The study was performed at the Department of Clinical Sciences of Companion Animals, Faculty of Veterinary Medicine, Utrecht University, the Netherlands. Presented in part at the 18th ECVIM-CA Congress, Ghent, Belgium, September 4–6, 2008.
Urinary Corticoid : Creatinine Ratios in Dogs with Pituitary-Dependent Hypercortisolism during Trilostane Treatment
Article first published online: 26 AUG 2009
Copyright © 2009 by the American College of Veterinary Internal Medicine
Journal of Veterinary Internal Medicine
Volume 23, Issue 6, pages 1214–1219, November/December 2009
How to Cite
Galac, S., Buijtels, J.J.C.W.M. and Kooistra, H.S. (2009), Urinary Corticoid : Creatinine Ratios in Dogs with Pituitary-Dependent Hypercortisolism during Trilostane Treatment. Journal of Veterinary Internal Medicine, 23: 1214–1219. doi: 10.1111/j.1939-1676.2009.0374.x
- Issue published online: 27 OCT 2009
- Article first published online: 26 AUG 2009
- Submitted February 20, 2009; Revised June 22, 2009; Accepted July 14, 2009.
- Cushing's disease;
Background: The adrenocorticotropic hormone (ACTH) stimulation test is used to evaluate trilostane treatment in dogs with hypercortisolism.
Hypothesis: The urinary corticoid : creatinine ratio (UCCR) is a good alternative to the ACTH stimulation test to determine optimal trilostane dose.
Animals: Eighteen dogs with pituitary-dependent hypercortisolism.
Methods: In this prospective study, the dose of trilostane was judged to be optimal on the basis of resolution of clinical signs of hypercortisolism and results of an ACTH stimulation test. The owners collected urine for determination of UCCR at 2-week intervals for at least 8 weeks after achieving the optimal trilostane dose.
Results: The UCCRs were significantly higher before treatment (11.5–202.0 × 10−6; median, 42.0 × 10−6) than at rechecks 2 months after optimal dosing, but they did not decrease below the upper limit of the reference range in the majority of dogs. The UCCRs of 11 dogs that initially were dosed insufficiently (range, 7.5–79.0 × 10−6; median, 31.0 × 10−6) did not differ significantly from UCCRs when the dosage was optimal (8.2–72.0 × 10−6; median, 33.0 × 10−6). Post-ACTH cortisol concentrations did not correlate significantly with UCCRs at rechecks during trilostane treatment. Long-term follow-up indicated that the decrease in UCCR below the upper limit of the reference was associated with hypocortisolism.
Conclusion and Clinical Importance: The UCCR cannot be used as an alternative to the ACTH stimulation test to determine the optimal dose of trilostane, but might be helpful in detecting dogs at risk for developing hypocortisolism during trilostane treatment.
urinary corticoid to creatinine ratio
Trilostane has become the medical treatment of choice for pituitary-dependent hypercortisolism (PDH) in dogs. Trilostane is a competitive inhibitor of the 3β-hydroxysteroid dehydrogenase/isomerase system (3β-HSD), an essential enzyme system for the synthesis of cortisol, aldosterone, and androstenedione.1 In dogs with PDH, trilostane has the potential to induce a substantial decrease in basal and adrenocorticotropic hormone (ACTH)-stimulated plasma cortisol concentrations.2–5 Consequently, a loss of negative feedback occurs, leading to an increase in endogenous plasma ACTH concentration.6–8 Trilostane also affects aldosterone synthesis.7,8
The effectiveness of trilostane treatment is judged by resolution of clinical signs associated with glucocorticoid excess and results of an ACTH stimulation test.2,3 The aim of performing an ACTH stimulation test in a dog on trilostane treatment is to evaluate whether sufficient adrenocortical reserve is present at the time of maximal effect of trilostane, which is 2–3 hours after administration.a The disadvantages of the test are that (1) it only provides information about suppression of cortisol production during a short interval, (2) it is invasive, (3) the post-ACTH cortisol concentration thought to indicate optimal dosage of trilostane is still arbitrary, and (4) there is concern about the availability and cost of the injectable ACTH9 required for the test.
The urinary corticoid : creatinine ratio (UCCR) provides an integrated measure of corticoid production over a given interval, thereby overcoming the problem of fluctuations in plasma concentrations.10 It is useful not only in the initial diagnosis of hypercortisolism but also in detection of persisting cortisol secretion after hypophysectomy or nonselective destruction of the adrenal cortex by mitotane treatment.11,12 Because the UCCR is an integrated measure of glucocorticoid production, it might be a more appropriate indicator of the therapeutic efficacy of trilostane than an ACTH stimulation test. In addition, it requires less time and is not invasive.
The aim of the present study was to determine whether the UCCR is a suitable alternative to the ACTH stimulation test for evaluating the dosage of trilostane and the efficacy of treatment of dogs with PDH.
Materials and Methods
Animals, Treatment, and Tests
Eighteen client-owned dogs with PDH were investigated. They consisted of 4 spayed females and 14 males (8 intact and 6 castrated). Their ages ranged from 8 to 17 years (median, 12 years) and their body weights from 5 to 46 kg (median, 12 kg). Four dogs were mongrels and the others were from 11 different breeds.
Suspicion of hypercortisolism was based on the history, physical examination, and routine laboratory findings. An ACTH stimulation test demonstrated hyperresponsiveness (ie, post-ACTH cortisol >552 nmol/L)13 of the adrenal cortex in all but 1 dog. In the latter dog, the post-ACTH plasma cortisol concentration was 352 nmol/L. This dog was included in the study, because based on the history, clinical findings, laboratory investigation, increased UCCRs that could not be suppressed by high doses of dexamethasone, and diagnostic imaging there were no doubts about the correctness of the diagnosis of hypercortisolism.
The diagnosis of hypercortisolism was considered to be confirmed by finding an increased UCCR (>8.3 × 10−6) in 2 consecutive morning urine samples collected at home.10,14,15 After collection of the 2nd urine sample, the dogs received 3 doses of dexamethasone 0.1 mg/kg PO q8h and the 3rd urine sample was collected on the following morning. PDH was diagnosed when the UCCR in the 3rd sample was <50% of the mean of the 1st 2 samples.16,17 In 4 dogs in which the 3rd UCCR was >50%, basal plasma ACTH concentrations were 35–144 pg/mL (median, 49 pg/mL) and were interpreted as nonsuppressed. Abdominal ultrasonography revealed no evidence of an adrenocortical tumor in any of these 4 dogs and in 3 of them a pituitary mass was detected by computed tomography. Treatment with trilostaneb was started at an initial dosage of 2–4 mg/kg body weight once daily at 8 am. The 1st recheck was at 3 weeks after the start of treatment. The history and physical examination were recorded and, at 3.5 hours (median, range 2–4 hours) after administration of trilostane, a blood sample was collected for routine hematologic and biochemical analyses and determination of plasma concentrations of ACTH and cortisol.8 Immediately after blood collection, 0.25 mg ACTHc was administered IV and a 2nd blood sample for measurement of plasma cortisol was collected 90 minutes later.13,16 The dose of trilostane was adjusted on the basis of clinical signs and results of the ACTH stimulation test. When post-ACTH plasma cortisol concentration was <165 nmol/L, the dose was considered to be optimal if the owner also reported resolution of polyuria, polydipsia, and polyphagia (the most common signs of hypercortisolism). When post-ACTH plasma cortisol concentration was >165 nmol/L, the clinical signs of hypercortisolism persisted or both, the dose was increased and the examination was repeated 3 weeks later. The dose was increased in steps as follows (in mg): 10–20, 20–30, 30–40, 40–60, 60–70, 70–80, 80–90, and 90–120. When the dose appeared to be optimal, the examination was repeated 12 ± 1 weeks later and if the results were still satisfactory, examinations were repeated at 12- to 24-week intervals. The dog was examined in the clinic at least once after the dose was considered to be optimal. When there were no clinical signs suggestive of hypocortisolism, but the post-ACTH plasma cortisol concentration was suppressed to <28 nmol/L, the dose was decreased. If the dog became lethargic and its appetite decreased and the post-ACTH plasma cortisol concentration was suppressed to <28 nmol/L, treatment with trilostane was stopped. Treatment was resumed, at a lower dose, when clinical signs of hypercortisolism reoccurred.
Blood samples were collected from the jugular or cephalic vein. Blood samples for measurement of ACTH were placed immediately in ice-chilled EDTA-coated tubes and centrifuged at 4°C for 10 minutes. Plasma was stored at −25°C until assayed.
The owners collected urine samples for determination of the UCCR at 2-week intervals from 2 weeks on after start of treatment and continued collecting for at least 8 weeks after the dose of trilostane was considered to be optimal. These samples were collected shortly before (basal sample) and 6 hours after trilostane administration (6 hours posttrilostane sample), all collected at home. Long-term follow-up was possible in 10 dogs, in all of which the UCCR was determined at least once every 6 months according to the above scheme. The longest follow-up was 2 years after the dose was judged to be optimal.
Plasma ACTH concentration was measured by an immunoradiometric assayd validated for the dog.18 The intra- and interassay coefficients of variation were 3.2 and 7.8%, respectively, and the sensitivity was 0.2 pg/mL.
Plasma cortisol concentration was measured by a radioimmunoassaye validated for the dog.18 The interassay coefficient of variation was 4.5 to 6.3%, the intraassay coefficient of variation was 4%, and the sensitivity was 5.5 nmol/L.
Urinary corticoid concentration was measured by a radioimmunoassay described by Rijnberk et al.10 The intra- and interassay coefficients of variation were 6 and 8%, respectively, and the sensitivity was 1 nmol/L. The urinary creatinine concentration was determined by the Jaffé kinetic method (initial rate reaction). From these, the UCCR was calculated.14
Statistical analysis was performed by commercial statistical software.f The Q-Q plots and the Kolmogorov-Smirnov test were used to test the data for normal distribution. Differences in basal values of plasma cortisol and ACTH concentration before and during treatment with trilostane were assessed by Wilcoxon's signed ranks test. Spearman's rank correlations were performed between post-ACTH cortisol concentration when the dose was considered optimal and UCCRs (basal and 6 hours posttrilostane) at 2, 4, 6, and 8 weeks thereafter, described as the 1st, 2nd, 3rd, and 4th UCCR recheck.
Differences between pretreatment UCCRs and the basal and 6 hours posttrilostane UCCRs of urine samples collected after the dose of trilostane was optimal were compared with Wilcoxon's signed ranks test. Also, the basal and 6 hours posttrilostane UCCRs at 2, 4, 6, and 8 weeks after achieving the optimal dose (1st, 2nd, 3rd, and 4th UCCR recheck) were compared with the same test. Finally, in 11 dogs in which adjustments of the trilostane dose were needed, the UCCRs although insufficiently dosed were compared with the UCCRs at 2 weeks after the optimal dose had been achieved with Wilcoxon's signed ranks test. Results are expressed as median and range. P < .05 was considered significant.
The median dose of trilostane at the time of good control was 3.4 mg/kg body weight, once daily (range, 2–8 mg/kg). In 7 of the 18 dogs this was the initial dose, whereas in 5 dogs the dose had to be increased twice, in 2 dogs 3 times, in 3 dogs 4 times, and in 1 dog 5 times.
The median basal plasma cortisol concentration at the optimal dose was 45 nmol/L (range, 21–114 nmol/L), which was significantly lower (P < .001) than that before treatment (median, 165 nmol/L; range, 145–655 nmol/L). The median post-ACTH plasma cortisol concentration at the optimal dose was 106 nmol/L (range, 38–165 nmol/L), which was significantly lower (P < .001) than that before treatment (median, 881 nmol/L; range, 352–1380 nmol/L) (Fig 1A).
The median plasma ACTH concentration in well-controlled dogs (88 pg/mL; range, 25–401 pg/mL) was significantly higher (P < .05) than that before treatment (median, 46 pg/mL; range, 10–144 pg/mL) (Fig 1B). The median plasma ACTH concentration in 11 dogs that were insufficiently dosed (103 pg/mL; range, 21–313 pg/mL) did not differ significantly from that before treatment (median, 51 pg/mL; range, 12–144 pg/mL).
Basal UCCRs before treatment were 11.5–202.0 × 10−6 (median, 42.0 × 10−6), which were significantly higher than basal UCCRs at 1st, 2nd, 3rd (P= .001), and 4th recheck (P < .05) after the optimal dose of trilostane had been reached (Fig 2). The 6 hours posttrilostane UCCRs were significantly lower than the basal UCCRs on the same day (P < .05) at each recheck.
The UCCRs of dogs (n= 11) that initially were inadequately dosed ranged from 7.5 to 79.0 × 10−6 (median, 31.0 × 10−6) and did not differ significantly from UCCRs when the dosage was considered optimal (8.2–72.0 × 10−6; median, 33.0 × 10−6). The 6 hours posttrilostane UCCRs of the 11 dogs that initially were dosed inadequately ranged from 5.1 to 97.0 × 10−6 (median, 24.0 × 10−6) and did not differ significantly from 6 hours posttrilostane UCCRs of dogs receiving the optimal dose (4.4–46.0 × 10−6; median, 13.0 × 10−6).
The post-ACTH cortisol concentrations did not correlate significantly with basal and 6 hours posttrilostane UCCRs at 1st, 2nd, 3rd, and 4th recheck after the optimal dose had been achieved.
Basal UCCRs in 13 dogs and 6 hours posttrilostane UCCRs in 12 dogs did not decrease below the upper limit of the reference range within 8 weeks after the dose of trilostane was judged to be optimal. Among the 6 dogs in which the UCCR did decrease below the upper limit of the reference range within 8 weeks after the dose was judged to be optimal, in 2 the basal UCCRs were below the upper limit of the reference range 3 of 4 times and 6 hours posttrilostane UCCRs were below the upper limit of the reference range all 4 times. In the other 4 dogs, the basal UCCRs were below the upper limit of the reference range only once in 3 dogs, whereas the 6 hours posttrilostane UCCRs were below the upper limit of the reference range once (1 dog), twice (2 dogs), or 3 times (1 dog).
Dogs with UCCRs below the Upper Limit of the Reference Range within 8 Weeks after Reaching the Optimal Trilostane Dose. In 2 dogs among the 6 dogs in which the UCCRs were below the upper limit of the reference range most of the times, the ACTH stimulation test revealed hypocortisolism at 15 and 22 weeks after the dose was judged to be optimal. Although neither of these 2 dogs had clinical signs of hypocortisolism, the dose was reduced.
In the other 4 dogs from this group, at 4, 6, and 21 months on the dose judged to be optimal 3 dogs developed anorexia, abdominal pain, and were lethargic, which was interpreted as mild clinical signs of hypocortisolism. Unfortunately, in none of these 3 dogs an ACTH stimulation test was performed at that time. Trilostane treatment was stopped temporarily and resumed at a lower dose when the dogs again developed clinical signs of hypercortisolism. There were no long-term follow-up UCCRs in these 3 dogs.
In the 4th dog, the ACTH stimulation test indicated hypocortisolism 40 weeks after the dose was judged to be optimal. In addition, both the basal and 6 hours posttrilostane UCCRs were below the upper limit of the reference range (5.0 and 4.5 × 10−6, respectively). Although there were no clinical signs of hypocortisolism, the dose of trilostane was decreased.
Dogs with UCCRs above the Upper Limit of the Reference Range within 8 Weeks after Reaching the Optimal Trilostane Dose. Among the 12 dogs in which the UCCRs were above the upper limit of the reference range within 8 weeks after the dose was considered optimal, the ACTH stimulation test indicated hypocortisolism in 4 of them at 40 weeks, and 1, 1.2, and 2 years. The basal UCCRs (5.5, 4.7, 3.0, and 4.3 × 10−6) and the 6 hours posttrilostane UCCRs (4.4, 2.7, 2.0, and 1.5 × 10−6) were below the upper limit of the reference interval a few weeks to months before hypocortisolism was indicated by the ACTH stimulation test. One of these dogs developed clinical signs of complete Addison's disease with vomiting, diarrhea, dehydration, hyponatremia, and hyperkalemia and needed supportive therapy with IV fluids, mineralocorticoids, and glucocorticoids. When stable, temporary oral supplementation with cortisone acetate,g fludrocortisone acetate,h and salti was continued. When clinical signs of hypercortisolism reoccurred, supplementation was stopped and trilostane treatment was resumed at a lower dose. The basal and 6 hours posttrilostane UCCRs in this dog at the time of good control on a decreased dose of trilostane were 26 and 4.5 × 10−6, respectively.
In 4 of the 12 dogs, it was not necessary to change the dose of trilostane during long-term follow-up. In 2 dogs at 1 year and in 1 dog at 2 years after starting treatment, basal UCCRs were above the upper limit of the reference range (23.0, 15.0, and 44.0 × 10−6) whereas 6 hours posttrilostane UCCRs were close to the upper limit of the reference range (5.9, 9.9, and 8.8 × 10−6).
In 1 of the remaining 4 dogs, the dose of trilostane had to be increased 6 months after it was first judged to be optimal. Before the dose was changed, both the basal and 6 hours posttrilostane UCCRs, were above the upper limit of the reference range (21.0 and 18.0 × 10−6, respectively). After the dose was increased, the values were 23.0 and 5.9 × 10−6. In another dog twice daily administration of trilostane was required 24 weeks after initial stabilization. The basal and 6 hours posttrilostane UCCRs before altering the dose were 17.0 and 12.0 × 10−6, respectively, but long-term follow-up UCCRs were not available. Two of the remaining dogs died because of causes unrelated to trilostane treatment. One dog developed severe neurologic problems because of pituitary macroadenoma and the 2nd dog had severe congestive heart failure and was euthanized.
The results of this study indicate that the UCCR cannot be used as an alternative to the ACTH stimulation test to determine the optimal dose of trilostane. In most dogs with PDH included in this study, UCCRs remained above the upper level of the reference range for at least 2 months after receiving the optimal dose of trilostane. The UCCRs also did not correlate with the post-ACTH plasma cortisol concentrations and thus the UCCR cannot be considered a reliable indicator of treatment control. Similarly, UCCRs also cannot be used as an alternative to the ACTH stimulation test in dogs in which mitotane is used to partially destroy the adrenal cortex.19,20 In contrast, the UCCR is a reliable indicator of persisting cortisol secretion after hypophysectomy or nonselective destruction of the adrenal cortex by mitotane treatment.11,12
In the measurement of urinary corticoids in unextracted urine by radioimmunoassay, not only free cortisol but also substances with a similar polarity are detected.14 Several of these are precursors and metabolites of cortisol. This lack of specificity does not constitute a problem in the use of the UCCR for the diagnosis of hypercortisolism or for monitoring treatment after hypophysectomy or nonselective destruction of the adrenal cortex by mitotane.11,12 It may, however, be important in evaluating the effect of trilostane treatment. When trilostane inhibits steroidogenesis by blocking 3β-HSD, the steroid precursors before the site of the blockade accumulate.7,21 Studies in dogs with PDH treated with trilostane have demonstrated significant increases in plasma 17α-OH-pregnenolone and dehydroepiandrostenedione, consistent with the inhibition of 3β-HSD.7 Trilostane may also inhibit other enzymes of the steroid hormone cascade, such as 11β-hydroxylase and 11β-HSD.7 17α-OH-pregnenolone and dehydroepiandrostenedione may also have been measured by the urinary corticoid assay in the present study, which could explain why the UCCR did not decrease in association with the low plasma cortisol concentrations. Quantitative determination of urinary steroids conceivably could provide the answer, but the urinary steroid profile has not yet been characterized in dogs. It is a useful tool in humans in the diagnosis of adrenocortical carcinoma and some congenital deficiencies of enzymes involved in steroidogenesis.22,23
Some authors have speculated that steroid precursors accumulating as a result of enzyme inhibition by trilostane might exert some glucocorticoid effect.3 This could explain why quite low post-ACTH cortisol concentrations appear to be necessary in some cases to avoid clinical signs of hypercortisolism. However, if the increased UCCRs in trilostane-treated dogs that are well-controlled reflect accumulated steroid precursors, it is unlikely that they have glucocorticoid effects. A better explanation for the large variation in the post-ACTH plasma cortisol concentration in well-controlled dogs might be interindividual differences in cortisol sensitivity. Although this effect has only been documented in humans,24,25 differences in cortisol sensitivity may also exist in dogs.
Decrease in the UCCR below the upper limit of the reference range by trilostane sometimes led to signs of hypocortisolism. In some cases, the ACTH stimulation test also demonstrated too little adrenocortical reserve, and the signs were reversed by stopping trilostane treatment. That the UCCRs were not as low as may be expected in dogs with hypocortisolism might be ascribed to detection of corticoids other than cortisol in the corticoid assay, but the simultaneous occurrence of clinical signs of hypocortisolism suggests that these corticoids have no clinically relevant intrinsic glucocorticoid effect.
Trilostane-induced hypocortisolism has been documented previously.7,8,26,27 Because the action of the drug is reversible, temporarily stopping trilostane is sufficient in most cases. Because adrenocortical insufficiency can be fatal if not treated promptly, some warning of this risk during trilostane treatment could be of great value. Lowering of the UCCR below the upper limit of the reference range might serve this purpose. Recently, necrosis and apoptosis of the adrenal glands have been reported to occur in dogs treated with trilostane.28 The severity appeared to be related to the dose and duration of trilostane administration as well as to individual sensitivity to the drug. In the present study, clinical signs of hypocortisolism occurred in few dogs several months after the dose of trilostane appeared to be optimal. In all of these, the UCCR decreased below the upper limit of the reference range before signs of hypocortisolism developed.
UCCR was measured both before and 6 hours after trilostane administration in order to determine which result would be more useful to evaluate treatment. In the majority of the dogs in which the dose was judged to be optimal, both of these UCCR values exceeded the upper limit of the reference range within 2 months, whereas in those dogs that developed hypocortisolism both results were below the upper limit of the reference range. Follow-up examinations indicated that the combination of a slightly increased basal UCCR and a 6 hours posttrilostane UCCR close to the upper limit of the reference range was associated with good control. This pattern was observed only after several months on the optimal dose of trilostane. Whether differences in clearance of various metabolites play a role14 needs to be examined before conclusions can be drawn.
It would be useful to have a protocol to determine the optimal dose of trilostane on the basis of resolution of the clinical signs of hypercortisolism and an ACTH stimulation test, together with use of the basal UCCR, to monitor treatment after clinical signs of hypercortisolism are satisfactory controlled. One benefit would be that measuring the UCCR is less time consuming, less invasive, and less expensive than performing ACTH stimulation tests. Perhaps even more important, patients at risk of developing hypocortisolism could be identified and given appropriate attention.
In conclusion, the results of this study demonstrate that the UCCR cannot be used to determine the optimal dose of trilostane, but could be helpful in detecting dogs that are at risk of developing hypocortisolism during trilostane treatment.
aNeiger R, Campbell L. 24 hours cortisol values after trilostane therapy in dogs with hypercortisolism. Proceedings 10th ESVIM Congress, Neuchatel, Switzerland, September 14–16,2000;31 (abstract)
bVetoryl, Dechra, Shrewsbury, UK
cSynacthen, Novartis Pharma BV, Arnhem, the Netherlands
dNichols Institute, Wijchen, the Netherlands
eCoat-A-Count Cortisol, Diagnostic Product Corporation, Los Angeles, CA
fSPSS 16.1 for Windows, SPSS, Chicago, IL
gCortisoni acetas, BV, Zaandam, the Netherlands
hFludrocortison, Aesculap BV, Boxtel, the Netherlands
iNatrii Chloridum 1000 mg, Genfarma BV
The study was supported by Dechra Veterinary Products, UK.
We thank Dechra Veterinary Products for financial support.
- 16Adrenals. In: RijnberkA, ed. Clinical Endocrinology of Dogs and Cats. Dordrecht, the Netherlands: Kluwer Academic Publishers; 1996:61–93.
- 20Use of urine cortisol:creatinine ratio versus adrenocorticotropic hormone stimulation testing for monitoring mitotane treatment of pituitary-dependent hyperadrenocorticism in dogs. J Am Vet Med Assoc 1997;15:1002–1004., , ,