Preventive Veterinary Medicine 55 (2002) 1–15

Diet and lifestyle variables as risk factors for chronic renal failure in pet cats K.L. Hughesa, M.R. Slatera,*, S. Gellera, W.J. Burkholderb,1, C. Fitzgeralda,2 a

Department of Veterinary Anatomy and Public Health, College of Veterinary Medicine, Texas A&M University, College Station, TX 77843-4458, USA b Department of Small Animal Medicine and Surgery, College of Veterinary Medicine, Texas A&M University, College Station, TX 77843-4474, USA Received 15 August 2001; accepted 19 July 2002

Abstract A case–control study examining diet and lifestyle variables to generate hypotheses of potential risk factors for chronic renal failure in pet cats was conducted in five private practices in Texas, USA and at the Texas A&M University Veterinary Medical Teaching Hospital. A telephone questionnaire was used to gather information from owners of 38 cats newly diagnosed with CRF between December 1994 and 1995 and from owners of 56 control cats. Factor analysis was used to determine whether composite variables should be constructed to summarize the nutritional predictors adequately. The composite variables and other lifestyle variables were analyzed with logistic-regression. Three final exploratory models were developed: ad libitum feeding with fiber; ad libitum with Factor-2 (a composite variable composed of fiber, magnesium, protein, sodium and ash); and fiber alone. Ad libitum feeding and increased ash intake were associated with increased odds of CRF; increased dietary fiber, magnesium, protein and sodium were associated with decreased odds of CRF. # 2002 Elsevier Science B.V. All rights reserved. Keywords: Cats; Chronic renal failure; Diet; Risk factors

1. Introduction Chronic renal failure (CRF) is an important cause of morbidity and mortality in pet cats (Dow and Fettman, 1992). CRF generally is defined as a progressive and irreversible loss of * Corresponding author. Tel.: þ1-979-845-3286; fax: þ1-979-847-8981. E-mail address: [email protected] (M.R. Slater). 1 Present address: Food and Drug Administration, Center for Veterinary Medicine, 7500 Standish Place, Rockville, MD 20855, USA. 2 Present address: 1216 Caldwell Street, Lexington, TX 78947, USA.

0167-5877/02/$ – see front matter # 2002 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 7 - 5 8 7 7 ( 0 2 ) 0 0 0 8 8 - 0

2

K.L. Hughes et al. / Preventive Veterinary Medicine 55 (2002) 1–15

renal function during a relatively long period of time (Dow and Fettman, 1992; Finco et al., 2000). Little has been written about the potential causes of CRF except to acknowledge that a variety of etiologies may initiate it. The self-perpetuating theory of CRF states that after one or more initial insults to the kidney (causing loss of some undetermined critical number of nephrons), the remaining nephrons undergo a series of compensatory changes to maintain homeostasis (Hostetter et al., 1982; Finco et al., 1999). These compensatory changes ultimately cause additional loss of nephrons—leading to the typical clinical signs (including weight loss) and laboratory changes associated with CRF after 2/3 to 4/5 of all nephrons are lost. Diagnosis of CRF in the earliest periods of the disease is problematic because the ongoing loss of nephrons is generally non-detectable during the sub-clinical phase; also, gross clinical signs do not always correlate with laboratory changes in later stages of the disease (Polzin et al., 1992; Elliott and Barber, 1998). Identifying the initiating cause(s) is often impossible by the time the diagnosis of CRF is made because of the dissociation or non-detection of events with time. Diet is an important component in the treatment of CRF, with the goal of reducing clinical signs associated with uremia and possibly slowing the progression of the disease. Restriction of both dietary protein and phosphorous have been recommended (Ross et al., 1982; Polzin et al., 1992; Rubin, 1997; Elliott et al., 2000; Finco et al., 2000). Although diet plays a role in the treatment of CRF, it is not clear what role it plays in the disease’s etiology. A few studies have examined the role of dietary factors in the pathogenesis of renal disease in cats. Initially, studies using renal ablation to induce CRF in cats seemed to indicate that more dietary protein was associated with more-severe renal pathology (Adams et al., 1993, 1994). However, on better isolation of protein intake from caloric intake, more calories—rather than more protein—was positively associated with development of renal lesions in cats subjected to renal ablation (Finco et al., 1998). Potassium depletion in particular has been associated with renal disease, although there is conflicting evidence and opinion about whether the electrolyte imbalance is the cause or result of renal dysfunction (Dow and Fettman, 1992). The association between potassium depletion and renal disease was examined in one study involving hypokalemic cats fed a diet deficient in potassium, in which evidence of nephropathy (based on increased serum-creatinine concentration) was observed (Dow et al., 1987). Increased excretion of potassium in the urine was observed in most of the cats. The authors hypothesized that the hypokalemia was caused by the urinary loss and the dietary deficiency. The increased urinary excretion of potassium was attributed to the renal dysfunction. The possibility that the hypokalemia initiated the renal disease also was considered. Dietary supplementation with potassium reversed the renal dysfunction in most of the cats. Several other studies have examined the effect of inadequate dietary potassium on renal function in healthy cats. One study found that cats fed a diet low in potassium—and supplemented with a dietary acidifier—developed more-severe hypokalemia than cats fed the same diet without the acidifier (Dow et al., 1990). Also, the cats receiving the acidifier developed renal dysfunction but the other cats did not. Renal function improved somewhat when a potassium supplement was added to the acidifying diet. The authors concluded that diet-induced systemic acidification combined with the potassium-deficient diet led to metabolic acidosis, which in turn might have had an additive effect with the hypokalemia

K.L. Hughes et al. / Preventive Veterinary Medicine 55 (2002) 1–15

3

on renal function. Two other studies (Buffington et al., 1991; DiBartola et al., 1993) examined the effects of a commercial acidifying diet containing inadequate potassium and excess protein content. Both studies found that healthy cats fed this diet developed renal disease. Our objective was to gather epidemiological data on the relationships between diet and lifestyle factors and the occurrence of spontaneous CRF in pet cats in order to generate hypotheses regarding potential risk factors.

2. Materials and methods 2.1. Subject selection A case–control study in five private practices in Texas, USA and at the Veterinary Medical Teaching Hospital (VMTH) at Texas A&M University was conducted using a telephone questionnaire to examine diet and lifestyle variables as risk factors for renal disease in pet cats. Cats newly diagnosed between December 1994 and 1995 with chronic renal disease were eligible for the study. Renal disease was defined as the presence of two or three of the following criteria: (1) increased BUN (>35 mg/dl) and/or increased serum creatinine (>1.9 mg/dl); (2) urine specific gravity 7 years old, and were free of detectable renal disease. This was defined as two or three of the following: (1) normal BUN (35 mg/dl) and/or normal serum creatinine (1.9 mg/dl); (2) urine specific gravity 1.035; and (3) normal kidneys on physical exam. Only physical exam was included for consideration for control cats because it was unlikely that these cats would have had radiography or ultrasound of their kidneys. Therefore, only relatively healthy cats with appropriate laboratory data obtained as a routine part of their health care could be considered as controls. The definitions of normal follow: on palpation, a normal right kidney extended from the first through the fourth lumbar vertebral transverse processes, while the left kidney extended from the second through the fifth lumbar vertebral transverse processes (Poffenbarger, 1991). The surface of the kidneys were smooth and had a uniform texture. For radiography, normal feline kidneys were 2.4 to 3 times the length of the second lumbar vertebra, with regular shape and smooth margins (Feeney and Johnston, 1998). For ultrasonography, reports of normal kidney size ranged from 3.8 to 4.4 cm in length, 2.7 to 3.1 cm in width, and 2 to 2.5 cm in height (Nyland et al., 2002). Cats >7 years old were chosen as controls because CRF is associated with middle-aged and older cats (Polzin et al., 1992; Rubin, 1997), but they were not matched on age with cases in the analysis. Exclusion criteria for all cats were: (1) died