Prognostic value of serum cystatin C concentration in dogs with myxomatous mitral valve disease

Abstract Background Impaired renal function is 1 of the poor prognostic factors in dogs with myxomatous mitral valve disease (MMVD). However, the value of cystatin C (Cys‐C), a marker of renal function, as a prognostic marker for MMVD in dogs has not yet been explored. Objective This study aims to investigate the prognostic value of Cys‐C in dogs with MMVD. Animals Fifty client‐owned small‐breed dogs with MMVD were included in this study. Methods This is a retrospective, cross‐sectional study. The prognostic value of serum Cys‐C concentration was assessed using univariable and multivariable Cox hazard regression analyses. Kaplan‐Meier survival curves for MMVD‐specific survival in dogs stratified into high and low Cys‐C groups were generated and analyzed using the log‐rank test. Results Serum Cys‐C concentrations were significantly associated with MMVD‐related death (P < .01) in both univariable (hazard ratio [HR], 5.086; 95% confidence interval [CI], 1.950‐13.270) and multivariable Cox hazard regression analysis (HR, 4.657; 95% CI, 1.767‐12.270). The high Cys‐C group (n = 14) had a significantly shorter MMVD‐specific survival time than the low Cys‐C group (n = 36; P < .01). In dogs with normal blood creatinine concentrations, the high Cys‐C group (n = 10) had a significantly shorter MMVD‐specific survival time than the low Cys‐C group (n = 36; P < .01). Conclusions and Clinical Importance High serum Cys‐C concentrations were associated with a worse prognosis of MMVD. Furthermore, serum Cys‐C could be a predictor of MMVD prognosis even in dogs with normal blood creatinine concentration.


| INTRODUCTION
Myxomatous mitral valve disease (MMVD) is the most common heart disease in small-breed dogs and can result in death because of pulmonary edema, syncope, and dyspnea caused by left ventricular volume overload. 1 The severity of MMVD is classified by the American College of Veterinary Internal Medicine (ACVIM) stages. The ACVIM stage was related to the prognosis of dogs with MMVD, and the higher the stage, the worse the prognosis. 1 Renal function is 1 of the important prognostic factors for dogs with MMVD. 1,2 Renal dysfunction could be a negative prognostic factor in dogs with MMVD. 2 The most common marker of renal function is serum creatinine (Cr), and the survival time of dogs with MMVD and high serum Cr concentrations is shorter than that of dogs with MMVD and low serum Cr concentrations. 2 However, serum Cr concentrations do not increase until the glomerular filtration rate (GFR) decreases by 75%. 3 Furthermore, their serum Cr concentrations tend to remain normal because smallbreed dogs have less skeletal muscle mass. 4 In addition, dogs with severe MMVD often develop cardiac cachexia, which can lead to lower serum Cr. Therefore, more sensitive markers that are independent of nonrenal factors and can detect renal dysfunction and prognosis earlier than serum Cr in dogs with MMVD are needed.
Serum cystatin C (Cys-C) is used in dogs as a GFR biomarker. 5,6 In humans, serum Cys-C is a better GFR marker for renal disease than serum Cr. [7][8][9][10] Serum Cys-C concentrations are significantly higher in dogs with renal failure than in clinically healthy individuals. 11,12 Despite that it is only useful in small-breed dogs, serum Cys-C is a better GFR marker than serum Cr. 13,14 Thus, serum Cys-C could be a promising sensitive renal marker in small-breed dogs. In humans, serum Cys-C concentrations are linked with prognosis not only in patients with renal insufficiency but also in patients with heart failure.
Higher serum Cys-C concentrations have been detected in human patients with heart disease and have been linked to heart failure and a poor prognosis, particularly in those with coronary artery disease. [15][16][17][18][19][20] Furthermore, serum Cys-C concentrations in human patients detect a poor prognosis of heart disease earlier than serum Cr concentrations. 17 However, the prognostic value of Cys-C in dogs with heart disease has not yet been reported. The present study aims to evaluate the prognostic value of serum Cys-C concentrations in dogs with MMVD.

| Case selection
This is a retrospective cross-sectional study. The data were acquired from the medical records of dogs diagnosed with MMVD at a primary care veterinary hospital between February 2015 and April 2021. Dogs weighting ≥15 kg were excluded from this study because serum Cys-C concentration is an inferior renal marker in larger breed dogs. 13 Of the dogs included, those with missing data required for this study were excluded from the study. Dogs who had received oral administration of prednisolone within the previous month were excluded because oral administration of prednisone increases serum Cys-C concentrations. 21  guidelines, the MMVD stage was classified as B1, B2, C, or D. 1 Survival was followed up, and dates of MMVD-related deaths and deaths from other causes were recorded. MMVD-related deaths were defined as deaths occurring as a result of progression of MMVD evidenced by 1 or more of pulmonary edema, syncope, or dyspnea. This study was approved by the local ethics committee for animal clinical research (approval no. E22002). Informed consent by dog owners was waived because of the study's retrospective nature. All dog owners were given the option to opt out of the present study, which was conveyed via the animal hospital's bulletin board.

| Echocardiography
The LA/AO ratio and LVIDDN were measured using echocardiographic images. M-mode, Doppler, and 2-dimensional (2D) echocardiography was performed by a veterinarian using an ultrasound unit (Noblus, FUJI-FILM Healthcare, Tokyo, Japan) with a 2.0-to 9.0-MHz sector probe (S-31 Probe, FUJIFILM Healthcare). During the examination, dogs were restrained on the right lateral recumbency. The LA/AO ratio was calculated from the right parasternal short-axis 2D view on the first frame after the aortic valve was closed. 1,25,26

| Blood biochemistry
Data on blood biochemistry was obtained during the initial visit. Blood was collected from the cephalic vein and delivered into plain, heparin, and aprotinin tubes. After allowing the samples in plain tubes to clot,

| Serum Cys-C concentration and prognosis of MMVD
The median follow-up period was 358 days (range,    Table 2). Furthermore, multivariable Cox hazard regression analysis with age as a confounding factor revealed that high serum Cys-C concentration (P < .01) remained significantly associated with MMVDrelated death (Table 3). Cys-C group had significantly shorter MMVD-specific and all-cause survival times than the low Cys-C group (P < .01; Figure 2). Even in dogs with normal blood Cr concentrations, the high Cys-C group had significantly shorter MMVD-specific and all-cause survival times than the low Cys-C group (P < .01; Figure 3).

| DISCUSSION
In this study, the high Cys-C group had a shorter MMVD-specific survival time than the low Cys-C group. Thus, high serum Cys-C concentrations in dogs with MMVD are suggested to be associated with MMVD-related deaths and poor prognosis for MMVD. This study establishes that a high serum Cys-C concentration is a negative prognostic factor for MMVD in dogs.
Serum Cys-C is a marker of renal function that reflects GFR. 13 Therefore, the current findings imply that impaired renal function might be associated with MMVD progression. Previous reports have also indicated that dogs with MMVD and chronic kidney disease (CKD) have a worse prognosis than those without CKD. 2 Cardiovascular-renal disease (CvRD), a condition in which both cardiac and renal dysfunction negatively affect each other, 2,32 is caused by sympathetic activation, renin-angiotensin-aldosterone system activation, hypertension, and GFR reduction. 32  CvRD. It is unknown whether Cys-C is associated with causes of death other than MMVD and kidney disease. However, only 6 dogs died from other causes and in these 6 dogs, serum Cys-C concentrations were low (data not shown). Therefore, it is unlikely that serum Cys-C concentrations are associated with causes of death other than MMVD and kidney disease.
In this study, high serum Cys-C concentrations tended to worsen MMVD prognosis even in dogs with normal blood Cr concentrations.
These findings are consistent with those reported in human cardiac patients, where high serum Cys-C concentrations were associated with a poor prognosis of cardiac disease even with normal serum Cr concentrations. 17,33 In small-breed dogs, serum Cys-C is more sensitive than serum Cr in identifying reduced GFR. 13 In dogs with normal Cr but high Cys-C, renal function was impaired, which might have contributed to a worse prognosis of MMVD. Thus, even in dogs with normal Cr, it could be useful to measure Cys-C to evaluate the prognosis of MMVD.
T A B L E 2 Results of univariable Cox proportional hazards analysis on MMVDrelated deaths. There is a correlation between MMVD severity and the International Renal Interest Society stage of CKD in dogs. 2 In addition, the higher the MMVD stage, the higher the serum Cr and Cys-C concentrations in dogs. 34 However, in this study, no significant differ- prognostic factor for cardiac disease. 16,17 Serum Cys-C, like in humans, has been suggested to be an independent predictor of MMVD prognosis in dogs.
In this study, the high Cys-C group was significantly older than the low Cys-C group. Previous studies have discovered that serum Cys-C concentrations correlate with age, 30,31,[35][36][37] and that aging reduces renal function while increasing Cys-C concentrations in humans. [35][36][37] Moreover, Cys-C concentrations are higher in elderly dogs than in younger ones. 30,31 Therefore, in this study, considering the possibility that older age might have contributed to the relationship between high Cys-C concentrations and a poorer prognosis of MMVD was necessary. Multivariable analysis using age as a confounding factor revealed that serum Cys-C concentration remained significantly associated with MMVD-related death. These findings T A B L E 4 Comparison of variables between the low and high Cys-C groups.  imply that the poor MMVD prognosis in dogs with elevated serum Cys-C concentrations was independent of age.
This study has some limitations. First, a limited number of samples exist. However, the prognosis between the high and low Cys-C groups differed markedly. Therefore, the present study's sample size was sufficient to demonstrate the difference. Second, data on GFR measurements were missing because of the retrospective nature of the study. If GFR had been assessed in this study, it might have demonstrated that the poor prognosis caused by elevated serum Cys-C concentrations was attributed to impaired renal function. Third, data on body condition scores and feeding status were missing. Obesity might affect serum Cys-C concentrations. 38 Although Cys-C is unaffected by diet in cats 39 and humans, 40,41 it can be affected by diet in humans. 42,43 Therefore, the possibility that the degree of obesity or the interval between feedings altered serum Cys-C concentrations in the present study cannot be ruled out.
In conclusion, the present study demonstrated that serum Cys-C was a prognostic factor for poor outcome for MMVD in small-breed dogs. High serum Cys-C concentrations were associated with a worse prognosis regardless of MMVD severity. Furthermore, serum Cys-C could be a predictor of MMVD prognosis even in dogs with normal blood Cr.

ACKNOWLEDGMENT
No funding was received for this study.

CONFLICT OF INTEREST DECLARATION
Validation data on measurement of canine serum cystatin C level was provided by FUJIFILM VET Systems Co, Ltd (Tokyo, Japan).

OFF-LABEL ANTIMICROBIAL DECLARATION
Authors declare no off-label use of antimicrobials.

INSTITUTIONAL ANIMAL CARE AND USE COMMITTEE (IACUC) OR OTHER APPROVAL DECLARATION
This study was approved by the ethics committee for animal clinical research of Gifu University (approval no. E22002). Informed consent by dog owners was waived because of the study's retrospective nature. All dog owners were given the option to opt out of the present study, which was conveyed via the bulletin board in Hashima Animal Hospital.

HUMAN ETHICS APPROVAL DECLARATION
Authors declare human ethics approval was not needed for this study.