Proteinuria and systemic hypertension are well recognised risk factors in chronic renal failure (crf). They are consequences of renal disease but also lead to a further loss of functional kidney tissue. The objectives of this study were to investigate the associations between proteinuria, systemic hypertension and glomerular filtration rate (gfr) in dogs with naturally occurring renal and non-renal diseases, and to determine whether proteinuria and hypertension were associated with shorter survival times in dogs with crf. Measurements of exogenous creatinine plasma clearance (ecpc), urine protein:creatinine ratio (upc), and Doppler sonographic measurements of systolic blood pressure (sbp) were made in 60 dogs with various diseases. There was a weak but significant inverse correlation between upc and ecpc, a significant inverse correlation between sbp and ecpc and a weak but significant positive correlation between upc and sbp. Some of the dogs with crf were proteinuric and almost all were hypertensive. Neoplasia was commonly associated with proteinuria in the dogs with a normal ecpc. crf was the most common cause leading to hypertension. In the dogs with crf, hypertension and marked proteinuria were associated with significantly shorter survival times.
Statistics from Altmetric.com
NATURALLY occurring chronic renal failure (crf) in dogs is progressive and typically ends in uraemia and death. Its primary cause is often undetermined. Its progression is attributed to the persistence of the original cause and/or to self-perpetuating mechanisms once the functional mass of the kidneys has been reduced to a critical value (Finco and others 1999). Developing effective interventions to slow down the progression of the disease requires the identification of risk factors. Proteinuria and systemic hypertension are consequences of renal disease but also lead to a further loss of functional kidney tissue. The level of proteinuria has been associated with the rate of progression of renal disease in both human beings and dogs, and persistent proteinuria is associated with increased mortality (Williams and Coles 1994, Jacob and others 2005). In human beings, a decrease in the level of proteinuria after treatment with angiotensin-converting enzyme (ace) inhibitors reduces the rate of progression of renal failure (Maschio and others 1996, Jacob and others 2005).
Proteinuria can be caused by changes in the vascular permeability of glomerular capillary walls and/or by impaired tubular handling of filtered proteins (Lees and others 2005). The integrity of the walls of the glomerular capillaries prevents macromolecules and cellular elements from passing into the filtrate, but disruptions in them allow various degrees of proteinuria and haematuria to occur. Changes in the membrane (size selectivity) and ionic charge abnormalities (charge selectivity) can be the cause.
Serum proteins may injure both the glomerular mesangial cells and the proximal tubular cells. Mesangial cell injury has been attributed to the accumulation of lipoproteins and their oxidation products, which cause increased production of matrix, leading to monocyte activation and the generation of growth factors that stimulate sclerosis. Overloading the tubular cells with large quantities of protein activates the proximal tubular epithelial cells to upregulate genes encoding endothelin, chemokines and cytokines. These vasoactive and inflammatory substances are released mainly into the basolateral compartment, leading to chemotaxis of inflammatory cells into the renal interstitium and subsequent renal scarring as a result of fibrogenic actions (Finco and others 1999, Zoja and others 1999).
The most accurate way to determine proteinuria is to collect a 24-hour urine sample and measure the total amount of protein excreted. However, this is often impractical and the urine protein:creatinine ratio (upc) can be used to estimate the 24-hour protein loss. Studies in human beings have shown that there is a good correlation between the upc and 24-hour protein excretion (Morales and others 2004, Xin and others 2004, Yamasmit and others 2004).
The association between crf and systemic hypertension is well recognised in several species, and a relationship between initial blood pressure and mortality has been reported (Finco 2004). Dogs with high systolic blood pressures (sbps) have a significantly greater risk of developing a uraemic crisis and dying than dogs with a lower sbp (Jacob and others 2003). A direct relationship between a high sbp and high levels of proteinuria has also been reported (Jacob and others 2005). Hypertension caused by crf leads to a deterioration in renal function and to renal lesions. A proposed mechanism is that systemic hypertension, in addition to causing abnormal glomerular morphology and function, may result in increased intraglomerular capillary pressure (Finco and others 1999). Studies in human beings and rats have indicated that treatment with ace inhibitors, which reduce blood pressure, have a beneficial effect by reducing the rate of decline of renal function (Anderson and others 1985, Maschio and others 1996).
To understand the aetiopathogenesis and treatment of hypertension, it is necessary to identify the factors that control arterial blood pressure. Arterial pressure is determined by the product of cardiac output and total peripheral resistance; cardiac output is determined by the heart rate and stroke volume. Disease processes that result in uncompensated increases in cardiac output and/or total peripheral resistance result in hypertension. Changes in the regulation of blood pressure that may occur as a result of crf include sodium retention, increases in the volume of extracellular fluid, activation of the renin-angiotensin-aldosterone system, increased levels of noradrenaline, increased vascular responsiveness to noradrenaline, decreased activity of vasodilatory substances, increased cardiac output, increased total peripheral vascular resistance, and secondary hyperparathyroidism (Bartges and others 1996).
The diagnosis of systemic hypertension in dogs is dependent on an accurate determination of blood pressure. Blood pressure is measured as systolic arterial pressure (sap) and diastolic arterial pressure (dap), both of which can be measured by direct and indirect methods. In veterinary medicine oscillometric and Doppler ultrasonic methods are established as indirect, non-invasive procedures (Bartges and others 1996). There is controversy about the relative importance of sap and dap in the pathogenesis of cardiovascular and renal disease, but in some studies sap has been considered to be more important and an increase in sap is usually accompanied by a corresponding increase in dap (Finco 2004).
The aims of this study were, first, to investigate whether proteinuria and hypertension are correlated with the level of renal excretory function, and whether proteinuria accompanies hypertension and vice versa, and secondly to determine whether non-renal diseases might be associated with proteinuria and/or hypertension, and whether proteinuria and hypertension are associated with shorter survival times in dogs with crf.
MATERIALS AND METHODS
An exogenous creatinine plasma clearance (ecpc) test was performed in 60 dogs with various renal and non-renal diseases. The dogs were six months to 15 years old (median five years). Twenty-five of them were female, 16 spayed, and 35 were male, eight of them neutered. Their bodyweights ranged from 7 to 51·8 kg (median 27·2 kg). There were six Bernese mountain dogs, four boxers, three beagles, three German shepherd dogs and two West Highland white terriers, 15 dogs of other breeds and 27 mixed-breed dogs.
The glomerular filtration rate (gfr) was determined in 27 of the dogs in which renal insufficiency was suspected, and in the other 33 to assess their renal function before they were treated with potentially nephrotoxic substances. The final diagnoses were based on histopathology and/or cytology in 24 of the dogs, on serology in 11, on their ecpc in 17, and other methods (endocrine testing, ultrasonography and echocardiography) in eight. Dogs with fever, haematuria, and pyuria were excluded to minimise prerenal and postrenal impacts on the upc ratio. Dogs treated with antihypertensive drugs were also excluded from the study.
Measurements of sbp, upc and ecpc
Measurements of sbp were made by Doppler ultrasonography (Ultrasonic Doppler Flow Detector; Parks Medical Electronics), before any other diagnostic procedures were performed. All measurements were made in a quiet environment. Neonatal cuffs with a width 40 per cent of the circumference of a forelimb were used, and the probe was applied to a clipped area ventral to the dewclaw. An average value was calculated from three consecutive measurements made daily over three days, and values up to 140 mmHg were considered normal.
The dogs were fasted for 12 hours. They were all clinically well hydrated. A complete blood count (cbc) and chemistry profile was performed. Blood samples were obtained by jugular venepuncture. edta blood for the cbc was analysed immediately (CellDyn; Abbott), and serum was separated within 30 minutes of collection and analysed with a commercial automatic analyser (Hitachi 911; Roche). Urine samples were collected by cystocentesis under ultrasonographic guidance and analysed immediately by dipstick, and for specific gravity, and the sediment was also analysed. Urine total protein was measured by a micro-turbidimetric method and urine creatinine was measured using an enzymatic procedure, both on the Hitachi 911 automatic analyser, and the upc ratio was calculated. The urine samples for the determinations of upc were stored at 4°C and analysed within 24 hours of collection. A upc ratio less than 0·5 was considered normal, a ratio between 0·5 and 1·0 was considered to be dubious, and a ratio of 1·0 or more was considered to be increased.
The ecpc test was carried out with a single intravenous injection of 5 per cent creatinine solution (Creatinine anhydrous 5 per cent solution; Sigma-Aldrich). A dose between 60 and 125 mg/kg was chosen depending on the bodyweight of the dog, lower concentrations being used in larger dogs. During the next 10 hours, 12 to 15 blood samples were taken and the serum creatinine concentration was measured enzymatically as described above. The area under the curve (auc) was calculated by the trapezoidal method using a non-compartmental model (Heiene and Moe 1998). The ecpc was calculated as the amount of creatinine injected auc × 100. According to recent studies, this is a valid method for assessing gfr in dogs (Watson and others 2002, Hochel and others 2004, Kerl and Cook 2005). An ecpc of 3 ml/minute/kg or more was considered normal; values ranging from 2·00 to 2·99 ml/minute/kg were considered to indicate a mild reduction in renal excretory function, and lower values were considered to indicate a moderate reduction.
The dogs with crf were followed up regularly, and a cbc, blood chemistry profile, urine analysis and measurements of upc and sbp were made every three months in the dogs that lived more than three months after the performance of the ecpc. Death or euthanasia of the dogs that did not survive was attributed to progressive renal failure if their clinical signs (for example, lethargy, anorexia and vomiting) could not be explained by other causes and, if their blood urea nitrogen (bun) was at least three times the upper limit of the reference range.
The data were analysed by using spss 13.0. Pearson's correlation analysis was used to assess the correlations between sbp, upc and ecpc, and to decide whether upc and sbp were influenced by non-renal diseases. Kaplan-Meier analysis with the log rank test was used to determine whether hypertension and proteinuria in the dogs with crf were associated with time to death from renal causes. Significance was defined as P≤0·05.
Measurements of sbp, upc and ecpc
The dogs' serum creatinine concentrations ranged from 42 to 590 μmol/l (median 91 μmol/l); 43 per cent of the values were above the reference range, that is, more than 106 μmol/l. Their bun concentrations ranged from 2·12 to 48·4 mmol/l (median 7·14 mmol/l), and 23 of the 60 dogs (38 per cent) had a value above 8·3 mmol/l. The dogs' upc ratios ranged from 0·05 to 21·7 (median 0·34), and 21 dogs (35 per cent) had a upc above 0·5. Their sbp ranged from 90 to 240 mmHg (median 130 mmHg), and 21 (35 per cent) of them had a high sbp. The results for the ecpc ranged from 1·15 to 5·03 ml/minute/kg (median 2·6 ml/minute/kg); 23 (38 per cent) of the dogs had normal values of 3 ml/minute/kg or above, and the other 37 had lower values; of these, 22 had a mild reduction (2·00 to 2·99 ml/minute/kg) and 15 had a moderate reduction (≤1·99 ml/minute/kg). Other abnormalities included a haematocrit less than 35 per cent in nine, thrombocytopenia (<150 × 109/l) in six, thrombocytosis (>500 × 109/l) in four, leucopenia (<5 × 109/l) in three, leucocytosis (>16 × 109/l) in 11, hypoalbuminaemia (<25 g/l) in four, high activities of alanine aminotransferase (alt) (>91 U/l) in seven, high activities of alkaline phosphatase (alp) (>225 U/l) in 18, hypophosphataemia (<0·97 mmol/l) in one, and hyperphosphataemia (>2·36 mmol/l) and hypocalcaemia (<2·3 mmol/l) in 10. None of the dogs was hypercalcaemic. The final diagnoses are listed in Table 1. All the dogs with ‘neoplasia’ suffered from malignant lymphoma, and the dogs with ‘neoplasia and crf’ had malignant lymphoma and crf. Two of the dogs with ‘infection’ were diagnosed with dirofilariosis, one was diagnosed with leishmaniosis, one with babesiosis, three with concurrent babesiosis and dirofilariosis, one with leishmaniosis and dirofilariosis, and two with babesiosis, ehrlichiosis and dirofilariosis. One of the two dogs with endocrinopathies in the group ‘other diseases’ had Cushing's disease and the other had hypothyroidism.
Correlations between upc, sbp and ecpc
Taking the results from all 60 dogs into account, there was a weak but significant inverse correlation between upc and ecpc (r=−0·284; P<0·05) (Fig 1), a stronger inverse correlation between sbp and ecpc (r=−0·515; P<0·01) (Fig 2), and a weak positive correlation between upc and sbp (r=0·387; P<0·01) (Fig 3).
Association of upc and sbp with the degree of renal function
The dogs were assigned to three groups according to the results of their ecpc; group 1 (23 dogs) had normal values (≥3 ml/minute/kg); group 2 (22 dogs) had a slightly reduced ecpc (2·00 to 2·99 ml/minute/kg), and group 3 (15 dogs) had a moderately reduced ecpc (≤1·99 ml/minute/kg). The upc ratios and sbp values were compared within these groups. The median upc ratio in group 1 and in group 2 was within the normal range, whereas it was mildly increased in group 3. The dogs in group 3 had the widest range of upc values (Fig 4a). The median sbp values in groups 1 and 2 were within the normal range, whereas the dogs in group 3 had significantly higher median sbp values (Fig 4b).
Association of upc and sbp with the final diagnosis
The dogs were assigned to five groups according to their final diagnosis (Table 1), and the upc ratios and sbp values in the different groups were compared to assess the influence of non-renal diseases on proteinuria and hypertension. The median upc ratios were within normal limits in all the groups, but the dogs with neoplasia had the widest range of upc ratios (Fig 5a). The dogs with crf (group B) had the highest median sbp measurements, and the median sbp values in all the other groups were within the normal range (Fig 5b).
Association between upc and death from renal causes
Of the 24 dogs with crf, 14 had a normal upc ratio, and 10 had a high upc ratio (≥0·5). The final outcomes in this group were as follows: 10 patients were alive, one had died and 13 had been euthanased at the time of the analysis owing to progressive renal failure and uraemia in all cases as defined previously. Kaplan-Meier analysis indicated that the dogs with a normal upc ratio lived for a mean (se) of 1006 (169) days (95 per cent confidence interval [ci] 676 to 1336 days), compared with 758 (186) days (95 per cent ci 394 to 1123 days) for the dogs with a high upc ratio. However, the log rank test indicated that these results were not significant (Fig 6a). The log rank test was highly significant (P<0·01) for the dogs with marked proteinuria (upc ≥1); these dogs lived for a mean of 390 (238) days (95 per cent ci 0 to 858 days), whereas the dogs with a upc ratio less than 1 lived for 1053 (139) days (95 per cent ci 781 to 1326 days) (Fig 6b). The results for the dogs in other disease groups were not analysed because their death or euthanasia may have been due to non-renal causes.
Association between sbp and death from renal causes
Of the 24 dogs with crf, 11 had a normal sbp, and 13 had a high sbp. Kaplan-Meier analysis of this group indicated that the dogs with a normal sbp lived for 1151 (140) days (95 per cent ci 876 to 1426 days), whereas the dogs with a high sbp lived for 634 (164) days (95 per cent ci 313 to 956 days). The log rank test indicated that these results were significant (P<0·05) (Fig 7). The results for the dogs in other disease groups were not analysed because their death or euthanasia may have been due to non-renal causes.
Proteinuria and hypertension are recognised risk factors in human patients with renal disease, and increasing proteinuria and hypertension are associated with declining renal function (Ross 1992, Burton and Harris 1996, Cortes and others 1996, Klahr 1996, 1999, Maschio and others 1996, Jerums and others 1997, Zoja and others 1999, Ots and others 2000, Walls 2001, Wolf and others 2003). However, there have been few studies of the relationships between proteinuria and hypertension and the renal excretory function of dogs (Finco and others 1999, Jacob and others 2003, 2005, Finco 2004). The results of this study show that there was a significant (P<0·05) but weak overall association (r=−0·284) between declining renal excretory function and increasing proteinuria. Some dogs with a moderately reduced ecpc excreted more protein in the urine than dogs with minor or no renal excretory dysfunction, but most of the dogs with crf were only mildly proteinuric (upc <1·0) or were not proteinuric. It has been suggested that the degree of proteinuria could decline once a critical number of nephrons have been lost (D'Amico and Bazzi 2003). However, clinical trials in human beings have not supported this theory; the level of proteinuria remained stable or even increased during the course of progressive renal disease (McMorrow and others 1982).
There are two possible explanations why there was no strong correlation between the dogs' proteinuria and their renal function. First, non-renal diseases such as neoplasia (especially lymphoproliferative cancers), chronic infections, such as ehrlichiosis, leishmaniosis and dirofilariosis, and endocrinopathies, such as Cushing's disease and diabetes mellitus, can be associated with urinary protein loss despite normal renal excretory function (Buoro and Atwell 1983, Leifer and Matus 1985, Codner and others 1992, Hurley and Vaden 1998, Struble and others 1998, Lautzenhiser and others 2003, Zatelli and others 2003). In this study, the dogs with neoplasia had the widest range of upc ratios; all of them were diagnosed with malignant lymphoma, and proteinuria can be a feature of this disease. The pathophysiology includes changes in glomerular permselectivity, probably due to a defect in charge selectivity. The glomerular sieving dysfunction may be associated with an inflammatory response to the malignancy (Pedersen and Johnsen 2005). Although chronic infections can be associated with urinary protein loss, none of the dogs in the ‘infection’ group was proteinuric. The proposed pathophysiology of infections includes minimal-change glomerulopathy in Ehrlichia species infection and glomerulonephritis in Leishmania species infection (Codner and others 1992, Zatelli and others 2003). The one dog with Cushing's disease that was categorised in the group ‘other diseases’ had a high upc ratio. The underlying mechanism is not known precisely, but in most cases the proteinuria resolves after the successful treatment and resolution of hypercortisolism (Hurley and Vaden 1998). The second possible explanation of the discrepancy between the gfr and proteinuria might be the broad spectrum of renal diseases that were included in the study. Some of the dogs had primarily glomerular lesions and others had tubular lesions. The early stages of glomerular diseases can be associated with normal renal excretory function (D'Amico and Bazzi 2003), but tubular diseases with a significant reduction in renal excretory function can be accompanied by minimal urinary protein loss (Zini and others 2004).
There were only 24 dogs with crf available to investigate the influence of proteinuria on survival times. However, the dogs with marked proteinuria (upc ≥1) had significantly shorter survival times (P<0·01); milder degrees of proteinuria (upc ≥0·5<1·0) did not lead to significantly shorter survival times.
The results of the present study show that there was a significant association (P<0·01) between increasing sbp and declining excretory renal function in all the dogs; however, when different degrees of renal function were compared hypertension was common only in advanced cases, in agreement with studies by Jacob and others (2003). A mild reduction in renal excretory function was not associated with an increase in blood pressure. Other diseases, such as neoplasia or infections, were not associated with hypertension, and only one dog in the group ‘other diseases’, which was diagnosed with Cushing's disease, was hypertensive. Endocrinopathies such as diabetes mellitus and Cushing's disease can be associated with hypertension, and increased pressor sensitivity to endogenous catecholamines is believed to be the main mechanism (Martinez and others 2005). Hypertension in the dogs with crf was associated with significantly shorter survival times (P<0·05).
There was a weak (r=0·387) but significant (P<0·01) correlation between increasing upc ratios and sbp values, indicating that hypertension was not always associated with proteinuria, and vice versa. Earlier studies have suggested a stronger relationship between hypertension and proteinuria (Anderson and others 1985, Cortes and others 1996, Finco and others 1999, D'Amico and Bazzi 2003, Finco 2004). There is no known pathophysiological mechanism to explain why severe proteinuria should cause hypertension, although it has been shown that dogs with severe glomerulonephritis tend to have higher blood pressure values than normotensive dogs (Langston 2003). Human patients with glomerulonephritis and hypertension are usually older than patients with glomerulonephritis who do not have hypertension (Corpa and Soares 2002), and children with glomerulonephritis are usually normotensive (Yagi and others 2005). These observations indicate that other factors such as arteriosclerosis may be involved in the pathogenesis of hypertension. Dogs with glomerulonephritis are often critically ill and suffer from syystemic inflammatory response syndrome (sirs). The observed hypertension is more likely to be a result of sirs than of the renal lesion itself when gfr is within the normal range. The proposed mechanism to explain why hypertension could lead to proteinuria is an increase in the intraglomerular capillary hydraulic pressure that exacerbates proteinuria. Abnormalities in glomerular endothelial, epithelial and mesangial cell morphology and function have been described in affected individuals (Schwartz and Bidani 1993, Cortes and others 1996, Finco and others 1999, Ots and others 2000, Ikee and others 2006).
The results of this study show that declining excretory renal function in dogs can be associated with increasing proteinuria, but many of the affected dogs were not proteinuric. The dogs with mildly reduced gfr were normotensive and the dogs with moderately reduced gfr had an increased likelihood of being hypertensive. Hypertension was not necessarily associated with proteinuria and vice versa. Hypertension and marked proteinuria in the dogs with crf were associated with a significantly higher mortality. crf was the most common disease category to cause hypertension. Proteinuria was observed in several of the dogs with neoplasia.
The authors would like to thank Dr Michael D. Willard for his comments on the manuscript.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.