Response and survival of dogs with proteinuria (UPC > 2.0) treated with angiotensin converting enzyme inhibitors

1 INTRODUCTION

Proteinuria is a hallmark of glomerular disease and when the urine protein-creatinine ratio (UPC) is >2.0 in the absence of pre- and post-renal causes, glomerular disease is often present.¹ While renal biopsy is required for a definitive pathological diagnosis thereby informing treatment choices, it is infrequently performed, and when the UPC is >2.0, a presumptive diagnosis of glomerular proteinuria is usually reached by noninvasive exclusion of pre- or post-renal causes of proteinuria.

Glomerulopathies are a major cause of chronic kidney disease (CKD) and eventual renal failure in dogs,² with proteinuria per se likely contributing to progressive renal disease.^3-5 Even in the absence of a histopathological diagnosis, management of cases with presumed glomerular proteinuria is aimed at reduction of proteinuria and management of clinically relevant complications.^{6, 7}

The 2013 ACVIM Consensus statement for standard therapy of glomerular disease in dogs states modification of the renin-angiotensin-aldosterone system (RAAS) should be standard of care for dogs with glomerular proteinuria and this is often achieved by use of an angiotensin converting enzyme inhibitor (ACEi).⁷ Benazepril and enalapril are commonly used ACEi. The proposed target of ACEi therapy is either a reduction in UPC to <0.5 or a >50% reduction from baseline.⁷ However, there is currently limited data on how frequently dogs treated with ACEi meet such treatment targets and whether achieving these targets improves disease progression and overall survival.

The aims of this retrospective study were to determine the proportion of dogs with a UPC > 2.0 achieving the proposed target reduction in UPC within 3 months of starting an ACEi and determine whether a positive response to treatment was associated with improved survival. To investigate for possible confounding factors, secondary aims were to evaluate whether clinicopathological markers of disease severity in the 30-days before treatment commencement had an association with either response to treatment or survival.

It was hypothesized that response to ACEi therapy within 3 months would be associated with improved 12-month survival.

2 MATERIALS AND METHODS

Clinical records of dogs referred to a University Teaching Hospital were searched to identify those with a UPC > 2.0 prescribed an ACEi for treatment of proteinuria between January 2006 and April 2021. Dogs were excluded if they were prescribed ACEi for any other condition; were receiving concurrent therapy known to affect renal protein loss (tyrosine kinase inhibitors, corticosteroids, or angiotensin receptor blockers [ARBs]) at the time of baseline urinalysis or did not have a baseline urinalysis and biochemistry performed within 30-days before starting ACEi therapy. UPC was measured as standard at the laboratory performing urinalysis. Dogs with incomplete records were not included. Additionally, those without at least 1 follow-up urinalysis (including UPC measurement) at the university laboratory within 3 months of starting ACEi therapy were excluded. Screening for infectious diseases was not required for inclusion as the prevalence of infectious diseases that can contribute to proteinuria is low in the UK. Cases with an active sediment on urinalysis or co-morbidities were not excluded.

Medical records were reviewed and signalment and baseline variables recorded. Baseline variables were defined as the creatinine, albumin, cholesterol, UPC, and systolic blood pressure (SBP) values obtained within 30 days before ACEi commencement. If several values were available within this time frame, the ones closest to the start of ACEi treatment were used. The choice and starting dose of ACEi (mg/kg/day), concurrent comorbidities and diagnoses were recorded, as were medications, dietary management and omega-3 fatty acid supplementation already being administered or started at the time of starting ACEi therapy. For clinical threshold analysis, azotemia was defined as a creatinine >1.4 mg/dL (>125 μmol/L); hypoalbuminemia an albumin <2.5 g/dL (<25 g/L) and hypercholesterolemia as cholesterol >348 mg/dL (>9 mmol/L).^{8, 9} Severe proteinuria was defined as a UPC > 3.5.¹⁰ Using the ACVIM Consensus Guidelines for Systemic Hypertension, normotensive or pre-hypertensive dogs were classed as “non-hypertensive” (SBP < 159 mm Hg) while those with hypertension or severe hypertension were classed as “hypertensive” (SBP ≥ 160 mm Hg).¹¹ The following data were collected from subsequent re-checks: date, UPC, new dose of ACEi if changed and whether ARBs or corticosteroids were started.

Dogs were categorized as responders or non-responders. Treatment response was defined as a reduction of UPC to ≥50% from baseline or to <0.5⁷ at any follow-up visit within 3 months of starting ACEi. If dogs did not achieve these targets by 3 months, they were classed as non-responders. Dogs were also classed as non-responders if they were started on an ARB or corticosteroid within 3 months and before the UPC targets being achieved or if ACEi therapy was withdrawn because of progressive azotemia or hyperkalemia at any time during the first 3 months before the UPC targets being achieved.

To perform survival analyses, the date of death/euthanasia, or the point at which the patient was lost to follow-up (last known to be alive) was recorded. If available, the reason for death/euthanasia was recorded. Dogs were then additionally classified based on their status 12 months after initiation of ACEi as either alive, dead, or lost to follow-up.

3 RESULTS

3.4 Outcome

3.4.1 Association between treatment response and survival

There was a statistically significant difference in MST between responders and non-responders (664 [95% CI: 459-869] vs 177 [95% CI: 131-223] days, respectively, P = .009; Figure 1). Of the 68 dogs with a known outcome at 12 months, 41% (n = 28/68) were responders. Of these, 79% (n = 22/28) were alive at 12 months while 28% (n = 11/40) of non-responders were alive at 12 months; this difference was statistically significant (χ²₁ = 17.2, P < .001), with the odds of death greatly reduced (0.1) in responders (Table 2).

TABLE 2. Odds ratios from the univariable analysis of the association between response to treatment with an angiotensin converting enzyme inhibitor and baseline clinicopathological parameters and death by 12 months in a cohort of 85 dogs with a UPC >2.0.

Variable	OR	OR CI	P value
Either UPC target achieved within 3 months	0.10	0.03-0.31	<.001
Baseline creatinine
Values (mg/dL)	2.53	1.38-5.61	.008
Azotemic	5.34	1.85-17.32	.003
Baseline albumin
Values (g/dL)	0.40	0.18-0.82	.02
Hypoalbuminemic	4.51	1.66-13.14	.004
Baseline cholesterol
Values (mg/dL)	1.00	1.00-1.00	.74
Hypercholesterolemia	0.95	0.36-2.51	.92
Baseline urine protein-creatinine ratio (UPC)
Values	1.18	1.05-1.34	.01
Severe proteinuria (UPC > 3.5)	2.91	0.84-11.79	.11
Baseline systolic blood pressure
Hypertensive (systolic blood pressure ≥ 160 mm Hg)	2.36	0.65-9.05	.2

• Note: Treatment response was defined as achieving a UPC of <0.5 or reduction of ≥50% from baseline within 3 months of starting treatment. Bold values represent statistically significance.
• Abbreviations: OR, odds ratio; OR CI, odds ratio 95% confidence interval; P value, associated statistical significance; UPC, urine protein-creatinine ratio.

3.4.2 Association between clinicopathological variables and hypertension on survival

The MST of dogs that were or were not azotemic at the time of diagnosis also differed significantly (175 [95% CI: 62-288] vs 586 [95% CI: 398-773] days, respectively), P < .001; Figure 2). While dogs that were hypoalbuminaemic at baseline had a shorter MST than those that were not (177 [95% CI: 115-239] vs 531 [95% CI: 373-689] days, respectively), this was not statistically significant (P = .06). There was no statistically significant difference in the MST for dogs that had hypercholesterolemia at baseline compared to those that did not (375 [95% CI: 28-722] vs 412 [95% CI: 145-679] days, respectively, P = .24) nor those with severe proteinuria compared to those that did not (300 [95% CI: 63-537] vs 531 [95% CI: 403-659] days, respectively, P = .92) or hypertension (hypertensive—251 [95% CI: 0 = 532] vs normotensive—454 [95% CI: 240 = 668] days, P = .37).

When assessed as continuous variables, creatinine, albumin, and UPC were found to be statistically significantly associated with 12-month outcome while cholesterol was not (Figure 3A-D). Furthermore, logistic regression estimated 2.53 and 1.18 increases in the odds of death with each 1.1 mg/dL (100 μmol/L) increase in creatinine and each unit increase in UPC, respectively, a reduction in odds (0.40) with each 0.1 g/dL increase in albumin and no change in odds associated with increasing cholesterol (1.00; Table 2). Qualitatively similar results were obtained when creatinine, albumin, cholesterol, UPC and SBP were assessed in terms of clinical thresholds. The presence of baseline azotemia (χ²₁ = 9.5, P = .003, OR = 5.34; Table 2) and hypoalbuminemia (χ²₁ = 8.7, P = .004, OR = 4.51) were both negatively associated with 12-month survival, whereas there was no difference in survival at 12 months in dogs with or without hypercholesterolemia (n = 17/36; χ²₁ < 0.01, P = .92, OR = 0.95) or those with or without severe proteinuria (χ²₁ = 2.756, P = .11, OR 2.91). Twelve-month survival data were known for 38 dogs with baseline SBP readings available; 60% (n = 12/20) of the dogs that were not hypertensive were alive at 12 months compared to 39% (n = 7/18) of hypertensive dogs. Hypertension at baseline was not associated with 12-month survival (χ²₁ = 1.689, P = .2, OR = 2.36).

To remove any impact of baseline azotemia and hypoalbuminemia on outcome, 12-month survival analysis was repeated on the sub-group of 25 dogs without these abnormalities present. Response to treatment remained significantly associated with survival at 12 months in this cohort of dogs (χ²₁ = 4.425, P = .04, OR = 0.01).

3.7 Categorical tree

Finally, to evaluate the interplay between the presence of a combination of factors in terms of clustering of clinical thresholds and response to treatment in particular dogs, a categorical tree was generated. Although 12-month survival data were available for 68 dogs, only 67 dogs were included in the categorical tree as cholesterol was unavailable for 1 dog. Hypertension was not included in this model as SBP was not available for all dogs. The categorical tree determined the principal factor influencing 12-month survival to be treatment response. The worst outcome was seen in dogs that failed to respond to ACEi within 3 months and that were azotemic; none of these dogs (n = 17) were alive at 12 months (Figure 4, [1]). Conversely, the best outcome was seen in dogs that responded to treatment and which had normal albumin and cholesterol at baseline; all dogs (n = 8) with this combination of factors were alive at 12 months (Figure 4, [2]).

4 DISCUSSION

We set out to evaluate the frequency that ACEi treatment resulted in a ≥50% reduction in UPC or a UPC < 0.5 within 3 months in a cohort of dogs with a UPC > 2.0, and to evaluate whether achieving this target conveyed survival benefit. Although only 41% of our study cohort achieved 1 of the targets for reduction in UPC within 3 months, those that did were significantly more likely to be alive 12 months after starting treatment with responders also having significantly longer MST. The presence of baseline azotemia or hypoalbuminemia, and magnitude of proteinuria, were also associated with a worse outcome, although to a lesser extent.

There are numerous studies in the human literature supporting the use of ACEi in proteinuric renal diseases, including diabetic nephropathy,^{13, 14} non-diabetic nephropathy,^15-18 and IgA nephropathy,^{19, 20} to slow the progression to end-stage renal failure. Similarly, there is evidence within veterinary literature to support their use in the management of proteinuric renal disease. Enalapril reduces proteinuria and slows disease progression in some dogs with idiopathic glomerulonephritis²¹ and delays the increase in serum creatinine and UPC in dogs with X-linked hereditary nephritis.²² What is less clear is whether the benefits of ACEi are associated with the magnitude of reduction in proteinuria. While ACEi reduce proteinuria via their effect on glomerular hemodynamics,^23-25 they also lead to a decrease in the production of vasoactive substances implicated in development of glomerulosclerosis, delay the growth and proliferation of mesangial cells, and reduce the degradation of bradykinin, all of which could influence the rate of disease progression independent of the magnitude of reduction of proteinuria.^{24, 26-30} Given that dogs achieving a 50% reduction in UPC were 10 times more likely to be alive at 12 months compared to those that did not and had a significantly longer MST, it appears that outcome is associated with the magnitude of reduction in UPC in response to ACEi. This strong association supports the currently recommended target of >50% reduction in UPC, justifying both the time and financial commitments needed to achieve this target.

Treatment response was not the only variable found to be associated with survival. The presence of baseline azotemia, hypoalbuminemia, and magnitude of baseline creatinine, albumin, and UPC were all negatively associated with 12-month survival. This is unsurprising as increasing IRIS stage, the presence of nephrotic syndrome and increased UPC are negative prognostic indicators in a variety of renal diseases of dogs.^{12, 31-34} Similarly, in human IgA nephropathy, baseline creatinine and magnitude of proteinuria are independent predictors of progression to end-stage renal disease.³⁵

While multivariable analysis confirmed that azotemia and hypoalbuminemia were independent risk factors for death before 12 months, response to therapy remained positively associated with 12-month survival in the subset of dogs with less severe disease (ie, with normal creatinine and albumin). Additionally, the categorical tree analysis provided a clear visual representation of the association between the categorical variables assessed and 12-month survival. Although the presence of azotemia had a negative association and lack of hypoalbuminemia a positive association, treatment response ranked highest in its association with survival at 12 months. Therefore, although the presence of markers of disease severity (azotemia and hypoalbuminemia) appears to be negative prognostic indicators, response to treatment is suggested to have the strongest association with 12-month survival and hence these results again support the currently recommended target of a 50% reduction in UPC regardless of disease severity.

When UPC was assessed as a continuous variable it was found to be associated with 12-month survival, however, when it was assessed as a categorical variable the same significance was not found. Severe proteinuria was defined as a UPC > 3.5; as per the ACVIM Consensus Guidelines.¹⁰ Eight percent of dogs had severe proteinuria and hence it possible that this study was underpowered to detect a difference between the 2 groups. The cut-off of 3.5 to define severe proteinuria might also be too low in cases of glomerular protein loss; additional studies are required to further interrogate this cut-off.

As reported in people with proteinuric renal disease³⁶ and in previous studies of dogs,^{21, 37} a variable response to ACEi was observed in our study cohort, with only about 40% of dogs reaching the target UPC. This is unsurprising as ACEi therapy does not address the underlying cause of glomerular disease and treatment with ACEi is unlikely to result in complete resolution of glomerular injury.⁶ Additionally, in our study, inadequate dose escalation could have been a factor in failure to achieve response. Other potential explanations for this inconsistent response include angiotensin converting enzyme (ACE) gene polymorphisms,^38-41 differing etiology for proteinuria,⁴² and disease severity or magnitude of the proteinuria at time of starting ACEi therapy.⁴³ Lack of response to ACEi could also occur because of incomplete suppression of angiotensin II synthesis either via incomplete inhibition of ACE or production via ACE-independent pathways.⁴⁴ Aldosterone breakthrough (increased aldosterone levels despite ACEi treatment) occurs in a subset of human patients.^45-47 While this phenomenon has not been widely investigated in dogs, 34%-59% of dogs with proteinuric CKD receiving RAAS inhibitors demonstrate aldosterone breakthrough; as do 32% of dogs treated with benazepril for cardiac disease.^{48, 49} Aldosterone breakthrough might also account for the subset of dogs that initially responded to treatment but had subsequent increases in UPC that did not meet either UPC target. The variability in response to ACEi and the clear survival benefit associated with achieving target reductions in UPC, support the recent focus on alternative or adjunctive methods of RAAS suppression when ACEi are insufficient.

In humans, combination therapy of an ACEi and ARB is reported to have a possible synergistic effect with a recent meta-analysis review suggesting combination therapy to be both safe and effective.^{50, 51} Combination therapy has not yet been widely studied in veterinary medicine, however, ARBs, specifically telmisartan, are becoming more frequently prescribed and represent an alternative to ACEi treatment. While literature on the use of telmisartan in dogs is sparse, promising data is now emerging. In a cohort of dogs with a UPC of ≥2.0 (if non-azotemic) or ≥0.5 (if azotemic) 3 months after telmisartan treatment 68% had reached target UPC (defined as per the current study) while a UPC of <0.5 was achieved in 9%-14% of dogs.⁵² A UPC < 0.5 was also achieved in 21% of dogs treated with telmisartan for proteinuric CKD.⁵³ The use of telmisartan was directly compared to enalapril with a change in UPC from baseline to day 30 significantly greater in dogs with proteinuric CKD (UPC > 0.5 and azotemic or ≥1.0 if non-azotemic) treated with telmisartan.³⁷

Response to ACEi in this study was not associated with disease severity as defined by the presence of azotemia or hypoalbuminemia. In the human literature, the use of ACEi slows progression of disease in patients with severe renal failure.⁵⁴ Our results suggest that, similarly to human medicine, dogs that are azotemic at the time of starting ACEi might still show a positive response and benefit from improved survival. This has possible clinical relevance as there is often hesitation to start ACEi therapy in dogs with pre-existing azotemia because of concerns that ACEi-associated alterations in renal hemodynamics will lead to worsening of azotemia. Response to ACEi was not found to be associated with the presence of baseline hypertension. When the human literature is reviewed this finding is perhaps not surprising with several studies showing ACEi to have beneficial effects that are only in part because of the reduction in blood pressure.^{15, 16, 36}

This study had several important limitations, including the decision to use a UPC of >2.0 as an inclusion criterion to recruit cases with presumed glomerular protein loss. While UPC of >2.0 has been widely accepted as an indicator of glomerular proteinuria,¹ several more recent studies have reported UPC > 2.0 in dogs with histopathologically confirmed tubulointerstitial disease^{25, 55} and therefore it is possible that some included dogs did not have primary glomerular disease. Similarly, dogs with confirmed glomerular disease can have a UPC < 2.0⁵⁵ and would have been excluded from our study, again potentially influencing our conclusions. Given the retrospective nature of this study, with case enrollment starting from 2006, the presence of glomerulonephropathy could not be definitively confirmed or characterized.

Due to the retrospective nature of this study, dogs did not undergo a standardized work-up and comprehensive screening for potential triggers was not performed in all cases. It is also possible that some dogs could have had pre- or post-renal conditions contributing to their proteinuria. Additionally, the starting dose of ACEi and timing of dose escalations was uncontrolled; a substantial proportion of dogs were started on a dose that was below the recommended starting dose and few dogs had their dose of ACEi increased to the upper limits of the dose range. The median dose at the time of response was 0.52 mg/kg/day which is the current recommended starting dose¹⁶ and median time to response was 33 days. The median maximum dose for dogs that failed to reach UPC targets was 0.55 mg/kg/day (range, 0.24-2.38 mg/kg/day). This may account for the low proportion of dogs reaching UPC targets and it is possible that the numbers of responders would have been higher if dogs had been treated more aggressively and further dose escalations pursued. Dose escalation of ACEi improves response in dogs,³⁷ although not always.²¹ The median starting UPC in the study of dogs that reported an improvement with dose escalation was 2.29, whereas in our study it was 6.61, and this might have increased the likelihood of achieving a treatment response.³⁷ In human medicine, ACEi dose optimization improves outcome.^{57, 58} The cohort of dogs in this study was heterogenous, particularly with respect to their concurrent diagnoses; numerous dogs had a co-morbidity that could have increased their UPC and if such co-morbidities were not well controlled then it is possible that regardless of the efficacy of the ACEi therapy the UPC might have remained abnormally high. Conversely, treatment of underlying diseases could decrease proteinuria independently of ACEi therapy. The study cohort was chosen to reflect the situation commonly seen in clinical practice, where standard therapies (eg, ACEi) for glomerular diseases are often started without performing renal biopsies and in the presence of co-morbidities and whereas, the presence of co-morbidities could also have affected the reliability of survival data, over 85% of the dogs with known cause of death had progression of renal disease listed as at least a contributing factor to the reason behind their euthanasia, supporting this decision.

Concurrent therapy for the treatment of glomerular proteinuria or its complications was also not controlled. Some enrolled dogs were already receiving a prescription renal diet whereas others were started on a renal diet at the time of starting ACEi therapy and others were never fed such a diet. Feeding a renal diet has previously been shown to reduce UPC in dogs with proteinuric CKD.⁵⁹ Omega-3 supplementation, antithrombotic and anti-hypertensive therapy were also not controlled. Whether diet and the aforementioned additional therapies contributed to decreasing UPC and/or improving survival in this study is unknown.

In conclusion, this study suggests response to treatment with ACEi in dogs with a UPC > 2.0 conveys a significant survival benefit, however, achieving target UPCs was only achieved in 40% of dogs on treatment with ACEi. The presence of baseline azotemia, hypoalbuminemia and the magnitude of proteinuria were found to be independent negative prognostic indicators in these dogs.

CONFLICT OF INTEREST DECLARATION

Authors declare no conflict of interest.

OFF-LABEL ANTIMICROBIAL DECLARATION

Authors declare no off-label use of antimicrobials.

INSTITUTIONAL ANIMAL CARE AND USE COMMITTEE (IACUC) OR OTHER APPROVAL DECLARATION

Approved by University of Glasgow, ethical approval reference EA22/21.

HUMAN ETHICS APPROVAL DECLARATION

Authors declare human ethics approval was not needed for this study.