Comparison of a digital and an optical analogue hand-held refractometer for the measurement of canine urine specific gravity

Urine specific gravity (USG) is used clinically as a measure of urine concentration, and is routinely assessed by refractometry. A comparison between optical analogue and digital refractometers for evaluation of canine urine has not been reported. The aim of this study was to compare a digital and an optical analogue hand-held refractometer for the measurement of canine USG, and to assess correlation with urine osmolality. Prospective study. Free-catch urine samples were collected from 285 hospitalised adult dogs, and paired USG readings were obtained with a digital and an optical analogue refractometer. In 50 dogs, urine osmolality was also measured using a freezing point depression osmometer. There was a small but statistically significant difference between the two refractometers (P<0.001), with the optical analogue refractometer reading higher than the digital refractometer (mean difference 0.0006, sd 0.0012). Paired refractometer measurements varied by <0.002 in 91.5 per cent of cases. The optical analogue and digital refractometer readings showed excellent correlation with osmolality (r=0.980 and r=0.977, respectively, P<0.001 in both cases). Despite statistical significance, the difference between the two refractometers is unlikely to be clinically significant. Both instruments provide an accurate assessment of USG in dogs.     

Comparison of digital and optical hand-held refractometers for the measurement of feline urine specific gravity

Measuring urine specific gravity (USG) is an important component of urine analysis as it evaluates renal concentrating capability. The objective of this study was to quantify the difference in USG values between a hand-held optical analogue refractometer and a cat-specific digital instrument. Urine samples from 55 cats were assessed. There was a statistically significant difference between these two refractometers (P<0.001), with the optical refractometer (mean USG=1.031) consistently reading higher than the digital refractometer (mean USG=1.027). Results for a random subset of the samples (n=10) were compared with urine osmolality and both the optical and digital instruments demonstrated excellent correlation. While an accurate USG reading is important, it is unlikely that the statistical significance between the two instruments is clinically significant and, therefore, unlikely to result in a change in patient evaluation or treatment plans. While both the digital and optimal refractometers are highly correlated to the urine osmolality, making both devices valid for assessment of USG in clinical practice, this digital device is easier to read and eliminates the variability of subjective interpretation.     

Clinical studies on canine dirofilarial hemoglobinuria: measured and calculated serum osmolalities and osmolar gap

Serum osmolalities and osmolar gap were determined in 43 normal healthy beagles (control group) and 40 dogs with dirofilarial hemoglobinuria (hemoglobinuria group). In the control group, the measured and calculated serum osmolality levels were in the means of 296 +/- 5 (SD) mOsm/kg and 293 +/- 6 mOsm/kg respectively, showing an osmolar gap less than 10 mOsm/kg. In the hemoglobinuria group, the measured serum osmolality ranged from 272 to 370 mOsm/kg. A considerable number of dogs had normal serum osmolalities in spite of severe intravascular hemolysis, suggesting that the changes in serum osmolality would not be the direct cause of intravascular hemolysis. The measured serum osmolality (331 +/- 28 mOsm/kg) was significantly higher in 11 dogs which died after a surgical removal heartworms than in 29 dogs which recovered after the removal (302 +/- 17 mOsm/kg). The calculated serum osmolality level was 296 +/- 16 mOsm/kg in 24 recovered cases, and 304 +/- 22 mOsm/kg in 10 fatal cases. The osmolar gap stayed in the normal ranges of 5.4 +/- 5.9 mOsm/kg in recovered cases, but it attained a higher level of 22.7 +/- 8.9 mOsm/kg in fatal cases, suggesting poor prognosis in cases with large osmolar gaps. There were significant positive correlations between the measured serum osmolality and osmolar gap, serum sodium, potassium, BUN, GOT, GPT, creatinine, bilirubin and plasma hemoglobin values, as well as between the osmolar gap and serum potassium, BUN, GOT, GPT, creatinine and bilirubin values. The plasma hemoglobin concentration fell markedly without significant change in serum osmolality 20 hr after the heartworm removal.     

Use of the advanced micro-osmometer model 3300 for determination of a normal osmolality and evaluation of different formulas for calculated osmolarity and osmole gap in adult dogs

Objective: To determine a reference interval of whole blood and plasma osmolalities for dogs using the Advanced Micro Osmometer Model 3300, to compare calculated osmolarity to measured osmolality, to determine a reference osmole gap, and to determine the best formula for calculated osmolaity. Design: Prospective, observational. Setting: Tertiary referral and teaching hospital. Animals: One hundred healthy adult dogs. Interventions: None. Measurements: Serum and whole blood biochemistry and osmolality assessments. Results: The mean and median of the measured whole blood osmolality were 323 and 320 mOsm/kg, respectively, with a standard deviation of 13.2 mOsm/kg. The mean and median of the measured plasma osmolality were 313 and 310 mOsm/kg, respectively, with a standard deviation of 13.2 mOsm/kg. The formula that was closest to predicting the measured whole blood and plasma osmolality was ((1.86(Na+K))+(BUN/2.8)+(Glucose/18))/0.93 followed closely by the traditional formula of (2(Na+K))+(BUN/2.8)+(Glucose/18). The mean calculated osmolarities using these formulas were 314.1 and 313.25 mOsm/L, respectively. The mean osmole gap using these formulas was 3.49 and 4.41 mOsm, respectively, for whole blood and ?2.01 and ?1.1 mOsm, respectively, for plasma. Conclusion: The Advanced Micro Osmometer Model 3300 was successful in measuring the osmolality in relative agreement with the current published reference intervals for osmolality. Measured osmolality correlated well with traditional calculated osmolarity.     

The Effect of Hetastarch (670/0.75) on Urine Specific Gravity and Osmolality in the Dog

Background: Urine specific gravity (USG) is used clinically to estimate urine osmolality (UOsm). Although USG has been shown to have a linear correlation with UOsm in dogs, the relationship is altered when there are significant numbers of high molecular weight (MW) molecules in the urine.
Hypothesis: USG would no longer predict UOsm in dogs given intravenous hetastarch (670/0.75)(HES).
Animals: Eight healthy employee-owned adult dogs.
Methods: Prospective, controlled experimental study. USG and UOsm were measured every 30 minutes from t=0 minutes to t=360 minutes. Dogs were administered 20mL/kg of either NaCl 0.9% (control group, n=4) or HES (treatment group, n=8) IV over 1 hour starting at t=90 minutes.
Results: There was a decrease in UOsm in both groups starting at t=120 minutes and continuing for the study duration, and there was no significant difference in UOsm between treatment and control groups across all time points. There was an appropriate decrease in USG from t=120 minutes for the control group. In the treatment group, USG increased significantly at t=120 minutes ( P = .0006), t=150 minutes ( P = .0002), and t=180 minutes ( P = .0044). The largest increase in USG occurred at t=150 minutes with a mean USG of 1.070 ± 0.021 (range 1.038-1.104).
Conclusions and clinical importance: Urine specific gravity should not be used to estimate urine solute concentration in dogs following the administration of 20mL/kg of HES. In a clinical setting, the evaluation of USG following this dose of HES may lead to an overestimation of urine concentration.     

Interobserver reliability of canine urine specific gravity assessed by analog or digital refractometers

Refractometry is utilized routinely to evaluate canine urine specific gravity (USG) in veterinary clinical settings. We aimed to determine if the magnitude of interobserver reliability when assessing canine USG via refractometry could impact clinical judgment. USG was determined in 38 dogs by 3 registered veterinary technicians (RVTs) using both an optical analog refractometer and a digital refractometer. Summary statistics were reported, interobserver reliability was assessed via intraclass correlation coefficient (ICC) analysis through a 2-way mixed-effects model, and agreement between RVT pairs was compared through Bland–Altman plots. The median analog refractometer USG measurement was 1.018 (range: 1.004–1.040) and for the digital refractometer was 1.0176 (1.0035–1.0357). The analog refractometer average measure ICC was 0.995 (95% CI: 0.992, 0.997; p < 0.001). The digital refractometer average measure ICC was 0.999 (95% CI: 0.999, 1.000; p < 0.001). Strong agreement between all pairs of RVTs was seen via Bland–Altman plots for both analog and digital refractometers, with 95% CIs spanning no more than 0.002 in either the positive or negative direction for all pairings. The interobserver variability in canine USG measurements by RVTs was trivial and did not impact clinical judgment and decision-making.     

Assessment of a pyrogallol red technique for total protein measurement in the cerebrospinal fluid of dogs

The measurement of protein concentration in the cerebrospinal fluid is a basic analytical method in neurology. In this study, a pyrogallol red technique using a human albumin calibrator previously validated in human medicine was tested for canine samples, and the results were compared with those obtained using urine test strips. Pyrogallol red significantly (P<0.05) but moderately underestimated purified dog albumin and globulins. The imprecision of the technique was low: intra- and between-series coefficients of variation were 1.6 and 4.3 per cent at protein concentrations of about 0.3 g/litre. Over 49 samples, there was good agreement between the pyrogallol red and test strip results (r=0.63), especially for low and high protein concentrations, but misclassifications were observed with '+' test strip readings.     

Vasopressin secretion in response to osmotic stimulation and effects of desmopressin on urinary concentrating capacity in dogs with pyometra

OBJECTIVE: To determine vasopressin (VP) secretory capacity during osmotic stimulation and the response to desmopressin treatment in dogs with pyometra and control dogs. ANIMALS: 6 dogs with pyometra before and after ovariohysterectomy and 6 control dogs. PROCEDURE: Urine osmolality (Uosm) was measured during 12 hours. Values measured on the first day defined the basal Uosm pattern. On the second day, dogs were given desmopressin to induce a desmopressin-stimulated Uosm pattern. On day 3, the VP response to osmotic stimulation was examined. RESULTS: Median Uosm on day 1 was 340 mOsm/kg (range, 104 to 1,273 mOsm/kg) and 807 mOsm/kg (range, 362 to 1,688 mOsm/kg) in dogs with pyometra before and after surgery, respectively, and 1,511 mOsm/kg (range, 830 to 1,674 mOsm/kg) in control dogs. Median Uosm during desmopressin treatment was 431 mOsm/kg (range, 168 to 1,491 mOsm/kg) and 1,051 mOsm/kg (range, 489 to 1,051 mOsm/kg) in dogs with pyometra before and after surgery, respectively, and 1,563 mOsm/kg (range, 1,390 to 2,351) in control dogs. In dogs with pyometra, threshold for VP secretion was lower before surgery (median, 340 mOsm/kg; range, 331 to 366 mOsm/kg) than after surgery (median, 358 mOsm/kg; range, 343 to 439 mOsm/kg) or in control dogs (median, 347 mOsm/kg; range, 334 to 360 mOsm/kg). Highest maximum plasma VP values were found in dogs with pyometra. CONCLUSIONS AND CLINICAL RELEVANCE: Dogs with pyometra had increased urine concentration in response to desmopressin but not to the degree of control dogs, whereas VP secretory ability was not reduced.     

Clinical pathologic alterations in horses during a water deprivation test

A 72-hour water deprivation test was performed in 12 horses to determine clinical pathologic changes. Reference values for electrolyte (X) clearance, expressed as a percentage of creatinine clearance (CLCR; %CLCRX), were also determined. A comparison was made between urine concentration measurement techniques. Results of %CLCRX determination in 12 horses before water deprivation were 0.034 +/- 0.095 %CLCRNa, 42.4 +/- 9.8 %CLCRK, 0.352 +/- 0.190 %CLCRCl, and 0.710 +/- 0.250 %CLCRP. During water deprivation, there was individual variation for electrolyte clearances, but Na excretion increased significantly (P less than 0.01) at 24 and 48 hours. After 48 hours' water deprivation, %CLCRNa decreased significantly, but was still greater than the initial clearance. Plasma protein was a better indicator of water deprivation (dehydration) in the horse than was PCV. Electrolyte concentrations in serum and urine were determined. Little significant (P less than 0.01) change in acid-base values was noticed after 72 hours' water deprivation. Urine osmolality (as determined by osmometry) was compared with sp gr (determined by refractometry) in determining urine concentration. Initially, sp gr correlated well with urine osmolality determinations, but this correlation decreased after 48 hours.     

Intra- and Interindividual Variation in Urine Osmolality and Urine Specific Gravity in Healthy Pet Dogs of Various Ages

Urine specific gravity (Usg) and urine osmolality (Uosm) are used routinely to assess renal concentrating ability, but limited data on these variables are available for healthy dogs. Consequently, we studied the intra- and interindividual variations in Usg and Uosm in healthy dogs as well as the influence of age and gender on these variables. Dogs were selected for health and anestrus in female dogs through the use of a detailed questionnaire. Eighty-nine owners collected morning and evening urine samples from their dogs on 2 consecutive days. In 8 dogs in which the Uosm of different samples varied more than 50%, owners collected urine for 24 hours at 2-hour intervals during the day and at 4-hour intervals at night. The possible effect of changes in adrenocortical function with age was assessed by measurements of urinary corticoid/creatinine (C/C) ratios. Among all samples, Uosm ranged from 161 to 2,830 mOsm/kg and Usg from 1.006 to > 1.050. In the morning, Uosm (1,541 ± 527 mOsm/kg, range 273–2,620 mOsm/kg) and Usg (1.035 ± 0.010, range 1.009- > 1.050) were higher than in the evening (Uosm 1,400 ± 586 mOsm/kg, range 161–2,830 mOsm/kg; Usg 1.031 ± 0.012, range 1.006- > 1.050). The interindividual coefficient of variation in Uosm was 34.2% for morning urine samples and 41.9% for evening samples. In 8 dogs with large differences in urine concentration, there were 2– to 3-fold increases or decreases in Uosm during the day, and the intraindividual coefficient of variation was 33.0%. There was no relation between gender and urine concentration. Urine concentration in both the morning and evening samples decreased with age. Urinary corticoid/creatinine ratios did not change with age. It can be concluded that Uosm and Usg vary widely among healthy dogs. Urine concentration is generally lower in the evening than in the morning and is not related to gender. Urine concentration decreases with age, and this cannot be ascribed to an associated increase in endogenous corticoids. In some dogs, Uosm varies widely during the day, with an intraindividual coefficient of variation approaching the interindividual coefficient of variation. This may be regarded as a biologic variation but also could represent an early undi-agnosed clinical abnormality.     

Serum and urine inorganic fluoride concentrations and urine oxalate concentrations following methoxyflurane anesthesia in the dog.

Plasma fluoride, urine fluoride and urine oxalate concentrations were measured before administering an anesthetic to 8 dogs, and at 0, 3, 9, 24, 48, and 72 hours following 1.5 hours of anesthesia with 1% methoxyflurane. Plasma and urine osmolalities were measured and compared with fluoride and oxalate values. Fluoride concentration increased in both plasma and urine following anesthesia when compared with the preanesthetic concentrations. Maximum mean plasma inorganic fluoride was 106.71 mumoles per liter (+/- 25.44 SE) at 9 hours after exposure to methoxyflurane was completed. By 72 hours after exposure to methoxyflurane the plasma fluoride concentration was 23.47 microM/L (+/- 5.74 SE). Mean urine inorganic fluoride concentration was highest at 9 hours after exposure to methoxyflurane and reached 6047.03 microM/L (+/- 1378.46 SE) as compared to the mean preanesthetic base-line concentration of 542.68 microM/L (+/- 132.93 SE), and the 72 hour mean urine fluoride concentration which was 1593.78 microM/L (+/- 579.46 SE). Urine oxalate concentrations, when compared with urine osmolality (mg/mOsm), increased throughout the study. The 72-hour concentration after exposure to methoxyflurane was 2.5 times the preanesthetic (mg/mOsm) oxalate concentration. Plasma osmolality did not change markedly during the study. Urine osmolalities varied between animals and collection times, but a consistent pattern did not occur. Clinical and laboratory signs of renal dysfunction were not observed in any animal during the study.     

Water deprivation test in the dog: maximal normal values.

Studies were undertaken to determine maximal urine osmolality and urine specific gravity following water deprivation for 20 dogs with normal renal function. In addition, the reliability of body weight, skin pliability, total plasma protein concentration, and packed cell volume as indices of negative water balance was assessed. Following water deprivation for periods sufficient to induce dehydration, the mean maximal urine osmolality was 2,289 mOsm/kg. The corresponding mean maximal urine specific gravity was 1.062 and ranged from 1.050 to 1.076. The ratio of mean maximal urine osmolality to mean serum osmolality at the time of peak urine concentration was 7.3. There was no detectable difference in urine concentration indices between males and females. Changes in skin pliability and packed cell volume proved unreliable as estimates of dehydration. Weight loss and increases in total plasma protein concentration proved to be more consistent indicators of hydration status. Abnormal increases in serum urea nitrogen and serum creatinine concentrations occurred rarely, even though some dogs had water withheld for periods of up to 96 hours.     

Use of protein-to-creatinine ratio in a single urine specimen for quantitative estimation of canine proteinuria

The daily excretion of urinary protein was evaluated in 8 conditioned research dogs and in 10 hospitalized, proteinuric dogs, using 24-hour urine collections. Concurrent with each 24-hour urine collection, a 5- to 10-ml urine specimen was obtained during midday. The ratio of urine protein to urine creatinine concentration was determined from the single urine specimen for each dog. Linear regression analysis was used to calculate the correlation between that ratio and the 24-hour urinary protein loss (mg/kg of body weight). The coefficient of determination was significant (r2 = 0.95, P less than 0.0001). Determination of the protein-to-creatinine ratio in a single urine specimen was found to be a sensitive, rapid, and dependable diagnostic technique for detection and quantitative estimation of proteinuria.     

Estimation of quantitative proteinuria in the dog, using the urine protein-to-creatinine ratio from a random, voided sample

The correlation between 24-hour urine protein excretion and the protein-to-creatinine ratio (U-P/C) from random, voided urine specimens was assessed in 16 healthy Beagles (9 to 11 months old) and in 14 dogs with suspected renal proteinuria. Initially, a voided urine specimen was obtained from each dog, and the U-P/C was determined. An attempt was not made to standardize the time of collection of the voided urine. Subsequently, each dog was placed in a metabolism cage, and 24-hour urine specimens were collected for quantitative protein analysis. The Coomassie blue technique was used to measure urine protein. The correlation between the U-P/C and the 24-hour urine protein excretion (mg/kg/24 hr), evaluated by linear-regression analysis, was found to be significant (r = 0.975, P less than 0.01). These results substantiate previous findings and indicate that random, voided urine specimens may be used to compute the ratio and to accurately reflect 24-hour urinary protein loss in the dog.     

Water absorption in relation to fermentation in the colon of the ostrich (Struthio camelus)

The colon is a major site for fermentation and water absorption in the ostrich. Water absorption along the colon was evaluated and its relationship to osmolality, Na+ concentration, short chain fatty acid (SCFA) concentration and carbohydrate content of digesta analysed. Mean water content decreased from 5.30 +/- 0.99 to 2.51 +/- 0.13 mf/g dry mass in the first 5 m of the colon. Correspondingly, mean carbohydrate content fell from 529.85 +/- 46.61 to 434.99 +/- 29.89 mg/g dry mass. A significant correlation was shown between the decreases in mean carbohydrate and water content along the colon (r2 = 0.997, P < 0.05). Changes in mean osmolality (+/- 10 mOsm/kg) and SCFA concentration (+/- 7 mmol/l) were minimal in comparison to the change in Na+ concentration (-54 mmol/l). These findings reflect a close coupling between SCFA production and absorption on the one hand and water absorption on the other.     

Effects of oral administration of a commercial activated charcoal suspension on serum osmolality and lactate concentration in the dog

The purpose of this investigation was to determine the effects of an activated charcoal (AC) suspension containing propylene glycol and glycerol on serum osmolality, osmolal gap, and lactate concentration in dogs. Six healthy adult dogs were administered 4 g/kg AC in a commercially available suspension that contained propylene glycol and glycerol as vehicles. Blood samples were taken before and 1, 4, 6, 8, 12, and 24 hours after the administration of the test suspension. Samples were analyzed for osmolality, blood gases, and concentrations of lactate, sodium, potassium, serum urea nitrogen, and glucose. Osmolal gaps were calculated for each time point. Mean serum osmolality, osmolal gap, and lactate concentration were significantly increased after suspension administration compared to baseline. Serum osmolality increased from 311 mOsm/kg at baseline to 353 mOsm/kg, osmolal gap increased from 5 to 52 mOsm/kg, and lactate concentration increased from 1.9 to 4.5 mmol/L after suspension administration (all P < .01). Three of the 6 dogs vomited between 1 and 3 hours after the administration of the test suspension, and 4 of 6 dogs were lethargic. All dogs drank frequently after AC administration. Commercial AC suspension administered at a clinically relevant dose increases serum osmolality, osmolal gap, and lactate concentration in dogs. These laboratory measures and the clinical signs of vomiting, lethargy, and increased frequency of drinking might complicate the diagnosis or monitoring of some intoxications (such as ethylene glycol) in dogs that have previously received AC suspension containing propylene glycol, glycerol, or both as vehicles.     

Influence of the feed base excess on urine parameters in cats

In this study the base excess (BE) was used as a method to predict the influence of the food on the urinary pH on cats. Nine cat foods (six dry and three canned) were consecutively fed to eight cats. The urine pH, volume, specific gravity and water and food intake were determined daily. The base excess [BE; mmol/kg dry matter (DM)] was calculated from the compounds in the food (BE = 49.9*Ca+82.3*Mg*+43.5*Na+25.6*K-64.6*P-13.4*Met-16.6*Cys-28.2*Cl). The BE of the tested foods was between -287.35 and 133.38 mmol/kg DM. The mean urine pH varied between 5.76 (SD = 0.13) and 7.16 (SD = 0.22). The BE correlated with the mean urine pH (pH = 6.25+0.0023*BE; r = 0.74**). The urine volume (ml/kg BW/day) correlated significantly positive with the K- (r = 0.71**) and significantly negative with the P-content (r = -0.67**), the Ca-content (r = -0.50**) followed by the Mg-content (r = -0.36**) of the food. The correlation coefficients between the anions/cations in the food and the urine pH was for K 0.36**, for P -0.61**, the Met+Cys -0.60** and Cl -0.27**. In practice the correlation between urine pH and BE would help to pre-estimate the effect of food on the urine pH and to prevent urolith formation.     

Evaluation of a colonic lavage solution to prepare the colon of the dog for colonoscopy

Ten dogs were given 3 different doses (60, 80, and 100 ml/kg of body weight) of a commercial colonic lavage solution in 2 equally divided doses, 2 to 4 hours apart. The degree of colonic cleanliness was evaluated subjectively, using a numbered scoring system (1 = clean to 4 = excessive fecal material). Systemic and metabolic effects were assessed by comparing body weight, PCV, total protein concentration, serum osmolality, and urine specific gravity before and 24 hours after administration of the high-dose solution. The lowest mean cleanliness score (1.6) was obtained with the 80 ml/kg dose (P less than 0.001). The solution had no effect on any measurement except urine specific gravity (P = 0.036). Oral administration of a colonic lavage solution in a divided total dose of 80 ml/kg is a safe and effective method of large-intestine preparation in the dog.     

Evaluation of a lithium dilution cardiac output technique as a method for measurement of cardiac output in anesthetized cats

OBJECTIVE: To evaluate the use of a lithium dilution cardiac output (LiDCO) technique for measurement of CO and determine the agreement between LiDCO and thermodilution CO (TDCO) values in anesthetized cats. ANIMALS: 6 mature cats. PROCEDURE: Cardiac output in isoflurane-anesthetized cats was measured via each technique. To induce different rates of CO in each cat, anesthesia was maintained at > 1.5X end-tidal minimum alveolar concentration (MAC) of isoflurane and at 1.3X end-tidal isoflurane MAC with or without administration of dobutamine (1 to 3 microg/kg/min, i.v.). At least 2 comparisons between LiDCO and TDCO values were made at each CO rate. The TDCO indicator was 1.5 mL of 5% dextrose at room temperature; with the LiDCO technique, each cat received 0.005 mmol of lithium/kg (concentration, 0.015 mmol/mL). Serum lithium concentrations were measured prior to the first and following the last CO determination. RESULTS: 35 of 47 recorded comparisons were analyzed; via linear regression analysis (LiDCO vs TDCO values), the coefficient of determination was 0.91. The mean bias (TDCO-LiDCO) was -4 mL/kg/min (limits of agreement, -35.8 to + 27.2 mL/kg/min). The concordance coefficient was 0.94. After the last CO determination, serum lithium concentration was < 0.1 mmol/L in each cat. CONCLUSIONS AND CLINICAL RELEVANCE: Results indicated a strong relationship and good agreement between LiDCO and TDCO values; the LiDCO method appears to be a practical, relatively noninvasive method for measurement of CO in anesthetized cats.     

Dehydration in stressed ruminants may be the result of a cortisol-induced diuresis

The effect on water and electrolyte balance of stress, simulated by intravenous infusion of cortisol, was studied using 24 18-mo-old Merino wethers (37.0 +/- 0.94 kg mean body weight [BW]) over 72 h. The sheep were allocated to one of four groups: 1) no water/no cortisol (n = 6); 2) water/no cortisol (n = 4); 3) no water/cortisol (n = 6); and 4) water/cortisol (n = 4). Animals allocated to the two cortisol groups were given 0.1 mg x kg BW(-1) x h(-1) of hydrocortisone suspended in isotonic saline to simulate stress for the duration of the experiment. Total body water, plasma cortisol, osmolality and electrolytes, and urine electrolytes were determined at 24-h intervals for 72 h. In the presence of cortisol, total body water was maintained in the face of a water deprivation insult for 72 h. Water deprivation alone did not induce elevated plasma concentrations of cortisol, in spite of a 13% loss of total body water between 48 and 72 h. Infusion of cortisol was found to increase urine output (P = 0.003) and decrease total urinary sodium output (P = 0.032), but had no effect on plasma electrolyte levels or water intake. Water deprivation was found to increase plasma sodium concentrations (P = 0.037). These results indicate that sheep given cortisol to simulate stress suffer from a loss of body water in excess of that associated with a loss of electrolytes, and support the hypothesis that elevated physiological concentrations of cortisol induce a diuresis in ruminants that contributes to dehydration.