This article was downloaded by: [University of Bristol] On: 28 February 2015, At: 12:01 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
New Zealand Veterinary Journal Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/tnzv20
Comparison of specific gravity analysis of feline and canine urine, using five refractometers, to pycnometric analysis and total solids by drying a
b
c
HW Tvedten , H Ouchterlony & IE Lilliehöök a
Clinical Sciences, Faculty of Veterinary Medicine and Animal Sciences, Swedish University of Agricultural Sciences b
Institution for Women and Children's Health, Uppsala University
c
University Animal Hospital, Swedish University of Agricultural Sciences, Uppsala, Sweden Accepted author version posted online: 27 Jan 2015.
Click for updates To cite this article: HW Tvedten, H Ouchterlony & IE Lilliehöök (2015): Comparison of specific gravity analysis of feline and canine urine, using five refractometers, to pycnometric analysis and total solids by drying, New Zealand Veterinary Journal, DOI: 10.1080/00480169.2014.1002553 To link to this article: http://dx.doi.org/10.1080/00480169.2014.1002553
Disclaimer: This is a version of an unedited manuscript that has been accepted for publication. As a service to authors and researchers we are providing this version of the accepted manuscript (AM). Copyediting, typesetting, and review of the resulting proof will be undertaken on this manuscript before final publication of the Version of Record (VoR). During production and pre-press, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal relate to this version also.
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
Publisher: Taylor & Francis & New Zealand Veterinary Association Journal: New Zealand Veterinary Journal DOI: 10.1080/00480169.2014.1002553
Scientific Article
rip
t
Comparison of specific gravity analysis of feline and canine urine, using five refractometers, to pycnometric analysis and total solids by drying
HW Tvedten*§, H Ouchterlony† and IE Lilliehöök‡
M an
us c
Clinical Sciences, Faculty of Veterinary Medicine and Animal Sciences, Swedish University of Agricultural Sciences. † Institution for Women and Children's Health, Uppsala University. ‡ University Animal Hospital, Swedish University of Agricultural Sciences, Uppsala, Sweden. § Author for correspondence: Email: [email protected] Abstract
AIMS: To compare the performance of five refractometers for determination of urine specific gravity in cats and dogs, with reference to weight of total solids and pycnometer analysis.
d
METHODS: Urine samples from 27 cats and 31 dogs submitted for routine urinalysis were
pt e
included. Urine specific gravity was determined with five refractometers. Four were optical, hand held refractometers with a temperature compensation method and one was a digital model. Urine was dried to determine the precise weight of total solids. The total solids (g/L) were converted to an estimated specific gravity by division with 2.33. Urine specific gravity of four feline and seven
ce
canine samples were analysed with a pycnometer. Limits of agreement analysis was used to evaluate the agreement between specific gravity (analysed as specific gravity minus 1) measured by the refractometers and estimated from dried total solids, or pycnometer results,
Ac
Downloaded by [University of Bristol] at 12:01 28 February 2015
*
RESULTS: The five refractometers reported clearly different results from each other. Proportional negative bias was noted for refractometer results compared to estimated specific gravity from total solids and a constant negative bias compared to pycnometer results. The two refractometers designed for cat urine reported similar and lowest specific gravity results with a mean negative bias of 0.007 and 0.008 units compared to estimated specific gravity from total solids and a mean negative bias of 0.006 units compared pycnometer results.
1
CONCLUSIONS: Refractometer results did not increase consistently with increasing urine specific gravity compared to reference methods or to other refractometers. Two feline refractometers reported consistently lower specific gravity results than reference methods and other refractometers. CLINICAL RELEVANCE: Because of this imprecision, veterinarians should not use precise cut off values such as 1.030 or 1.035 for evaluation of renal concentrating ability in dogs and cats. Veterinarians should consider the variability of refractometric specific gravity results in their clinical assessment. Two feline refractometers appeared to report falsely low specific gravity
rip
us c
KEY WORDS: Urine specific gravity, refractometer, total solids, pycnometer, cats, dogs
M an
Introduction
Determination of urine specific gravity is commonly used as a sensitive test of renal concentrating ability for canine and feline patients (Watson 1998). Previous comparison of a cat refractometer and a standard medical refractometer suggested the cat refractometer reported falsely low results, that may interfere with clinical diagnosis (Tvedten and Norén 2014). This observation needed to be
d
better validated.
pt e
The objective of the current study was to compare the performance of several currently available refractometers, including two cat refractometers, to each other and to two reference methods over a wide range of specific gravities. The two reference methods used were determining the weight of
ce
total solids from feline and canine urine samples after drying, and pycnometer analysis of specific gravity (George 2001). A pycnometer is a special glass flask which can be filled with a very exact volume of fluid. The weight of an identical volume of urine divided by the weight of the same
Ac
Downloaded by [University of Bristol] at 12:01 28 February 2015
t
results.
volume of water equals the specific gravity. The reference methods should indicate which refractometers best reflected these reference values at various urine specific gravities.
Materials and methods Sample and patient selection
Urine samples for the study included 27 feline and 31 canine samples which were submitted for patient diagnosis to the University Animal Hospital, Swedish University of Agricultural Sciences between January and April 2013. All feline urine samples that did not appear red (bloody) and had a supernatant which was not turbid and had a volume of >2 mL were included. A similar number of 2
canine samples were selected with the same criteria and with a preference for more concentrated urines. More concentrated urines should have a greater weight of dry matter after drying and therefore measurements of the weight of empty glass tubes compared to those with dried urine material may be more precise. Urine samples were centrifuged at 2000g for 5 minutes. Supernatants from urine samples were analysed with refractometers that day. The samples were stored in a refrigerator (about 4 °C) if the rest of the analysis could be completed within two days. Otherwise samples were frozen (about −20 °C) until analysis with the same methods as fresh urine samples.
t
Instruments and methods
rip
Five refractometers were used to analyse specific gravity. Selection was only based on easy
us c
Goldberg type refers to an optical, hand held refractometer with a temperature compensation
method developed by Herbert Goldberg. One was the Uricon refractometer (Atago Corp, Tokyo Japan); termed Atago 1. Two refractometers appeared to be the same but one was labelled as a Reichert TS meter, termed Rei 2, and one was labelled Leica TS meter, termed Rei 3; both were
M an
Model 10400A Reichert TS meters (Cambridge Instruments Inc, Buffalo NY, USA). Because Rei 2 and Rei 3 appeared identical and gave the same results only results from Rei 2 are presented. One refractometer had two scales, for cats and dogs; the Model 10436 AO Veterinary refractometer (AO Scientific Instruments, Buffalo NY, USA), referred to as Vet Cat and Vet Dog, respectively. One
pt e
termed Atago Cat.
d
refractometer was a digital model; the Atago ‘Pocket’ PAL-USG Cat refractometer (Atago Corp)
Urine samples that had a specific gravity above the range of analysis of a refractometer were diluted with an equal volume of distilled water to obtain a result greater than the refractometer reported.
ce
The upper values for specific gravity reported by various refractometers without requiring dilution were 1.035 for Rei 2 and Rei 3, 1.050 for Atago 1, 1.060 for Vet Cat and Vet Dog, and 1.080 for Atago Cat.
Ac
Downloaded by [University of Bristol] at 12:01 28 February 2015
availability from two local laboratories. Four optical refractometers were of the "Goldberg type".
Refractometers were calibrated with distilled water at room temperature to give a zero result of 1.000. Analysis of data was based on measured specific gravity minus 1.000. The effect of substances dissolved in urine on specific gravity is called the specific gravity increment and is the measured specific gravity minus the specific gravity of water (1.000) (Price et al. 1940). For example, a solution with a specific gravity of 1.040 is twice as concentrated as 1.020. Drying of the urine samples was used to determine total solids of the urine supernatants. Weight of the dry material after drying of 2 mL of urine was used to calculate total solids. The tubes were weighted empty, when they contained urine and after drying. Freeze drying of urine with a freeze 3
drying instrument (Scanvac Coolsafe 55-4, LaboGene ApS, Lynge Denmark) was attempted initially, but urine did not dry with the usual freeze drying procedure even overnight. Subsequently a special evaporator centrifuge (SpeedVac Plus SC250DDA; Savant, Thermo Fisher Scientific Stockholm Sweden) was attached to the vacuum system of the freeze drying instrument and urine was centrifuged in 50 mL plastic tubes to evaporate all fluid at room temperature. Drying in the SpeedVac was usually performed for 16 hours. Samples which had a positive glucose result (1+ to 4+) on dipstick analysis were excluded from evaluation of total solids data because three urine samples with 3+ to 4+ glucose never dried completely. Some matter remained as a gel or syrup in
rip
t
those three feline samples after drying 16–32 hours. The total solids from drying the feline and
canine urine samples were converted to estimated specific gravities by dividing the weight of total
us c
A pycnometer (Pycnometer 2 mL; Joan Lab Equipment, Zhejiang, China) was used to determine specific gravity of eleven of the previously analysed urine samples (four feline and seven canine) stored in a −20 ºC freezer. Samples with higher specific gravity results and >2 mL volume were
M an
selected and thawed to room temperature prior to analysis. The weights of the dry pycnometer, the pycnometer filled with distilled water and the pycnometer filled with the urine sample were determined with an analytic grade balance (Mettler Toledo PB 303, Stockholm Sweden). Specific
Statistical analysis
d
gravity was the weight of an equal volume of urine divided by that of water.
pt e
Limits of agreement analysis (Bland and Altman 1999) was used to evaluate the agreement between specific gravity measured by the refractometers and estimated from dried total solids, or
ce
pycnometer results, using Analyze-it v. 3.80 (Analyze-it Software, Ltd., Leeds United Kingdom).
Results
The specific gravity results for the refractometers compared to the estimated specific gravity based on total solids from drying of feline and canine urine samples are shown in Figure 1. The Rei 2
Ac
Downloaded by [University of Bristol] at 12:01 28 February 2015
solids by 2.33 (Price et al. 1940).
refractometer reported results only up to 1.035 and required dilutions to obtain greater values for comparison in this study. Comparison of the results for the Vet Dog and Rei 2 refractometer showed that values for the Rei 2 had a shift to falsely high results above 1.035 after dilution. Therefore results for Rei 2 are only reported up to 1.035. The estimated specific gravity from dried total solids was similar to most refractometer results at low values but greater than all results at higher specific gravities. Atago 1 results were generally greater than other refractometer results at lower urine specific gravities up to about 1.040. The two 4
feline refractometers (Vet Cat and the Atago Cat) consistently gave the lowest and almost identical results. Pycnometer results from four feline and seven canine urine samples are shown with results from four refractometers in Figure 2. Results from the Rei 2 refractometer are not shown because only two results were ≤1.035 and reported by the instrument. The results from the two feline refractometers were consistently lower than pycnometer results, while the other refractometer results were more similar to pycnometer results.
rip
t
Results for the limits of agreement analysis comparing the refractometers and estimated specific
gravity from dried total solids are given in Table 1. The results from two feline refractometers (Vet
us c
from the Atago 1 and Atago Cat refractometers with specific gravity estimated from dried total solids. An increasing negative bias with increasing mean specific gravity is shown for both refractometers.
M an
Results for the limits of agreement analysis comparing four refractometers and pycnometer specific gravity are given in Table 2. Again the two feline refractometers (Vet Cat and Atago Cat) had the greatest mean negative bias. Figure 4 shows the comparison for the Atago 1 and Atago Cat
pt e
d
refractometers; a constant error is noted especially with the Atago Cat refractometer.
Discussion
Determination of urine specific gravity with a refractometer is simple, quick and inexpensive.
ce
Therefore refractometers are routinely used in most veterinary clinics and laboratories and will continue to be used in the future. However, there were prominent and inconsistent differences in results reported by five currently available refractometers over the range of results seen with feline and canine patient urine samples. Because of this imprecision, veterinarians should not use precise
Ac
Downloaded by [University of Bristol] at 12:01 28 February 2015
Cat and Atago Cat) had the greatest mean negative bias. Figure 3 shows the comparison of results
cut off values such as 1.030 or 1.035 for evaluation of renal concentrating ability in dogs and cats. Veterinarians should consider this variability of refractometric specific gravity results in diagnosis. In order to evaluate the results from the different refractometers, the definition of specific gravity should be considered. Urine specific gravity is a ratio, without units, of the weight of urine divided by the weight of an equal volume of water at a certain temperature. Density is similar but is weight/volume (kg/m3 or g/L). Pycnometer analysis most directly determines urine specific gravity
by precise measurement of the weight of exactly the same volume of urine and water and determining the ratio of the weight of urine to water. The other reference method used in this study 5
determined the total solids (g/L) which indicates the density of urine, but which is not the same as specific gravity results. Drying of known volumes of urine and precise measurement of dry material remaining objectively measured the amount of total solids (g/L). However, adding the same weight of different substances to 1 L of water causes different increments of change in specific gravity due to different densities of the substances, as shown in Supplementary Table 11.. Dissolving 50 g of NaCl in 1 L of water did not give a specific gravity of 1.050. The five refractometers reported specific gravities between
t
1.018 and 1.029 for that solution A solution with an equal weight of 50 g of urea in 1 L of water
rip
resulted in specific gravities of 1.014–1.025 with the five refractometers. These are very different
us c
A conversion factor was needed to adjust results of total solids (g/L) to results which could be compared to refractometer specific gravity results. One conversion factor is called the Häser formula or Trapp-Häser formula (Price et al. 1940). Multiplying the third and fourth decimal figures
M an
of the specific gravity by 2.33 is used to estimate total solids of human 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. Urine is a complex mixture of substances so this conversion does not correct the total solids of each individual urine to a correct specific gravity. However use of the 2.33 factor provided consistent comparison values of similar magnitude for
d
evaluation in this study. Other conversion factors of 2.18–2.60 have been determined from various
pt e
studies (Price et al. 1940). Use of a different factor could affect comparisons and conclusions of the current study. Other potential errors include that only the urine supernatant was used for the drying method and thus particulate material in the sediment such as crystals was not included. Not
ce
including crystals in the sediment, such as stuvite, would affect phosphate, calcium, magnesium etc. There are also volatile substances in urine that would be lost during evaporation in the SpedVac system. Other substances in urine besides glucose (such as protein) could affect completeness of drying in the SpedVac system.
Ac
Downloaded by [University of Bristol] at 12:01 28 February 2015
results, for 1.025 is 79% more concentrated than 1.014.
Based on results from the current study, none of the five refractometers matched closely or consistently results from the two reference methods in terms of magnitude over the entire range of increasing urine specific gravities. Different types of errors were noted for the two reference methods. For example, the feline refractometer (Atago Cat) had a greater negative bias with increasing specific gravity (proportional error) compared to estimated specific gravity from total solids . However, this refractometer had a constant negative error compared to pycnometer results.
1
DOI: [production editor to insert]
6
Thus the relationship and type of error (proportional or constant) of refractometric specific gravity compared to the two reference methods was different. The two feline refractometers (Vet Cat and Atago Cat) were consistently lower than reference methods and consistently gave the lowest specific gravity results of all refractometers. There is no apparent reason to use a feline refractometer or to use a conversion formula to convert the urine specific gravity of a feline urine specimen analysed with a routine medical refractometer to a “corrected” urine specific gravity. The other refractometers and Vet Dog refractometer appeared
rip
t
most similar to pycnometer results and estimated specific gravity from total solids. Special refractometers designed for feline urine have been recommended and sold for many years
us c
refractive) than dog and human urine and a conversion formula has been suggested to correct feline urine specific gravity results (Wolf et al. 1990). The AO Vet Cat and Vet Dog refractometer apparently measures refractive index and then uses two different conversion factors for reporting a
M an
canine or feline specific gravity for its two scales. The Atago Cat refractometer and the Vet Cat refractometer appeared to use the same conversion factor, because they reported similar to identical results, despite one being digital (Atago Cat) and the other a Goldberg type (Vet Cat). This indicates that differences among refractometers were not because they are digital or traditional Goldberg type. The formulas for predicting feline urine specific gravity from refractive index have
d
not been revalidated since 1956 but are currently being used by refractometer manufacturers or as a
pt e
correction formula by veterinarians (George 2001). This belief about unique aspects of feline specific gravity determination by refractometric analysis has been based on one article (Rubini and Wolf 1956). That article evaluated refractive index,
ce
specific gravity and weight of dried total solids from190 human urine samples, 22 feline and 21 canine samples. The change in refractive increment determined along the scale of their refractometer for feline urine samples was actually linear with increasing total solids of human and
Ac
Downloaded by [University of Bristol] at 12:01 28 February 2015
(Bennett et al. 2011). It has been stated that feline urine refracts light differently (greatly more
canine urine samples. This should suggest that feline urine was similar to human and canine urine in regards to refractometric analysis. However, specific gravity (by determining the ratio of the weight of urine samples divided by an equal volume of water) of feline samples appeared lower than expected compared to human and canine urine. An aspect in that study that may be important to consider was that all of the 22 feline urine samples were more concentrated than the human or
canine urines. Feline samples had specific gravities >1.030 while canine and human urine had more dilute urines. Human urine is considered concentrated if >1.020 (Price et al. 1940) while feline urine is considered concentrated if >1.035 or even >1.040 (Watson 1998). 7
Our study was similar to the study of Rubini and Wolf (1956) in comparing refractometric specific gravity to total solids and pycnometer type analysis. Inconsistencies were demonstrated among refractometers in our study, especially at higher specific gravities and it may be the variation reported by Rubini and Wolf relate to the higher urine concentrations of their 22 cat urine samples compared to the human or canine samples. It may not be that feline urine causes a different refractive index reading but only that they are often very concentrated. Estimated specific gravity from total solids appeared greater than all refractometric results at the highest urine specific gravities of feline and canine urine samples in our study. Therefore all refractometers underestimate
rip
t
specific gravity at highest concentrations if total solids are used as the reference method.
us c
specific gravity. Most veterinarians expect that all refractometers give equivalent results, but this was not observed. Atago 1 reported the highest specific gravity results of all refractometers and had least mean bias compared with estimated specific gravity from total solids. It reported higher specific gravity results compared with total solids at lower specific gravities of about 1.030 or less,
M an
suggesting a false positive bias in more dilute urines. Above a urine specific gravity of about 1.040, the Atago 1 results were less than total solids suggesting a false negative bias in more concentrated urines. A similar pattern of changing biases over increasing concentrations was noted for both feline and canine urine specimens. The Atago 1 results were most similar to pycnometer results and
d
increased in parallel with pycnometer results.
pt e
The Rei 2 refractometer had a distinct upward shift in results (false positive bias) compared to Vet Dog at specific gravity >1.035. For this refractometer dilution of urine was performed for urines with specific gravity >1.035, and these results indicate that refractive specific gravity results are
ce
neither linear nor accurate after dilution. It has been previously reported that if urine is diluted with water the change in specific gravity (specific gravity increment) is an inverse linear function of the degree of dilution (Price et al. 1940). If veterinarians want results >1.035 they should check that their refractometer accurately reports results above this value.
Ac
Downloaded by [University of Bristol] at 12:01 28 February 2015
Results from different refractometers in our study did not show consistent bias with increasing urine
Other inconsistencies were noted with various refractometer results with increasing concentrations of NaCl or urea (Supplementary Table 12). This imprecision among refractometers should be
recognised and indicates veterinarians should not interpret specific gravity results exactly according to recommended cut off values such as urine specific gravity of >1.030 or >1.035. The diagnostic principles are correct, but refractometric specific gravity results are imprecise and inconsistent.
2
DOI: [production editor to insert]
8
Recommended cut off values such as >1.030 should be more approximate guidelines than precise numerical limits. All refractometers were calibrated to 1.000 with distilled water but reported very different and variable results. Calibrating all refractometers with water to give a result of 1.000 is the standard procedure. However, this did not assure that results in the clinically diagnostic range of 1.030–1.035 were accurate or consistent among instruments as documented in this study. The method of calibration of refractometers should be re-evaluated.
rip
t
Variation among refractometer results have been reported earlier, but Bennett et al. (2011)
concluded that clinical interpretation was not affected by a negative bias of 0.003 with a cat
us c
specific gravity and refractive index". The mean negative bias with feline refractometers in our study was about 0.004–0.007 with even greater negative bias noted in individual samples and with different refractometers. If a diagnostic cut off value of >1.030 is used to indicate adequate renal
M an
concentrating ability for dogs, then seven of our 31 dogs (22%) would have had appeared adequate by the Atago 1 refractometer but not with the Atago cat refractometer. Similarly if a diagnostic cut off value of >1.035 is used to indicate adequate concentrating ability for cats, then three of 27 cats (11%) would have had appeared adequate by the Atago 1 refractometer but not the Atago Cat refractometer.
d
In conclusion, different refractometers reported inconsistent and clinically different results.
pt e
Refractometer results did not increase consistently with increasing urine specific gravity compared to reference methods or to other refractometers. Two feline refractometers reported consistently lower specific gravity results than reference methods and other refractometers for both feline and
ce
canine urine samples and pure chemical solutions. The feline refractometer results appeared falsely low. Routine medical refractometers and a canine refractometer reported results more similar to specific gravity results from reference methods based on total solids and pycnometer analysis for
Ac
Downloaded by [University of Bristol] at 12:01 28 February 2015
refractometer "programmed with a mathematical representation of the relationship of feline urine
both feline and canine urine. Refractometric specific gravity analysis is still recommended for urine concentration analysis, but consideration of imprecision and negative biases must be applied.
References Bennett AD, McKnight GE, Dokin SJ, Simpson KE, Schwartz AM, Gunn-Moore DA. Comparison of a digital and optical analogue hand-held refractometer for the measurement of feline urine specific gravity. Journal of Feline Medicine and Surgery 13, 152–4, 2011 9
Bland JM, Altman DG. Measuring agreement in method comparison studies. Statistical Methods in Medical Research 8, 135–60, 1999 George J. The usefulness and limitations of hand-held refractometers in veterinary laboratory medicine: an historical and technical review. Veterinary Clinical Pathology 30, 201–10, 2001 Price JW, Miller M, Haymen JM. The relation of specific gravity to composition and total solids in normal human urine. Journal of Clinical Investigations 19, 537–53, 1940
t
Rubini ME, Wolf AV. Refractometric determination of total solids and water of serum and urine.
rip
Journal of Biological Chemistry 225, 869–76, 1956
us c
USG Cat refractometer for determination of urine specific gravity in dogs and cats. Veterinary Clinical Pathology 43, 63–6, 2014
M an
Watson AD. Urine specific gravity in practice. Australian Veterinary Journal 76, 392–8, 1998 *Wolf AV, Brown MG, Prentis PG. Concentrative properties of aqueous solutions: conversion tables. In: Weast RC (ed). The CRC Handbook of Chemistry and Physics. 70th Edtn. Pp D252–
pt e
* Non-peer-reviewed
d
D253. CRC Press, Boca Raton, Florida, USA, 1990
ce
Submitted 10 February 2014
Accepted for publication 08 December 2014 First published online [insert date]
Ac
Downloaded by [University of Bristol] at 12:01 28 February 2015
Tvedten HW, Norén Å. Comparison of a Schmidt and Haensch refractometer and Atago PAL-
10
Table 1. Results of limits of agreement analyses comparing results for specific gravity (analysed as specific gravity minus 1) of feline (n=23) and canine (n=31) urine samples, measured using refractometers, with specific gravity estimated from dried total solids in the urine samples divided by 2.33. See Materials and methods for details of the refractometers. Mean Difference
95% limits of agreement
Atago 1
0.001
−0.010 to 0.012
Rei 2
−0.002
−0.004 to 0.001
Vet Dog
−0.004
−0.012 to 0.003
Vet Cat
−0.008
−0.016 to 0.001
Atago Cat
−0.007
−0.016 to 0.002
Atago 1
0.002
−0.008 to 0.013
Rei 2
−0.001
−0.003 to 0.004
Vet Dog
−0.004
−0.012 to 0.005
Vet Cat
−0.006
Atago Cat
−0.006
Refractometer
t rip us c
M an
Dog urine
−0.017 to 0.004
−0.016 to 0.004
pt e
d
Table 2. Results of limits of agreement analyses comparing results for specific gravity (analysed as specific gravity minus 1) of four feline and seven canine urine samples, measured using refractometers, with pycnometer results. See Materials and methods for details of the refractometers. Refractometer Mean Difference 95% limits of agreement Atago 1
0.000
−0.008 to 0.007
−0.004
−0.011 to 0.003
Vet Cat
−0.008
−0.015 to 0.001
Atago Cat
−0.007
−0.014 to 0.000
ce
Vet Dog
Ac
Downloaded by [University of Bristol] at 12:01 28 February 2015
Cat urine
11
Figure 1. Specific gravity of (a) feline urine samples (n=23) and (b) canine urine samples (n=31), reported as specific gravity minus 1. Results are arranged from lowest to highest concentration based on total solids (g/L) of urine samples after drying. Total solids were divided by 2.33 to estimate urine specific gravity, with results shown as a dashed gold line with solid circles. Refractometer results are for the Atago 1 (dark blue line with squares), Rei 2 (light blue line with X), Vet Dog (green line with triangles), Atago Cat (long dashed yellow line and diamonds) and Vet Cat (red line with star). See Materials and methods for details of the refractometers.
rip
M an
us c
Figure 3. Relationship between the difference between specific gravity of urine samples measured using refractometers and total solids in the urine samples divided by 2.33 (TS), and the mean of the two measures. Results are shown for feline samples (n=23) measured using the (a) Atago 1 (a Goldberg-type medical refractometer) and (b) Atago Cat (a digital feline refractometer), refractometers, and for canine samples (n=31) measured using the (c) Atago 1 and (d) Atago Cat refractometers. Results are reported as specific gravity minus 1, The dotted line is the mean difference and the dashed lines are the 95% limits of agreement.
ce
pt e
d
Figure 4. Relationship between the difference between specific gravity of four feline and seven canine urine samples measured using a pycnometer and the (a) Atago 1 and (b) Atago Cat refractometers, and the mean of the two measures. Results are reported as specific gravity minus 1. The dotted line is the mean difference and the dashed lines are the 95% limits of agreement.
Ac
Downloaded by [University of Bristol] at 12:01 28 February 2015
t
Figure 2. Specific gravity of four feline and seven canine urine samples, reported as specific gravity minus 1, measured with a pycnometer (gold dashed line with solid circles) and with refractometers; Atago 1 (dark blue line with squares), Vet Dog (green line with triangles), Atago Cat (long dashed yellow line and diamonds) and Vet Cat (red line with star). See Materials and methods for details of the refractometers.
12
13
ce
Ac d
pt e M an us c
t
rip
Downloaded by [University of Bristol] at 12:01 28 February 2015
14
ce
Ac d
pt e M an us c
t
rip
Downloaded by [University of Bristol] at 12:01 28 February 2015
15
ce
Ac d
pt e M an us c
t
rip
Downloaded by [University of Bristol] at 12:01 28 February 2015
16
ce
Ac d
pt e M an us c
t
rip
Downloaded by [University of Bristol] at 12:01 28 February 2015
New Zealand Veterinary Journal Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/tnzv20
Comparison of specific gravity analysis of feline and canine urine, using five refractometers, to pycnometric analysis and total solids by drying a
b
c
HW Tvedten , H Ouchterlony & IE Lilliehöök a
Clinical Sciences, Faculty of Veterinary Medicine and Animal Sciences, Swedish University of Agricultural Sciences b
Institution for Women and Children's Health, Uppsala University
c
University Animal Hospital, Swedish University of Agricultural Sciences, Uppsala, Sweden Accepted author version posted online: 27 Jan 2015.
Click for updates To cite this article: HW Tvedten, H Ouchterlony & IE Lilliehöök (2015): Comparison of specific gravity analysis of feline and canine urine, using five refractometers, to pycnometric analysis and total solids by drying, New Zealand Veterinary Journal, DOI: 10.1080/00480169.2014.1002553 To link to this article: http://dx.doi.org/10.1080/00480169.2014.1002553
Disclaimer: This is a version of an unedited manuscript that has been accepted for publication. As a service to authors and researchers we are providing this version of the accepted manuscript (AM). Copyediting, typesetting, and review of the resulting proof will be undertaken on this manuscript before final publication of the Version of Record (VoR). During production and pre-press, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal relate to this version also.
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
Publisher: Taylor & Francis & New Zealand Veterinary Association Journal: New Zealand Veterinary Journal DOI: 10.1080/00480169.2014.1002553
Scientific Article
rip
t
Comparison of specific gravity analysis of feline and canine urine, using five refractometers, to pycnometric analysis and total solids by drying
HW Tvedten*§, H Ouchterlony† and IE Lilliehöök‡
M an
us c
Clinical Sciences, Faculty of Veterinary Medicine and Animal Sciences, Swedish University of Agricultural Sciences. † Institution for Women and Children's Health, Uppsala University. ‡ University Animal Hospital, Swedish University of Agricultural Sciences, Uppsala, Sweden. § Author for correspondence: Email: [email protected] Abstract
AIMS: To compare the performance of five refractometers for determination of urine specific gravity in cats and dogs, with reference to weight of total solids and pycnometer analysis.
d
METHODS: Urine samples from 27 cats and 31 dogs submitted for routine urinalysis were
pt e
included. Urine specific gravity was determined with five refractometers. Four were optical, hand held refractometers with a temperature compensation method and one was a digital model. Urine was dried to determine the precise weight of total solids. The total solids (g/L) were converted to an estimated specific gravity by division with 2.33. Urine specific gravity of four feline and seven
ce
canine samples were analysed with a pycnometer. Limits of agreement analysis was used to evaluate the agreement between specific gravity (analysed as specific gravity minus 1) measured by the refractometers and estimated from dried total solids, or pycnometer results,
Ac
Downloaded by [University of Bristol] at 12:01 28 February 2015
*
RESULTS: The five refractometers reported clearly different results from each other. Proportional negative bias was noted for refractometer results compared to estimated specific gravity from total solids and a constant negative bias compared to pycnometer results. The two refractometers designed for cat urine reported similar and lowest specific gravity results with a mean negative bias of 0.007 and 0.008 units compared to estimated specific gravity from total solids and a mean negative bias of 0.006 units compared pycnometer results.
1
CONCLUSIONS: Refractometer results did not increase consistently with increasing urine specific gravity compared to reference methods or to other refractometers. Two feline refractometers reported consistently lower specific gravity results than reference methods and other refractometers. CLINICAL RELEVANCE: Because of this imprecision, veterinarians should not use precise cut off values such as 1.030 or 1.035 for evaluation of renal concentrating ability in dogs and cats. Veterinarians should consider the variability of refractometric specific gravity results in their clinical assessment. Two feline refractometers appeared to report falsely low specific gravity
rip
us c
KEY WORDS: Urine specific gravity, refractometer, total solids, pycnometer, cats, dogs
M an
Introduction
Determination of urine specific gravity is commonly used as a sensitive test of renal concentrating ability for canine and feline patients (Watson 1998). Previous comparison of a cat refractometer and a standard medical refractometer suggested the cat refractometer reported falsely low results, that may interfere with clinical diagnosis (Tvedten and Norén 2014). This observation needed to be
d
better validated.
pt e
The objective of the current study was to compare the performance of several currently available refractometers, including two cat refractometers, to each other and to two reference methods over a wide range of specific gravities. The two reference methods used were determining the weight of
ce
total solids from feline and canine urine samples after drying, and pycnometer analysis of specific gravity (George 2001). A pycnometer is a special glass flask which can be filled with a very exact volume of fluid. The weight of an identical volume of urine divided by the weight of the same
Ac
Downloaded by [University of Bristol] at 12:01 28 February 2015
t
results.
volume of water equals the specific gravity. The reference methods should indicate which refractometers best reflected these reference values at various urine specific gravities.
Materials and methods Sample and patient selection
Urine samples for the study included 27 feline and 31 canine samples which were submitted for patient diagnosis to the University Animal Hospital, Swedish University of Agricultural Sciences between January and April 2013. All feline urine samples that did not appear red (bloody) and had a supernatant which was not turbid and had a volume of >2 mL were included. A similar number of 2
canine samples were selected with the same criteria and with a preference for more concentrated urines. More concentrated urines should have a greater weight of dry matter after drying and therefore measurements of the weight of empty glass tubes compared to those with dried urine material may be more precise. Urine samples were centrifuged at 2000g for 5 minutes. Supernatants from urine samples were analysed with refractometers that day. The samples were stored in a refrigerator (about 4 °C) if the rest of the analysis could be completed within two days. Otherwise samples were frozen (about −20 °C) until analysis with the same methods as fresh urine samples.
t
Instruments and methods
rip
Five refractometers were used to analyse specific gravity. Selection was only based on easy
us c
Goldberg type refers to an optical, hand held refractometer with a temperature compensation
method developed by Herbert Goldberg. One was the Uricon refractometer (Atago Corp, Tokyo Japan); termed Atago 1. Two refractometers appeared to be the same but one was labelled as a Reichert TS meter, termed Rei 2, and one was labelled Leica TS meter, termed Rei 3; both were
M an
Model 10400A Reichert TS meters (Cambridge Instruments Inc, Buffalo NY, USA). Because Rei 2 and Rei 3 appeared identical and gave the same results only results from Rei 2 are presented. One refractometer had two scales, for cats and dogs; the Model 10436 AO Veterinary refractometer (AO Scientific Instruments, Buffalo NY, USA), referred to as Vet Cat and Vet Dog, respectively. One
pt e
termed Atago Cat.
d
refractometer was a digital model; the Atago ‘Pocket’ PAL-USG Cat refractometer (Atago Corp)
Urine samples that had a specific gravity above the range of analysis of a refractometer were diluted with an equal volume of distilled water to obtain a result greater than the refractometer reported.
ce
The upper values for specific gravity reported by various refractometers without requiring dilution were 1.035 for Rei 2 and Rei 3, 1.050 for Atago 1, 1.060 for Vet Cat and Vet Dog, and 1.080 for Atago Cat.
Ac
Downloaded by [University of Bristol] at 12:01 28 February 2015
availability from two local laboratories. Four optical refractometers were of the "Goldberg type".
Refractometers were calibrated with distilled water at room temperature to give a zero result of 1.000. Analysis of data was based on measured specific gravity minus 1.000. The effect of substances dissolved in urine on specific gravity is called the specific gravity increment and is the measured specific gravity minus the specific gravity of water (1.000) (Price et al. 1940). For example, a solution with a specific gravity of 1.040 is twice as concentrated as 1.020. Drying of the urine samples was used to determine total solids of the urine supernatants. Weight of the dry material after drying of 2 mL of urine was used to calculate total solids. The tubes were weighted empty, when they contained urine and after drying. Freeze drying of urine with a freeze 3
drying instrument (Scanvac Coolsafe 55-4, LaboGene ApS, Lynge Denmark) was attempted initially, but urine did not dry with the usual freeze drying procedure even overnight. Subsequently a special evaporator centrifuge (SpeedVac Plus SC250DDA; Savant, Thermo Fisher Scientific Stockholm Sweden) was attached to the vacuum system of the freeze drying instrument and urine was centrifuged in 50 mL plastic tubes to evaporate all fluid at room temperature. Drying in the SpeedVac was usually performed for 16 hours. Samples which had a positive glucose result (1+ to 4+) on dipstick analysis were excluded from evaluation of total solids data because three urine samples with 3+ to 4+ glucose never dried completely. Some matter remained as a gel or syrup in
rip
t
those three feline samples after drying 16–32 hours. The total solids from drying the feline and
canine urine samples were converted to estimated specific gravities by dividing the weight of total
us c
A pycnometer (Pycnometer 2 mL; Joan Lab Equipment, Zhejiang, China) was used to determine specific gravity of eleven of the previously analysed urine samples (four feline and seven canine) stored in a −20 ºC freezer. Samples with higher specific gravity results and >2 mL volume were
M an
selected and thawed to room temperature prior to analysis. The weights of the dry pycnometer, the pycnometer filled with distilled water and the pycnometer filled with the urine sample were determined with an analytic grade balance (Mettler Toledo PB 303, Stockholm Sweden). Specific
Statistical analysis
d
gravity was the weight of an equal volume of urine divided by that of water.
pt e
Limits of agreement analysis (Bland and Altman 1999) was used to evaluate the agreement between specific gravity measured by the refractometers and estimated from dried total solids, or
ce
pycnometer results, using Analyze-it v. 3.80 (Analyze-it Software, Ltd., Leeds United Kingdom).
Results
The specific gravity results for the refractometers compared to the estimated specific gravity based on total solids from drying of feline and canine urine samples are shown in Figure 1. The Rei 2
Ac
Downloaded by [University of Bristol] at 12:01 28 February 2015
solids by 2.33 (Price et al. 1940).
refractometer reported results only up to 1.035 and required dilutions to obtain greater values for comparison in this study. Comparison of the results for the Vet Dog and Rei 2 refractometer showed that values for the Rei 2 had a shift to falsely high results above 1.035 after dilution. Therefore results for Rei 2 are only reported up to 1.035. The estimated specific gravity from dried total solids was similar to most refractometer results at low values but greater than all results at higher specific gravities. Atago 1 results were generally greater than other refractometer results at lower urine specific gravities up to about 1.040. The two 4
feline refractometers (Vet Cat and the Atago Cat) consistently gave the lowest and almost identical results. Pycnometer results from four feline and seven canine urine samples are shown with results from four refractometers in Figure 2. Results from the Rei 2 refractometer are not shown because only two results were ≤1.035 and reported by the instrument. The results from the two feline refractometers were consistently lower than pycnometer results, while the other refractometer results were more similar to pycnometer results.
rip
t
Results for the limits of agreement analysis comparing the refractometers and estimated specific
gravity from dried total solids are given in Table 1. The results from two feline refractometers (Vet
us c
from the Atago 1 and Atago Cat refractometers with specific gravity estimated from dried total solids. An increasing negative bias with increasing mean specific gravity is shown for both refractometers.
M an
Results for the limits of agreement analysis comparing four refractometers and pycnometer specific gravity are given in Table 2. Again the two feline refractometers (Vet Cat and Atago Cat) had the greatest mean negative bias. Figure 4 shows the comparison for the Atago 1 and Atago Cat
pt e
d
refractometers; a constant error is noted especially with the Atago Cat refractometer.
Discussion
Determination of urine specific gravity with a refractometer is simple, quick and inexpensive.
ce
Therefore refractometers are routinely used in most veterinary clinics and laboratories and will continue to be used in the future. However, there were prominent and inconsistent differences in results reported by five currently available refractometers over the range of results seen with feline and canine patient urine samples. Because of this imprecision, veterinarians should not use precise
Ac
Downloaded by [University of Bristol] at 12:01 28 February 2015
Cat and Atago Cat) had the greatest mean negative bias. Figure 3 shows the comparison of results
cut off values such as 1.030 or 1.035 for evaluation of renal concentrating ability in dogs and cats. Veterinarians should consider this variability of refractometric specific gravity results in diagnosis. In order to evaluate the results from the different refractometers, the definition of specific gravity should be considered. Urine specific gravity is a ratio, without units, of the weight of urine divided by the weight of an equal volume of water at a certain temperature. Density is similar but is weight/volume (kg/m3 or g/L). Pycnometer analysis most directly determines urine specific gravity
by precise measurement of the weight of exactly the same volume of urine and water and determining the ratio of the weight of urine to water. The other reference method used in this study 5
determined the total solids (g/L) which indicates the density of urine, but which is not the same as specific gravity results. Drying of known volumes of urine and precise measurement of dry material remaining objectively measured the amount of total solids (g/L). However, adding the same weight of different substances to 1 L of water causes different increments of change in specific gravity due to different densities of the substances, as shown in Supplementary Table 11.. Dissolving 50 g of NaCl in 1 L of water did not give a specific gravity of 1.050. The five refractometers reported specific gravities between
t
1.018 and 1.029 for that solution A solution with an equal weight of 50 g of urea in 1 L of water
rip
resulted in specific gravities of 1.014–1.025 with the five refractometers. These are very different
us c
A conversion factor was needed to adjust results of total solids (g/L) to results which could be compared to refractometer specific gravity results. One conversion factor is called the Häser formula or Trapp-Häser formula (Price et al. 1940). Multiplying the third and fourth decimal figures
M an
of the specific gravity by 2.33 is used to estimate total solids of human 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. Urine is a complex mixture of substances so this conversion does not correct the total solids of each individual urine to a correct specific gravity. However use of the 2.33 factor provided consistent comparison values of similar magnitude for
d
evaluation in this study. Other conversion factors of 2.18–2.60 have been determined from various
pt e
studies (Price et al. 1940). Use of a different factor could affect comparisons and conclusions of the current study. Other potential errors include that only the urine supernatant was used for the drying method and thus particulate material in the sediment such as crystals was not included. Not
ce
including crystals in the sediment, such as stuvite, would affect phosphate, calcium, magnesium etc. There are also volatile substances in urine that would be lost during evaporation in the SpedVac system. Other substances in urine besides glucose (such as protein) could affect completeness of drying in the SpedVac system.
Ac
Downloaded by [University of Bristol] at 12:01 28 February 2015
results, for 1.025 is 79% more concentrated than 1.014.
Based on results from the current study, none of the five refractometers matched closely or consistently results from the two reference methods in terms of magnitude over the entire range of increasing urine specific gravities. Different types of errors were noted for the two reference methods. For example, the feline refractometer (Atago Cat) had a greater negative bias with increasing specific gravity (proportional error) compared to estimated specific gravity from total solids . However, this refractometer had a constant negative error compared to pycnometer results.
1
DOI: [production editor to insert]
6
Thus the relationship and type of error (proportional or constant) of refractometric specific gravity compared to the two reference methods was different. The two feline refractometers (Vet Cat and Atago Cat) were consistently lower than reference methods and consistently gave the lowest specific gravity results of all refractometers. There is no apparent reason to use a feline refractometer or to use a conversion formula to convert the urine specific gravity of a feline urine specimen analysed with a routine medical refractometer to a “corrected” urine specific gravity. The other refractometers and Vet Dog refractometer appeared
rip
t
most similar to pycnometer results and estimated specific gravity from total solids. Special refractometers designed for feline urine have been recommended and sold for many years
us c
refractive) than dog and human urine and a conversion formula has been suggested to correct feline urine specific gravity results (Wolf et al. 1990). The AO Vet Cat and Vet Dog refractometer apparently measures refractive index and then uses two different conversion factors for reporting a
M an
canine or feline specific gravity for its two scales. The Atago Cat refractometer and the Vet Cat refractometer appeared to use the same conversion factor, because they reported similar to identical results, despite one being digital (Atago Cat) and the other a Goldberg type (Vet Cat). This indicates that differences among refractometers were not because they are digital or traditional Goldberg type. The formulas for predicting feline urine specific gravity from refractive index have
d
not been revalidated since 1956 but are currently being used by refractometer manufacturers or as a
pt e
correction formula by veterinarians (George 2001). This belief about unique aspects of feline specific gravity determination by refractometric analysis has been based on one article (Rubini and Wolf 1956). That article evaluated refractive index,
ce
specific gravity and weight of dried total solids from190 human urine samples, 22 feline and 21 canine samples. The change in refractive increment determined along the scale of their refractometer for feline urine samples was actually linear with increasing total solids of human and
Ac
Downloaded by [University of Bristol] at 12:01 28 February 2015
(Bennett et al. 2011). It has been stated that feline urine refracts light differently (greatly more
canine urine samples. This should suggest that feline urine was similar to human and canine urine in regards to refractometric analysis. However, specific gravity (by determining the ratio of the weight of urine samples divided by an equal volume of water) of feline samples appeared lower than expected compared to human and canine urine. An aspect in that study that may be important to consider was that all of the 22 feline urine samples were more concentrated than the human or
canine urines. Feline samples had specific gravities >1.030 while canine and human urine had more dilute urines. Human urine is considered concentrated if >1.020 (Price et al. 1940) while feline urine is considered concentrated if >1.035 or even >1.040 (Watson 1998). 7
Our study was similar to the study of Rubini and Wolf (1956) in comparing refractometric specific gravity to total solids and pycnometer type analysis. Inconsistencies were demonstrated among refractometers in our study, especially at higher specific gravities and it may be the variation reported by Rubini and Wolf relate to the higher urine concentrations of their 22 cat urine samples compared to the human or canine samples. It may not be that feline urine causes a different refractive index reading but only that they are often very concentrated. Estimated specific gravity from total solids appeared greater than all refractometric results at the highest urine specific gravities of feline and canine urine samples in our study. Therefore all refractometers underestimate
rip
t
specific gravity at highest concentrations if total solids are used as the reference method.
us c
specific gravity. Most veterinarians expect that all refractometers give equivalent results, but this was not observed. Atago 1 reported the highest specific gravity results of all refractometers and had least mean bias compared with estimated specific gravity from total solids. It reported higher specific gravity results compared with total solids at lower specific gravities of about 1.030 or less,
M an
suggesting a false positive bias in more dilute urines. Above a urine specific gravity of about 1.040, the Atago 1 results were less than total solids suggesting a false negative bias in more concentrated urines. A similar pattern of changing biases over increasing concentrations was noted for both feline and canine urine specimens. The Atago 1 results were most similar to pycnometer results and
d
increased in parallel with pycnometer results.
pt e
The Rei 2 refractometer had a distinct upward shift in results (false positive bias) compared to Vet Dog at specific gravity >1.035. For this refractometer dilution of urine was performed for urines with specific gravity >1.035, and these results indicate that refractive specific gravity results are
ce
neither linear nor accurate after dilution. It has been previously reported that if urine is diluted with water the change in specific gravity (specific gravity increment) is an inverse linear function of the degree of dilution (Price et al. 1940). If veterinarians want results >1.035 they should check that their refractometer accurately reports results above this value.
Ac
Downloaded by [University of Bristol] at 12:01 28 February 2015
Results from different refractometers in our study did not show consistent bias with increasing urine
Other inconsistencies were noted with various refractometer results with increasing concentrations of NaCl or urea (Supplementary Table 12). This imprecision among refractometers should be
recognised and indicates veterinarians should not interpret specific gravity results exactly according to recommended cut off values such as urine specific gravity of >1.030 or >1.035. The diagnostic principles are correct, but refractometric specific gravity results are imprecise and inconsistent.
2
DOI: [production editor to insert]
8
Recommended cut off values such as >1.030 should be more approximate guidelines than precise numerical limits. All refractometers were calibrated to 1.000 with distilled water but reported very different and variable results. Calibrating all refractometers with water to give a result of 1.000 is the standard procedure. However, this did not assure that results in the clinically diagnostic range of 1.030–1.035 were accurate or consistent among instruments as documented in this study. The method of calibration of refractometers should be re-evaluated.
rip
t
Variation among refractometer results have been reported earlier, but Bennett et al. (2011)
concluded that clinical interpretation was not affected by a negative bias of 0.003 with a cat
us c
specific gravity and refractive index". The mean negative bias with feline refractometers in our study was about 0.004–0.007 with even greater negative bias noted in individual samples and with different refractometers. If a diagnostic cut off value of >1.030 is used to indicate adequate renal
M an
concentrating ability for dogs, then seven of our 31 dogs (22%) would have had appeared adequate by the Atago 1 refractometer but not with the Atago cat refractometer. Similarly if a diagnostic cut off value of >1.035 is used to indicate adequate concentrating ability for cats, then three of 27 cats (11%) would have had appeared adequate by the Atago 1 refractometer but not the Atago Cat refractometer.
d
In conclusion, different refractometers reported inconsistent and clinically different results.
pt e
Refractometer results did not increase consistently with increasing urine specific gravity compared to reference methods or to other refractometers. Two feline refractometers reported consistently lower specific gravity results than reference methods and other refractometers for both feline and
ce
canine urine samples and pure chemical solutions. The feline refractometer results appeared falsely low. Routine medical refractometers and a canine refractometer reported results more similar to specific gravity results from reference methods based on total solids and pycnometer analysis for
Ac
Downloaded by [University of Bristol] at 12:01 28 February 2015
refractometer "programmed with a mathematical representation of the relationship of feline urine
both feline and canine urine. Refractometric specific gravity analysis is still recommended for urine concentration analysis, but consideration of imprecision and negative biases must be applied.
References Bennett AD, McKnight GE, Dokin SJ, Simpson KE, Schwartz AM, Gunn-Moore DA. Comparison of a digital and optical analogue hand-held refractometer for the measurement of feline urine specific gravity. Journal of Feline Medicine and Surgery 13, 152–4, 2011 9
Bland JM, Altman DG. Measuring agreement in method comparison studies. Statistical Methods in Medical Research 8, 135–60, 1999 George J. The usefulness and limitations of hand-held refractometers in veterinary laboratory medicine: an historical and technical review. Veterinary Clinical Pathology 30, 201–10, 2001 Price JW, Miller M, Haymen JM. The relation of specific gravity to composition and total solids in normal human urine. Journal of Clinical Investigations 19, 537–53, 1940
t
Rubini ME, Wolf AV. Refractometric determination of total solids and water of serum and urine.
rip
Journal of Biological Chemistry 225, 869–76, 1956
us c
USG Cat refractometer for determination of urine specific gravity in dogs and cats. Veterinary Clinical Pathology 43, 63–6, 2014
M an
Watson AD. Urine specific gravity in practice. Australian Veterinary Journal 76, 392–8, 1998 *Wolf AV, Brown MG, Prentis PG. Concentrative properties of aqueous solutions: conversion tables. In: Weast RC (ed). The CRC Handbook of Chemistry and Physics. 70th Edtn. Pp D252–
pt e
* Non-peer-reviewed
d
D253. CRC Press, Boca Raton, Florida, USA, 1990
ce
Submitted 10 February 2014
Accepted for publication 08 December 2014 First published online [insert date]
Ac
Downloaded by [University of Bristol] at 12:01 28 February 2015
Tvedten HW, Norén Å. Comparison of a Schmidt and Haensch refractometer and Atago PAL-
10
Table 1. Results of limits of agreement analyses comparing results for specific gravity (analysed as specific gravity minus 1) of feline (n=23) and canine (n=31) urine samples, measured using refractometers, with specific gravity estimated from dried total solids in the urine samples divided by 2.33. See Materials and methods for details of the refractometers. Mean Difference
95% limits of agreement
Atago 1
0.001
−0.010 to 0.012
Rei 2
−0.002
−0.004 to 0.001
Vet Dog
−0.004
−0.012 to 0.003
Vet Cat
−0.008
−0.016 to 0.001
Atago Cat
−0.007
−0.016 to 0.002
Atago 1
0.002
−0.008 to 0.013
Rei 2
−0.001
−0.003 to 0.004
Vet Dog
−0.004
−0.012 to 0.005
Vet Cat
−0.006
Atago Cat
−0.006
Refractometer
t rip us c
M an
Dog urine
−0.017 to 0.004
−0.016 to 0.004
pt e
d
Table 2. Results of limits of agreement analyses comparing results for specific gravity (analysed as specific gravity minus 1) of four feline and seven canine urine samples, measured using refractometers, with pycnometer results. See Materials and methods for details of the refractometers. Refractometer Mean Difference 95% limits of agreement Atago 1
0.000
−0.008 to 0.007
−0.004
−0.011 to 0.003
Vet Cat
−0.008
−0.015 to 0.001
Atago Cat
−0.007
−0.014 to 0.000
ce
Vet Dog
Ac
Downloaded by [University of Bristol] at 12:01 28 February 2015
Cat urine
11
Figure 1. Specific gravity of (a) feline urine samples (n=23) and (b) canine urine samples (n=31), reported as specific gravity minus 1. Results are arranged from lowest to highest concentration based on total solids (g/L) of urine samples after drying. Total solids were divided by 2.33 to estimate urine specific gravity, with results shown as a dashed gold line with solid circles. Refractometer results are for the Atago 1 (dark blue line with squares), Rei 2 (light blue line with X), Vet Dog (green line with triangles), Atago Cat (long dashed yellow line and diamonds) and Vet Cat (red line with star). See Materials and methods for details of the refractometers.
rip
M an
us c
Figure 3. Relationship between the difference between specific gravity of urine samples measured using refractometers and total solids in the urine samples divided by 2.33 (TS), and the mean of the two measures. Results are shown for feline samples (n=23) measured using the (a) Atago 1 (a Goldberg-type medical refractometer) and (b) Atago Cat (a digital feline refractometer), refractometers, and for canine samples (n=31) measured using the (c) Atago 1 and (d) Atago Cat refractometers. Results are reported as specific gravity minus 1, The dotted line is the mean difference and the dashed lines are the 95% limits of agreement.
ce
pt e
d
Figure 4. Relationship between the difference between specific gravity of four feline and seven canine urine samples measured using a pycnometer and the (a) Atago 1 and (b) Atago Cat refractometers, and the mean of the two measures. Results are reported as specific gravity minus 1. The dotted line is the mean difference and the dashed lines are the 95% limits of agreement.
Ac
Downloaded by [University of Bristol] at 12:01 28 February 2015
t
Figure 2. Specific gravity of four feline and seven canine urine samples, reported as specific gravity minus 1, measured with a pycnometer (gold dashed line with solid circles) and with refractometers; Atago 1 (dark blue line with squares), Vet Dog (green line with triangles), Atago Cat (long dashed yellow line and diamonds) and Vet Cat (red line with star). See Materials and methods for details of the refractometers.
12
13
ce
Ac d
pt e M an us c
t
rip
Downloaded by [University of Bristol] at 12:01 28 February 2015
14
ce
Ac d
pt e M an us c
t
rip
Downloaded by [University of Bristol] at 12:01 28 February 2015
15
ce
Ac d
pt e M an us c
t
rip
Downloaded by [University of Bristol] at 12:01 28 February 2015
16
ce
Ac d
pt e M an us c
t
rip
Downloaded by [University of Bristol] at 12:01 28 February 2015
