J Vet Intern Med 2013

A Retrospective Study of Acute Kidney Injury in Cats and

Development of a Novel Clinical Scoring System for Predicting
Outcome for Cats Managed by Hemodialysis
G. Segev, R. Nivy, P.H. Kass, and L.D. Cowgill
Background: Information regarding acute kidney injury (AKI) in cats is limited, and there are no reliable tools to
objectively assess disease severity and predict outcome.
Objectives: To assess clinical signs, clinicopathologic abnormalities, etiology, and outcome of cats with AKI, and to
develop models using clinical metrics and empirically derived scores to predict outcome.
Animals: One hundred and thirty-two client-owned cats.
Methods: Retrospective study. Bivariate logistic regression analyses were performed to identify variables predictive of
30-day survival. Continuous variables outside the reference range were divided into quartiles to yield quartile-specic odds
ratios (OR) for survival. Models were developed incorporating weighting factors assigned to each quartile based on the
OR. A predictive score for each model was calculated for each cat by summing all weighting factors. A second, multivari-
able logistic regression model was created from actual values of the same variables. Receiver operating characteristic curve
analyses were performed to determine the models performance. Models were further tested using a subset of cases not
used in initial assessment.
Results: Fifty ve of 132 cats (42%) remained dialysis-independent for at least 30 days after discharge, and the remain-
ing 77 cats either died (n = 37, 28%) or were euthanized (n = 40, 30%). The most common etiology was ureteral obstruc-
tion (n = 46, 35%). Higher scores were associated with decreased probability of survival (P < .001). Models correctly
classied outcomes in 7577% of the cases and 8489% of the cases in the subsequent evaluation.
Conclusions and Clinical Importance: Models can provide objective guidance in assessing AKI prognosis and severity,
but should be validated in other cohorts of cats.
Key words: Acute renal failure; Model; Risk factor; Score; Survival.

cute kidney injury (AKI) represents a spectrum of

A disease associated with a sudden onset of renal
parenchymal injury most typically characterized by
Abbreviations:
AKI acute kidney injury
generalized failure of the kidneys to meet the excre- ALT alanine aminotransferase
tory, metabolic, and endocrine demands of the body, AU acute uremia
resulting in acute uremia (AU).1 At advanced stages of AUC area under curve
AKI, retained uremic solutes accumulate with dysregu- BUN blood urea nitrogen
lation of acid-base homeostasis, metabolic processes, CBC complete blood count
and uid and electrolyte balance.1 AU results from CKD chronic kidney disease
volume and hemodynamic abnormalities, intrinsic OR odds ratio
RBC red blood cells
injury to the kidney, and postrenal causes. Prerenal
ROC receiver operating characteristic
azotemia or volume-responsive AKI manifests from
VMTH Veterinary Medical Teaching Hospital
decreased renal perfusion, most typically from dehy-
dration or hypovolemia; however, it generally is
limited in severity, resolves rapidly with timely correc-
tion of the underlying cause, and does not typically Acute kidney injury develops from multiple etiolo-
predispose to AU.1 gies including prolonged ischemia, nephrotoxins,
intrinsic kidney diseases, infectious causes, and
obstruction or rupture of urinary outow.1,2 Of those,
ureteral obstruction is a leading cause of kidney injury
From the Internal Medicine Department, School of Veterinary in cats in North America with variable clinical presen-
Medicine, The Hebrew University of Jerusalem, Rehovot, Israel tations resembling AKI and manifest as AU or chronic
(Segev, Nivy); the Department of Population Heath and kidney disease (CKD).2,3,a
Reproduction (Kass) and the Department of Medicine and
The management of AU is aimed at eliminating the
Epidemiology, School of Veterinary Medicine (Cowgill),
University of California, Davis, CA. The patients were treated at initiating cause of kidney injury and controlling clini-
the William R. Pritchard Veterinary Medical Teaching Hospital, cal consequences of uremia until recovery of kidney
University of California, Davis, Davis, CA. function. In the severest stages, medical management
Corresponding author: G. Segev, School of Veterinary Medicine, often is ineective, and animals might die within days
The Hebrew University of Jerusalem, P.O. Box 12, Rehovot from the consequences of uremia before recovery of
76100, Israel; e-mail: gsegev@agri.huji.ac.il. renal function can occur. Renal replacement therapies,
Submitted September 14, 2012; Revised March 12, 2013;
including intermittent hemodialysis, can restore eec-
Accepted April 11, 2013.
Copyright 2013 by the American College of Veterinary Internal tively electrolyte, acid-base, and uid balance, and
Medicine eliminate retained uremic solutes (uremia toxins),
10.1111/jvim.12108 excessive water loads, and endogenous and exogenous
2
toxins, thereby prolonging survival and improving the
potential for recovery.4 Despite the extended survival
aorded by intermittent hemodialysis compared with
conventional medical treatment, renal injury might be
too severe for recovery. The principal factors that
determine renal recovery, and thus short- and long-
term prognosis, are the inciting cause, the extent of
kidney damage, and the presence of comorbid disor-
ders.3 Despite advances in both human and veterinary
medicine, the case fatality for AKI remains unaccept-
ably high,511 and there is limited information in the
veterinary literature forecasting the prognosis of cats
experiencing AU and their outcome before recom-
mending expensive and invasive therapies.2,7,10,1214
Staging and scoring systems are based on clinical
and laboratory ndings, commonly used in human
medicine, and might help dene disease severity and
predict outcome. Many scoring systems have become
available in human medicine over the years, predomi-
nantly for critically ill patients.1517 In veterinary
medicine, the use of scoring systems has been limited
to the assessment of trauma, critical illness, and
surgery.6,18,19 Recently, a prognostic index has been
proposed for cats with AKI,11 but to date, only a sin-
gle scoring system has been developed to help predict
outcome for dogs with AKI managed with intermit-
tent hemodialysis.8 Developed using a retrospective
clinical setting, the scoring strategy was simple yet
accurate in predicting therapeutic outcome.8 An
objective scoring system for cats with AU might like-
wise help assess disease severity and prognosis, and
facilitate decision-making of the initiation of costly
renal replacement treatment.
The objective of this study was to assess clinical
severity of AKI in cats presenting with AU and man-
aged with intermittent hemodialysis. A companion
objective was to more precisely characterize the clinical
signs, clinicopathologic abnormalities, common etiolo-
gies, and outcome of these same cats to develop clini-
cally applicable predictive models that could aid the
clinician in objectively assessing disease severity and
prognosis for recovery.
Materials and Methods
Animals and Data Collection
The medical records of all cats presented with AU to the
University of California, William R. Prichard Veterinary Medical
Teaching Hospital (VMTH) between January 1993 and February
2007 and managed with intermitted hemodialysis were reviewed.
Cats with urinary system rupture, with urethral obstruction, or
that underwent renal transplantation were excluded.
Acute uremia was dened by the following: (1) acute onset of
clinical signs; (2) history and physical examination consistent
with AKI (anuria, oliguria, vomiting, and inappetence) or ure-
teral obstruction; and (3) azotemia (serum creatinine >3 mg/dL
and urine specic gravity <1.020). Data retrieved from the elec-
tronic medical records included signalment, history, physical
examination ndings, blood pressure (measured by indirect
Doppler or oscillometry), complete blood count (CBC), serum
chemistry, urinalysis, venous blood gas analysis, number and
identity of extrarenal organs involved, etiology, and outcome.
Surviving cats were dened as those that remained dialysis inde-
pendent for at least 30 days after discharge. Nonsurviving cats
were dened as those that either died or were euthanized because
of poor prognosis. Cats that were euthanized at their owners
request, before being treated for at least 14 days with
hemodialysis, were excluded.
Laboratory Findings
Blood and urine specimens for CBC, biochemistry prole,
blood gas analysis, and urinalysis were collected at initial presen-
tation, and analyses were performed by routine methods at the
diagnostic laboratories of the VMTH.
Etiology
Etiology was classied when known as lily intoxication,
ethylene glycol intoxication, ureteral obstruction, pyelonephri-
tis, renal lymphoma, multicentric lymphoma, and hemodynamic
dyscrasia. Ethylene glycol intoxication was documented on the
basis of known exposure to antifreeze, detection of serum eth-
ylene glycol or glycolic acid, or histologic evidence of tubular
necrosis and calcium oxalate crystals within the renal tubules.
Hemodynamic dyscrasia was dened as a hypotensive episode
documented before the development of AU and suspected as
the cause of kidney injury. Lily toxicosis was diagnosed based
on a history of exposure, clinical signs, and characteristic
histopathologic lesions. Imaging surveys (radiography, ultra-
sonography, nephropyelogram, and computed tomography) or
surgery (when imaging was inconclusive) was used to document
ureteral obstruction.
Organ System Involvement
Extrarenal organ involvement was classied based on clinical
signs, laboratory abnormalities, or radiographic or ultrasono-
graphic evidence of organ damage. Respiratory system involve-
ment was considered when either increased respiratory eort was
observed at the presenting physical examination or signicant
radiographic changes such as alveolar pattern were evident on
thoracic radiographs. Neurological system involvement was diag-
nosed when seizures were part of the cats current history or
when signicant neurological abnormalities such as ataxia were
noted at presentation. Hepatic involvement was dened when
serum alanine aminotransferase (ALT) activity or bilirubin con-
centrations were above 200 U/L (RR: 27101) and 1.0 mg/dL
(RR: 00.2), respectively. Evidence of cardiovascular involvement
was based on echocardiography, the presence of heart murmurs,
with or without arrhythmias.
Statistical Analysis
The statistically derived scoring system described for cats was
patterned after that described previously and characterized to
predict outcome in dogs with AKI undergoing hemodialysis.8
Because of the long study span, an exact CochranArmitage test
of the successive (over time) binomial proportions was performed
to assess changes in case fatality over the study period.
Stage I AnalysisVariable Selection. A series of logistic
regression analyses were performed initially to screen a series of
variables present at initial presentation for potential association
with 30 days of survival (dialysis-independent) after discharge;
continuous variables were entered as linear terms regardless of
the nature of their functional relationships (eg, linear, quadratic,
 Acute Kidney Injury in Cats 3

etc.). Among these, but not exclusively, were temperature, heart Statistical analyses were performed by the SPSS statistical
rate, respiratory rate, CBC parameters, serum biochemistry software.b For all tests, unless specied otherwise, P < .05 was
parameters, venous blood gas analysis measures, and extrarenal considered statistically signicant.
organ involvement. Variables with P  .10 were considered for
Stage II analysis.
Stage II AnalysisGeneration of Weighting Factors for Models
Results
Using Scores. Continuous variables identied in Stage I (eg, red A total of 132 cats fullled the inclusion criteria. Of
blood cells [RBC] count) were partitioned into normal or abnor- these, ve (3.8%) were intact males, three (2.3%) were
mal ranges as either increased or decreased from the reference
intact females, 76 (57.6%) were castrated males, and
range. For variables whose values above the reference range were
detrimental to survival, any outlying value below the reference
48 (36.3%) were spayed females. The median age was
range was included with reference range values. Conversely, for 83 months (range, 4271). Median body weight was
variables whose values below the reference range were considered 5 kg (range, 2.411.5).
detrimental to outcome, any outlying value above the reference The most common breeds were mixed breed (115 cats,
range also was included with reference values. The abnormal 87%), Siamese (11 cats, 8%), Abyssinian, Himalayan,
range for each variable was further partitioned into quartiles, Maine coon, American shorthair, Persian, Bombay, and
and the quartiles were subsequently treated as categorical vari- Burmese (1 each, 0.8%).
ables. Logistic regression models were then constructed using the
reference range as the reference category, and quartile-specic
weighting factors were calculated based on the odds ratios (OR) Clinical Signs and Blood Pressure
for survival. For variables for which there was no reference
range (eg, body weight), the rst quartile was used as the refer-
Median body temperature was 37.7C (range, 34.2
ence category. 39.7); median heart rate was 160 beats/min (range, 66
The weighting factors for each quartile were dened as the 240); and median respiratory rate was 40 breaths/min
OR where the relationship with survival was direct, or, as the (range, 2080). The most common abnormalities in
reciprocal of the OR when the relationship with survival was vital signs included hypothermia (75 cats, 64%), bra-
inverse. Quartile-specic categories with similar OR were com- dycardia (<160 beats/min, 42 cats, 35%), and increased
bined to reduce the number of estimated categories and to add respiratory rate (>30 breaths/min, 75 cats, 70%)
precision without adversely aecting validity. Similarly, when no (Table 1). Notably, 73 (55%) cats were anuric at pre-
statistically signicant dierence (P > .10) was observed between sentation, 66 (50%) showed signs indicative of abdom-
the quartile-specic categories and the reference range category,
inal pain, 62 (47%) had a recent history of vomiting,
they were combined.
Stage III AnalysisModel Development. Predictive models
and 61 (46%) suered from uid overload.
were generated contingent upon the previous analyses using Median systolic blood pressure was 144 mmHg
logistic regression. In Model A, the OR for each quartile or (range, 70250). A systolic blood pressure >150 mmHg
combined quartiles of a variable (from Stage II analysis above) was documented in 38 cats (38%). Only 1 cat pre-
was rounded to its integer value and assigned as the weighting sented with concurrent hypothermia, bradycardia, and
factors for that quartile. For Model B, the weighting factor for hypotension.
each quartile was assigned as the exact OR determined from
Stage II analysis. Models C and D were constructed in a simi-
lar manner except an etiology, weighted with the rounded or Extrarenal Organ Involvement
exact OR, respectively, was included additionally when known.
Sixty-ve cats (49%) had gastrointestinal tract
Model E was a multivariable model that included the same
variables as Models C and D, but because of model nonconver-
involvement, 26 (20%) had respiratory involvement, 22
gence caused by sparse data, (1) the number of known etiolo- (17%) liver involvement, 20 (15%) nervous system
gies was reduced to three, with all other known and unknown involvement, 14 (11%) cardiovascular involvement, 4
etiologies combined into a reference category; and (2) ataxia (3%) endocrine system involvement, and 2 (1.5%) had
was removed. pancreatic involvement.
Stage III AnalysisModel Assessment. In Models A through
D, the sum of the weighting factors of each variable produced
a nal predictive score for each cat. In all 4 models, the value
1 was assigned as the weighting factor to the reference cate- Table 1. History and physical examination ndings
gory. Models were assessed initially on cats with complete data in 132 cats with acute uremia at presentation.
(Models A though D, 94 cats; Model E, 101 cats). Receiver
operating characteristic (ROC) curve analysis was performed to Clinical Sign n (%)
determine sensitivities and specicities for outcome prediction at
Lethargy 117 (89)
dierent cuto points. The optimal cuto point was chosen as
Inappetence 105 (80)
the value associated with the fewest misclassications (ie, maxi-
Anuria 73 (55)
mizing Youdens index).20 Area under the ROC curve (AUC)
Abdominal pain 66 (50)
was calculated as an additional assessment of its respective per-
Vomiting 62 (47)
formance. To further evaluate the models, cats with missing
Fluid overload 61 (46)
data, which were not assessed initially, were used (38 cats for
Dehydration 25 (19)
Models A though D and 31 cat for Model E). Missing data
Distended urinary bladder 15 (11)
were replaced with the average value of the variable based on
Ataxia 11 (8)
the outcome group (survivor versus nonsurvivor). Evaluation
Polyuria/polydipsia 8 (6)
was performed using the established cuto points based on the
Diarrhea 3 (2)
initial assessment.
4

Clinicopathologic Data Etiology and Outcome

Most common hematologic abnormalities included
anemia and leukocytosis (Table 2). Serum biochemis-
try abnormalities included azotemia, hyperkalemia,
hypochloremia, and hyperphosphatemia. Median ALT
activity, aspartate aminotransferase activity, and total
serum bilirubin concentration were above their refer-
ence ranges in 26, 35, and 64% of the cats, respectively
(Table 3).
Forty-six cats (38.4%) were diagnosed with ureteral
obstruction, and 28 (21.2%) underwent surgery (eg, ur-
eterotomy, neoureterocystostomy) during hospitaliza-
tion to correct the obstruction (Table 4).
Fifty-ve cats (42%) remained dialysis independent
for at least 30 days post discharge, and the remaining
77 cases either died (n = 37, 28%) or were euthanized
(n = 40, 30%). There was no signicant dierence

Table 2. Complete blood count parameters in 132 cats with acute uremia at presentation.
n (%)

Parameter n Median (range) Below Ref Rng Above Ref Rng Ref Rng
6
Red blood cells (910 /lL) 112 6.1 (1.511.5) 80 (71.4) 1 (0.9) 710.5
Hemoglobin (g/dL) 112 9.1 (2.217.1) 68 (60.7) 1 (0.9) 1016
Hematocrit (%) 113 26.8 (8.251.1) 68 (60.2) 1 (0.9) 3050
MCV (fL) 112 45.5 (35.360.7) 15 (13.4) 8 (7.1) 4253
MCH (pg/cell) 112 15.2 (12.419) 2 (1.8) 9 (8) 1317
MCHC (g/dL) 112 33.3 (26.838.2) 3 (2.7) 48 (42.9) 3033.5
White blood cells (9103/lL) 112 12.4 (3.847.1) 1 (0.9) 44 (39.3) 4.514
Bands (9103/lL) 112 0 (04.3) 0 (0) 18 (16) 00.2
Neutrophils (9103/lL) 112 10.2 (2.740.4) 0 (0) 72 (64.3) 29
Lymphocytes (9103/lL) 112 0.9 (05.7) 68 (60.7) 0 (0) 17
Monocytes (9103/lL) 112 0.24 (01.84) 7 (6.3) 18 (16.1) 0.050.6
Eosinophils (9103/lL) 112 0.059 (06) 72 (64.3) 1 (0.9) 0.151.1
Basophils (9103/lL) 112 0 (00.36) 0 (0) 1 (0.9) 00.2
Platelets (9103/lL) 110 233 (34600) 28 (25.5) 1 (0.9) 180500

Ref Rng, reference range; MCV, mean corpuscular volume; MCH, mean corpuscular hemoglobin; MCHC, mean corpuscular
hemoglobin concentration.

Table 3. Serum biochemistry parameters in 132 cats with acute uremia at presentation.
n (%)

Parameter n Median (range) Below Ref Rng Above Ref Rng Ref Rng
Albumin (g/dL) 118 2.4 (1.53.5) 32 (27.1) 0 (0) 2.24.6
Albumin/globulin ratio 105 0.7 (0.41.2) 7 (6.7) 0 (0) 0.51.7
ALP (U/L) 106 23 (171) 19 (17.9) 0 (0) 1471
ALT (U/L) 105 64 (2840) 9 (8.6) 27 (25.7) 27101
Anion gap (mEq/L) 117 36 (1947) 0 (0) 106 (90.6) 1327
AST (U/L) 105 47 (51098) 6 (5.7) 37 (35.2) 1758
Bilirubin (mg/dL) 124 0.5 (016.1) 0 (0) 79 (63.7) 00.2
BUN (mg/dL) 131 228 (68456) 0 (0) 131 (100) 1833
BUN/creatinine ratio 130 13 (6.833.7) 16 (12.3) 40 (30.8) 1015
Calcium (mg/dL) 122 9.4 (3.915) 45 (36.9) 12 (9.8) 910.9
Chloride (mmol/L) 118 108 (88131) 106 (89.8) 1 (0.8) 117126
Cholesterol (mg/dL) 105 134 (65314) 10 (9.5) 1 (1) 89258
CO2 (mmol/L) 122 12 (431) 85 (69.7) 8 (6.6) 1521
Creatinine (mg/dL) 131 17.2 (544.2) 0 (0) 131 (100) 1.12.2
Globulin (g/dL) 105 3.7 (25.7) 11 (10.5) 3 (2.9) 2.85.4
Glucose (mg/dL) 106 117 (38996) 4 (3.8) 50 (47.2) 63118
Phosphorus (mg/dL) 126 15.2 (1.127.7) 12 (9.5) 108 (85.7) 6.68.4
Potassium (mmol/L) 123 6 (3.110.9) 5 (4.1) 93 (75.6) 3.64.9
Sodium (mmol/L) 120 150 (122169) 65 (54.2) 9 (7.5) 151158
Total protein (g/dL) 108 6.1 (3.58.9) 70 (64.8) 2 (1.9) 6.68.4

Ref Rng, reference range; BUN, blood urea nitrogen; ALT, alanine aminotransferase; AST, aspartate aminotransferase; ALP, alkaline
phosphatase.
 Acute Kidney Injury in Cats 5

Table 4. Etiology and outcome of 132 cats with acute ureteral obstruction, followed by 34% with unknown
uremia. etiology, 8% with EG intoxication, 6.4% with lily
intoxication, and only 3% with pyelonephritis.
Etiology n % of Total Survival (%) When cases with ureteral obstruction were excluded,
Unknown 53 40.1 26.4 vomiting was still positively associated with survival
Ureteral obstruction 46 34.8 67.4 (P = .003).
Ethylene glycol intoxication 12 9.1 8.3 The following variables were included for subse-
Pyelonephritis 7 5.3 57.1 quent analyses: body weight, total protein, RBC, body
Lily intoxication 6 4.5 16.7
temperature, ataxia, uid overload, respiratory involve-
Renal lymphoma 3 2.3 2.3
Hemodynamic 2 1.5 66.7
ment, vomiting, ethylene glycol intoxication, lily intoxi-
Polycystic kidney disease 1 0.8 0.0 cation, and ureteral obstruction. Ninety-four animals
Transplant rejection 1 0.8 0.0 records had complete data for all the variables iden-
Multi-centric lymphoma 1 0.8 0.0 tied in Stage I analysis and were used to assess
performance of each model.
Some variables (eg, RBC) showed a threshold phe-
nomenon, in which the 1st 3 quartiles yielded the same
(P = .73) in case fatality over the study period. Exclud- OR as the reference group, and only the 4th quartile
ing hemodynamic causes (an etiology documented in was detrimental for survival. Other continuous vari-
only 2 cats), ureteral obstruction was the etiology with ables demonstrated a more progressive association
the most favorable outcome, with a 67% survival rate with survival, with greater deviations from the refer-
(Table 4). Pyelonephritis also had a favorable outcome ence range yielding lower OR for survival and there-
(57% survival), whereas lily and ethylene glycol intoxi- fore higher weighting factors.
cations carried grave prognoses (17% and 8% survival, Initially, 4 models were generated (Tables 5, 6). Out-
respectively). come scores ranged from 4 to 28 for Model A (aver-
Odds ratios and their 95% condence intervals were age, 11.6; SD, 5.3), 4.5 to 28 for Model B (average, 12;
calculated to compare the likelihood of survival SD, 5.1), 1 to 39 for Model C (average, 13.1; SD, 7.3),
between each etiology and ureteral obstruction. Ani- and 1.9 to 39.1 for Model D (average, 13.7; SD, 7). In
mals with ethylene glycol and lily intoxications had all models, a higher score was associated with lower
odds of not surviving 23 times (P = .004, 95% CI probability of survival.
2.7200) and 10 times (P = .040, 95% CI 1.1100)
greater than animals with ureteral obstruction, respec-
tively. Cats whose etiology remained unknown had 6
times lower odds of survival compared with cats Table 5. Models A and C: Outcome prediction for 95
with ureteral obstruction (P = .001, 95% CI 2.413.7). cats with acute uremia managed with hemodialysis,
No signicant dierences in survival were identied based on the integer value of the OR, with (Model C)
between ureteral obstruction and the remaining etiologies. or without including etiology (Model A).
Variable Range
Scoring System Model A
Body weight (kg) >5 5
Of all the variables (~70) screened initially for a pos- Weighting factor 1 3
sible relationship with survival in Stage I analysis, 18 Total protein (g/dL) >5 5
had a P  .10. Ten of these were continuous vari- Weighting factor 1 5
ables, 5 categorical variables reected extrarenal sys- RBC (9106/lL) >4.86  4.86
tem involvement, and the remaining 3 were underlying Weighting factor 1 7
etiologies. Neither sex nor breed was associated with Body temperature (F) >99.35 98 < X  99.35  98
survival. Two of the continuous variables deemed Weighting factor 1 3 7
detrimental to survival in Stage I (osmolality, colloid Ataxia Yes No
oncotic pressure) could not be used in subsequent Weighting factor 8 1
Fluid overload Yes No
analyses or formal modeling because their data were
Weighting factor 2 1
missing from a large number of patient records. To Respiratory involvement Yes No
avoid collinearity, only RBC was used as a hematolog- Weighting factor 3 1
ic parameter despite the additional Stage I signicance Vomiting Yes No
of hematocrit and hemoglobin. Similarly, instead of Weighting factor ( )3 1
using both albumin and total protein, only the latter Model C: when the etiology is known, an additional weighting
was employed in further analyses. Neither overall factor is added
survival nor survival within the various etiologies EG intoxication Yes No
was associated with either plasma creatinine or BUN Weighting factor 9 1
concentrations. Lily intoxication Yes No
Weighting factor 4 1
Vomiting and ureteral obstruction were positively
Ureteral obstruction Yes No
associated with survival. Of the 62 cases in which Weighting factor ( )5 1
vomiting had been documented, nearly 50% had
6

Table 6. Models B and D: Outcome prediction for 95 83%, 90%/86%, and 71%/88%, respectively, with cor-
cats with acute uremia managed with hemodialysis, rect classication of 8489% of the cases.
based on exact odds ratios, with (Model D) or without Variables selected from the Stage I analysis were
including etiology (Model B). also used to create the following multivariable logistic
regression model (Model E): logit (p) = 0.31 9 (body
Variable Range weight) + 0.59 9 (temp) 0.7 9 (uid overload) +
Model B 0.46 9 (RBC) 2.11 9 (Lily intoxication) 2.11 9
Body weight (kg) >5 5 (ethylene glycol intoxication) + 2.3 9 (ureteral obstruc-
Weighting factor 1 3.36 tion) 63.95. When applying the model to the 101 cats
Total protein (g/dL) >5 5 that had data on all variables in the model, the optimal
Weighting factor 1 4.6
probability cuto point was 0.41, corresponding to sen-
RBC (9106/lL) >4.86  4.86
Weighting factor 1 6.7
sitivity and specicity of 82% and 77%, respectively,
Body temperature (F) >99.35 98 < X  99.35  98 and 80/101 (79%) were classied correctly. When apply-
Weighting factor 1 3.2 6.6 ing the model to cats after missing data analysis (31
Ataxia Yes No cats), using the same optimal probability cuto point,
Weighting factor 8.1 1 sensitivity and specicity were 80% and 91%, respec-
Fluid overload Yes No tively, and 27/31 (87%) were classied correctly.
Weighting factor 2.3 1 The relationship between prior probability of
Respiratory involvement Yes No survival and positive predictive value after applying
Weighting factor 2.9 1 Models C and E is presented in Figure 2.
Vomiting Yes No
Weighting factor ( ) 2.5 1
Model D: when an etiology is known, an additional weighting Discussion
factor is added
EG intoxication Yes No This study describes the clinical signs, clinicopatho-
Weighting factor 9.1 1 logic features, and the outcomes of a large population
Lily intoxication Yes No of cats with severe AKI requiring hemodialysis. In
Weighting factor 3.7 1 addition, dierent models (eg, a multivariable regres-
Ureteral obstruction Yes No sion model and a clinically applicable scoring system)
Weighting factor ( ) 4.6 1 that can aid in the objective assessment of disease
severity and prognosis prediction of cats with AU
undergoing hemodialysis treatment are proposed.
Models A, B, C, and D yielded sensitivities/specici- The clinical presentation of the cats in this study
ties of 74%/75%, 74%/75%, 62%/88%, and 66%/ was consistent with AU. The majority of cases were
83%, respectively (Tables 7, 8). Optimal cuto scores lethargic and inappetent. Hypothermia, a consistent
(ie, those that minimized overall misclassication) with nding in AU, was common and most likely resulted
their respective sensitivities, specicities, number of from retained uremia solutes.21
correctly classied cases, and AUC values for each The prevalence of increased blood pressure in this
model are presented in Table 7 and Figure 1. All mod- population of uremic cats was generally lower than
els performed similarly with regard to classication reported for dogs with AKI and is in accordance with
(7577% correctly classied). When etiology was previously published reports of blood pressure in cats
known, specicity increased at the expense of sensitiv- with CKD.2224 Fifty-ve percent of cases were anuric.
ity (Models C and D). When applying the models to Oliguria could not be adequately assessed as urine pro-
cats not used for initial assessment (38 cats with miss- duction, a prognostic factor in other studies,5,25 was
ing data) using the same cutos, Models A, B, C, and not measured because urinary catheters were not
D yielded sensitivities/specicities of 82%/85%, 89%/ implanted routinely.

Table 7. Cuto scores and their respective sensitivities and specicities for Models AD. Models A and C were
generated based on the integer-equivalent value of the odds ratio for each variable. Models B and D were
generated based on the exact odds ratio. In Models C and D, etiology was included.
Model A Model B Model C Model D

Cuto Sen Spe Cuto Sen Spe Cuto Sen Spe Cuto Sen Spe
6 38 98 6.8 31 100 7 45 98 5.4 26 100
8 52 83 9 50 88 9 52 94 10 54 92
10 74 75 10.4 74 75 10 62 88 12 66 83
12 83 54 12.5 83 60 12 74 72 12.4 74 75
15 95 41 17.3 100 28 17 100 43 17 97 47

Sen, sensitivity value; Spe, specicity value.

Bolded font represents optimal cuto points with the fewest misclassications.
 Acute Kidney Injury in Cats 7

Table 8. Performance characteristics of 5 predictive models. Models A and C utilized scores generated from the
rounded integer value of the odds ratio for each variable. Models B and D were generated from the exact odds
ratio value. Etiology was included in Models C and D. Model E was a multivariable logistic regression model.
Ninety-four and 101 cats had complete data and were used for initial model assessment of Models AD and
Model E, respectively. Numbers in parentheses are results for 38 and 31 cats with missing data that were used for
further model assessment of Models AD and Model E, respectively.
Model Optimal Cuto Point Sensitivity Specicity Correctly Classied (%) AUC
A 10 74 (82) 75 (85) 75 (87) 0.81
B 10.4 74 (89) 75 (83) 75 (86) 0.81
C 10 62 (90) 88 (86) 77 (89) 0.85
D 12 66 (71) 83 (88) 76 (84) 0.86
E 0.41a 82 (80) 77 (91) 77 (87) 0.86

AUC, area under the ROC curve.

a
Probability cuto point.

A B

C D

Fig 1. Receiver operating characteristics (ROC) analyses for Models A though D. Models A and C were generated based on the
integer value of the odds ratio for each variable. Models B and D were generated based on the exact odds ratio. In Models C and D,
etiology was included. AUC, area under the curve.

Anemia was the most common abnormality revealed
on the CBC and was characterized as nonregenerative
in most cats based on the absence of characteristics of
regeneration (ie, reticulocytes, polychromasia). This
nding is more characteristic of CKD, which might
have coexisted in some cats. The appearance of nonre-
generative anemia also might have resulted from
inammation, acute blood loss (eg, gastrointestinal
tract), and aggressive uid treatment with subsequent
overhydration. Similar results were observed in a
review of 136 cats with ureteral urolithiasis.26 How-
ever, in a dierent review of acute intrinsic renal fail-
ure in cats, the mean (SD) hematocrit at initial
diagnosis was 37  7.5%.10 Many cats with nephro-
lithiasis and ureteral obstruction that present with a
seemingly AU have acute-on-CKD with anemia
proportionate to the severity and duration of the
underlying CKD component.
A few clinicopathologic ndings at the time of pre-
sentation were not consistent with AKI. Surprisingly,
8
Fig 2. Relationship between prior probability of survival and positive predictive value for Models C and E.
a large proportion of cats demonstrated increased
ALT activity as well as increased serum bilirubin
concentrations. Increased ALT activity also has been
reported in a previous study in a high percentage of
cats with CKD.27 The increase in liver enzyme activity
and serum bilirubin concentration in this study might
reect liver damage secondary to ischemic or hypoxia
injury, toxic insults, or pancreatitis.
The most commonly identied etiology of AKI was
ureteral obstruction (35%), with ethylene glycol intoxi-
cation the next most frequent etiology representing
only 9% of cases. This observation represents a shift
in the most prevalent cause of AKI in cats requiring
hemodialysis since the mid-1990s. In the 5-year period
between 1993 and 1998, only 10% of cats were pre-
sented with ureteral obstruction as the cause of AKI,
and 35% of cats had toxic etiologies (primarily ethyl-
ene glycol). In the subsequent 5-year period from 1999
to 2003, the prevalence of ureteral obstruction
increased to 51%, and toxic etiologies decreased to
11%.a A steady increase in the incidence of ureteral
calculi from the late 1990s was previously described,
and paralleled with the change documented in the inci-
dence of calcium oxalate and struvite urolithiasis in
the last 3 decades.26,2830
The case fatality was 58.3%, with the remaining 55
cases (41.7%) surviving independent of dialysis for
longer than 30 days. This agrees with other studies of
uremic cats treated with hemodialysis.8,10,11,31 It should
be noted, however, that denitions of survival as well
as duration of follow-up vary among the dierent
studies, as might severity of kidney injury and extent
of pre-existing CKD. These limitations further under-
score the need for objective staging and scoring sys-
tems that can be used to index disease severity and
allow more valid comparisons between studies per-
formed in dierent clinical settings and at dierent
times.
Etiology had a signicant impact on the survival
in cats managed with hemodialysis for AU as it does
for dogs.8 Ureteral obstruction and pyelonephritis
were associated with favorable outcomes (67% and
57% survival, respectively), whereas lily and ethylene
glycol intoxication had exceptionally poor prognoses
(17% and 8% survival, respectively). In animals
managed by hemodialysis, the reversibility of the
injury, rather than its severity, is considered a more
important indicator of prognosis, because the clinical
and clinicopathologic consequences of uremia that
inuence mortality with conventional treatment can
be managed with extracorporeal renal replacement
treatment. The relatively positive outcome for ure-
teral obstruction was likely associated with surgical
correction of the inciting cause. Nevertheless, prog-
nosis for a favorable outcome for AKI after ureteral
obstruction is highly dependent on available and
experienced surgical expertise, opportunity for ure-
teral stenting, and degree of pre-existing (chronic)
kidney injury.
For several etiologies, the number of cases was con-
spicuously low, and the reported survival rates should
be interpreted cautiously. The severity of ethylene gly-
col and lily toxicoses, the delayed presentation for
treatment, and the irreparable damage to the kidneys
probably account for their grave prognoses.7,10,32 Eth-
ylene glycol negatively aects survival in dogs with
AKI in a similar evaluation.8 In a dierent study, 3 of
9 cats with ethylene glycol-induced AKI survived after
hemodialysis treatment for at least 4 months.31 How-
ever, these uncharacteristically high survival rates in
cats with ethylene glycol intoxication were associated
with prolonged courses of hemodialysis, and incom-
plete recovery of renal function upon cessation of
treatment. In this study, 5 of 17 cats with ethylene gly-
col intoxication underwent renal transplantation and
were excluded from analyses. The timing of diagnosis
 Acute Kidney Injury in Cats
and initiation of hemodialysis treatment in cats with
ethylene glycol intoxication was also critical for sur-
vival and might explain the dierences in outcomes
between studies.33,34
Scoring System
The scoring systems were developed to facilitate pre-
diction of 30-day postdischarge survival of cats with
AU managed with hemodialysis. Cats with severe AKI
are subject to prolonged hospitalization and intensive
treatment, including hemodialysis, associated with high
treatment costs. Consequently, the objective assessment
of survival outcome at the time of presentation could
play a pivotal role in the decision to pursue such a
treatment or opt for euthanasia. An AKI scoring sys-
tems also has potential application to facilitate quanti-
tative comparison of the severity of AKI in cats from
dierent studies. Many such scoring systems have
become available in human medicine over the years to
predict prognosis in critically ill patients. In veterinary
medicine, however, the use of scoring systems is lim-
ited, and only 1 scoring system for dogs with AKI has
been developed.8,19,35
Of the variables considered to potentially inuence
survival screened in Stage I, only 18 had a P  .10.
Hyperkalemia, a historical cause of mortality in AKI,
was not one of these, in contrast to a previous report10
and in agreement with a recent one.11 This counterintui-
tive nding relates to the ecacy of hemodialysis to
control potassium homeostasis, so that hyperkalemia
no longer compromises the animal. Similarly, the degree
of azotemia was not signicantly associated with sur-
vival, in agreement with previous studies.8,10,11,31 These
observations in cats and comparable observations in
dogs undergoing hemodialysis contravene the hypothesis
that severity of azotemia is a risk factor for AKI in
dogs managed without hemodialysis.5
Vomiting was positively associated with survival. Of
the 62 cases in which vomiting had been documented,
nearly 50% had ureteral obstruction as an etiology. The
positive predictive eect of vomiting might be partly
ascribed to its association with ureteral obstruction,
which had a favorable outcome in this study. Similarly,
in the scoring system for dogs with AKI, increased
ALT activity was positively associated with survival
likely attributable to its association with leptospirosis as
an etiology, which typically had a good outcome for
survival.8 Nevertheless, when cases with ureteral
obstruction were excluded from the analysis, vomiting
was still positively associated with survival (P = .003),
suggesting that an association between vomiting and
ureteral obstruction cannot explain entirely the positive
association of vomiting with survival.
Four statistics-based scoring systems and a multivar-
iate model were developed to test performance at pre-
dicting survival outcome in cats presenting with AKI.
The multivariate approach provided the most accurate
predictive t; however, the performance of all models
was comparable. Nonetheless, we specically sought to
create an integer-based scoring system that could easily
9
be implemented in an ICU context, as has been done
in human scoring systems. When using the AUC of
the ROC analysis for assessing model performances,
models were considered good (ie, >0.80) at discriminat-
ing survivors from nonsurvivors. Whenever the etiology
was known, the specicity increased at the expense of
sensitivity: fewer animals were classied incorrectly
as survivors, but model performance only slightly
improved. The association between specic etiologies and
specic clinical signs or clinicopathological parameters
(eg, ureteral obstruction and vomiting, hypocalcemia,
and EG intoxication) might have obscured dierences
between the models as these parameters served as
proxies for etiologies.
Models A or B, which were independent of etiology,
might be more pragmatic as often etiological cause is
unknown at case presentation, and decisions to initiate
hemodialysis must be made early in the disease course.
Etiology might remain unknown throughout hospitaliza-
tion and hence could not facilitate prognostic projections.
The overall classication performance of all models
was very good, but there were dierences between
models predictions of survivors and nonsurvivors, and
the use of scoring systems should be used cautiously
when predicting the outcome in individual cats. Mod-
els should be used in conjunction with other labora-
tory tests, clinical assessment, and clinician experience,
rather than serve as the sole means by which decisions
are made. Discretionary selection of outcome classi-
cation cuto points can maximize sensitivities and
specicities according to clinicians (and owners) pref-
erences. The higher the cuto point, the higher the
sensitivity (fewer false negatives) and the lower the
specicity (greater false positives).
A similar scoring system, developed for dogs with
AKI, showed better results with an AUC of 0.880.91.8
These dierences might be explained by the dierences
in etiologies of AKI between dogs and cats. Perhaps,
the most important factor that accounts for species dif-
ferences between scoring systems is the observation that
AU in dogs represents bona fide acute intrinsic kidney
injury, whereas many cats presenting for hemodialysis
had considerable degrees of pre-existing CKD before
the development of AU, ie, so-called acute-on-chronic
kidney disease. Therefore, cats with AU in this study
probably represented a more heterogeneous group than
dogs with intrinsic AKI with no chronic component.
There are several drawbacks to this study. First,
medical records were reviewed retrospectively, and
data for several key variables were missing from a sub-
stantial number of cases that could not be included in
the initial statistical analyses. For example, urine pro-
duction, a reportedly important prognostic factor,
could not be included in the models. Second, hemodi-
alysis is only available in a small number of referral
centers, and most cats typically received treatment
before arrival. This might have resulted in changes in
some clinicopathologic ndings at presentation. Third,
in the development of the scoring system, both surgical
cases and cases managed solely by medical means were
included. Surgical cases (eg, ureteral obstruction)
10
might have had a better prognosis as the inciting cause
could potentially be resolved. Nevertheless, ureteral
obstruction constituted a substantial cause of AU in
cats, and surgical cases were therefore included in this
study. Fourth, this study considered euthanized cats as
a nonsurvivor group. To minimize this eect, eutha-
nized cats with no meaningful attempt at management
were excluded from the study, and euthanasia was
performed because of poor prognosis for recovery.
Footnotes
a 19. King G, Stevence M, Ostro E, et al. A model of predic-
Pantaleo V, Francey T, Fischer JR, Cowgill LD. 2004. Applica-
tion of hemodialysis for the management of acute uremia in
cats: 119 cases (19932003). ACVIM Forum 2004, Minneapolis,
MN (abstract)
b 1950;3:3235.
SPSS 15.0 for Windows; SPSS Inc, Chicago, IL
Acknowledgment
Conflict of Interest Declaration: Authors disclose no
conict of interest.
References
1. Langston CE. Acute uremia. In: Ettinger SJ, Feldman EC,
eds. Textbook of Veterinary Internal Medicine. Philadelphia, PA:
Saunders WB; 2010:19552115.
2. Hardie EM, Kyles AE. Management of ureteral obstruc-
tion. Vet Clin North Am Small Anim Pract 2004;34:9891010.
3. Segev G. Outcome Prediction of Acute Uremia in Dogs
and Cats. Porto, Portugal: Proceedings of the European College
of Veterinary Internal Medicine; 2009.
4. Cowgill L, Francey T. Hemodialysis. In: DiBartola S, ed.
Fluid Therapy in Small Animal Practice, 3rd ed. Philadelphia,
PA: Saunders WB; 2006:650677.
5. Behrend EN, Grauer GF, Mani I, et al. Hospital-acquired
acute renal failure in dogs: 29 cases (19831992). J Am Vet Med
Assoc 1996;208:537541.
6. Bernieh B, Al Hakim M, Boobes Y, et al. Outcome and
predictive factors of acute renal failure in the intensive care unit.
Transplant Proc 2004;36:17841787.
7. Langston CE. Acute renal failure caused by lily ingestion in
six cats. J Am Vet Med Assoc 2002;220:4952, 36.
8. Segev G, Kass HP, Francey T, et al. Novel clinical scoring
system for outcome prediction in dogs with acute kidney injury
managed by hemodialysis. J Vet Intern Med 2008;22:301308.
9. Vaden SL, Levine J, Breitschwerdt EB. A retrospective
case-control of acute renal failure in 99 dogs. J Vet Intern Med
1997;11:5864.
10. Worwag S, Langston CE. Acute intrinsic renal failure in
cats: 32 cases (19972004). J Am Vet Med Assoc 2008;232:728732.
11. Lee YJ, Chan JP, Hsu WL, et al. Prognostic factors and a
prognostic index for cats with acute kidney injury. J Vet Intern
Med 2012;26:500505.
12. Gookin JL, Riviere JE, Gilger BC, et al. Acute renal fail-
ure in four cats treated with paromomycin. J Am Vet Med Assoc
1806;1999:18211823.
13. Fulton RB Jr, Fruechte LK. Poisoning induced by admin-
istration of a phosphate-containing urinary acidier in a cat.
J Am Vet Med Assoc 1991;198:883885.
14. Mealey KL, Boothe DM. Nephrotoxicosis associated with
topical administration of gentamicin in a cat. J Am Vet Med
Assoc 1994;204:19191921.
15. Knaus WA, Draper EA, Wagner DP, et al. APACHE II:
A severity of disease classication system. Crit Care Med
1985;13:818829.
16. Knaus WA, Wagner DP, Draper EA, et al. The APACHE
III prognostic system. Risk prediction of hospital mortality for
critically ill hospitalized adults. Chest 1991;100:16191636.
17. Knaus WA, Zimmerman JE, Wagner DP, et al. APACHE-
acute physiology and chronic health evaluation: A physiologically
based classication system. Crit Care Med 1981;9:591597.
18. Hardie E, Jayawickrama J, Du L, et al. Prognostic
indicators of survival in high risk canine surgery patients.
J Vet Emerg Crit Care 1995;5:4249.
tion of survival in critically ill dogs. J Vet Emerg Crit Care
1994;4:8598.
20. Youden WJ. Index for rating diagnostic tests. Cancer
21. Eiger SM, Kluger MJ. Cyanate and temperature regula-
tion in anephric rabbits. Am J Physiol 1985;249:F69F73.
22. Acierno MJ, Labato MA. Hypertension in renal disease:
Diagnosis and treatment. Clin Tech Small Anim Pract
2005;20:2330.
23. Syme HM, Barber PJ, Markwell PJ, et al. Prevalence of
systolic hypertension in cats with chronic renal failure at initial
evaluation. J Am Vet Med Assoc 2002;220:17991804.
24. Geigy CA, Schweighauser A, Doherr M, et al. Occurrence
of systemic hypertension in dogs with acute kidney injury and
treatment with amlodipine besylate. J Small Anim Pract
2011;52:340346.
25. Segev G. Prognosis of acute renal failure. In: Bartges J,
Polzin D, eds. Nephrology and Urology of Small Animals. Ames,
IA: Blackwell Publishing; 2012:524527.
26. Kyles AE, Hardie EM, Wooden BG, et al. Clinical, clini-
copathologic, radiographic, and ultrasonographic abnormalities
in cats with ureteral calculi: 163 cases (19842002). J Am Vet
Med Assoc 2005;226:932936.
27. King JN, Tasker S, Gunn-Moore DA, et al. Prognostic
factors in cats with chronic kidney disease. J Vet Intern Med
2007;21:906916.
28. Kyles AE, Hardie EM, Wooden BG, et al. Management
and outcome of cats with ureteral calculi: 153 cases (19842002).
J Am Vet Med Assoc 2005;226:937944.
29. Low WW, Uhl JM, Kass PH, et al. Evaluation of trends in
urolith composition and characteristics of dogs with urolithiasis:
25,499 cases (19852006). J Am Vet Med Assoc 2010;236:193200.
30. Cannon AB, Westropp JL, Ruby AL, et al. Evaluation of
trends in urolith composition in cats: 5,230 cases (19852004).
J Am Vet Med Assoc 2007;231:570576.
31. Langston CE, Cowgill LD, Spano JA. Applications and
outcome of hemodialysis in cats: A review of 29 cases. J Vet
Intern Med 1997;11:348355.
32. Rumbeiha WK, Francis JA, Fitzgerald SD, et al. A com-
prehensive study of Easter lily poisoning in cats. J Vet Diagn
Invest 2004;16:527541.
33. Connally HE, Thrall MA, Hamar DW. Safety and ecacy
of high-dose fomepizole compared with ethanol as therapy for
ethylene glycol intoxication in cats. J Vet Emerg Crit Care
(San Antonio) 2012;20:191206.
34. Dial SM, Thrall MA, Hamar DW. Comparison of ethanol
and 4-methylpyrazole as treatments for ethylene glycol intoxica-
tion in cats. Am J Vet Res 1994;55:17711782.
35. Rockar R, Drobatz K, Shofer F. Development of a scor-
ing system for the veterinary trauma patient. J Vet Emerg Crit
Care 1994;4:7783.