### Tvedten, Harold

- Institutionen för kliniska vetenskaper, Sveriges lantbruksuniversitet

2014

Tvedten, Harold

Update on Specific Gravity Measurement,

Harold Tvedten DVM PhD dACVP dECVCP

It is useful to reconsider long held beliefs to see if they are as true with today's knowledge as they were when they were established. Veterinarians today often purchase "veterinary specific" refractometers with different scales for feline urine than urine of dogs and other animals. Other veterinarians use conversion formulas to calculate the "true" specific gravity of feline urine analyzed with standard medical refractometers.1 This belief comes from a 1956 article in which urine specific gravity was determined from190 human, 22 feline and 21 canine urine samples.2-3 In that study the increase on the refractive scale of their refractometer compared to increase in total solids in the urine was linear for all three species, but the calculated specific gravity for the feline samples was relatively lower compared to total solids than for canine and human samples. This has led to the belief that feline urine has different refractive characteristics and feline refractometers are calibrated to report lower specific gravity than routine refractometers. We reported on how often use of specific gravity results from a feline refractometer would give a different clinical conclusion about renal function than results from a routine medical refractometer.4 We continued to work on the question "Is feline urine really different from canine or other species' urine?" A gold standard was needed and initially we used drying of urine to determine total solids as Rubini and others have used. This method seems simple but there were problems in adequately drying the urine and then using total solids results to indicate which of currently available refractometers gave the most accurate specific gravities. The urine specific gravity of a solution with 50 g/L of total solids is not 1.050. The specific gravity depends on the density of the solute or solutes in solution. Refractometers also vary in results despite all being calibrated (zeroed) to 1.000 with water. Five refractometers reported specific gravity of 1.014 to 1.025 with a solution of 50 g urea/l which is much less than 1.050. A conversion factor was used to adjust total solids in urine samples to a reasonable specific gravity result. Because urine is a complex and variable solution one conversion factor is not exact but provided comparison values of similar magnitude for evaluation. A commonly used conversion is called Häser formula or Trapp-Häser formula. Multiplying the third and fourth decimal figures of the specific gravity x 2.33 is used to estimate total solids of urine. In reverse, the total solids from drying the feline and canine urine samples were converted to estimated specific gravities by dividing the weight of total solids by 2.33. For example 0.030 grams total solids per liter urine divided by 2.33 suggested a urine specific gravity of about 1.013. Based on total solids/2.33 in feline and canine urines it was shown that the refractometers reporting the highest specific gravities appeared more accurate. Two feline refractometers always reported the lowest specific gravity results and lower results that total solids/2.33. Therefore feline refractometer results appeared erroneous in both cats and dogs. Feline results mirrored canine results and no difference was seen. Refractometers reported statistically and diagnostically different results from each other. Dipstick specific gravity results were rarely > 1.020 in cats and did not accurately reflect true specific gravity. Use of a pycnometer to determine a true specific gravity was used in a limited number of samples with the same pattern of results. The routine medical refractometers always reported higher results and were more similar to pycnometer results. Pycnometer results were always greater than the 2 feline refractometers which reported the lowest and very similar specific gravities. Osmolality was also determined on the samples but osmolality measures the number of particles in solution and not weight of solutes and reports concentration in different units. Osmolality was highly correlated (r = 0.99) to results from all refractometers, despite statistical difference among refractometers, and even to total solids which were about 2.33 times greater than respective specific gravity results. Correlation did not indicate which method was more accurate. This demonstrates why correlation is not a useful statistical method for method validation. Correlation shows an association between tests but not what test is most accurate if there is a consistent error (bias).

urine specific gravity

Utgivare: European Society of Veterinary Clinical Pathology

### Tvedten, Harold

- Institutionen för kliniska vetenskaper, Sveriges lantbruksuniversitet

- Institutionen för kliniska vetenskaper, Sveriges lantbruksuniversitet

Annan medicinsk bioteknologi

https://res.slu.se/id/publ/61711