Chronic kidney disease (CKD) is a major cause of morbidity and death in dogs and, given that it typically diagnosed relatively late in the course of the disease, dogs with advanced changes represent only a fraction of all dogs with CKD.[1, 2] Human obesity is a risk factor for the development and progression of CKD, and reducing body fat mass, either through dietary energy restriction or bariatric surgery, can reverse many of the associated clinical and nephropathologic manifestations.
Obesity is a common medical disorder in dogs, and can predispose to associated diseases such as osteoarthritis, respiratory disease, neoplasia, and insulin resistance. However, it is less clear as to whether there is an association between excess adiposity and renal functional or disease alterations in this species. When monitoring renal disease, clinicians rely primarily on serum creatinine and urea concentrations, urine specific gravity (USG), and urine protein:creatinine ratio (UPCR). Further, microalbuminuria is considered to be an excellent marker of early kidney disease, being a sign of mildly altered glomerular permeability. However, all of these tests are insensitive, with experimental studies suggesting that azotemia and impaired urine concentrating ability are not seen until functional nephron mass is reduced by at least two thirds. Non-invasive and simple methods that have the ability to detect renal damage in CKD prior to functional nephron impairment (eg, inadequate urine concentrating ability or azotaemia) are limited. Given that there is a high risk of progression to irreversible renal damage in patients with CKD, there is a current need both to develop markers that enable early detection of renal dysfunction, as well as to identify causal factors that might predispose to such dysfunction. In response, new serum biomarkers for early renal injury have started to gain attention in human and veterinary medicine, and examples include homocysteine (Hcy), cystatin C (CysC), and clusterin (Clu).
Homocysteine is an amino acid, the majority of which (98%) circulates in an oxidized form bound to protein. In humans, total plasma homocysteine (tHcy) concentration is inversely correlated with the glomerular filtration rate (GFR), and positively correlated with circulating creatinine concentration. In dogs, significantly greater tHcy concentrations are reported in dogs with cardiac and renal diseases. CysC is a 13 kDa, 122-amino acid, cysteine protease inhibitor, and is produced at a constant rate by all nucleated cells in the body.[15, 16] As a result, circulating concentrations correlate with GFR. Dogs appear to be similar to humans, whereby circulating CysC concentration might be a better marker of renal functional impairment than serum creatinine concentration.[10, 18, 19] Clu is a glycoprotein which is composed of two 40 kD subunits (NA1, NA2) bound by disulfide groups. Reports in human medicine indicate that Clu is upregulated and released into the urine after nephron damage. Serum clusterin is also increased after renal injury, possibly because it is upregulated. Therefore, circulating clusterin concentrations can also be used as an early marker of renal injury.
The aim of this study was to investigate the possible influence of weight loss, in dogs with naturally occurring obesity, on biomarkers of renal status, as described in humans. As a result, we chose to examine the behavior of three novel biomarkers of renal functional impairment and/or disease (tHcy, CysC, and CLU), in addition to traditional markers of CKD (serum urea and creatinine, USG, UPCR, and urine albumin:creatinine [UAC]) before and after weight loss in dogs with naturally occurring obesity.
- Top of page
- Material and Methods
The current study has investigated putative renal biomarkers in obese dogs undergoing a weight management program, that lead to a marked reduction in body fat mass. Both conventional biomarkers in current clinical use (serum urea, serum creatinine, USG, UAC, and UPCR) and novel biomarkers (tHcy, CysC, and Clu) were assessed before and after weight loss. While it is tempting to speculate that the results might be the result of altered renal function, alternative explanations are possible for many of the changes noted. Most notably, the differences identified might have been the result of other alterations occurring concurrently during the weight loss program, for example the feeding of a high protein diet for weight loss (for urea) loss of lean tissue mass (for creatinine), as described in detail below. As a result, further studies would be required to confirm these findings and determine their significance.
The observed changes in urine biomarkers used in routine clinical practice (USG, UPCR, and UAC) with weight loss could imply improved renal function, through an increase in tubular concentrating ability (increased USG) and a decrease in protein filtered by the glomerulus (decreased UPCR and UAC). Experimentally-induced obesity in dogs is known to alter renal function (eg, glomerular hyperfiltration with an associated increase in GFR) and cause histologic changes such as expansion of Bowman's capsule, cell proliferation in the glomeruli, thickening of glomerular and tubular basement membranes, and increased mesangial matrix. Similar changes could be the reason why USG was less (eg, due to an increase in GFR) and UPCR and UAC more commonly abnormal (eg, due to the glomerular lesions causing protein leakage) for the dogs of the current study, when in an obese state. Other diagnostic modalities such as kidney biopsy could have helped to determine the significance of the changes in this study. However, since this study was performed under clinical conditions it was not ethically possible to perform invasive procedures such as serial kidney biopsy in client-owned dogs.
Although the concentrations of urea and creatinine were within laboratory reference intervals, an increase in serum urea and a decrease in serum creatinine were observed after weight loss. The increase in plasma urea concentration in obese dogs undergoing weight loss has been previously reported in some, but not all previous studies. The reason for such an increase is not clear, but it might either have resulted from a decrease in GFR as a result of weight loss, or from feeding a high-protein weight loss diet. Increased urea concentration after weight loss has also been described in obese humans consuming a moderate protein diet for weight loss, but not those receiving a high carbohydrate diet.[36, 37] Measurement of GFR before and after weight loss might have helped to differentiate between these possibilities. The decrease in serum creatinine concentration is contradictory to previous studies in obese dogs, whereby increased serum creatinine concentration has been documented following weight loss.[34, 35] This decrease could be the result of loss of muscle mass during weight management. If a genuine decrease in GFR were to have occurred with weight loss, as suggested by the increase in urea concentration, it could counteract the effect of lean tissue loss on serum creatinine concentration.
Systolic blood pressure (SBP) is another clinical parameter of critical importance in CKD, and is used in an internationally accepted clinical staging scheme. Whilst canine obesity is associated with hypertension,[38, 39] the effect is relatively minor and does not usually warrant antihypertensive therapy. Although the focus of the current study was to assess urinary and plasma biomarkers, SBP has been measured in another study from our clinic, which included dogs from the current study. As with previous work, SBP is often marginally increased in the obese state, and decreases significantly after successful weight loss. These findings support the possibility of renal structural and functional changes in canine obesity.
The current study also investigated 3 novel biomarkers putatively associated with renal functional impairment or disease: tHcy, CysC, and Clu. Increased tHcy concentration has been associated with renal disease in dogs and humans whilst, in humans, moderate hyperhomocysteinemia has been noted in early stages of chronic renal failure, becoming more prominent as renal function deteriorates. Furthermore, other human studies have identified greater tHcy concentrations in overweight and obese patients when compared with normal weight patients. Thus, the decrease in tHcy concentration after weight loss in the obese dogs of the current study might indicate altered renal structure or function in the obese state with subsequent improvement with weight loss. However, given that dietary folate intake can influence plasma tHcy concentration, folate should ideally have been measured in all study dogs. This is a limitation of the current study and further assessment is required. However, in a previous human study, changes in folate concentration did not influence serum tHcy concentration in weight loss programs.
Dogs are similar to humans in that circulating CysC concentration is reportedly a better and more accurate marker of renal function than creatinine or creatinine-based equations. This is thought to be because the influence of nonrenal factors (such as body composition, age, gender, or dietary protein intake) on circulating CysC concentration is less than for creatinine concentration.[10, 15, 16, 19, 43, 44] Further, given that increases in circulating CysC concentration with progressive renal compromise parallel one another in obese and nonobese humans, changes in CysC concentration are thought to reflect renal function whatever the degree of obesity.[45, 46] However, circulating CysC is consistently increased in obese subjects independent of GFR, and adipose tissue expression of CysC is increased in the obese state. This suggests that adipose tissue might contribute directly to circulating CysC concentration through increased adipose tissue synthesis in the obese state. Given that it is not currently known whether canine adipose tissue can synthesize CysC, it is feasible that the decrease in plasma CysC concentration is either the result of improved renal function or to decreased synthesis of CysC by adipose tissue after weight loss. Also, with regard to CysC, it should be emphasized that the median change from baseline concentrations was only 18%, and it is not known as to whether or not such a change would be clinically important. Further studies would be recommended to determine the reasons for and significance of such changes.
Weight loss also resulted in a decrease in plasma Clu concentration, which could have been the result of 2 different mechanisms: (1) improvement in renal injury, given that increased serum Clu concentration is seen after nephron damage in humans; (2) improvement in lipid profile after successful weight loss, since circulating Clu reportedly acts as an apolipoprotein, by partially associating with high density lipoprotein (HDL). In dogs, HDL is the predominant lipoprotein and this decreases after weight loss. Furthermore, a positive correlation between changes of Clu with body fat mass has been described in humans independent of age, gender, HbA1c, and fasting plasma insulin concentrations. It is again advisable to consider further studies to determine the true significance of this finding.
The official National Institute of Health definition of a biomarker is “a characteristic that is objectively measured and evaluated as an indicator of normal biologic processes, pathogenic processes, or pharmacologic responses to a therapeutic intervention”. Since obesity is easy to diagnose, requiring only a physical examination and body condition score to be performed, the authors believe that biomarkers are applied to assessing either disease associations or in predicting outcomes of weight loss. Such information might help when advising owners of likely weight loss outcomes, and might, ultimately, enable the clinician to tailor their weight loss therapy accordingly. For this reason, we used simple and multiple regression analysis to determine whether the biomarkers studied, when measured prior to weight loss, could predict either rate of weight loss or change in lean tissue. In this respect, preservation of lean tissue mass during weight loss is a key outcome for successful obesity management and, in humans, loss of skeletal muscle mass is correlated with physical impairment and disability as well as being associated with an increased incidence of death. In the present study, Clu concentration before weight loss correlated with amount of lean tissue lost during subsequent weight management (greater Clu, more lean tissue loss). The reasons for this association are not clear but, in humans, widespread upregulation of Clu gene expression and protein synthesis is seen in diseases where either abnormal cell death or proliferation occurs, including atherosclerosis, myocardial infarction, and muscle damage.[54, 55] Thus, it could be hypothesized that Clu acts as a surrogate marker of deranged metabolic function in canine obesity, and this improves after successful weight loss. Whatever the reason, this intriguing study finding raises the possibility that Clu concentration before weight loss could be used as a biomarker to identify those dogs at risk of excessive loss of lean tissue mass.
The main limitation of this study would be that the studied animals were a population of client-owned dogs with variable living conditions, family environment, husbandry, and medical care. This made it impossible to evaluate influence of changes in diet composition on studied analytes, to perform GFR measurements, or to perform renal biopsies in order to evaluate the weight loss on histologic changes in the kidneys. Nonetheless, the results are arguably more representative of the true clinical picture since the obesity is naturally occurring. On a related note, a 2nd limitation was the fact that a control group of healthy dogs was not included. The main reason for this was that our institution's ethics committee would not have allowed it. Although, on the face of it, blood and urine sampling a group of healthy pet dogs would be straightforward, in the UK there needs to be a clear benefit to the animal in order to justify such a procedure or it is classed as an experimental act. An alternative would have been to consider utilizing surplus frozen plasma from a ‘hospital’ control population of dogs, ie, dogs referred for reasons other than obesity and renal disease. Whilst this would certainly have circumvented the ethical dilemma, such an approach would have created concerns of its own. In this respect, a DEXA scan would arguably have been required to ensure that their weight status was known, which would again not have been allowed by our ethics committee. Further, it would have been difficult to guarantee that such controls were free from subclinical renal disease, since their underlying disease might arguably have caused such problems. This would then have meant that more detailed (but invasive) tests would have been required such as measurement of GFR and/or renal biopsy. Once again, such tests are invasive and would not be allowed by our ethics committee. Given these limitations, we chose to take the alternative approach of using the enrolled dogs as their own controls, ie, by assessing them before and after weight loss.
A 3rd limitation was the fact that an a priori power calculation was not performed and, as a result, raising the possibility that genuine findings might have been missed. Against this, however, a number of highly significant differences were identified. This suggests that the study had adequate power to detect the major differences of clinical importance.
A final limitation was the fact that the study mainly studied circulating biomarkers so that metabolic effects of obesity and subsequent weight loss could be assessed in addition to renal effects. Nonetheless, for future studies, it would be of interest to evaluate urinary changes of these analytes since, recently, assays for cystatin C and clusterin measurements in dog urine have been validated,[56, 57] and seem to have a high sensitivity for evaluating kidney function.
In summary, the current study has demonstrated changes in a variety of renal function biomarkers in obese dogs undergoing weight loss. Although these results might suggest possible subclinical alterations in renal function in canine obesity, other explanations are possible for many of the changes observed. Therefore, further studies are now necessary to determine whether renal functional changes or injury occur in canine obesity and whether or not they improve after weight loss. Finally, the reason for the association between plasma clusterin before weight loss and subsequent lean tissue loss during weight management is intriguing, and warrants additional investigation.