) or mean ± SD | Spearman ρ | P value | ||
| BUN (mg/dL) | 37 | 14 (3 to 69) | 0.73 | < 0.001 | 0.331 | 0.032 | 34 | 49 (9 to 160) | 0.646 | 0.002 | 0.502 | < 0.001 |
| Creatinine (mg/dL) | 37 | 2 (0.7 to 6.1) | 0.63 | 0.002 | 0.352 | 0.022 | 34 | 2.9 (0.5 to 12.2) | 0.578 | 0.008 | 0.442 | 0.003 |
| Phosphorus (mg/dL) | 37 | 0.7 (−0.8 to 4.1) | 0.31 | 0.169 | 0.083 | 0.600 | 34 | 1.6 (0.5 to 7.8) | 0.489 | 0.029 | 0.601 | < 0.001 |
| Sodium (mmol/L) | 26 | −1.5 ± 5.7 | −0.64 | 0.002 | −0.314 | 0.118 | 34 | 0.0 6 ± 4.7 | −0.561 | 0.010 | −0.132 | 0.457 |
| Potassium (mmol/L) | 26 | 0.9 ± 1.3 | 0.60 | 0.004 | 0.433 | 0.027 | 34 | 0.25 (−0.6 to 1.8) | 0.257 | 0.275 | 0.347 | 0.030 |
| Albumin (g/dL) | 37 | 0.6 ± 0.4 | −0.37 | 0.098 | 0.042 | 0.805 | 34 | 0.40 ± 0.54 | −0.493 | 0.027 | −0.178 | 0.300 |
| pH | 36 | −0.0 6 ± 0.11 | −0.68 | < 0.001 | 0.139 | 0.449 | 0 | — | — | — | — | — |
| Bicarbonate (mEq/L) | 37 | −4.2 ± 5.1 | −0.80 | < 0.001 | −0.059 | 0.747 | 0 | — | — | — | — | — |
| Standard base excess (mEq/L) | 37 | −4.7 ± 5.7 | −0.82 | < 0.001 | −0.024 | 0.897 | 0 | — | — | — | — | — |
Values of P ≤ 0.05 were considered significant.
— = Analyte was not assessed immediately before hospital discharge.
When the absolute changes in clinicopathologic findings (findings ≤ 24 hours before surgery vs immediately before hospital discharge) were assessed, the maximum severity of POD had significant positive correlations with absolute changes in BUN, creatinine, and phosphorus concentrations, but significant negative correlations with sodium and albumin concentrations (Table 3). Similarly, the duration of POD had significant positive correlations with absolute changes in BUN, creatinine, phosphorus, and potassium concentrations. In addition, duration in the ICU correlated positively and significantly (ρ = 0.392, P = 0.012; and ρ = 0.498, P = 0.001; respectively) with absolute changes in BUN and phosphorus concentrations. None of the absolute changes in clinicopathologic findings for analytes evaluated before surgery and immediately before hospital discharge were meaningfully associated with survival to hospital discharge, duration of hospitalization after surgery, or resolution of azotemia.
Discussion
Results of the present study indicated that high BUN, creatinine, phosphorus, and potassium concentrations ≤ 24 hours before decompressive surgery in cats with ureteral obstruction were associated with longer and more severe POD following surgery. In cats with ureteral obstruction (unilateral or bilateral), many of these abnormal clinicopathologic findings could have indicated more severe or longer-standing disease in affected cats, compared with disease in cats without these abnormalities. Further, cats with these abnormalities could require more intense postoperative management owing to the greater likelihood of longer and more severe POD with accompanying electrolyte fluid balance abnormalities. In addition, preoperative anuria was associated with a longer duration of POD in cats of the present study, which was consistent with findings in humans for which high serum creatinine concentration and urine retention at initial evaluation are predictive of POD.12
In the cats of the present study, longer and more severe POD was associated with greater absolute changes in BUN, creatinine, and potassium concentrations from ≤ 24 hours before surgery to 1 day after surgery as well as with greater absolute changes in BUN and creatinine concentrations from ≤ 24 hours before surgery to immediately before discharge. There were several likely reasons for these findings. First, cats in the present study with higher BUN concentrations had higher solute burdens that needed to be excreted. As this osmotically active BUN concentration was excreted through the nephrons, ultrafiltrate was carried with it, and POD resulted. As discussed previously, the higher BUN, creatinine, and potassium concentrations were likely indicators for more profound and prolonged renal injury that resulted in more severe nephron and tubular damage in affected cats, compared with cats without these high concentrations. Damaged renal tubules are not able to properly function, including the tubular reabsorption necessary to concentrate urine. As such, the prolonged POD in cats of the present study was likely caused in part by damaged nephrons that required repair and recovery before clinically normal renal function and water reabsorption could occur. Another likely contributor to POD in cats with more severe renal injury in the present study was the loss of the hyperosmotic medullary interstitium necessary for urine concentration.
Although we believe that many cats with ureteral obstruction have underlying CKD, results of the present study suggested that a more promising outcome could be suspected for cats that have a greater initial response to renal decompression, compared with cats that have a lesser initial response. When discussing postoperative expectations with clients, a clinician could use the absolute changes in these parameters as a guide for prognosis because cats in the present study that had greater decreases in BUN and creatinine concentrations in the first 24 hours following surgery were more likely to have had resolution of azotemia while hospitalized. Further, for all cats in the present study that survived to hospital discharge, the median BUN and creatinine concentrations prior to discharge were within reference ranges.
Preoperative hyperkalemia is largely associated with oliguria or anuria owing to a profound decrease in potassium excretion; however, preoperative hyperkalemia in the present study was not associated with any of the outcomes measured, including survival to hospital discharge. It is therefore important to note that for cats with postrenal azotemia, more severe metabolic and electrolyte changes do not necessarily predict the degree of renal recovery that could occur after renal decompression, and this should be communicated with owners. Severe hyperkalemia, however, can be a life-threatening emergency, and in the present study, the immediate identification of cats with hyperkalemia combined with the rapid and effective treatment of hyperkalemia in the preoperative period and renal decompression as soon as possible could have played a role in outcomes observed. In cats with hypokalemia in the present study, potassium supplementation was provided in accordance with a standard protocol of the UCD-VMTH; therefore, it was possible that the changes observed in serum potassium concentrations were underestimated. The rapid excretion of potassium combined with the need for aggressive supplementation in cats of the present study likely indicated distal tubule dysfunction, especially the principal cells, which are responsible for fine-tuning potassium excretion for maintenance of potassium balance.13 Although results of preoperative blood gas analyses in the present study were not associated with duration or maximum severity of POD, greater changes in serum pH, bicarbonate concentrations, and base excess from ≤ 24 hours before surgery to 1 day after surgery were associated with a less severe POD. Acidemia is typically encountered with ureteral obstructions in cats because of uremic acid accumulation and an inability of the impaired kidney to compensate with bicarbonate production. Further, a previous study9 in cats showed that acidemia has a positive correlation with diuresis following relief of urethral obstruction. The changes in acid-base status could reflect kidney function because resolution of metabolic acidosis and normalization of pH require active renal mechanisms. Serum bicarbonate concentration, pH, and base excess could be markers of nephron and tubular damage, and in the present study, greater changes in these parameters after surgery could have reflected faster return to clinically normal kidney function and therefore less POD. Some cats in the present study did receive sodium bicarbonate during their initial hospitalization period, and because of such, some of the acid-base improvements noted may have resulted from medical management, not intrinsic kidney function.
Interestingly, results of the present study suggested that neither preoperative hydration status nor preoperative acid-base status was associated with duration or maximum severity of POD in cats. These findings contrasted with findings noted previously in cats9,14 with urethral obstructions. One reason for the disagreement could have been misclassification of hydration status for cats in the present study because hydration status is difficult to estimate in veterinary species. We believe that many cats treated for ureteral obstruction have fluid overload that is not identified. Cats in the present study were regarded as overhydrated only if such was noted in the medical record for their initial physical examination at the UCD-VMTH. The fact that 6 cats in the present study received < 1 mL/kg/h of fluid support during the anesthetic procedure and that the median fluid rate administered to all cats during anesthesia in the present study was 3.9 mL/kg/h (a rate less than maintenance fluid administration rates previously used at our institution for cats undergoing general anesthesia) suggested that overhydration was potentially underdiagnosed on initial examination or that cats in the present study became overhydrated during the preoperative period after admission. However, the median fluid rate of the present study was in accordance with more recently recommended15 rates of IV fluid administration for patients undergoing general anesthesia and was relatively consistent with the UCD-VMTH's maintenance fluid rate of 5 mL/kg/h (2.3 mL/lb/h) for cats undergoing general anesthesia; therefore, findings may not have reflected an underreported volume overload issue.
The authors believe that ureteral obstructions in cats are more commonly being diagnosed and surgically treated in veterinary medicine. The present study did not subcategorize causes of ureteral obstruction; however, 92% of cats in the present study survived to hospital discharge, which compared favorably with perioperative mortality rates of 7.5% to 19% reported by investigators who had evaluated interventional and surgical treatment options for ureteral obstructions.4,16–19 No preoperative variable or absolute changes in results of serum biochemical or blood gas analyses were predictive of survival to hospital discharge. Although the present study did not evaluate long-term survival rates, the high percentage of cats that survived to hospital discharge was information that could help owners make more informed decisions before pursuing renal decompressive surgery for cats with ureteral obstruction.
In the present study, we found no substantial influence of the type of surgery performed (ie, stent vs SUB, unilateral vs bilateral) on the duration or maximum severity of POD, and surgery type did not affect the likelihood of survival to hospital discharge. We feel it is important to note that the need for bilateral surgery should not affect the decision to move forward with surgery because in the present study, outcomes for cats undergoing bilateral surgery were as favorable as those for cats undergoing unilateral surgery. In addition, although most surgeries performed on cats in the present study were unilateral, we believe it could be assumed that all cats with azotemia are experiencing some dysfunction of the contralateral kidney. This was further supported by our findings that all cats in the present study, regardless of unilateral or bilateral obstruction, developed POD, and by the fact that POD is thought to occur in cats only when the function of the nonobstructed kidney is also not clinically normal.20,21 Interestingly, despite the effect of CKD on urine concentration, cats in which CKD had been diagnosed previously did not develop more severe or longer POD in the present study. However, duration of POD in the present study was underestimated because all but 4 cats had their urinary catheters removed while POD was still present.
Interestingly, the median duration from surgery to maximum urine output in the study reported here was 12 hours (range, 6 to 60 hours). This delay supported the idea that the initial postrenal obstruction could have led to development of intrinsic renal damage and AKI. In addition, intrarenal mechanisms, including vasoconstriction and decreased renal blood flow, likely occurred and needed to be reversed before renal function and glomerular filtration could be maximally reestablished.
The duration of POD in the present study was associated with hours in the ICU and days hospitalized after surgery, factors for which veterinarians should adequately prepare owners regarding the financial undertaking, especially when cats are expected to have prolonged POD and subsequently require long-term intensive monitoring. The profound diuresis (up to 33 mL/kg/h [15 mL/lb/h] on the basis of findings in the present study) that could affect cats speaks to the intensive monitoring required after surgery to ensure electrolyte balance and maintenance of appropriate hydration. It should also be stressed that management of cats with POD requires careful fluid prescriptions to ensure adequate hydration yet prevent overhydration in the postoperative period because the latter is a phenomenon that, in the authors' experiences, could cause serious morbidity and death.
Owing to the retrospective nature of the present study, there were several limitations. As mentioned previously, there was not a complete data set available for all cats. For example, the absence of a urinary catheter for the entire duration of POD in some cats made it impossible to precisely determine the duration and maximum severity of POD in all cats in the study population. In an attempt to describe a homogeneous population of cats that had similar surgeries and similar postoperative care, cats undergoing more traditional surgical procedures (eg, ureterotomy, ureteral resection and anastomosis, ureteronephrectomy, or ureteroneocystostomy) to treat ureteral obstructions were not included in the present study. Further, no control group was used to account for potential confounding variables. However, to ensure consistency of treatment for all patients, only cats treated under the guidance of a single surgeon were included in the present study, and although this decreased variables, we had little doubt that increasing surgical skill and comfort with procedures for ureteral stent and SUB placement over time by the surgeon may have influenced outcomes. Therefore, results of the present study may not have been representative of outcomes for all surgeons. In addition, although the variables analyzed in the present study were chosen at the discretion of the authors and on the basis of previous literature on POD, there are many other variables that could be of interest, such as concentrations of serum globulins, duration of general anesthesia, and patient variables (eg, body temperature, blood pressure, pulse oximetry, or serial evaluations of acid-base status) during anesthesia that were not included in the present study. Additionally, it was possible that cats in the study population were misclassified on the basis of the information available in their medical records. An additional limitation was the use of a case series with exploratory analyses study design for the assessment and identification of variables that could be associated with POD. Case series studies are not ideal for identifying associations between exposures and outcomes because of low internal validity. Although some associations were identified in the present study, a well-designed prospective study should be performed to further identify true associations between these variables and outcomes. Additionally, variables associated with the likelihood of developing POD after decompressive surgery to treat ureteral obstruction could not have been tested in the present study because all cats in this study developed POD. Lastly, because surgical procedures involving placement of stents and SUBs for treatment of ureteral obstruction in cats were relatively new in veterinary medicine at the time of the present study, sample size was small, which may have resulted in a type II error.
Results of the present study highlighted several factors associated with duration and maximum severity of POD and documented a high percentage of cats to have survived to hospital discharge, regardless of the extent of azotemia at the initial evaluation, owing in part to their intensively treated fluid volume and electrolyte derangements. Additionally, resolution of azotemia was not related to the duration or maximum severity of POD in the present study; therefore, this parameter should not be used to predict which cats will regain clinically normal renal function after relief of ureteral obstructions. Further prospective investigation into management of POD and the long-term outcomes of cats following POD is warranted.
Acknowledgments
Presented in abstract form at the Surgical Summit of the American College of Veterinary Surgeons, Nashville, Tenn, October 2015.
ABBREVIATIONS
| AKI | Acute kidney injury |
| BCS | Body condition score |
| CKD | Chronic kidney disease |
| ICU | Intensive care unit |
| IQR | Interquartile (25th to 75th percentile) range |
| POD | Postobstructive diuresis |
| SUB | Subcutaneous ureteral bypass system |
| UCD-VMTH | University of California-Davis William R. Prichard Veterinary Medical Teaching Hospital |
Footnotes
Vet Stent-Ureter, Infiniti Medical LLC, Redwood City, Calif.
SUB system, Norfolk Vet Products, Skokie, Ill.
Stata, version 14.0 for Mac, StataCorp LLC, College Station, Tex.
Plasmalyte 148, Baxter Healthcare Corp, Deerfield, Ill.
References
1. Klahr S. Pathophysiology of obstructive nephropathy. Kidney Int 1983;23:414–426.
2. Narins RG. Post-obstructive diuresis: a review. J Am Geriatr Soc 1970;18:925–936.
3. Atamer T, Artim-Esen B, Yavuz S, et al. Massive post-obstructive diuresis in a patient with Burkitt's lymphoma. Nephrol Dial Transplant 2005;20:1991–1993.
4. Kyles AE, Hardie EM, Wooden BG, et al. Management and outcome of cats with ureteral calculi: 153 cases (1984–2002). J Am Vet Med Assoc 2005;226:937–944.
5. Zaid MS, Berent AC, Weisse C, et al. Feline ureteral strictures: 10 cases (2007–2009). J Vet Intern Med 2011;25:222–229.
6. Kyles AE, Hardie EM, Wooden BG, et al. Clinical, clinicopathologic, radiographic, and ultrasonographic abnormalities in cats with ureteral calculi: 163 cases (1984–2002). J Am Vet Med Assoc 2005;226:932–936.
7. Westropp JL, Ruby AL, Bailiff NL, et al. Dried solidified blood calculi in the urinary tract of cats. J Vet Intern Med 2006;20:828–834.
8. Baum N, Anhalt M, Carlton CE Jr, et al. Post-obstructive diuresis. J Urol 1975;114:53–56.
9. Francis BJ, Wells RJ, Rao S, et al. Retrospective study to characterize post-obstructive diuresis in cats with urethral obstruction. J Feline Med Surg 2010;12:606–608.
10. Fröhlich L, Hartmann K, Sautter-Louis C, et al. Postobstructive diuresis in cats with naturally occurring lower urinary tract obstruction: incidence, severity and association with laboratory parameters on admission. J Feline Med Surg 2016;18:809–817.
11. Schmiedt C, Tobias KM, Otto CM. Evaluation of abdominal fluid: peripheral blood creatinine and potassium ratios for diagnosis of uroperitoneum in dogs. J Vet Emerg Crit Care (San Antonio) 2001;11:275–280.
12. Hamdi A, Hajage D, Van Glabeke E, et al. Severe post-renal acute kidney injury, post-obstructive diuresis and renal recovery. BJU Int 2012;110:E1027–E1034.
13. Pearce D, Soundararajan R, Trimpert C, et al. Collecting duct principal cell transport processes and their regulation. Clin J Am Soc Nephrol 2015;10:135–146.
14. Feldman EC. Polyuria and polydipsia. In: Ettinger SF, Feldman EC, eds. Textbook of veterinary internal medicine. 7th ed. St Louis: Saunders Elsevier, 2010;156–159.
15. Davis H, Jensen T, Johnson A, et al. 2013 AAHA/AAFP fluid therapy guidelines for dogs and cats. J Am Anim Hosp Assoc 2013;49:149–159.
16. Garcia de Carellan Mateo A, Brodbelt D, Kulendra N, et al. Retrospective study of the perioperative management and complications of ureteral obstruction in 37 cats. Vet Anaesth Analg 2015;42:570–579.
17. Kulendra NJ, Syme H, Benigni L, et al. Feline double pigtail ureteric stents for management of ureteric obstruction: short- and long-term follow-up of 26 cats. J Feline Med Surg 2014;16:985–991.
18. Berent AC, Weisse CW, Todd K, et al. Technical and clinical outcomes of ureteral stenting in cats with benign ureteral obstruction: 69 cases (2006–2010). J Am Vet Med Assoc 2014;244:559–576.
19. Horowitz C, Berent A, Weisse C, et al. Predictors of outcome for cats with ureteral obstructions after interventional management using ureteral stents or a subcutaneous ureteral bypass device. J Feline Med Surg 2013;15:1052–1062.
20. Harris RH, Yarger WE. The pathogenesis of post-obstructive diuresis. The role of circulating natriuretic and diuretic factors, including urea. J Clin Invest 1975;56:880–887.
21. Sophasan S, Sorrasuchart S. Factors inducing post-obstructive diuresis in rats. Nephron 1984;38:125–133.
