Results of this study were presented at the ECVIM annual congress, Porto 2009.
Survival and the Development of Azotemia after Treatment of Hyperthyroid Cats
Article first published online: 1 JUL 2010
Copyright © 2010 by the American College of Veterinary Internal Medicine
Journal of Veterinary Internal Medicine
Volume 24, Issue 4, pages 863–869, July/August 2010
How to Cite
Williams, T.L., Peak, K.J., Brodbelt, D., Elliott, J. and Syme, H.M. (2010), Survival and the Development of Azotemia after Treatment of Hyperthyroid Cats. Journal of Veterinary Internal Medicine, 24: 863–869. doi: 10.1111/j.1939-1676.2010.0550.x
- Issue published online: 1 JUL 2010
- Article first published online: 1 JUL 2010
- Submitted November 19, 2009; Revised March 26, 2010; Accepted April 29, 2010.
- Chronic kidney disease;
- Thyroid gland
Background: Hyperthyroidism complicates the diagnosis of chronic kidney disease (CKD) as it increases glomerular filtration rate. No practical and reliable means for identifying those cats that will develop azotemia after treatment for hyperthyroidism has been identified. Hyperthyroidism is associated with proteinuria. Proteinuria has been correlated with decreased survival of cats with CKD and with progression of CKD.
Hypothesis: Proteinuria and other clinical parameters measured at diagnosis of hyperthyroidism will be associated with the development of azotemia and survival time.
Animals: Three hundred client owned hyperthyroid cats treated in first opinion practice.
Methods: Retrospective, cohort study relating clinical parameters in hyperthyroid cats at diagnosis to the development of azotemia within 240 days of diagnosis and survival time (all cause mortality). Multivariable logistic regression analysis was used to identify factors that were predictive of the development of azotemia. Multivariable Cox regression analysis was used to identify factors associated with survival.
Results: Three hundred cats were eligible for survival analysis and 216 cats for analysis of factors associated with the development of azotemia. The median survival time was 417 days, and 15.3% (41/268) cats developed azotemia within 240 days of diagnosis of hyperthyroidism. Plasma concentrations of urea and creatinine were positively correlated with the development of azotemia. Plasma globulin concentration was negatively correlated with the development of azotemia. Age, urine protein : creatinine ratio, and the presence of hypertension were significantly correlated with decreased survival time. Urine specific gravity and PCV were significantly correlated with increased survival time.
Conclusions and Clinical Importance: The proteinuria associated with hyperthyroidism is not a mediator of progression of CKD; however, it does correlate with all cause mortality.
Hyperthyroidism is the most frequently diagnosed endocrinopathy of cats.1 There are few studies that examine the long-term survival of cats with hyperthyroidism, and most previous studies have included primarily cats referred for radioiodine treatment.
Chronic kidney disease (CKD) is highly prevalent in the geriatric feline population. The prevalence of CKD increases with advancing age, with 31% of cats over 15 years reported to have CKD.2 It is therefore not uncommon for CKD and hyperthyroidism to occur in geriatric cats concurrently. The presence of hyperthyroidism can complicate the diagnosis of CKD because it results in an increased glomerular filtration rate (GFR) and decreased creatinine concentration.3 Often the diagnosis of CKD is only made after treatment for hyperthyroidism, once the GFR has normalized. Currently, there is no single test that can reliably predict renal function after treatment for hyperthyroidism. A predictive biomarker would be of value because it would allow identification of those cats with underlying CKD before treatment so that management strategies, including closer monitoring of renal function, could be instituted.
Recently, there has been increased interest in proteinuria as a prognostic indicator in cats with CKD, with the finding that both urine protein : creatinine ratio (UPC) and urine albumin : creatinine ratio (UAC) are of prognostic significance.4 Hyperthyroid cats have proteinuria before treatment, which resolves within 4 weeks of radioiodine treatment.5 The prognostic significance of proteinuria in hyperthyroid cats is uncertain. Given the association between proteinuria and survival in cats with CKD, and the relatively frequent occurrence of proteinuria in hyperthyroidism, evaluation of proteinuria as a prognostic marker is warranted. We postulated that proteinuria might be involved in the progression of CKD6 therefore proteinuric hyperthyroid cats could be at an increased risk of developing CKD or of progressing to end-stage renal disease.
This aim of this retrospective study was to evaluate the survival of hyperthyroid cats treated in first opinion practice with antithyroid medication alone or in combination with surgery, in order to identify prognostic indicators. Factors that predict the development of azotemia after treatment of hyperthyroidism were also investigated, in particular to assess the relationship between proteinuria and the development of azotemia.
Records from 2 London-based first opinion practices (People's Dispensary for Sick Animals, Bow, and the Beaumont Animals' Hospital, Camden) between January 1, 1999 and December 31, 2007 were reviewed and newly diagnosed hyperthyroid cats identified. Diagnosis of hyperthyroidism was based on a plasma total thyroxine concentration (TT4) greater than the laboratory reference range (>55 nmol/L, 4.26 μg/dL) or a failure of the TT4 concentration to suppress to <20 nmol/L after administration of 20 μg liothyronine sodium (T3) 3 times daily for 2 days, then 20 μg liothyronine sodium (T3) on day 3, given 2–4 hours before sampling.
The clinical records were reviewed and the following data collected from the time of diagnosis: clinic to which the cat presented, age, breed, sex, systolic arterial blood pressure (SBP), plasma TT4 concentration, PCV, routine plasma biochemical parameters (total protein, albumin, globulin, urea, creatinine, total bilirubin, cholesterol, sodium, potassium, chloride, inorganic phosphorous, total calcium concentrations and activities of alanine aminotransferase and alkaline phosphatase), urine specific gravity (USG), and urine culture result (if available). The time to development of azotemia (defined below) after treatment of hyperthyroidism was also documented. Cats were excluded from the analysis of factors associated with the development of azotemia if they were azotemic at, or before, diagnosis of hyperthyroidism.
Cats included in the study were treated for hyperthyroidism with antithyroid medication (carbimazole or methimazole) alone or in combination with thyroidectomy. Two cats were also treated with radioiodine therapy.
Blood Pressure Measurement
SBP measurements were made with an 8.1 MHz Doppler ultrasound probe following the protocol described previously.7 Fundic examination by indirect ophthalmoscopy was performed in cats with an average SBP > 160 mmHg to assess for evidence of hypertensive retinopathy. Hypertension was diagnosed if cats had an average SBP > 170 mmHg with evidence of hypertensive retinopathy, or if the average SBP was >170 mmHg on 2 consecutive occasions.
Categorization of Hypertension, Azotemia, and Control of Hyperthyroidism
Cats were categorized as hypertensive at diagnosis if they were already receiving treatment for hypertension or were diagnosed as hypertensive at the time of diagnosis of hyperthyroidism. Cats that had a mean SBP > 170 mmHg without evidence of retinopathy at the time of diagnosis of hyperthyroidism, but which were subsequently diagnosed as hypertensive at the follow-up visit (within 1–2 weeks) were also classified as hypertensive at diagnosis. In cases of concurrent hypertension and hyperthyroidism, antithyroid medication was withheld until hypertension was successfully controlled by antihypertensive medication (usually amlodipine).
Renal azotemia was defined as a plasma creatinine concentration >177 μmol/L (2.0 mg/dL) in conjunction with inadequate urine concentrating ability (USG < 1.035), or persistent azotemia on 2 or more consecutive occasions without evidence of a prerenal cause.
Hyperthyroidism was categorized as well controlled if cats appeared euthyroid on clinical examination (no tachycardia or weight loss) and had serial TT4 measurements <40 nmol/L (3.1 μg/dL) for a 6-month period. This time point was used as the 6-month postcontrol time point. The number of cats that were well controlled, based on serial TT4 measurements <40 nmol/L in the period 2–6 months after the start of treatment was also recorded.
Blood and Urine Sampling and Processing
Blood and urine samples were collected as part of a geriatric screening and healthcare program at the time of diagnosis with the consent of the owner. The Ethics and Welfare Committee of the RVC approved the diagnostic protocol. Jugular venous blood samples were collected and placed in heparinized tubes, and urine samples collected by cystocentesis. Samples were kept at 4°C before sample processing which occurred within 6 hours of sample collection. Blood samples were placed in a centrifuge at 2,016 × g for 10 minutes to enable separation of plasma from cellular components. Heparinized plasma was submitted to a single external laboratorya for biochemical analysis including TT4.
Urine samples underwent full in-house urinalysis including measurement of USG by refractometry, dipstick analysis, and urine sediment examination. If bacteria or pyuria was identified on sediment examination, urine was submitted for bacterial culture and sensitivity. Urine samples that were positive on urine culture or which were grossly hematuric were excluded from UPC and UAC analysis. Urine samples were centrifuged at 2,016 × g for 10 minutes and the supernatant separated from any sediment. This was stored at −80°C until batch analysis of UPC and UAC. Urine protein concentration was measured by a colorimetric pyrogallol red method and urine creatinine measured by a colorimetric picric acid method. Urine albumin concentrations were measured by a sandwich ELISA as described previously.4
Cats were re-examined at approximately 8-week (if treated with antithyroid medication) or 3-month (if treated surgically) intervals after diagnosis. Blood and urine samples were obtained approximately every 3–4 months.
Survival times were calculated from the date of diagnosis of hyperthyroidism to the date of death or euthanasia. If these data were not available, the owner was contacted by telephone to request follow-up information. If the cat had died and the exact date of death was unknown, the month of death was recorded and it was assumed that the cat died on the 15th of that month for the purposes of the survival analysis. Cats were censored in the survival analysis if they were still alive at the end of the follow-up period (January 1, 2009) or if they were lost to follow-up. Cats were categorized as lost to follow up if they had not attended the clinic for 6 months and their owners were not contactable by telephone.
Statistical analyses were performed by computerized statistical software.b Data were assessed graphically for normality. Results are reported as median [range]. The survival times for azotemic cats and cats with previously diagnosed CKD at diagnosis of hyperthyroidism were compared with nonazotemic cats by the log rank test. The Wilcoxon signed rank test was used to compare median UPC and UAC before treatment and at the 6-month postcontrol time point.
For the survival analysis, available continuous data for the following variables were entered into a univariable Cox regression model to determine if they were associated with survival: clinic to which cat presented, hypertensive status at diagnosis, age, sex, baseline TT4, routine plasma biochemical parameters, PCV, USG, UPC, and UAC. The aforementioned quantitative and qualitative data were also assessed with the Mann–Whitney U and Fisher's exact tests, respectively, to determine variables correlated with the subsequent development of azotemia within 240 days of diagnosis of hyperthyroidism. Variables with P < .2 were included in a manual forward selection, stepwise multivariable Cox regression analysis for survival, or a manual forward selection, stepwise multivariable logistic regression model for the development of azotemia. For the survival analysis, variables with missing values were recoded into categorical variables, including a category for missing data. Available continuous data for each variable were further divided into quartiles. Plasma creatinine and urea, and total protein and globulin, were highly correlated to one another; therefore they were not included in the same statistical models. For the logistic regression analysis, USG data were divided into quartiles and converted into categorical variables, including a category for the missing data. The models were then validated according to the methods outlined by Dohoo et al.8 First-order interactions were assessed and the relationship between the continuous predictors and outcome of interest in the logistic regression model was examined by categorizing continuous data into quartiles and computing and plotting the log odds of outcome against the category means.8 Statistical significance was set at P < .05.
Three hundred and thirteen cats were diagnosed with hyperthyroidism during the study period. Of these 313 cats, 13 were excluded from the analysis for the following reasons: lack of treatment for hyperthyroidism (7), furosemide therapy (3), concurrent diagnosis of diabetes mellitus (2), and concurrent diagnosis of hepatocellular carcinoma (1). This left 300 cats eligible for inclusion in the survival analysis. Of these 300 cats, 32 cats were azotemic at diagnosis or had previously been diagnosed with azotemic CKD. These cats were excluded from the analysis of factors associated with the development of azotemia after treatment of hyperthyroidism.
There were 166 female cats (55%, 2 entire) and 134 male cats (45%, 3 entire) eligible for inclusion in the survival analysis. Available biochemical and urinalysis data at diagnosis are summarized in Table 1. The median age at diagnosis was 14.3 [6.2–25.0] years. Seventeen cats were acquired as adults and their age was unknown. Two hundred and forty cats (80%) had died or were euthanased at or before the end of the follow-up period (January 1, 2009) with a median survival time of 417 [0–2,541] days. Sixty cats (20%) were censored from the analysis because they were still alive (n = 35) or had been lost to follow-up (n = 25). The median follow-up time for censored cases was 555 [0–2,312] days.
|TT4 (μg/dL)||300||9.03 [3.38–23.25]|
|Age (years)||285||14.3 [6.2–25.0]|
|Urea (mg/dL)||246||29.4 [16.2–135.2]|
|Creatinine (mg/dL)||246||1.17 [0.47–4.46]|
|Total protein (g/dL)||245||7.0 [4.8–9.0]|
|Albumin (g/dL)||245||3.1 [1.8–4.1]|
|Globulin (g/dL)||245||3.8 [2.5–6.3]|
|ALT (IU/L)||244||167.6 [3.0–1,879.7]|
|ALP (IU/L)||244||103.9 [5–997.1]|
|Phosphate (mg/dL)||244||5.02 [2.17–15.7]|
|Calcium (mg/dL)||244||9.64 [8.1–12.4]|
|PCV (%)||285||36 [18–55]|
|UAC (mg/g)||186||38.7 [2.7–7,277.4]|
Cats with azotemia or previously diagnosed CKD at the time of diagnosis of hyperthyroidism had significantly shorter median survival times (178 [0–1,505] days) than nonazotaemic hyperthyroid cats (612 [0–2,541] days; P < .001).
Thirteen cats (4.3%) had been diagnosed as hypertensive before diagnosis of hyperthyroidism and 30 cats (10%) were classified as hypertensive at the time of diagnosis of hyperthyroidism.
Of the 300 cats studied, 123 were documented to have a 6-month period of good control, and paired data for UPC and UAC before treatment and at the 6-month post control time point were available in 42/123 of these well-controlled cats. Of the 177 cats that were not documented to have achieved 6 months of good control, 74 had died or were lost to follow-up within this 6-month period, 70 cats were documented to be euthyroid after instituting treatment but had no sample taken at 6–9 months after treatment commenced, and 33 cats were never well controlled. UPC decreased significantly with treatment (pretreatment 0.32 [0.01–3.40], posttreatment 0.18 [0.05–3.65]; P < .001, n = 42), while UAC did not (pretreatment 33.9 [3.13–3,256.47] mg/g, posttreatment 21.3 [2.47–1,539.40] mg/g; P= .105, n = 42).
There were 150 female cats (2 entire) and 118 male cats (3 entire) eligible for inclusion in the analysis of factors associated with development of azotemia. The median age at diagnosis of hyperthyroidism in these cats was 14.1 [6.2–24.0] years. Forty-one cats (15.3%) developed azotemia within 240 days of diagnosis and initiation of treatment of hyperthyroidism. There were 106 initially nonazotemic cats that were defined as well-controlled in the 2–6-month period after treatment, of which 28 (26.4%) became azotemic. Out of 39 cats that were defined as poorly controlled in the 2–6-month period after treatment, 3 (7.7%) became azotemic. The other 123 initially nonazotemic cats either had variable control over this period, or their degree of control could not be assessed because of lack of follow-up. Ten of these 123 cats (8.1%) became azotemic.
Age, hypertension, urea, creatinine, USG, total protein, globulin, phosphate, PCV, and UPC were correlated with survival at the 20% level (P < .2) and were taken forward for consideration in the multivariable model (Table 2). The following factors at diagnosis were excluded from further analysis because they were not significantly correlated with survival (P > .2): clinic at which presented, sex, TT4, UAC, plasma concentrations of albumin, total bilirubin, cholesterol, sodium, potassium, chloride, and total calcium and plasma activities of ALT and ALP. In the final multivariable model (Table 3), survival was negatively correlated with age (P < .001), UPC (P= .007) and the presence of hypertension (P= .026). Survival was also positively correlated with USG (P < .001) and PCV (P < .001). Plasma urea concentration tended toward a significant association with survival (P= .059), however, plasma creatinine did not (P= .175). The missing data category for age and USG was significantly associated with survival.
|Variable||n||Sig.||Hazard Ratio (HR)||95% CI for HR|
|Age (years)||285||< 0.001||1.128||1.078–1.182|
|Urea (mg/dL)||246||< 0.001||1.028||1.020–1.036|
|Creatinine (mg/dL)||246||< 0.001||1.763||1.363–2.282|
|Total protein (g/dL)||245||0.062||1.213||0.990–1.485|
|PCV (%)||285||< 0.001||0.933||0.913–0.954|
|Hypertensive status (normotensive)||300||< 0.001||0.470||0.335–0.66|
|n||B||SE||Sig.||Hazard Ratio||95% CI for Hazard Ratio|
|Age (< 12.3 years)||71||<0.001|
|Plasma urea (< 23.8 mg/dL)||63||0.059|
|PCV (< 32%)||77||<0.001|
|USG (< 1.024)||62||<0.001|
|UPC (< 0.31)||46||0.007|
|Hypertensive status (normotensive)||257||0.026|
First order interactions were assessed between age, USG, PCV, UPC, hypertensive status, and plasma urea concentration. There were no significant interactions (P < .05) identified.
Graphical assessment of categorical variables by plotting survival time against the log cumulative hazard confirmed that the assumption of proportional hazards was satisfied.
Development of Azotemia
Univariable analysis identified urea, creatinine, USG, total protein, TT4, globulin, UPC, ALT activity, the presence of hypertension (P= .172; OR 1.92; 95% CI for OR 0.68–5.21) and ALP activity as factors correlated (P < .2) with the development of azotemia after treatment (Table 4).
|Variable||Azotemic Cats||Non-Azotemic Cats||Sig.|
|n||Median [25th, 75th Percentile]||n||Median [25th, 75th Percentile]|
|TT4 (μg/dL)||41||8.14 [5.78, 11.2]||227||9.84 [6.86, 15.0]||0.018|
|Urea (mg/dL)||34||34.3 [28.42, 42.06]||183||27.4 [23.0, 32.8]||< 0.001|
|Creatinine (mg/dL)||34||1.31 [1.18, 1.69]||183||1.07 [0.90, 1.28]||< 0.001|
|Total protein (g/dL)||33||6.57 [6.30, 7.19]||183||6.98 [6.66, 7.50]||0.011|
|Globulin (g/dL)||33||3.55 [3.16, 4.11]||183||3.84 [3.50, 4.32]||0.027|
|ALT (IU/L)||33||155.0 [103.7, 218.6]||182||173.9 [118.7, 289.5]||0.151|
|ALP (IU/L)||33||92.2 [63.9, 127.7]||182||108.9 [67.3, 156.8]||0.188|
|USG||34||1.030 [1.020, 1.049]||159||1.040 [1.028, 1.060]||0.002|
|UPC||32||0.37 [0.27, 0.61]||142||0.54 [0.34, 0.82]||0.05|
Urinalysis data had the most missing values, so to allow inclusion of as many cases as possible, UPC and USG were included in the model as categorical variables and plasma globulin and creatinine as continuous variables. Because of missing biochemistry data before treatment, 52 out of 268 cases were excluded. Within the 240-day follow-up period, 33/216 cats became azotemic. The development of azotemia was positively correlated with pretreatment plasma creatinine concentration (P < .001) and negatively correlated with plasma globulin concentration (P= .006, Table 5). When pretreatment plasma urea concentration was substituted for plasma creatinine concentration in the model then the development of azotemia was positively correlated with plasma urea concentration (P < .001) and negatively correlated with plasma globulin concentration (P= .031) (model not shown). The coefficients (B) and odds ratios for USG and plasma globulin were similar for models that included plasma creatinine or urea concentration. Plots of log odds of development of azotemia against category means for plasma creatinine concentration revealed that the log odds of developing azotemia increased with increasing categorical plasma creatinine concentration and was approximately represented by a linear association. The log odds of developing azotemia decreased with decreasing categorical plasma globulin concentration and was approximately represented by a linear association.
|Explanatory Variable||B||SE||Sig.||Odds Ratio (OR)||95% CI for OR|
|Creatinine (mg/dL)||2.863||0.668||< 0.001||17.513||4.733–64.805|
There were no significant (P < .05) first order interactions between plasma creatinine or urea concentration and plasma globulin concentration. The Hosmer-Lemenshow goodness of fit test confirmed good model fit (P= .439).
The results of this study indicate that age, UPC, and the presence of hypertension are significantly and independently correlated with decreased survival time, whereas USG and PCV are significantly and independently correlated with increased survival time after diagnosis and treatment of hyperthyroidism.
The presence of hypertension was significantly correlated with a reduced survival time in this study. Previous studies of cats with CKD4 and hypertension9 did not report an association between the presence of hypertension and reduced survival time. It was suggested that this was because of adequate treatment of hypertension in these cases.9 It may be that adequate treatment of hypertension in hyperthyroid cats is more difficult to achieve, because of the effects of hyperthyroidism on systemic vascular resistance and cardiac output. Alternatively, the presence of hypertension at diagnosis of hyperthyroidism could be related to the presence of concurrent CKD.
Recent studies suggest that proteinuria is of prognostic significance in cats with azotemic CKD and hypertension,4,9 and the results of this study confirm that proteinuria is correlated with reduced survival in cats with hyperthyroidism. However, UAC was not related to survival time, unlike in cats with CKD. The median UPC decreased after treatment of hyperthyroidism, in agreement with previous studies,5,c however, the UAC did not decrease significantly with treatment, although a trend towards a decrease was evident. This suggests that the proteinuria that occurs in hyperthyroidism is because of increased excretion of proteins other than albumin. Detailed analysis of the urinary proteome from untreated hyperthyroid cats would be necessary to investigate this further.
In many cats treated for hyperthyroidism, pre-existing CKD only becomes apparent once euthyroidism is achieved and the GFR normalizes. In this study, 15.3% of cats that were nonazotemic at diagnosis of hyperthyroidism became azotemic within 240 days of diagnosis. Previous studies have reported an incidence of azotemia after treatment of hyperthyroidism of 17–49%.10–15 Many of the cats in the present study were inadequately treated. Inadequate control may not have normalized GFR sufficiently for azotemia to occur, despite underlying CKD. However, in this study the incidence of azotemia in cats that were defined as having good control of hyperthyroidism in the 2–6-month period after treatment was 26.4%, which is also less than in some previous studies. Alternatively, the lower incidence of posttreatment azotemia in this study could be attributed to a low incidence of biochemical hypothyroidism in these, predominantly medically treated, cats. Iatrogenic hypothyroidism has been reported in cats after radioiodine treatment and bilateral thyroidectomy,d which have been the predominant treatment modalities in previous studies. Hypothyroidism has been correlated with reduced GFR in humans and dogs.16,17 If a cat had mild CKD (IRIS stage I or IIa),18 hypothyroidism could lead to the development of azotemia.
The median survival time for hyperthyroid cats in this study was 417 days, which is shorter than reported previously. Previous studies examining cats treated with radioiodine identified a median survival time of 25 months,12 24 months,19 and 4 years,20 however, the cats referred for radioiodine treatment are generally younger and healthier than those that are not. Previous studies may only represent a selected population, whereas the present study is more representative of all hyperthyroid cats diagnosed in first opinion practice.
The present study demonstrated that cats with higher pretreatment plasma urea or creatinine concentrations, or lower plasma globulin concentrations, were more likely to develop azotemia within 240 days after treatment of hyperthyroidism. USG and UPC were not correlated with the development of azotemia.
Hyperthyroid cats with lower plasma globulin concentrations were more likely to develop azotemia after treatment of hyperthyroidism. This was an unexpected result and to the authors' knowledge has not been reported previously, though this could be explained by a type-1 error. Further, studies of the relationship between plasma globulin and the development of azotemia in cats with and without hyperthyroidism are warranted.
UPC was not found to be correlated with the development of azotemia in hyperthyroid cats, despite the relatively high number of cats considered to have significant proteinuria (>0.4) according to the IRIS staging guidelines. Proteinuria has been reported to be associated with the development of azotemia in cats,21 although the reason for the relationship between proteinuria and development of azotemia is still the subject of conjecture. Proteinuria may be a marker of more rapidly progressive renal disease, for example because of more glomerular injury being present, or could be a marker of glomerular hypertension. Alternatively, the presence of proteins in the urine could be a mediator of renal disease progression.4 In experimental models, the exposure of proximal tubular cells to albumin at high concentrations results in the activation of an inflammatory cascade which may stimulate interstitial fibrosis and progression of CKD.6 The results of the present study suggest that proteinuria is not a mediator of renal disease progression. This could be because the urinary proteins associated with hyperthyroidism are different to those associated with CKD. This study demonstrated that there was no change in the UAC with treatment of hyperthyroidism despite a significant decrease in UPC, suggesting that albumin may not be the component of proteinuria, which is increased in hyperthyroidism. Further investigations to establish if there are differences in the nature of the proteinuria associated with these conditions are warranted.
One of the main limitations of the present study was the lack of availability of a full data set for all cats because of the retrospective nature of the study. Stored samples were used wherever possible to increase the data available; however, in many cases no stored samples were available. Cats with missing data could represent a subpopulation of hyperthyroid cats, and so exclusion of these cases may bias the population in the study. For example, cats with good renal function are likely to be less polyuric and therefore more difficult to obtain a urine sample from by cystocentesis. Exclusion of these data would therefore bias the study toward cats with poorer renal function. The results of the present study would suggest that this is the case, as cats with missing data had a reduced hazard of death when compared with cats in the reference category (USG < 1.024), and had a similar hazard of death as cats with USG > 1.046. This suggests that cats with missing urinalysis data had good renal concentrating ability and renal function.
The majority of cats in the present study were treated with antithyroid medication alone or in combination with thyroidectomy. The choice of treatment modality may affect the survival of cats; however, to the authors' knowledge, no studies have compared the survival times of cats treated with antithyroid medication with those treated by thyroidectomy. One study did demonstrate that hyperthyroid cats treated with radioiodine had a significantly longer median survival time than cats treated with methimazole alone,20 however, it is possible that cats treated with radioiodine were selected based on age and renal status. The study was not a randomized trial and therefore it may be inappropriate to draw conclusions as to the effect of treatment type on survival of hyperthyroid cats.20 Equally, the present study was performed in a retrospective manner, and the choice of treatment modality used was likely to be biased according to the age and renal status of the patient, and in some cases the costs associated with performing surgery. Therefore, in the present study, cats treated for hyperthyroidism by thyroidectomy are likely to represent a selected subgroup of hyperthyroid cats. It is likely that younger and healthier cats, which are more likely to survive, would have been selected for thyroidectomy, and older cats with renal insufficiency were more likely to be maintained on medical therapy. Therefore the effect of treatment type on survival could not be adequately addressed by the present study.
This study did not seek to identify the cause of death of these animals, which may have been of value when interpreting the results of survival analysis. Hyperthyroidism is a disease of geriatric cats, a population which often have multiple medical problems resulting in comorbidity. Identifying a single cause of death is frequently not possible. In most cases cats were euthanized, and did not die of natural causes. The decision for euthanasia is made by the owner and may be prompted by a number of factors including decreased quality or life, financial considerations, or the ability of the owner to manage the cat's the condition at home. The reason for euthanasia was often not recorded and so the cause of death cannot be ascertained in these cases easily.
In summary, UPC was identified in the present study as a factor associated with survival in hyperthyroid cats at diagnosis; however, it was not related to the development of azotemia after treatment.
aIDEXX Laboratories, Wetherby, UK
bSPSS 16.0 for Windows, SPSS Inc, Chicago, IL
cSyme HM. Evaluation of proteinuria in hyperthyroid cats. J Vet Intern Med 2001;2001:299 (abstract)
dGraham P. Measurement of feline thyrotropin using a commercial canine-specific immunoradiometric assay. J Vet Intern Med 2000; 14:342 (abstract)
- 2Feline renal-failure—questions, answers, questions. Compend Contin Educ Pract Vet 1992;14:127–152., , , et al.
- 8Veterinary Epidemiologic Research. Charlottetown, PEI: AVC Inc; 2003; 706pp., ,
- 18ElliottJ, GrauerGF eds. BSAVA Manual of Canine and Feline Nephrology and Urology, 2nd ed. Quedgeley, UK: British Small Animal Veterinary Association; 2007; 298pp.