Papers & Articles

Retrospective study of the survival of cats

with acquired chronic renal insufficiency
offered different commercial diets
E. A. Plantinga, H. Everts, A. M. C. Kastelein, A. C. Beynen

A retrospective study was carried out on the efficacy of seven commercial diets designed to be fed to cats
with chronic renal failure. The median survival time of 175 cats that received conventional diets was seven
months, whereas the median survival time of 146 cats given one of the seven diets was 16 months. The cats
on the most effective of the diets had a median survival time of 23 months and those on the least effective
diet had a median survival time of 12 months. The composition of the seven diets was comparable, except
that the most effective diet had a particularly high content of eicosapentaenoic acid.

CHRONIC renal failure is a common clinical condition in MATERIALS AND METHODS

cats, and the prevalence of impaired kidney function increases
with age (Burkholder 2000), reaching 30 per cent in cats over Data
15 years old (Krawiec and Gelberg 1989). The condition is The database used in this study contained information
characterised by an irreversible and progressive loss of kidney on 88,037 cats, derived from 31 veterinary clinics in the
function, leading to azotaemia, uraemia and clinical signs Netherlands. The clinics used a computer-based practice
associated with the kidneys’ decreasing ability to perform management system (DAISY; Micpoint) and their consultation
their normal functions. Affected cats have a poor prognosis reports were collected electronically. Table 1 shows the com-
because the renal dysfunction generally progresses to its end positions of the kidney diets that were prescribed.
stage (Polzin and Osborne 1995).
Nutritional management is essential in the treatment of Data processing
chronic renal failure. In cats with induced renal failure, a Reports from deceased cats that had been at least eight years
restriction of phosphorus intake, as the only dietary variable, old were selected. The data were further selected with the help
has been shown to reduce the extent of lesions observed in the of search arguments (‘cre*’, ‘krea*’, ‘BUN’, ‘urea’ and ‘ureum’),
kidneys postmortem (Ross and others 1982). It has been sug- the asterisk indicating that all words beginning with ‘cre’ or
gested that further dietary changes, such as reductions in the ‘krea’ were eligible. The reports obtained were subjected to the
intakes of protein and sodium, increases in potassium, and a inclusion criteria listed in Table 2. Thus, cats diagnosed with
moderate base excess may contribute to reducing the pro- diseases other than chronic renal failure were excluded, these
gression of the condition (Plantinga and Beynen 2004). In diseases being principally hyperthyroidism, diabetes mellitus
dogs with experimental renal disease, the feeding of fish oil and urolithiasis. For inclusion, the cause of death had to be
instead of either safflower oil or beef tallow diminished the chronic renal failure, diagnosed on the basis of the cat’s his-
rate of decrease in renal function (Brown and others 2000). It tory, clinical chemical data, clinical signs and the absence of
has also been suggested that the eicosapentaenoic acid (EPA) other disease. Survival after diagnosis was in the period from
in fish oil alters eicosanoid metabolism and results in an 1999 to 2003.
increase in the production of vasodilatory mediators and a From the reports selected, the following variables were
reduction in the production of proinflammatory eicosanoids, extracted and classified: clinic code, breed, age, sex, plasma
these changes being beneficial for renal function (Barcelli and urea and creatinine concentrations at diagnosis, the use of a
Pollak 1985). Arachidonic acid is a competitive inhibitor of kidney diet (0 No, 1 Yes), the type of kidney diet (1 to 7), the
the conversion of EPA into eicosanoids (Siess and others 1980, use of an angiotensin-converting enzyme (ACE) inhibitor
Fischer and Weber 1983). It can be hypothesised that diets for (0 No, 1 Yes) and the survival time after diagnosis (months).
the management of chronic renal failure in cats should be rich The kidney diets were as follows: 1 Hill’s KD wet, 2 Hill’s KD
in EPA and satisfy their minimum requirement for arachidonic dry, 3 Leo FKW, 4 Leo FKD, 5 Royal Canin renal programme
Veterinary Record (2005) acid (Plantinga and Beynen 2004). Both of these fatty acids dry, 6 Waltham low phosphorus/protein wet, 7 Waltham low
157, 185-187 are essential for cats because cats lack the enzymatic activity phosphorus/protein dry. The cats were classified according to
to produce adequate amounts of them (Rivers and others the type of diet they were offered on the basis of the reports,
E. A. Plantinga, DVM, 1975). that is, the diet prescribed by the veterinarians and/or the
PhD, The use of specially designed diets to manage cats with owner’s choice not to feed a kidney diet. The diet had to be
H. Everts, DVM, PhD, chronic renal failure is now standard in veterinary practice. consumed for at least 75 per cent of the cat’s survival time.
A. C. Beynen, PhD, Elliott and others (2000) have shown that affected cats fed No record was kept of the cats’ daily food intake. The use of
Department of Nutrition, a special diet instead of a normal cat food survived longer, an ACE inhibitor was considered as a possible confounding
Faculty of Veterinary the median survival times being 20·8 and 8·7 months, factor.
Medicine, Utrecht respectively. On the market there is a wide variety of kid-
University, PO Box 80.152, ney diets for cats, produced by different manufacturers, Statistical analysis
3508 TD Utrecht, which differ in their composition (Plantinga and Beynen The data were analysed by using the statistical package
The Netherlands 2004); they may therefore differ in their efficacy. The aim GENSTAT 5 (Payne and others 1993). The initial model con-
A. M. C. Kastelein, DVM, of this retrospective study was to compare the survival times tained all the variables and the survival time as the dependent
Micpoint, Vlierweg 38a, of cats newly diagnosed with chronic renal failure that had variable; each variable was removed successively and then
3991 BD Houten, been fed either a normal diet or different brands of kidney added again to the model, to identify variables that signifi-
The Netherlands diets. cantly affected the variance in survival time. The procedure

The Veterinary Record, August 13, 2005

Papers & Articles

was carried out either with all the kidney diets pooled (kid- TABLE 1: Composition of the seven brands of kidney diet
ney diet 0 or 1) or with specified brands (kidney diet 0 or 1
Protein Phosphorus Sodium Potassium AA EPA Base excess*
to 7), and single and multiple regression analyses were carried Diet Type (g/MJ) (g/MJ) (g/MJ) (g/MJ) (g/MJ) (g/MJ) (mmol/kg DM)
out with the selected variables. Kaplan-Meier survival curves
were constructed and the mean and median survival times 1 Wet 12·6 0·20 0·13 0·48 0·25 0·05 168
were calculated. 2 Dry 13·1 0·24 0·12 0·42 0·03 ND 61
3 Wet 12·6 0·19 0·08 0·60 0·18 0·47 46
4 Dry 12·7 0·22 0·09 0·67 0·03 0·02 32
5 Dry 14·9 0·34 0·21 0·57 0·06 0·11 154
RESULTS 6 Wet 11·8 0·17 0·10 0·40 0·14 ND 35
7 Dry 12·8 0·21 0·11 0·49 0·02 ND 48
After data processing, 321 reports were considered suitable. * Base excess was calculated according to Kienzle and others (1991). Data derived from Plantinga
Table 3 shows the characteristics of the cats. Table 4 shows the and Beynen (2004)
types of diet offered in relation to the age and initial plasma AA Arachidonic acid, EPA Eicosapentaenoic acid, DM Dry matter, ND Not detected
urea and creatinine concentrations of the cats.
In the statistical model with the normal diet and all the
TABLE 2: Criteria for the inclusion of the cats with chronic renal
kidney diets pooled, the variables clinic code, breed, age, sex, failure (CRF) in the study
and the use of an ACE inhibitor did not significantly affect sur-
vival time, but diet, plasma creatinine and plasma urea did Criterion
have a significant effect. The regression models were: survival Age at diagnosis >8 years
time (months) = (22·00 – 0·454) x plasma urea (mmol/l) Survival time after diagnosis >2 months
(R2adj=0·064, P<0·001); survival time (months) = (22·86 – Plasma urea >14 mmol/l
0·027) x plasma creatinine (µmol/l) (R2adj=0·102, P<0·001); Plasma creatinine >175 µmol/l
Diet Normal cat food or one type of
and survival time (months) = (8·483 + 10·50) x (kidney diet kidney diet for at least 75 per cent
= 1) (R2adj=0·229, P<0·001). The Kaplan-Meier survival curves of survival time
for the last model are given in Fig 1. After combining the Health status No diseases other than CRF
three significant predictors, the regression model with the Cause of death Related to CRF
highest predictive value was: survival time (months) = (15·47
+ 9·51) x (kidney diet = 1 – 0·339) x urea (R2adj=0·283,
P<0·001). TABLE 3: Characteristics of the 321 cats used in the study
In further analyses, the brand of kidney diet was specified. Characteristic
Again, the variables clinic code, breed, age, sex and the use of
an ACE inhibitor had no significant effect on survival time, but Age (mean [sd]) 15·08 (2·91) years
the variables plasma creatinine, plasma urea and the type of Breed (%) European shorthair 82·9, Persian
8·4, Siamese 3·1, other 5·6
kidney diet did have a significant effect. Table 5 shows the esti- Sex (%) Male 2·2, male neutered 51·1,
mated prolongation of the mean and median survival times female 6·8, female neutered 39·9
by each kidney diet when compared with no dietary inter- Plasma urea (mean [sd]) 19·3 (5·8) mmol/l
vention. The incorporation of the variables plasma urea and Plasma creatinine (mean [sd]) 269·8 (90·0) µmol/l
Kidney diet (%)* 0 = 54·5, 1 = 45·5
creatinine into the model increased the percentage of variance ACE inhibitor (%)* 0 = 89·4, 1 = 10·6
accounted for, but did not affect the prolongation of survival
time. The Kaplan-Meier survival curves for each type of kid- * 0 No, 1 Yes
ney diet are shown in Fig 2. ACE Angiotensin-converting enzyme

DISCUSSION 2004). Thus, diet 5 may be too rich in phosphorus (Table 1).
It has been suggested that the ideal concentrations of sodium,
For the interpretation of this retrospective study it is important potassium and arachidonic acid may be 0·04 to 0·06, 0·4 to 0·6
to understand its limitations. First, its validity depends on the and 0·01 to 0·02 g/MJ, respectively (Plantinga and Beynen
accuracy of the reports received from the practitioners. Different 2004). On this basis, all the diets contain too much sodium and
veterinarians may diagnose and treat chronic renal failure in cats too much arachidonic acid, but are optimal in potassium. On
differently, and it could be suggested that the observed effect of
the diets was biased by the attitude and expertise of the practi- 1·0
tioners. However, it is reassuring that in the statistical analysis
the clinic code had no effect on the cats’ survival time. Other
limitations are the assumption that the kidney diet bought by
0·8 Group 0
the cat’s owner was indeed offered to the cat and the fact that
Proportion of cats surviving

various characteristics of the cats in the study were not taken Group 1
into account. Such characteristics could be confounding factors.
The median survival time of the cats that did not receive 0·6
one of the seven special diets was seven months, and the FIG 1: Kaplan-Meier
median survival time of the cats that received any of the seven survival curves for the
diets was on average nine months longer. These data agree model: survival time =
0·4
with those of Elliott and others (2000); the validity of the (8·483 + 10·50) x
study may thus be acceptable. (kidney diet = 1)
(R2=0·229). The
Table 5 and Fig 2 show that the type of kidney diet influ-
0·2 difference in median
enced the survival time. Diet 3 was the most effective and diet survival between
6 was the least effective when compared with normal cat food. groups 0 and 1 is nine
The restriction of phosphorus intake is the mainstay of the months. Group 0 No
treatment of cats with chronic renal failure, and a phosphorus 0 kidney diet used,
level of 0·1 to 0·2 g/MJ in wet diets and 0·13 to 0·27 g/MJ in 0 10 20 30 40 50 60 70 Group 1 Kidney diet
dry diets may be considered ideal (Plantinga and Beynen Survival time (months) used

The Veterinary Record, August 13, 2005

 Papers & Articles

TABLE 4: Mean ages and plasma urea and creatinine tive in modulating progressive renal injury, which was associ-
concentrations of the cats offered a normal diet and the seven ated with a reduction of glomerular and systemic hypertension
kidney diets and proteinuria, but not glomerular hypertrophy. The efficacy
Plasma of ACE inhibitors in cats has not been reported, but these drugs
Number Age Urea Creatinine are frequently prescribed for cats with chronic renal failure.
Diet of cats (years) (mmol/l) (µmol/l) The results of this study provide further evidence that com-
Normal 175 15·5a 20·6a 277·1a
mercial kidney diets can prolong the survival time of cats with
1 5 16·6ab 19·2ab 313·2ab chronic renal failure. The data indicate that diet 3 prolonged
2 56 14·2b 16·5b 242·1b the survival time most; it is relatively low in phosphorus and
3 24 14·7ab 17·3b 269·5ab sodium, but high in potassium, arachidonic acid and EPA. The
4 9 15·4ab 18·5ab 257·7ab
5 10 14·5ab 16·7ab 209·2b
superior performance of diet 3 may be due to the complex
6 35 14·7ab 20·2ab 278·6ab combined effects of its constituent parts, but interpreting the
7 7 14·0ab 15·4ab 276·3ab results together with earlier findings, it appears that the low
a, b Different superscripts within the column indicate significant
level of phosphorus and the high levels of potassium and EPA,
differences between the diets (P<0·05)
particularly the latter, may be its most important characteris-
tics. This result is clearly important for assisting manufactur-
ers in the formulation, and veterinarians in the selection, of
TABLE 5: Mean (sd) and median survival times of the cats suitable diets for treating cats with chronic renal failure.
offered a normal diet and the seven kidney diets

Survival time (months) Number

Diet Mean (sd) Median of cats P* ACKNOWLEDGEMENTS
Normal 8·47 (5·49) 7·0 175
1 18·8 (9·61) 13·5 5 <0·10
The authors thank the co-workers from Micpoint Automotive
2 18·1 (10·6) 16·0 56 <0·001 Services and Automatisering, Houten, the Netherlands, espe-
3 29·7 (10·6) 23·0 24 <0·001 cially Mr J. Leemkuil, for their cooperation, the use of facili-
4 16·8 (6·22) 16·0 9 <0·05 ties and their useful input. Without them this research would
5 17·4 (8·98) 17·0 10 <0·05
6 14·0 (10·0) 12·0 35 <0·005
not have been possible. The authors also wish to thank the
7 21·9 (8·49) 19·0 7 <0·01 veterinary clinics for making their practice records available
for this research; their help was greatly appreciated.
* Difference versus normal diet

References
the assumption that the optimal value for the base excess is 150 BARCELLI, U. & POLLAK, V. E. (1985) Is there a role for polyunsaturated
to 350 mmol/kg dry matter (Plantinga and Beynen 2004), the fatty acids in the prevention of renal disease and renal failure? Nephron 41,
base excess of diets 1 and 5 may be considered appropriate. 209-212
The ideal content of EPA may be 0·2 to 1·0 g/MJ (Plantinga and BROWN, S. A., BROWN, C. A., CROWELL, W. A., BARSANTI, J. A., KANG,
Beynen 2004), which would imply that only diet 3 has the ideal C. W., ALLEN, T., COWELL, C. & FINCO, D. R. (2000) Effects of dietary
polyunsaturated fatty acid supplementation in early renal insufficiency in
level. A major difference between diet 3, which gave the longest
dogs. Journal of Laboratory and Clinical Medicine 135, 275-286
increase in survival time, and diet 6, which gave the shortest, BROWN, S. A., FINCO, D. R., BROWN, C. A., CROWELL, W. A., ALVA, R.,
appears to be their EPA content. The authors suggest that the ERICSSON, G. E. & COOPER, T. (2003) Evaluation of the effects of inhibi-
efficacy of diet 3 may be explained by its high concentration tion of angiotensin converting enzyme with enalapril in dogs with induced
of EPA. Controlled research is necessary to investigate the role chronic renal failure. American Journal of Veterinary Research 64, 321-327
of EPA in the survival of cats with chronic renal failure. BURKHOLDER, W. J. (2000) Dietary considerations for dogs and cats with renal
The use of an ACE inhibitor had no significant effect on the disease. Journal of the American Veterinary Medical Association 216, 1730-1734
survival time of the cats, whereas there is evidence that ACE ELLIOTT, J., RAWLINGS, J. M., MARKWELL, P. J. & BARBER, P. J. (2000)
inhibitors may be beneficial in dogs with renal disease. A study Survival of cats with naturally occurring chronic renal failure, effect of dietary
management. Journal of Small Animal Practice 41, 235-242
by Brown and others (2003) showed that enalapril was effec-
FISCHER, S. & WEBER, P. C. (1983) Thromboxane A3 (TXA3) is formed in
human platelets after dietary eicosapentaenoic acid (C20:5 omega 3).
Biochemical and Biophysical Research Communications 116, 1091-1099
1·0
KIENZLE, E., SCHUKNECHT, A. & MEYER, H. (1991) Influence of food com-
Group 0 position on the urine pH in cats. Journal of Nutrition 121, S87-S88
Group 1 KRAWIEC, D. R. & GELBERG, H. B. (1989) Chronic renal disease in cats.
Group 2 In Current Veterinary Therapy X: Small Animal Practice. Ed R. W. Kirk.
0·8
Philadelphia, W. B. Saunders. pp 1170-1173
Group 3
Proportion of cats surviving

PAYNE, R. W., LANE, P. W., DIGBY, P. G. N., HARDING, S. A., LEECH, P. K.,
Group 4
MORGAN, G. W., TODD, A. D., THOMPSON, R., TUNNICLIFFE WILSON,
Group 5 G., WELHAM, S. J. & WHITE, R. P. (1993) Genstat 5 Release 3 Reference
0·6 Group 6 Manual. Oxford, Oxford University Press
Group 7 PLANTINGA, E. A. & BEYNEN, A. C. (2004) Vergelijkende analyse van com-
merciële nierdiëten voor katten, met bijzondere aandacht voor de vetzuur-
samenstelling. Tijdschrift voor Diergeneeskunde 129, 656-662
0·4 POLZIN, D. J. & OSBORNE, C. A. (1995) Pathophysiology of renal failure and
uremia. In Canine and Feline Nephrology and Urology. Eds C. A. Osborne,
D. R. Finco. Philadelphia, Williams & Wilkins. pp 335-367
RIVERS, J. P., SINCLAIR, A. J. & CRAWFORD, M. A. (1975) Inability of the cat
0·2
to desaturate essential fatty acids. Nature 258, 171-173
ROSS, L. A., FINCO, D. R. & CROWELL, W. A. (1982) Effect of dietary phos-
FIG 2: Kaplan-Meier phorus restriction on the kidneys of cats with reduced renal mass. American
survival curves for the Journal of Veterinary Research 43, 1023-1026
seven types of kidney 0 SIESS, W., ROTH, P., SCHERER, B., KURZMANN, I., BOHLIG, B. & WEBER,
diet (groups 1 to 7 refer 0 10 20 30 40 50 60 70 P. C. (1980) Platelet-membrane fatty acids, platelet aggregation, and throm-
to the diet codes) Survival time (months) boxane formation during a mackerel diet. Lancet i, 441-444

The Veterinary Record, August 13, 2005