Retrospective evaluation of risk factors for worsening renal function after angiotensin‐converting enzyme inhibitor treatment in dogs

Abstract

Background

Angiotensin‐converting enzyme inhibitors (ACEi) have the potential to cause worsening renal function (WRF). Therefore, reevaluation of renal function is recommended 1‐2 weeks after starting ACEi therapy.

Objectives

To identify risk factors for WRF in dogs receiving ACEi for cardiac diseases, proteinuria, or systemic hypertension.

Animals

A total of 156 client‐owned dogs that received ACEi were included.

Methods

Serum creatinine concentration was determined at the initial presentation and first reevaluation to detect and grade WRF (increase in sCr ≥ 0.3 mg/dL). Grade 1 (nonazotemic), 2 (mild), and 3 (moderate to severe) WRF were characterized by sCr remaining ≤1.6 mg/dL, 1.7‐2.5 mg/dL increase, and 2.6‐5.0 mg/dL increase, respectively. Demographic and serum chemistry data, such as total protein, albumin, blood urea nitrogen, creatinine, symmetric dimethylarginine, glucose, triglyceride, total cholesterol concentrations, and serum electrolyte concentrations at first presentation, were evaluated. Multivariable modeling was performed to identify risk factors for WRF after treatment with ACEi.

Results

Worsening renal function was identified in 27/156 (17%, 95% confidence interval [CI], 0.11‐0.23) dogs after ACEi treatment. It was classified as Grades 1, 2, and 3 in 17, 2, and 8 dogs, respectively. The only significant factors associated with WRF in dogs receiving ACEi were concurrent administration of furosemide (odds ratio, 5.05; 95% CI, 2.05‐12.4; P < .001) and pre‐existing azotemia (odds ratio, 3.21; 95% CI, 1.28‐8.03; P = .01).

Conclusions and Clinical Importance

Although WRF is uncommon and mild, ACEi should be cautiously prescribed in dogs receiving furosemide or those with pre‐existing azotemia.

Abbreviations

ACEi

angiotensin‐converting enzyme inhibitors

BP

blood pressure

BUN

blood urea nitrogen

CHF

congestive heart failure

CI

confidence interval

GFR

glomerular filtration rate

IRIS

International Renal Interest Society

KI

kidney injury

MMVD

myxomatous mitral valve disease

OR

odds ratio

RAAS

renin‐angiotensin‐aldosterone system

RCHF

right‐sided congestive heart failure

RI

reference interval

sCr

serum creatinine

SDMA

symmetric dimethylarginine

TP

time point

UPC

urine protein‐to‐creatinine ratio

WRF

worsening renal function

1. INTRODUCTION

Angiotensin‐converting enzyme inhibitors (ACEi) are frequently prescribed in veterinary medicine for treating cardiac diseases, proteinuria, and systemic hypertension. ¹ , ² , ³ ACEi inhibit the production of angiotensin II by ACE, reducing the resistance of both arteries and veins. This action increases renal blood flow by causing an increased cardiac output and a decreased renovascular resistance. Furthermore, decreased angiotensin II production leads to a decrease in aldosterone production, which enhances sodium and water excretion through the kidneys, thereby reducing venous return to the heart and preload. ⁴

Despite their efficacy and tolerability, unwanted side effects have been reported in dogs receiving ACEi. The main side effect is worsening renal function (WRF) in dogs receiving captopril, benazepril, or enalapril because of insufficient glomerular perfusion pressure as the effect of angiotensin II diminishes, which is generally well tolerated. ⁵ , ⁶ , ⁷ , ⁸ Therefore, checking the hydration status before ACEi administration and evaluating renal function 1‐2 weeks after integrating ACEi into the treatment regimen is recommended because of the potential detrimental effects of the drugs on renal function. ³ , ⁵

While veterinary literature acknowledges the risk of WRF associated with ACEi, information regarding this risk remains limited. Given the potential severe worsening of renal function in some dogs, caution is warranted when using ACEi. However, the specific risk factors for WRF in dogs receiving ACEi remain unknown. Therefore, this study aimed to identify the prevalence and risk factors for WRF in dogs prescribed ACEi for treating cardiac diseases, proteinuria, and systemic hypertension. We hypothesized that WRF occurs more often in dogs with pre‐existing renal impairment or those that are receiving diuretics.

2. MATERIALS AND METHODS

2.1. Study cohort

This retrospective cohort study did not require approval from the Institutional Animal Care and Use Committee. We included 156 client‐owned dogs that received ACEi for cardiac diseases, proteinuria, systemic hypertension, or a combination of the conditions between January 1, 2018 and January 1, 2024 at the Chungbuk National University Veterinary Teaching Hospital. The inclusion criteria were as follows: (a) dogs diagnosed with cardiac disease and prescribed ACEi for treatment (confirmed and characterized with an echocardiogram); (b) dogs prescribed ACEi for antihypertensive treatment; (c) dogs prescribed ACEi for antiproteinuric treatment; (d) record of dose and frequency of ACEi at the time of the initial prescription and at least 1 subsequent reevaluation; and (e) assessment of serum creatinine (sCr) and blood urea nitrogen (BUN) concentrations before the prescription of ACEi and within 1 month of the reevaluation. Dogs were excluded from the study if they had received ACEi before presentation or did not undergo kidney function tests within 1 month after the ACEi prescription.

2.2. Data collection

Data extracted from the medical records at the time of the initial prescription of ACEi included date, signalment, physical examination findings, cardiac disease diagnosis, systolic arterial blood pressure (BP) determined by the Doppler method, urine protein‐to‐creatinine ratio (UPC), urine specific gravity, type and dosage of ACEi administered, and concurrent medications prescribed with ACEi. Fasting serum biochemistry panel data before ACEi administration comprised serum concentrations of creatinine, BUN, symmetric dimethylarginine (SDMA), sodium, potassium, total calcium, chloride, phosphorus, glucose, triglyceride, and cholesterol. Data collected at the first reevaluation after ACEi prescription included physical examination findings, BP, UPC, and serum biochemistry results as determined before ACEi administration.

Two time points (TPs) were defined for data analysis. Time point 1 (TP1) was the time of initial sCr evaluation before ACEi administration, whereas Time point 2 (TP2) was the time of sCr reevaluation within 1 month of ACEi administration. Variables analyzed for the time interval (TP1 to TP2) included time elapsed, daily ACEi dosage, percentage of WRF occurrence after ACEi administration, and data collected from physical examination, BP, UPC, and serum biochemistry results.

The definition (sCr of ≥0.3 mg/dL between TP1 and TP2) and grading of WRF were adapted from the International Renal Interest Society (IRIS) guidelines (http://www.iris-kidney.com/pdf/4_ldc-revised-grading-of-acute-kidney-injury.pdf) for acute kidney injury (KI), which are similar to those used to define WRF in human patients. ⁹ , ¹⁰ Grade 1 (nonazotemic) was characterized by sCr remaining ≤1.6 mg/dL, Grade 2 (mild) by 1.7‐2.5 mg/dL increase, and Grade 3 (moderate to severe) by 2.6‐5.0 mg/dL increase. Worsening renal function was defined and graded identically to acute KI but occurred over a period exceeding 48 hours. Azotemia was defined as an increase in serum BUN (>25 mg/dL), sCr (>1.6 mg/dL) or both concentrations above reference interval (RI).

2.3. Statistical analyses

Statistical analyses were performed using commercially available statistical programs (Prism 6.01; GraphPad Software Inc., La Jolla, California, USA; IBM SPSS Statistics Version 22; SPSS Inc., Chicago, Illinois, USA). The Shapiro‐Wilk test was conducted to determine normal distributions. Categorical data are expressed as frequencies (proportions), and quantitative data are expressed as median (range) because of the non‐normal distribution of many variables. The paired t test (for normally distributed data) or Wilcoxon sign‐rank test (for non‐normally distributed data) was used to compare the quantitative data between the 2 TPs (TP1 vs TP2). Quantitative data were compared between the WRF and non‐WRF groups at hospital presentation (TP1) using Student's t test or Mann‐Whitney U test, whereas categorical data were compared using Pearson's chi‐square or Fisher's exact tests.

After conducting univariable analyses, variables with a significance level of P < .1 were included in subsequent multivariable backward stepwise regression analyses to assess their predictive value for WRF following ACEi treatment. Odds ratios (OR) and 95% confidence intervals (CIs) were estimated using multiple logistic regression analysis, with OR > 1 indicating an increased risk of WRF and OR < 1 indicating a decreased risk. A subset of demographic and clinicopathologic data was selected for their clinical relevance and entered into the multiple logistic regression model. Backward stepwise regression, including variables with entry probability ≤.05 and removal probability ≥.1, was used for modeling. A significance level of P < .05 was considered statistically significant for all analyses. A final multiple logistic regression model was assessed for goodness‐of‐fit with the Hosmer‐Lemeshow test.

3. RESULTS

3.1. Study cohort

Medical records were searched using the following criteria: records of dogs only; data range from January 1, 2018 to January 1, 2024; and discharge summary terms “enalapril,” “benazepril,” or “ramipril.” This search initially identified 287 dogs, of which 131 were excluded for the following reasons: 43 dogs received ACEi before presentation, and 88 dogs did not undergo renal function reevaluation within 1 month after the ACEi prescription. Consequently, 156 dogs were included in this study.

Of the 156 included dogs, 76 (49%) were female and 80 (51%) were male. Among them, 36 were sexually intact (16 females and 20 males). The median age was 11 years (1‐17 years), and the median weight was 4.4 kg (1.6‐59 kg). The included breeds were Maltese (n = 59), Shih Tzu (n = 18), Yorkshire Terrier (n = 12), Miniature Poodle (n = 12), mixed breed (n = 11), Pomeranian (n = 8), Chihuahua (n = 4), Miniature Schnauzer (n = 4), Spitz (n = 4), Miniature Pinscher (n = 3), Cocker Spaniel (n = 3), Golden Retriever (n = 3), Labrador Retriever (n = 2), Japanese Chin (n = 2), and 1 each of 11 other breeds. Indications for ACEi included cardiac disease (113 dogs, 72.4%), proteinuria (35 dogs, 22.4%), and systemic hypertension (22 dogs, 14.1%). Among the dogs with cardiac disease, 102 were diagnosed with myxomatous mitral valve disease (MMVD), 5 with dilated cardiomyopathy, 4 with right‐sided congestive heart failure (RCHF), and 1 each with other cardiac diseases, including mitral valve stenosis, subaortic stenosis, and arrhythmogenic right ventricular cardiomyopathy. Dogs had concurrent diseases, except for the indications for ACEi (tracheobronchomalacia [n = 19], gallbladder sludge [n = 5], pancreatitis [n = 5], canine atopic dermatitis [n = 4], chronic bronchitis [n = 4], intervertebral disc disease [n = 4], idiopathic epilepsy [n = 4], otitis externa [n = 4], primary hyperlipidemia [n = 4], renal calculi [n = 4], bladder cystitis [n = 3], cutaneous adverse food reaction [n = 3], superficial pyoderma [n = 2], diabetes mellitus [n = 3], and hypothyroidism [n = 3]). Dogs received concurrent medications (pimobendan [n = 108], furosemide [n = 44], antibiotics [n = 33], sildenafil [n = 18], steroids [n = 10], spironolactone [n = 7], trilostane [n = 6], insulin [n = 3], desoxycorticosterone pivalate [n = 2], and nonsteroidal anti‐inflammatory drugs [n = 1]) other than ACEi.

3.2. ACEi treatment

Of the 156 dogs receiving ACEi during the study period, 120 received enalapril, 36 received benazepril, and none received ramipril. Information regarding the elapsed time and daily ACEi dosage administered between the study TPs is presented in Table 1. Of the 156 dogs, 133 (85%) received ACEi twice daily, while 23 (15%) received them once daily. Additionally, 108 dogs (69%) received pimobendan, 44 (28%) received furosemide, 7 (4%) received spironolactone, and 18 (18%) received sildenafil. Forty‐four dogs were receiving median (range) furosemide dosage of 4.0 (1.2‐4.3) mg/kg/day. Of the 44 dogs receiving furosemide, 16 showed WRF at TP2. The median (range) of daily furosemide dosage did not different between dogs with WRF (4.0 [2.0‐4.3] mg/kg/day) and those without WRF (3.5 [1.2‐4.3] mg/kg/day; P = .39).

TABLE 1.

Time elapsed, daily dosage of ACEi prescribed, and occurrence of worsening renal function in 156 dogs treated with ACEi parenterally.

	TP1 to TP2
Time elapsed (days)	14 (4‐29)
Daily ACEi dosage (mg/kg/day)	1.0 (0.3‐1.1)
Worsening renal function (n, %)	27 (17%)
	Grade 1: 17
	Grade 2: 2
	Grade 3: 8

Note: Data are expressed as median (range).

Abbreviations: ACEi, angiotensin‐converting enzyme inhibitors; TP, time point.

A median period between TP1 and TP2 was 14 (4‐29) days. Worsening renal function was observed in 27 dogs (17%, 95% CI, 0.11‐0.23) at the time of reevaluation. Worsening renal function was classified as Grade 1 (nonazotemic) in 17 dogs (63% of dogs with WRF and 11% of all dogs), Grade 2 (mild) in 2 dogs (7% of dogs with WRF and 1% of all dogs), and Grade 3 (moderate to severe) in 8 dogs (30% of dogs with WRF and 5% of all dogs). The frequency and severity of WRF are presented in Table 2, and sCr values for each TP in dogs with WRF are illustrated in Figure 1.

TABLE 2.

Clinical data and selected clinicopathological results at hospital presentation (TP1) and first reevaluation (TP2) in 156 dogs treated with angiotensin‐converting enzyme inhibitors.

	TP1		TP2		TP1 vs TP2
	n	Value	n	Value	n	P value
Blood urea nitrogen (mg/dL)	156	22.1 (5.0‐170.9)	156	24.1 (6.5‐283.2)	156	.38
Creatinine (mg/dL)	156	0.9 (0.1‐6.7)	156	0.9 (0.3‐7.9)	156	.24
SDMA (mg/dL)	88	11 (5‐67)	15	9 (5‐21)	12	.30
Sodium (mmol/L)	149	146 (129‐157)	142	145 (121‐153)	137	<.001
Potassium (mmol/L)	148	4.8 (3.1‐6.8)	142	4.9 (3.4‐6.2)	136	.01
Chloride (mmol/L)	146	114 (89‐123)	142	112 (82‐122)	135	<.001
Total calcium (mmol/L)	123	9.8 (5.1‐12.6)	77	9.8 (6.5‐12.0)	68	<.001
Phosphorus (mmol/L)	127	4.0 (2.0‐16.8)	77	3.8 (2.1‐21.3)	70	.18
Glucose (mg/dL)	109	113 (72‐706)	46	116 (69‐398)	41	.88
Total protein (g/dL)	153	6.3 (4.3‐9.0)	138	6.3 (4.5‐8.3)	137	.46
Albumin (g/dL)	153	3.1 (1.6‐4.4)	138	3.0 (1.6‐4.4)	137	.60
Triglyceride (mg/dL)	97	78 (14‐1486)	25	108 (36‐1402)	13	.05
Total cholesterol (mg/dL)	103	215 (64‐502)	26	223.5 (31‐440)	15	.15
UPC	101	1.60 (0.20‐13.70)	36	1.34 (0.21‐12.60)	18	.89
Urine specific gravity	106	1.022 (1.003‐1.046)	30	1.015 (1.003‐1.038)	32	.09

Note: Data are expressed as median (range). The number of dogs with available data is shown for each variable.

Abbreviations: SDMA, symmetric dimethylarginine; TP, time point; UPC, urine protein‐to‐creatinine ratio.

FIGURE 1.

Serum creatinine concentrations (sCr) at the time of hospital presentation (TP1) and first reevaluation (TP2) in 27 dogs treated parenterally with angiotensin‐converting enzyme inhibitors and had worsening renal function (≥0.3 mg/dL increase in sCr between time points). Blue dotted lines indicate Grade 1 (nonazotemic) worsening renal function (n = 17); red solid lines indicate Grade 2 worsening renal function (1.7‐2.5 mg/dL increase in sCr, n = 2); and black solid lines indicate Grade 3 worsening renal function (2.6‐5.0 mg/dL increase in sCr, n = 8).

3.3. Clinical pathology and WRF

Selected clinicopathological results for dogs at TP1 and TP2 are presented in Table 2. Significant changes in serum electrolyte concentrations (sodium, potassium, chloride, and total calcium) were observed between TP1 and TP2. However, other clinicopathological variables (BUN, sCr, SDMA, phosphorus, glucose, total protein, albumin, triglyceride, total cholesterol, UPC), and urine specific gravity did not exhibit significant changes after ACEi administration.

3.4. Univariable analysis

A significant difference was observed in BUN concentration between the WRF and non‐WRF groups (Table 3). Additionally, significant differences were observed in the proportions of pre‐existing azotemia and the use of furosemide and spironolactone between the groups (Table 4). However, no significant difference was observed in the proportion of the type or total daily dose of ACEi.

TABLE 3.

Comparisons of the clinical data and selected clinicopathological results at hospital presentation (TP1) between the non‐worsening renal function and worsening renal function groups after administration of angiotensin‐converting enzyme inhibitors.

	Non‐WRF group		WRF group		P‐value
	n	Value	n	Value
Age (year)	129	11 (2‐17)	27	12 (1‐16)	.14
Body weight (kg)	129	4.23 (1.58‐59.0)	27	5.18 (1.96‐21.00)	.15
Body condition score	129	4 (1‐9)	27	5 (2‐8)	.12
Heart rate (bpm)	129	150 (78‐208)	27	152 (126‐180)	.34
Systolic blood pressure (mm Hg)	128	143.5 (80‐260)	26	132 (86‐210)	.19
Blood urea nitrogen (mg/dL)	129	21.1 (5.0‐170.9)	27	32.2 (6.1‐142.7)	.04
Creatinine (mg/dL)	129	0.9 (0.4‐6.7)	27	1.0 (0.5‐4.5)	.07
SDMA (mg/dL)	81	12 (5‐67)	7	11 (7‐17)	.68
Sodium (mmol/L)	124	146 (129‐157)	25	145 (134‐152)	.10
Potassium (mmol/L)	123	4.8 (3.5‐6.8)	25	4.8 (3.1‐5.7)	.90
Chloride (mmol/L)	122	114 (95‐123)	24	113.5 (89‐122)	.62
Total calcium (mmol/L)	103	9.80 (5.10‐12.6)	20	9.75 (7.40‐11.6)	.66
Phosphorus (mmol/L)	107	4.0 (2.0‐16.8)	20	4.8 (2.3‐12.3)	.12
Glucose (mg/dL)	95	114 (72‐706)	14	110 (90‐126)	.35
Total protein (g/dL)	126	6.3 (4.3‐9.0)	27	6.4 (4.8‐7.7)	.46
Albumin (g/dL)	126	3.1 (1.7‐4.0)	27	3.0 (1.6‐4.4)	.37
Triglyceride (mg/dL)	84	81 (21‐1486)	13	75 (14‐367)	.45
Total cholesterol (mg/dL)	88	215 (75‐502)	15	238 (64‐324)	.72
UPC	90	0.34 (0.20‐13.7)	11	0.55 (0.20‐4.17)	.71
Urine specific gravity	93	1.022 (1.003‐1.045)	13	1.023 (1.010‐1.046)	.85

Note: Data are expressed as median (range). The number of dogs with available data is shown for each variable and timepoints with incomplete datasets.

Abbreviations: SDMA, symmetrical dimethylarginine; TP, time point; UPC, urine protein‐to‐creatinine ratio; WRF, worsening renal function.

TABLE 4.

Comparisons of the proportions of pre‐existing azotemia, drugs and the daily doses of angiotensin‐converting enzyme inhibitors at hospital presentation (TP1) between the non‐ worsening renal function and worsening renal function groups after administration of angiotensin‐converting enzyme inhibitors.

	Non‐WRF group (n = 129)	WRF group (n = 27)	P‐value
	n (%)	n (%)
Pre‐existing azotemia	48 (37%)	18 (67%)	.01
ACEi indications
Cardiac diseases	92 (71%)	21 (78%)	.50
Proteinuria	31 (24%)	4 (15%)	.30
Systemic hypertension	17 (13%)	5 (19%)	.54
Drugs
Pimobendan	88 (68%)	20 (74%)	.55
Furosemide	28 (22%)	16 (59%)	<.001
Spironolactone	3 (2%)	4 (15%)	.02
Sildenafil	13 (10%)	5 (19%)	.20
Trilostane	6 (5%)	0 (0%)	.59
NSAIDs	0 (0%)	1 (3%)	.17
Steroids	8 (6%)	2 (7%)	.68
Antibiotics	24 (19%)	9 (33%)	.09
Insulins	3 (2%)	0 (0%)	1.00
DOCP	2 (2%)	0 (0%)	1.00
ACEi
Enalapril	101 (78%)	19 (70%)	.37
Benazepril	28 (22%)	8 (30%)
ACEi total daily dose
≤0.5 mg/kg/day	25 (19%)	4 (15%)	.58
>0.5 mg/kg/day	104 (81%)	23 (85%)

Abbreviations: ACEi, angiotensin‐converting enzyme inhibitor; DOCP, desoxycorticosterone pivalate; KI, kidney injury; NSAIDs, nonsteroidal anti‐inflammatory drugs; TP, time point; WRF, worsening renal function.

Dogs were further categorized into 2 groups: those with Grade 2 and 3 WRF were classified into the clinically relevant WRF group and the remainder into the non‐clinically relevant WRF group. Therefore, 25 (93%) of 27 dogs with WRF (16 of 17 with Grade 1 WRF, 2 of 2 with Grade 2 WRF, and 7 of 8 with Grade 3 WRF) were classified into the clinically relevant WRF group. Serum BUN and sCr concentrations at TP1 differed significantly between the clinically relevant and non‐clinically relevant WRF groups, as shown in Table S1. The proportion of pre‐existing azotemia, ACEi indication for systemic hypertension, and type of ACEi (enalapril) also differed significantly between the groups, as shown in Table S2.

3.5. Multivariable risk factor analysis

Seven variables at TP1 with adjusted P values < .1 in the univariable analysis were included in building multivariable models to identify risk factors for WRF after ACEi treatment: serum BUN concentration, sCr concentration, serum sodium concentration, pre‐existing azotemia, and concurrent drug administration (furosemide, antibiotics, or spironolactone). Finally, variables entered into the multivariable model were serum sodium concentration, pre‐existing azotemia, and concurrent drug administration (furosemide, antibiotics, or spironolactone; Table 5). Other significant variables (serum BUN and sCr concentrations) were excluded because of significant multicollinearity. After backward stepwise regression, the OR of WRF in dogs receiving ACEi with concurrent administration of furosemide was 5.05 times higher than in dogs without furosemide administration (P = <.001; 95% CI, 2.05‐12.4). Additionally, the OR of WRF in dogs receiving ACEi with pre‐existing azotemia was 3.21 times higher than in dogs without pre‐existing azotemia (P = .01; CI, 1.28‐8.03). The goodness‐of‐fit statistics suggested that the data adequately fit the final model (Hosmer‐Lemeshow test χ ² = 0.66, P = .72).

TABLE 5.

Variables predictive of worsening renal function in the multivariable model in 156 dogs treated with angiotensin‐converting enzyme inhibitors.

Variable	Odds ratio	95% CI lower	95% CI upper	P value
Concurrent administration of Furosemide	5.05	2.05	12.4	<.001
Pre‐existing azotemia	3.21	1.28	8.03	.01

Note: Hosmer‐Lemeshow test χ ² = 0.66, P = .72. Variables considered in the model include serum sodium concentration, pre‐existing azotemia, and concurrent drug administration (furosemide, antibiotics, or spironolactone). Odds ratios from the logistic regression model were provided for the occurrence of worsening renal function after treatment with angiotensin‐converting enzyme inhibitors. Odds ratios >1 indicate an increased risk of worsening renal function, whereas ratios <1 indicate a decreased risk.

Abbreviation: CI, confidence interval.

A further multivariable model was built to identify risk factors for clinically relevant WRF (Grade 2 and 3 WRF) after ACEi treatment (Table S3). Seven variables at TP1 with adjusted P values <.1 in the univariable analysis were included in building multivariable models to identify risk factors for clinically relevant WRF after ACEi treatment: serum BUN concentration, sCr concentration, serum sodium concentration, serum phosphorus concentration, pre‐existing azotemia, angiotensin‐converting enzyme inhibitors indication for systemic hypertension, and type of angiotensin‐converting enzyme inhibitors (enalapril). Variables entered into the multivariable model were serum sodium concentration, serum phosphorus concentration, pre‐existing azotemia, angiotensin‐converting enzyme inhibitors indication for systemic hypertension, and type of angiotensin‐converting enzyme inhibitors (enalapril). Other significant variables (serum BUN and sCr concentrations) were excluded because of significant multicollinearity. After backward stepwise regression, the model showed the OR of clinically relevant WRF in dogs receiving ACEi with pre‐existing azotemia was 14.05 times higher than that in dogs without pre‐existing azotemia (P = 0.01; 95% CI, 1.73‐114) in the model.

4. DISCUSSION

The present study investigated the occurrence and risk factors for WRF in dogs receiving ACEi. In this study, 17% of the dogs receiving ACEi developed WRF, although clinically relevant WRF (Grade 2 and 3) occurred in only 6.4% of the dogs receiving ACEi. Moreover, pre‐existing azotemia and concurrent furosemide administration were associated with a higher risk of WRF in dogs receiving ACEi. Only the presence of pre‐existing azotemia at the time of ACEi prescription was a risk factor for clinically relevant WRF (Grade 2 and 3) in dogs. Therefore, our findings suggest that caution is necessary when administering ACEi to dogs under these conditions.

The occurrence of ACEi‐induced WRF has been reported in previous human studies. ⁶ , ¹¹ , ¹² In addition to the risk associated with furosemide administration, pre‐existing azotemia increased the odds of ACEi‐induced WRF by approximately 3‐fold. However, furosemide administration was not a risk factor when considering only more severe and clinically relevant WRF (Grades 2 and 3) as an outcome. For these clinically relevant WRF grades, pre‐existing azotemia increased the odds of ACEi‐induced WRF by a larger amount of 14‐fold. These findings are consistent with those of a previous study, which found that chronic renal insufficiency is a risk factor for WRF in human patients admitted to the intensive care unit with acute KI while being treated for hypertension and heart failure with ACEi. ¹¹ According to that study, human patients with chronic renal insufficiency experienced a significant increase in sCr concentrations and had a poor prognosis compared to those with normal renal function after receiving ACEi. This phenomenon could be attributed to the reversal of glomerular hyperfiltration. When renal function is compromised in chronic renal insufficiency, the remaining nephrons undergo glomerular hyperfiltration to compensate for the renal insufficiency. However, ACEi administration in this condition induces efferent arteriolar vasodilation and reduces glomerular capillary pressure, resulting in the loss of glomerular hyperfiltration. Consequently, the glomerular filtration rate (GFR) decreases, accompanied initially by an increase in BUN and sCr concentrations, as the compensatory mechanism diminishes. ¹²

According to human studies, an elevation in sCr of approximately 10%‐20% can occur in ACEi‐induced WRF, with stabilization or recovery of sCr observed within the first 2 months because of the renoprotective effect of long‐term ACEi administration. ¹³ However, the present study revealed that all dogs with WRF experienced a rise in sCr concentration exceeding 20%. Furthermore, the presence of pre‐existing azotemia at the time of ACEi prescription was identified as a risk factor for clinically relevant WRF (Grade 2 and 3) after receiving ACEi. Therefore, closely monitoring and addressing WRF in dogs with pre‐existing azotemia is crucial. However, it remains uncertain whether stabilization or recovery of sCr concentration can be achieved through the renoprotective effect in dogs with ACEi‐induced WRF, as observed in human patients.

A secondary hypothesis of the present study was that the probability of developing ACEi‐induced WRF would be high in dogs receiving both ACEi and diuretics. Previous studies have indicated a rare occurrence of azotemia when loop diuretics and ACEi were used concurrently. ³ , ⁷ , ¹⁴ , ¹⁵ However, the present study found that 16 of 44 dogs receiving furosemide were classified into the WRF group, indicating that 36% of dogs receiving both ACEi and furosemide developed WRF. This finding is similar to the results from human studies, where the incidence of WRF was higher in patients receiving both ACEi and diuretics (33%) than in those receiving ACEi alone (2.4%) in cases of hypertension, congestive heart failure (CHF), and diabetes mellitus. ¹⁶

The daily furosemide dosage did not differ between dogs with and without WRF in the present study. This was somewhat surprising because PO furosemide dosage was associated with KI after hospital discharge in dogs treated with furosemide for left‐sided CHF. ¹⁰ The furosemide dosage of our study is consistent with the ACVIM consensus guideline and is similar to a previous study. ¹⁰ , ¹⁷ However, the previous study was focused on the furosemide administration, not on the ACEi co‐administration. In our study, all dogs receiving furosemide had also received with ACEi. Furosemide or ACEi dosages are generally adjusted considering renal function and clinical signs at first reevaluation, so the initial dosage of furosemide might be maintained until TP2 (first reevaluation) in the present study. Therefore, the dosage of furosemide might be not different between dog with WRF and without WRF. Furosemide administration was not a risk factor for clinically relevant WRF in our study. However, we suggest caution when prescribing ACEi and furosemide concurrently since it was a risk factor for any degree of WRF.

The effects of diuretics on WRF can be elucidated by sodium and volume depletion and renin‐angiotensin‐aldosterone system (RAAS) activation. ¹⁸ , ¹⁹ Renal blood flow and sodium homeostasis are autoregulated by angiotensin II. When ACEi inhibits the synthesis of angiotensin II by ACE in human patients experiencing sodium volume loss because of diuretic use, renal perfusion and GFR may be severely reduced. ²⁰ Furthermore, diuretic use can activate RAAS despite concurrent ACEi application in some dogs. ²¹ Despite RAAS suppression, increased angiotensin II and aldosterone may induce nephrotoxic effects through glomerular damage, glomerular dysfunction, intraglomerular pressure elevation, and tubulointerstitial injury. ¹⁹

In a previous clinical trial conducted in human medicine, the incidence of WRF increased when all types of diuretics, including loop diuretics and potassium‐sparing diuretics, were administered with ACEi. ²² In the present study, we hypothesized that all types of diuretics would be associated with the risk of ACEi‐induced WRF in dogs. However, the results regarding spironolactone were inconsistent with those of the human study. While a significant difference in the number of dogs receiving spironolactone was observed between the WRF and non‐WRF groups in the univariable analysis, subsequent multivariable regression analysis failed to confirm any association between spironolactone use and WRF occurrence. This finding is consistent with another study conducted in dogs with MMVD and CHF. ²³ Furthermore, no significant difference in sCr concentration was observed when comparing the group receiving benazepril and spironolactone simultaneously with the group receiving benazepril alone. However, in our study cohort, all dogs receiving spironolactone were concurrently prescribed furosemide. Therefore, even if spironolactone was associated with ACEI‐induced WRF, the effect of furosemide may have masked this relationship.

In our study cohort, 1 dog in the WRF group experienced a significant elevation in sCr concentration after ACEi administration. The sCr concentration at the first assessment (TP1) was 1.5 mg/dL. However, at the second assessment (TP2), occurring 9 days after administering enalapril at 0.3 mg/kg twice daily, the sCr concentration increased to 7.9 mg/dL. This dog, a 1‐year‐old castrated male Labrador Retriever, presented with abdominal distension because of ascites. It was diagnosed with dilated cardiomyopathy, right‐sided CHF, and cardiogenic pulmonary edema and was prescribed furosemide, pimobendan, and digoxin with enalapril as treatments. This dog had both risk factors identified in our study, including concurrent furosemide administration and pre‐existing azotemia. However, another dog in the WRF group also possessed both risk factors, with an average sCr increase of 150%, contrasting with the 427% increase observed in the abovementioned dog. Another risk factor for WRF, such as cardiorenal syndrome caused by dilated cardiomyopathy and RCHF, may have contributed to the significant increase in sCr concentration in this dog. Moreover, as RCHF improved, the dog experienced a reduction in azotemia.

The present study observed significant differences in clinicopathological variables, such as sodium, potassium, chloride, and total calcium concentrations, after ACEi administration. Electrolyte imbalance, particularly hyperkalemia, is usually caused by ACEi administration because ACEi reduces the GFR and prevents potassium excretion. Moreover, ACEi inhibits the sodium and water delivery to the distal nephron, which leads to hyperkalemia when combined with hypoaldosteronism. ²⁴ However, electrolyte imbalances tend to be mild. A human study reported that when ACEi was administered to patients with normal kidney function, potassium concentration rarely increased by more than 0.5 mmol/L. ²⁵ Similarly, in our study, although changes in potassium concentration were observed between TP1 and TP2, these changes remained within the RI. Furthermore, other variables such as sodium, chloride, and total calcium concentrations also showed slight fluctuations within the RI. Therefore, the electrolyte concentration change because of ACEi administration was not clinically significant in dogs.

The beneficial effects of ACEi have been demonstrated in dogs with CHF caused by MMVD and proteinuric renal disease. ⁷ , ²⁶ , ²⁷ , ²⁸ , ²⁹ In addition, ACEi are often prescribed as a first‐line antihypertensive agent in dogs, considering their antiproteinuric effect and the high prevalence of chronic kidney disease in hypertensive dogs. ² Therefore, ACEi are widely used for managing cardiovascular and renal diseases in dogs because of their beneficial effects. ¹⁹ However, there is a relatively mild risk of WRF with ACEi use; in a retrospective study, only 4 (2.8%) of 144 dogs with CHF had their ACEi dose decreased or discontinued because of WRF, although the severity of WRF was not presented. ³ Furthermore, in a prospective study, 2 (10.5%) of 19 dogs with renal proteinuria experienced a >30% increase in sCr after receiving enalapril at the starting dose. ²⁹ These results are consistent with our results, which showed that Grade 3 WRF occurred in only 5% of the dogs receiving ACEi. Therefore, clinicians should weigh these relatively small potential adverse effects against the clear benefits of ACEi when administering ACEi to dogs with risk factors for WRF. The ongoing STOP‐ACEi trial (https://www.hra.nhs.uk/planning-and-improving-research/application-summaries/research-summaries/stop-acei/) in people aims to determine the outcome effect of discontinuing vs continuing ACEi in azotemic patients who might obtain cardiorenal benefit from continuation despite decreases in GFR. A similar trial might be necessary to elucidate the benefit of ACEi administration in dogs with risk factors for WRF.

The main limitation of the present study was its retrospective nature. Owing to the intrinsic nature of this design, we could not apply standardized treatment and reevaluation schedules to the study cohort. Additionally, as this study was conducted at a single veterinary hospital, prescribing practices might have followed the preferences of a few practitioners. Therefore, the ACEi dose might have been reduced empirically, or the type of ACEi might have been prescribed differently, considering its pharmacokinetics in some portion of the study dogs with pre‐existing azotemia. However, when analyzing these factors, the dose or types of ACEi did not affect WRF occurrence (Table 4). Furthermore, according to the IRIS guidelines, acute KI was defined as an elevation in sCr by more than 0.3 mg/dL within 48 hours. However, the time between TP1 and TP2 was within 1 month (median, 14 days) in the present study. Therefore, if WRF had occurred and resolved before TP2 in some dogs, it may have influenced the study results. Besides the medications analyzed, underlying disease progression may have also contributed to WRF development. A published guideline suggests an increase in sCr of >30% from pre‐treatment sCr as an indicator for changing ACEi therapy in dogs. ¹ The criterion has also been suggested in human medicine. ³⁰ In the present study, 25 (93%) of 27 dogs with WRF (16 of 17 with Grade 1 WRF, 2 of 2 with Grade 2 WRF, and 7 of 8 with Grade 3 WRF) met the criterion (>30% change from pre‐treatment sCr), suggesting our definition of WRF (an elevation in sCr by more than 0.3 mg/dL) might be stricter than the criterion (>30% change from pre‐treatment sCr). Therefore, our results could be reliable in clinical practice.

Pre‐existing azotemia and concurrent furosemide administration were risk factors for WRF in dogs receiving ACEi in the present study. Pre‐existing azotemia might be caused by volume depletion related to various etiology including furosemide administration, and thus hydration status could affect GFR and could be a potential cause of WRF after ACEi administration. ²⁰ However, the hydration status was not considered as a factor for WRF because ACEi was cautiously prescribed in dehydrated dogs, although some dogs with pre‐existing azotemia and furosemide administration were inevitably dehydrated at the time of ACEi initiation. In addition, clinical assessment of hydration status is relatively subjective, which limits the clinical conclusions concerning the relationship of WRF and hydration status after ACEi administration. Therefore, we did not include the hydration status as a factor for WRF in the present study, and potential readers should take this account.

In conclusion, while ACEi‐induced WRF is rare and typically mild, ACEi should be cautiously prescribed in dogs receiving furosemide or in those with pre‐existing azotemia. Especially, clinically relevant WRF could occur in dogs with pre‐existing azotemia at the time of ACEi prescription. Furthermore, other unidentified factors might induce unexpectedly severe WRF after ACEi administration in dogs with both risk factors.

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

Authors declare no IACUC or other approval was needed.

HUMAN ETHICS APPROVAL DECLARATION

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

Supporting information

Table S1: Comparisons of the clinical data and selected clinicopathological results at hospital presentation (TP1) between dogs with and without clinically relevant worsening renal function (Grade 2 and 3) after administration of angiotensin‐converting enzyme inhibitors.

Table S2: Comparisons of the proportions of pre‐existing azotemia, drugs and the daily doses of angiotensin‐converting enzyme inhibitors at hospital presentation (TP1) between dogs with and without clinically relevant worsening renal function (Grade 2 and 3) after administration of angiotensin‐converting enzyme inhibitors.

Table S3: Variables predictive of clinically relevant worsening renal function (Grade 2 and 3) in the multivariable model in 156 dogs treated with angiotensin‐converting enzyme inhibitors.

Lee Y, Baek M, Lee D, et al. Retrospective evaluation of risk factors for worsening renal function after angiotensin‐converting enzyme inhibitor treatment in dogs. J Vet Intern Med. 2025;39(1):e17252. doi: 10.1111/jvim.17252