This work was performed at Ghent University in cooperation with Utrecht University and the Ecole Nationale Vétérinaire de Toulouse, and supported by a grant from the Fund for Scientific Research Flanders (FWO-Vlaanderen).
Long-Term Follow-Up of Renal Function in Dogs after Treatment for ACTH-Dependent Hyperadrenocorticism
Article first published online: 30 MAR 2012
Copyright © 2012 by the American College of Veterinary Internal Medicine
Journal of Veterinary Internal Medicine
Volume 26, Issue 3, pages 565–574, May-June 2012
How to Cite
Smets, P.M.Y., Lefebvre, H.P., Meij, B.P., Croubels, S., Meyer, E., Van de Maele, I. and Daminet, S. (2012), Long-Term Follow-Up of Renal Function in Dogs after Treatment for ACTH-Dependent Hyperadrenocorticism. Journal of Veterinary Internal Medicine, 26: 565–574. doi: 10.1111/j.1939-1676.2012.00915.x
- Issue published online: 2 MAY 2012
- Article first published online: 30 MAR 2012
- Manuscript Accepted: 8 FEB 2012
- Manuscript Revised: 20 JAN 2012
- Manuscript Received: 9 JUN 2011
- Glomerular filtration rate;
- Urinary marker
Systemic hypertension and proteinuria are frequent complications in dogs with Cushing's syndrome and do not always resolve after treatment of hypercortisolism. Therefore, dogs with Cushing's syndrome may be at risk for renal dysfunction before and after treatment.
To assess renal function in dogs with ACTH-dependent hyperadrenocorticism (ADHAC) before and after treatment.
A total of 19 dogs with ADHAC and 12 control dogs.
Renal function was assessed before and at 1, 3, 6, and 12 months after treatment. Twelve dogs were treated with trilostane and 7 dogs by transsphenoidal hypophysectomy. Routine renal markers were measured and urinary albumin (uALB), immunoglobulin G (uIgG), and retinol-binding protein (uRBP) were assessed by ELISA. Urinary N-acetyl-β-D-glucosaminidase (uNAG) was determined colorimetrically. All urinary markers were indexed to urinary creatinine concentration (c). Plasma clearance of creatinine (Clcreat), exo-iohexol (Clexo), and endo-iohexol (Clendo) was used to measure glomerular filtration rate (GFR). Data were analyzed using a general linear model.
Serum creatinine and urea concentrations increased post-treatment, but remained within reference ranges. Plasma Clcreat and Clendo were significantly lower post-treatment, whereas Clexo was not different. Urinary protein-to-creatinine ratio (UPC), uALB/c, uIgG/c, and uRBP/c were decreased post-treatment, but at 12 months 5/13 dogs remained proteinuric. Urinary NAG/c did not change significantly.
Conclusions and clinical importance
A decrease in GFR and persistent proteinuria post-treatment may warrant the clinician's attention. Future research including renal histopathology of dogs with persistent proteinuria or low GFR is needed to further assess renal outcome.
plasma clearance of exogenous creatinine
plasma clearance of endo-iohexol
plasma clearance of exo-iohexol
glomerular filtration rate
systolic blood pressure
urinary albumin-to-creatinine ratio
urinary immunoglobulin G-to-creatinine ratio
urinary N-acetyl-β-D-glucosaminidase-to-creatinine ratio
urinary protein-to-creatinine ratio
urinary retinol-binding protein-to-creatinine ratio
urine specific gravity
Cushing's disease, or ACTH-dependent hyperadrenocorticism (ADHAC), is a frequent endocrine disorder in middle-aged and older dogs, and can be treated with trilostane, a reversible inhibitor of cortisol synthesis, or by transsphenoidal hypophysectomy.[2, 3] Up to 86% of dogs with untreated ADHAC have systemic arterial hypertension and 44–46% have proteinuria, which may persist despite successful treatment of hypercortisolism.[4-6] Proteinuria and systemic hypertension are major factors in development and progression of chronic kidney disease (CKD) and are associated with increased morbidity and mortality in dogs with CKD.[7-9] Therefore, dogs with ADHAC may be at risk for renal complications. In human patients with Cushing's syndrome, decreased glomerular filtration rate (GFR), focal segmental glomerulosclerosis, and increased prevalence of nephrolithiasis have been reported.[10-12] However, the effects of spontaneous Cushing's syndrome on renal function are poorly documented in humans and dogs.[13, 14]
Sensitive and direct methods are needed to detect subtle changes in renal function. Measurement of GFR is considered the best overall indicator of renal function. In addition, urinary markers may allow early and site-specific detection of renal dysfunction.[16-19] Urinary markers include high molecular weight (HMW) proteins (eg, immunoglobulin G [IgG]), intermediate weight proteins (eg, albumin [ALB]), low molecular weight (LMW) proteins (eg, retinol-binding protein [RBP]), and urinary enzymes (eg, N-acetyl-β-D-glucosaminidase [NAG]). Glomerular dysfunction causes increased filtration of IgG and ALB, whereas tubular dysfunction leads to increased urinary excretion of RBP and NAG. These markers allow detection of subclinical renal dysfunction secondary to infectious or endocrine disorders in dogs and cats (eg, pyometra,[21, 22] leishmaniasis, and hyperthyroidism). Urine (u) ALB-to-creatinine (c) ratio, uIgG/c, uRBP/c, and uNAG/c were increased in untreated dogs with ADHAC or ACTH-independent hypercortisolism when compared with 12 healthy controls.
This prospective longitudinal study was designed to assess renal function in dogs with ADHAC before and after medical or surgical treatment of hypercortisolism, using routine renal variables as well as GFR, and glomerular and tubular urinary markers.
Materials and Methods
Animals and Study Design
This study was performed at the Faculty of Veterinary Medicine of Ghent University (Belgium) and the University Clinic for Companion Animals of Utrecht University (The Netherlands) between January 2008 and April 2011. After approval by the Ethical Committee of both institutions (EC2008/066 and 2008.III.06.052) and informed owner consent, 19 client-owned dogs with ADHAC were included in the study. Cushing's disease was suspected based on history, clinical signs, physical examination, biochemical changes, and a result consistent with hypercortisolism on at least one of these screening tests: low dose dexamethasone suppression test (LDDST, cortisol concentration in 8-hour blood sample >40 nmol/L or 1.4 μg/dL) urinary corticoid-to-creatinine ratio (UCCR) >8.3 × 10−6 in 2 consecutive morning urine samples or an ACTH stimulation test (post-ACTH cortisol concentration >600 nmol/L or 22 μg/dL). Diagnosis of ADHAC was confirmed when at least 2 of the following 5 test results were present: LDDST results indicative for ADHAC (cortisol concentration in 4-hour sample suppressed to <40 nmol/L or 1.4 μg/dL, or suppression of baseline cortisol concentration of ≥50% at 4 hours or throughout the test), UCCR suppression of 50% or more after oral administration of a high dose of dexamethasone (0.1 mg/kg, q8h × 3 doses), increased plasma ACTH concentration (>19.8 pmol/L), ultrasonographic evidence of 2 equal-sized adrenal glands or presence of a pituitary mass on computed tomography. Exclusion criteria were presence of concurrent systemic infections, neoplastic, or endocrine diseases, and treatment with drugs potentially affecting kidney function. Cardiac disease was an exclusion criterion except for dogs with subclinical disease (International Small Animal Cardiac Health Council class Ia and Ib).
Twelve dogs (Mx group) were treated medically with trilostane1 at a starting dosage of 2 mg/kg q24h or 1 mg/kg q12h. Response to treatment was based on control of clinical signs and an ACTH2 stimulation test, and was assessed after 10–14 days and additionally at 1 (T1), 3 (T3), 6 (T6), and 12 months (T12), after initiation of trilostane. Dogs were considered to be adequately controlled when clinical signs resolved and post-ACTH cortisol was between 1.4 and 5.4 μg/dL (40 and 150 nmol/L). Seven dogs (Hx group) underwent transsphenoidal hypophysectomy and received replacement therapy with cortisone acetate (1 mg/kg q12h starting dosage tapered over 4 weeks to 0.25 mg/kg q12h), levothyroxine (15 μg/kg), and desmopressin (0.01%, 1 drop into the conjuctival sac q8h, tapered and tailored to the individual patient). Response to treatment was assessed based on clinical signs and UCCR with additional evaluation of serum total thyroxine concentration at recheck appointments at T1, T3, T6, and T12 after surgery. Hypercortisolism was considered to be in remission when clinical signs were resolved and UCCR results were below 8.3 × 10−6. Choice of treatment was made by the owners and randomization of the 2 treatment groups was not performed.
All dogs were evaluated 1 day before treatment (T0) and at T1, T3, T6, and T12 after treatment. Tested variables at all time points were body weight (BW), systolic blood pressure (SBP), serum creatinine (sCr) concentration, serum urea concentration, urine specific gravity (USG), urine protein-to-creatinine ratio (UPC), uALB/c, uIgG/c, uRBP/c, and uNAG/c. Complete blood count (CBC), biochemistry profile and urine cultures were performed at T0, T6, and T12. Urine culture was performed more often when indicated based on sediment analysis. At the end of the study, owners were interviewed by telephone to document the clinical outcome of the patients.
Because control values for plasma clearance of creatinine (Clcreat), exo- and endo-iohexol (Clexo and Clendo), currently are not well established and often different analytical methods are used, 12 healthy BW matched control dogs ≥7 years old were included for measurement of GFR. For Clcreat a lower limit of 2 mL/min/kg previously has been suggested.[32-34] Dogs were judged healthy based on history, physical examination, CBC, biochemistry profile, and urinalysis (negative bacterial culture). Recent administration of drugs that could affect renal function was an exclusion criterion. When age and body weight were compared between control dogs and dogs with ADHAC using a Mann-Whitney U-test, ADHAC dogs were significantly older than healthy controls (P = .03) and there was no significant difference in body weight (P = .45).
Systolic blood pressure was measured indirectly using Doppler method,3 according to the American College of Veterinary Internal Medicine guidelines. Morning urine samples were taken by cystocentesis. Urinary dipstick analysis,4 USG, UPC, sediment analysis, and bacterial culture were performed. After centrifugation, urine supernatant was stored at −80°C until assayed.
Glomerular filtration rate was simultaneously measured by Clcreat, Clexo, and Clendo, as previously described.[36, 37] Iohexol5 (64.7 mg/kg) was injected, immediately followed by a creatinine solution (40 mg/kg of an 80 mg/mL solution) via a cephalic catheter. Blood was collected from the jugular vein before and 5, 15, 60, 120, 240, 360, and 480 minutes postinjection in EDTA tubes. Samples were centrifuged within 2 hours and stored in aliquots of 300 μL at −20°C until assayed.
Serum urea concentration (reference interval, 3.2–23.8 mg/dL, 1.2–8.5 mmol/L) was measured using an enzymatic method.6 Urinary protein concentration was determined with a turbidimetric method using benzethonium chloride and urinary creatinine with a modified Jaffé reaction. Urinary ALB, IgG, and RBP were determined with an ELISA7 and uNAG with a colorimetric assay,8 as previously validated, and indexed to urinary creatinine (c).[19, 38]
Enzymatic analysis of plasma creatinine and measurement of iohexol stereo-isomers (exo- and endo-iohexol) with high performance liquid chromatography were performed as previously described.[32, 37]
A noncompartmental analysis was performed using software9 as described previously. The area under the plasma Cr, exo- and endo-iohexol concentration versus time curve (AUC) was calculated using the trapezoidal rule with extrapolation to infinity. Plasma Clcreat, Clexo, and Clendo were determined by dividing the administered dose by the corresponding AUC, and indexing to BW (mL/min/kg).
Analyses were performed using commercial software.10 Level of significance was set at 5% (P < .05). A general linear model was used to test the effect of time on the selected variables. When a significant effect of time was detected, a Tukey post hoc hypothesis test was performed to analyze which time points differed significantly. Relationships between the most relevant tested variables (ie, age at T0, SBP, UPC, uRBP, uNAG, uALB, uIgG, and plasma clearance of creatinine, exo- and endo-iohexol) were tested 2 × 2. The statistical link between each covariable was tested by means of the following general linear model (Variable 2 = μ + [a× Variable 1] + ε, where μ is the mean, a is the regression coefficient for the variable, and ε is the model error).
All 19 dogs were followed for at least 6 months. Thirteen of 19 dogs underwent the final evaluation at T12 (7 dogs in the Mx group and 6 dogs in the Hx group). Between T6 and T12, 3 dogs died or were euthanized because of cardiac disease (n = 1), central nervous signs (n = 1), or severe respiratory disease (n = 1), and 3 dogs were lost to follow-up because of owners withdrawing them from the study (n = 3). At the time of submission, 10/19 dogs had died or were euthanized because of nonrenal related causes, with a mean (range) survival time of 523 (192–1193) days.
At T0, mean ± standard deviation (SD) age was 10.1 ± 1.6 years in the complete group, 10.5 ± 1.9 years in the Mx group, and 9.4 ± 0.6 years in the Hx group, respectively. Body weight of the complete group at the different time points, and of the Mx and Hx subgroups, is presented in Tables 1 and 2. Age and BW in the control group were 8.7 ± 1.6 years and 27.9 ± 14.2 kg. Body weight and SBP did not change significantly post-treatment in any of the patient groups (Tables 1 and 2). Systolic BP was ≥160 mmHg in 7/19 ADHAC dogs at T0, in 6/19 at T6, and in 5/13 dogs at T12.
|Tested Variables||Time Point|
|BW (kg)||28.2 (9.5–47.0)||28.0 (9.6–47.7)||28.0 (9.6–48.6)||29.3 (9.2–48.7)||30.2 (13.9–45.2)|
|SBP (mmHg)||155 (103–230)||138 (122–180)||155 (124–207)||140 (110–220)||155 (120–195)|
|sCr (μmol/L)||53 (37–126)||62 (46–117)||63 (44–119)||65a (48–132)||67 (56–100)|
|urea (mmol/L)||3.8 (2.0–8.3)||4.3a (3.2–8.0)||4.0 (2.3–7.3)||4.7a (2.7–6.7)||4.5 (1.3–6.2)|
|USG||1.009 (1.003–1.022)||1.011 (1.003–1.025)||1.013a (1.004–1.039)||1.023a (1.004–1.039)||1.020a (1.002–1.045)|
|UPC||1.66 (0.01–16.32)||0.41a (0.06–4.96)||0.27a (0.06–4.01)||0.25a (0.09–4.7)||0.28a (0.08–7.47)|
|uALB/c (mg/g)||951.44 (31.19–13713)||263.39 (2.03–6659.04)||131.65a (1.45–4483.08)||129.60a (3.47–5236.51)||159.54 (4.73–12903.48)|
|uIgG/c (mg/g)||189.35 (0.71–6186.11)||17.59 (0–1410.03)||12.1a (0.75–700.33)||24.74 (1.18–712.93)||27.03 (1.29–1365.10)|
|uRBP/c (mg/g)||0.63 (0.21–37.87)||0.10a (0–1.59)||0.12a (0–1.35)||0.11a (0–0.82)||0.10a (0–4.77)|
|uNAG/c (mg/g)||4.03 (0–58.30)||2.27 (0–58.45)||2.39 (0–22.57)||3.19 (0–13.56)||3.21 (0–11.78)|
|Clcreat (mL/min/kg)||3.2 (1.8–4.3)||NA||NA||2.2a (1.6–2.8)||2.2a (1.4–3.2)|
|Clexo (mL/min/kg)||3.0 (1.8–4.5)||NA||NA||2.5 (1.7–3.9)||2.5 (1.9–3.0)|
|Clendo (mL/min/kg)||2.8 (1.9–4.9)||NA||NA||2.6 (1.6–4.3)||2.3a (1.8–3.2)|
|Mx (n = 12)||Hx (n = 7)||Mx (n = 12)||Hx (n = 7)||Mx (n = 12)||Hx (n = 7)||Mx (n = 12)||Hx (n = 7)||Mx (n = 7)||Hx (n = 6)|
|BW (kg)||23.5 (9.5–44.5)||29.1 (14.8–47.0)||23.2 (9.6–44.0)||30.4 (14.8–47.7)||22.6 (9.6–43.2)||31.0 (14.8–48.6)||22.3 (9.2–45.0)||31.0 (15.5–48.7)||21.5 (13.9–41.5)||30.8 (16.8–45.2)|
|SBP (mmHg)||160 (135–230)||138 (103–161)||151 (131–180)||130 (122–138)||159 (131–187)||148 (124–207)||155 (118–220)||128 (110–165)||147 (120–195)||156 (127–168)|
|sCr (μmol/L)||56 (37–80)||50 (42–126)||60 (46–82)||65 (50–117)||61 (44–73)||66a (53–119)||66 (48–132)||65 (52–105)||67 (58–78)||70 (56–100)|
|urea (mmol/L)||3.41 (2.00–5.66)||4.16 (2.66–8.30)||4.33a (3.16–6.16)||4.83 (3.16–7.99)||4.15 (2.33–6.16)||4.00 (3.16–7.33)||5.16a (3.16–6.66)||4.50 (2.66–6.33)||4.16 (3.66–5.00)||4.98 (1.33–6.16)|
|USG||1.009 (1.003–1.022)||1.008 (1.003–1.015)||1.012 (1.005–1.025)||1.011 (1.003–1.024)||1.015a (1.010–1.039)||1.012 (1.004–1.037)||1.026a (1.007–1.038)||1.010 (1.004–1.039)||1.033a (1.019–1.045)||1.011 (1.002–1.028)|
|UPC||1.53 (0.14–16.32)||1.66 (0.01–5.26)||0.51 (0.13–4.96)||0.23a (0.06–0.73)||0.35a (0.12–4.01)||0.13a (0.06–047)||0.27a (0.10–4.70)||0.10a (0.09–0.72)||0.3 (0.23–7.47)||0.19a (0.08–0.64)|
|uALB/c mg/g)||935.93 (35.03–13713.00)||1218.41 (31.19–5081.65)||387.25 (12.19–6659.04)||100.24a (2.03–720.87)||200.61 (10.34–4483.08)||26.77a (1.45–396.47)||240.01 (8.32–5236.51)||31.4a (3.47–494.61)||1206.04 (106.23–12903.48)||78.16a (4.73–336.91)|
|uIgG/c (mg/g)||166.48 (0.71–6186.11)||189.35 (2.31–430.73)||52.84 (1.62–1410.03)||5.32a (0–17.59)||28.07 (2.09–700.33)||2.45a (0.75–24.91)||38.98 (1.83–712.93)||2.21a (1.18–46.98)||94.53 (11.93–1365.10)||4.39a (1.29–38.46)|
|uRBP/c (mg/g)||0.61 (0.21–37.87)||1.22 (0.21–27.21)||0.28 (0–1.59)||0.17a (0–1.08)||0.14 (0–1.35)||0.09a (0–0.94)||0.12 (0–0.82)||0.10a (0–0.46)||0.13 (0.08–4.77)||0.00 (0–0.14)|
|uNAG/c (mg/g)||5.6 (0–58.3)||1.36 (0–25.6)||6.1 (0–58.5)||0 (0–1.4)||4.0 (1.0–22.6)||1.2 (0–2.4)||3.9 (0.4–13.6)||0 (0–3.8)||3.2 (1.2–11.8)||3.0 (0–11.5)|
|Clcreat (mL/min/kg)||2.5 (2.0–3.4)||3.3 (1.8–4.3)||NA||NA||NA||NA||2.2a (1.6–2.8)||2.1a (1.8–2.4)||2.2 (1.9–3.2)||2.2a (1.4–2.3)|
|Clexo (mL/min/kg)||2.55 (1.97–4.25||3.24 (1.82–4.45)||NA||NA||NA||NA||2.56 (2.08–3.85)||2.04 (1.73–2.72)||2.51 (1.92–2.95)||2.56 (2.32–3.01)|
|Clendo (mL/min/kg)||2.95 (1.89–4.28)||2.83 (1.88–4.98)||NA||NA||NA||NA||2.63 (1.55–4.29)||2.57 (1.63–3.28)||2.34 (1.82–3.21)||2.48 (1.76–3.05)|
In the Mx group, adequate control of ADHAC was achieved by T1. At T3, 1/12 dogs was poorly controlled and at T12 1/7 dogs had clinical signs caused by concurrent diabetes mellitus. Total mean starting dose ± SD of trilostane was 1.8 ± 0.5 mg/kg q12h in 9/12 dogs and 2.4 ± 0.2 mg/kg q24h in 3/12 dogs. At T12, mean trilostane dose was 1.6 ± 0.7 mg/kg q12h in 6/7 dogs and 3 mg/kg q24h in 1 dog.
In the Hx group, control of ADHAC in all dogs was achieved by T3. At T6, 2/7 dogs had recurrence of ADHAC with increased UCCRs (45.0 and 44.8 × 10−6; 26.0 and 26.2 × 10−6). The latter 2 dogs subsequently were treated with trilostane with good clinical control at T12.
Post-Treatment Course of Renal Variables
Descriptive statistics for the complete group are presented in Table 1. Table 2 shows the results for the Mx and Hx groups separately. sCr was increased post-treatment in the complete group at T6 (P = .011) and in the Hx group at T3 (P = .037), but not in the Mx group (Table 2). One dog had a mildly increased sCr (>1.4 mg/dL, >125 μmol/L) at T6 (1.5 mg/dL, 132 μmol/L). Urea was higher in the complete group and the Mx group at T1 (P = .020 and P = .041, respectively) and T6 (P = .019 and P = .007), but did not differ post-treatment in the Hx group (Tables 1 and 2).
Plasma Clcreat, Clexo, and Clendo in the control dogs were 2.6 ± 0.5, 2.4 ± 0.6, 2.6 ± 0.9 mL/min/kg, respectively. Plasma Clcreat was unavailable at baseline in 3/19 ADHAC patients. Pretreatment Clcreat, Clexo, and Clendo are presented in Table 1 and were higher than plasma clearance values of the controls in 10/16, 11/19, and 14/19 ADHAC patients, respectively. Post-treatment Clcreat was significantly lower than pretreatment values at T6 and T12 in the complete group (P < .001) and the Hx group (P = .001 and P = .002, respectively) (Fig 1). In the Mx group, Clcreat was decreased at T6 (P = .014), but not at T12 (Fig 1). At T0 and T12, respectively, 1/16 and 4/13 dogs had a Clcreat below the previously suggested lower limit of normal of 2.0 mL/min/kg. There was no significant post-treatment decrease in Clexo, whereas Clendo was lower in the complete group at T12 (P = .024) (Fig 1).
Post-treatment USG increased in the complete group and the Mx group at T3 (P = .005 and P = .004), T6 (P < .001) and T12 (P < .001), but there was no significant change in the Hx group. In all groups, UPC (Fig 1) decreased significantly after treatment (P < .001) at all time points except for T1 and T12 in the Mx group. At T0, T6, and T12, UPC was above the cut-off (>0.5) in 13/19 (9/12 Mx, 4/7 Hx), 7/19 (5/12 Mx, 2/7 Hx), and 5/13 (3/7 Mx, 2/6 Hx) dogs, respectively. At T12, the 3 proteinuric dogs in the Mx group had UPCs of 0.78, 1.72, and 7.74. The latter dog was diagnosed with concurrent diabetes mellitus at T12. Previous blood and urinary examinations did not show hyperglycemia and glucosuria. The 2 dogs in the Hx group with mild proteinuria at T6 (0.62 and 0.72) and T12 (0.64 and 0.52) were those with recurrence of ADHAC at T6. At T12 they were treated with trilostane with good clinical control and optimal dosage as assessed by an ACTH stimulation test.
A decrease in uALB/c at T3 (P = .015) and T6 (P = .013), in uIgG/c at T3 (P = .0049), and in uRBP/c (P = .033) in the complete group was observed at every time point (Fig 2), but urine NAG/c did not change significantly. Although uALB/c (P = .004) and uIgG/c (P = .002) were decreased in the Hx group at all time points and uRBP/c (P = .021) was decreased at T1, T3, and T6, there was no statistically significant difference in these markers post-treatment in the Mx group.
Statistically significant positive relationships between UPC and uRBP (P < .001, R²=0.727), UPC and uAlb (P = .042, R²=0.176), uRBP and uALB (P = .011, R²=0.283), and uALB and uIgG (P < .001, R²=0.671) were found. For GFR estimates and other variables, no statistically significant effect of any tested covariable was observed.
Postmortem kidney histopathological examination was available for 2 dogs that had died between T6 and T12. The 1st dog was an 11-year-old Boxer that presumably died of ventricular tachycardia. Light microscopic evaluation identified mild thickening of Bowman's capsule, intraluminal tubular deposits (presumably protein), and focal accumulation of lymphocytes, plasma cells, and macrophages. The 2nd dog was a 13-year-old Jack Russell Terrier euthanized because of severe respiratory distress caused by diffuse interstitial pneumonia. Kidney histopathology showed diffuse glomerular lesions including mild hypercellularity, thickening of the glomerular basement membrane, and granular deposits in the glomerular tuft. Multifocal cystic glomerular atrophy and glomerulosclerosis also were found. Tubulointerstitial lesions included tubular atrophy, thickening of the tubular basement membrane, mild interstitial fibrosis, and sparse mononuclear cell infiltrates.
This study is the first to prospectively assess long-term effects of ADHAC on renal function in dogs, using both routine renal markers and more sensitive tests, such as GFR and urine markers. The main findings were that GFR is increased pretreatment and decreases post-treatment. Furthermore, UPC, uALB/c, uRBP/c, and uIgG/c significantly decreased post-treatment, although after 12 months of treatment, proteinuria persisted in 5/13 (38%) dogs.
Pretreatment Clcreat, Clexo, and Clendo in ADHAC patients were above the mean values of healthy controls in 10/16 (63%), 14/19 (74%), and 11/19 (58%) dogs, respectively. Post-treatment values were 27 and 26% lower after 6 and 12 months for Clcreat, 13 and 11% for Clexo, and 15 and 23% for Clendo. These changes are less pronounced than GFR changes observed in hyperthyroid cats. Between-day coefficient of variation for Clcreat in our department in healthy dogs is 14.1%, for Clexo it is 9.9%, and for Clendo 13.1% (data not shown). Thus, the significant decrease of Clcreat at T6 and T12 and of Clendo at T12 are real changes in GFR and not because of between-day measurement variability.
High pretreatment GFR may be because of an increased renal blood flow (RBF) and decreased renal vascular resistance (RVR), as described in dogs treated with exogenous glucocorticoids.[40, 41] Glucocorticoids may attenuate renal action of angiotensin II and suppress angiotensin converting enzyme activity in the renal cortex. Furthermore, nitric oxide-mediated vasodilatation of the pre- and postglomerular arterioles, and glucocorticoid interaction with prostaglandins also were suggested to increase GFR in rats treated with ACTH and sheep receiving glucocorticoids.[42, 43] In the only study on RVR in dogs with spontaneous Cushing's syndrome, renal resistive index and pulsatility index were calculated as indices of RVR using Doppler ultrasound. Only 2 of 12 dogs had higher indices, suggesting that RVR is not increased in the majority of these patients.
Secondly, the catabolic effects of glucocorticoids increase plasma amino acid concentrations, which may be another mechanism leading to increased GFR. Intravenous infusion of plasma to dogs caused hyperproteinemia, leading to an increase in GFR. This increase in GFR can be explained by renal vasodilatation, higher RBF, either no change or a moderate increase in glomerular hydrostatic pressure, and probably an increase in the glomerular ultrafiltration coefficient.
In the present study, 4/13 dogs had Clcreat <2.0 mL/min/kg at T12, which is compatible with renal impairment. After trilostane treatment, plasma renin activity in dogs is increased, which may lead to activation of the renin-angiotensin-aldosterone system and renal vasoconstriction. Renal vasoconstriction causes contraction of mesangial cells, which lowers available filtration surface in the glomerulus, decreases the ultrafiltration coefficient and consequently lowers GFR. An age-related decrease in GFR, as described in aging humans, and suggested in aging dogs, might also have contributed to the observed post-treatment decrease in GFR during the 12-month study period.11 However, no dog developed azotemia within 12 months. Although pretreatment sCr was low and increased post-treatment, as in previous studies, it did not exceed the reference interval.[25, 49] Also, of the 10 dogs that died or were euthanized, none died of renal failure, although 2 dogs did have glomerular and tubulo-interstitial renal lesions on postmortem examination. Nevertheless, it is inconclusive whether these lesions were caused by concomitant renal disease or by ADHAC.
In human patients with Cushing's syndrome, post-treatment GFR is decreased, leading to clinical CKD in some patients. Monitoring GFR therefore seems mandatory in these patients. Contrary to the situation in humans, cardiovascular risk factors do not seem to play an important role in development of renal dysfunction in dogs with Cushing's disease.[11, 50, 51] As in hypothyroid dogs, the clinical relevance of GFR changes in dogs with ADHAC currently remains unclear.[32, 34]
Discrepancies between Clcreat, Clexo, and Clendo were observed, as previously reported in healthy dogs, dogs with ADHAC or ACTH-independent hyperadrenocorticism and in healthy cats. Potential explanations could be the disposition of the marker, laboratory technique, the animal's characteristics, and the clinical setting. Simultaneous injection of both markers limits animal-dependent influences. Whatever the explanation, dispositions of exo- and endo-iohexol differ from each other and from creatinine so that they should be considered as different markers.
After hypophysectomy, dogs were supplemented with cortisone acetate and levothyroxine, because of very low endogenous ACTH and thyroid-stimulating hormone-production. Exogenous glucocorticoids and levothyroxine[32, 34] affect GFR, but a physiologic dosage (0.25 mg/kg q12h) of cortisone was given and none of the dogs had a serum thyroxine concentration below the reference range at the time GFR was measured. Therefore, these factors are unlikely to influence GFR in the Hx group.
In agreement with a previous study, SBP did not change significantly after treatment. At T3 and all later time points, USG was significantly higher in the Mx group, but not in the Hx group. This may be explained by impaired release of vasopressin posthypophysectomy. All Hx dogs were supplemented with desmopressin and urine samples were taken after the morning dose.
Pretreatment UPC was increased in 68% of dogs, which is a higher percentage than observed in earlier studies (44–46%)[5, 6], but in accordance with more recent reports (71%). Different cut-off values for proteinuria partly explain this difference (UPC >1.0 in initial studies and >0.5 in recent reports). Post-treatment proteinuria previously has been reported in 21–31% of cases with well-controlled ADHAC and is found in 38% of dogs in the current study at T12.[5, 6]
Because no randomization was performed, Mx and Hx groups differ at T0 and it is difficult to compare between-group post-treatment changes. In addition, both groups consist of a small number of dogs. In the Mx group, however, UPC tended to decrease more slowly than in the Hx group. At T12, 3/7 Mx dogs had increased UPC, whereas the 2 dogs in the Hx group with proteinuria at T6 and T12 were dogs with recurrence of ADHAC. These findings suggest that recurrence of ADHAC is associated with recurrence of proteinuria, and that surgically treated dogs may be less likely to remain proteinuric post-treatment than medically treated dogs. Therefore, proteinuria in a dog after hypophysectomy either suggests recurrence of ADHAC or concurrent disease and definitely warrants the clinician's attention.
Paralleling the UPC, uALB/c, uIgG/c, and uRBP/c decreased after treatment in the complete group. In a previous study comparing untreated dogs with Cushing's syndrome to healthy controls, these markers are increased, suggesting altered glomerular and tubular function. The correlation among different urinary proteins in the present study suggests that proteinuria in dogs with ADHAC is not selective. Combined HMW (IgG) and LMW (uRBP) proteinuria indicates increased glomerular permeability with decreased tubular protein reabsorption. The latter may be because of protein overload in the tubular lumen with saturation of the tubular reabsorptive capacity. Results of the present study also indicate that these changes are reversible in most dogs. As for UPC, post-treatment course of uALB/c, uIgG/c, and uRBP/c may differ in the Mx and Hx groups. In the Hx group, uALB/c, uIgG/c, and uRBP/c were lower after treatment, whereas no significant difference was found in the Mx group. In medically treated dogs with ADHAC, clinical signs of hypercortisolism are controlled without removing its primary cause. Thus, serum cortisol still may exceed physiological concentrations during a certain period of the day. This “temporary hypercortisolism” hypothetically may contribute to proteinuria in these patients, as opposed to dogs that underwent hypophysectomy and were treated with a low dose of cortisone.
In the current study, pretreatment uNAG/c was increased and did not decrease significantly post-treatment. In humans with glomerulonephritis, in geriatric cats with azotemia and in dogs with CKD, uNAG/c is correlated with UPC, albuminuria or both, but not with sCr (or GFR in humans). Therefore, high uNAG/c may reflect high lysosomal activity rather than active tubular damage and remain high in dogs with persistent proteinuria or albuminuria because of increased tubular processing. This hypothesis is supported by a study indicating that uNAG/c is significantly increased in cats with borderline proteinuria (UPC 0.2–0.4) compared with nonproteinuric cats (UPC < 0.2). Some of the dogs in the current study with a UPC < 0.5 also had mild increases in uNAG/c.
Urinary tract infection (UTI) is known to affect UPC and also may influence urinary marker concentrations.[57, 58] However, it is unlikely that UTI affected results of the present study. Urinary ALB for example was found to be increased only in urine samples with macroscopic hematuria, and not in the majority samples with pyuria. In the present study, 1 dog had a positive urine culture before treatment, but urine sediment red and white blood cell numbers were normal. A second dog had a UTI at T3 and T6, but only sediment white cell numbers were increased. In humans, there is no evidence that asymptomatic UTI cause proteinuria or albuminuria. In this study, both dogs with positive urine cultures were asymptomatic.
Limitations of the present study are the low number of dogs in both groups (Mx and Hx) and not using the gold standard of urinary inulin clearance for measurement of GFR. However, this is an impractical method for use in a clinical setting, requiring a metabolic cage or an indwelling urinary catheter. Moreover, both Clcreat and Cliohexol show good correlation with urinary inulin clearance.[15, 60] Furthermore, small sample size may have contributed to the lack of statistically significant effects of investigated covariables (eg, age, blood pressure, UPC) on GFR.
In conclusion, although none of the patients developed azotemia over 12 months, post-treatment decrease in GFR and persistent proteinuria in 38% of dogs warrant the clinician's attention. Future research with longer follow-up and kidney histopathology of patients with persistent proteinuria or low GFR is needed to further assess renal outcome.
The authors thank Dr H.S. Kooistra, Dr S. Galac, and Dr J.J.C.W.M. Buijtels for their help with case recruitment and Dr S. De Baere, J. Lambrecht, and M. Kramer for their excellent technical assistance.
Vetoryl, Dechra Veterinary Products, Shropshire, UK
Synacthen, tetracosactide hexa-acetate, Novartis Pharma, Vilvoorde, Belgium
Parks Medical Electronics Inc, Aloha, OR
Combur stick, Roche Diagnostics, Burgess Hill, UK
Omnipaque 300, GE Healthcare, Amersham Health, Wemmel, Belgium
Roche hitachi, Roche Diagnostics
Immunology Consultants Laboratory, Newberg, OR
β-N-Acetyl-glucosaminidase Assay kit, Sigma–Aldrich, St Louis, MO
WinNonlin version 4.0.1, Pharsight, Cary, NC
Systat, version 12.00.08, Chicago, IL
Queau Y, Biourge V, Germain C, et al. Effect of aging on plasma exogenous creatinine clearance in dogs. J Vet Intern Med 2007;21:93 (abstract)
- 1Epidemiological aspects of clinical hyperadrenocorticism in dogs (canine Cushing's syndrome). J Am Anim Hosp Assoc 1982;18:717–724., .
- 27Protocols for function tests. In: Rijnberk A, Kooistra HS, eds. Clinical Endocrinology of Dogs and Cats. Hannover, Germany: Schlütersche; 2010:305–314., .
- 39Chronic kidney disease. In: Ettinger SJ, Feldman EC, eds. Textbook of Veterinary Internal Medicine, 7th ed. St Louis, MO: Saunders Elsevier; 2010:1990–2021..
- 47Renal physiology. In: Cunningham J, Klein G, eds. Textbook of Veterinary Physiology, 4th ed. St. Louis, MO: Saunders Elsevier; 2007:527–563..