Absorption of maternal antibodies in neonatal calves is commonly evaluated by measuring immunoglobulin G (IgG) as it is the primary immunoglobulin in bovine colostrum.1 Radial immunodiffusion assays (RID) are the reference test for measuring IgG concentrations within neonatal calf serum.2,3 However, RID are time-consuming, with results reported 24 to 48 hours after sampling. Refractometers are portable devices with on-farm application and a convenient turnaround time producing results within minutes of sampling.4 Refractometers indirectly measure IgG by measuring serum total protein (STP) that is comprised of albumin and globulins with globulins accounting for approximately 60% of the total protein content.5 In neonatal calves, immunoglobulins account for approximately 70% of the globulin fraction with 80% to 90% of the immunoglobulin portion consisting of IgG.6,7 Calves that consume high-quality colostrum shortly after birth will have an increased concentration of IgG within their serum that results in an increased STP.
Neonatal calves with low levels of IgG and STP have an increased risk of disease before weaning.8 Immunoglobulin G threshold values to evaluate adequate absorption of maternal antibodies are often used to predict a calf’s disease risk before weaning. These values are commonly extrapolated from IgG concentrations to establish STP cutoffs for rapid assessment of absorption of maternal immunity, but previous research has noted imprecision in IgG concentrations measured by RID.9,10 Therefore, variability in IgG concentrations may affect the accuracy of STP threshold values in determining disease risk.2,11 Additionally, the repeatability and precision of STP values in neonatal calves using an optical refractometer have not been reported and may also contribute to misclassification of disease risk.
The primary objective of this study was to determine the precision of STP measured using an optical refractometer. The secondary objective was to compare the precision of STP measured using an optical refractometer and IgG concentration measured using RID in neonatal calf serum by evaluating the amount of variance produced by each test. We hypothesized that STP measured using an optical refractometer would produce the same amount of variance as a commercial RID assay used to measure IgG concentrations.
Methods
A method-comparison replication study was conducted between Dec. 19, 2020, and Nov. 10, 2021, to evaluate the precision of a commercially available optical refractometer (RHC-300ATC, Grand Index). Sera were repeatedly measured to observe the potential variance that the test would produce. The variance produced by the optical refractometer was compared to the variance produced by a commercially available bovine IgG RID (Triple J Farms).
Sample selection
Serum, stored frozen at −80 °C, from a serum bank of blood collected from 865 neonatal calves between the age of 2 and 7 days of age was used in this study.10 A random number generator in a spreadsheet software (Excel 2016, Microsoft Corp) was used to randomly select serum within the serum bank from the following strata: low-serum IgG stratum with previously measured IgG concentration using a commercially available bovine IgG RID (Triple J Farms) between >5.0 and <15.0 g/L; moderate-serum IgG stratum previously measured IgG concentration of 35.0–45.0 g/L; high-serum IgG stratum previously measured IgG concentration of 60.0–70.0 g/L.10 The low-, moderate-, and high-serum IgG strata had 39, 182, and 109 sera to sample from, respectively. Initially, 1 serum was selected from each stratum, however, the initial serum selected from the low-serum IgG stratum required a dilution. Therefore, 3 additional sera were selected from the low-serum IgG stratum to ensure that the observations were correct for the low-serum IgG stratum. There were a total of 4 sera from the low-serum IgG stratum, 1 serum from the moderate-serum IgG stratum, and 1 serum from the high-serum IgG stratum.
A commercially available optical refractometer (RHC-300ATC, Grand Index) was used to quantify STP. The STP scale ranged from 20 to 140 g/L and was marked in 1.0 g/L intervals. Serum total protein was measured to the nearest 1.0 g/L. A commercially available bovine IgG RID (lots 7284A10, 7284B09, 7284B20, 7284B30, 7284B40; Triple J Farm) was used as the reference test to quantify the serum IgG concentration. Each commercial kit included 3 IgG standards, with IgG concentrations of 28.0, 14.7, and 1.8 g/L (lots 7286G3, 7286G2, 7286G1, respectively), an anti–IgG antibody impregnated RID plate with 24 wells, and a package insert with instructions on how to perform the test. Sera that produced a precipitin ring greater than the highest standard were diluted 1:1 with phosphate-buffered saline and re-evaluated until the serum precipitin ring was within the range of the provided standards.
Refractometry procedure
To mask the reader to the identity of the sera, the 6 sera were randomly ordered in a spreadsheet software (Microsoft Excel) so that each sera would be read 13 times in a random fashion. The sera were thawed to room temperature for 1 hour and vortexed (3S) before the first reading and every 4th reading. The sera and refractometer were maintained at room temperature during the experiment. Approximately 1 drop (20 µL) of serum was placed on the measuring prism of the refractometer and the values on the refractometer were read by a reviewer masked to the serum identification number and order. After each sample, the measuring prism was washed with distilled water and dried with a tissue (Kimtech Science Kimwipes, Kimberley-Clark Professional). The sera were evaluated until 1 serum was depleted, resulting in 13 observations per serum.
Radial immunodiffusion assay procedure
Quantification of serum IgG by a commercial RID was performed as previously described.10 Based on preliminary data, an a priori sample size calculation was performed to detect a 1.0 g/L difference in a serum’s mean with a SD of 1.5 g/L between 2 lots, an α = 0.05, and a power of 0.95.10 The required number of replications was 14 per lot.12 The 6 sera were evaluated 28 times per sera for a total of 168 observations. Each serum was measured 7 times on a single plate. This arrangement was replicated a total of 4 times, with 2 plates run from each of the 2 separate RID lots (lots 7284A10, 7284B40 for sera 891, B026, 2987; lots 7284B20, 7284B30 for sera A084, 1912, B020) for a total of 28 observations per serum. Serum that produced a precipitin ring greater than the highest standard were consecutively serially diluted with phosphate-buffered saline and re-evaluated on a plate from the same lot until the precipitin ring was within the range of the standards. IgG concentrations were calculated from the precipitin ring diameters squared using a linear standard curve generated from standards from all plates performed within the same business day.10
Statistical analysis
Descriptive statistics for the results of the test were generated in a spreadsheet software (Microsoft Excel). To determine the precision of the tests, the coefficient of variation (CV) was separately calculated for the STP and IgG concentration for each serum. The equality of variance between serum samples was separately evaluated for STP and IgG concentrations using the Levene test for homogeneity of variance (Proc GLM, SAS v9.4; SAS Institute) and was assessed to determine if the CV could be averaged by test. The level of significance was set at α = 0.10 to reduce the chance of a type II error. Homogeneity of variance was present for STP but was not present between the 6 sera for IgG concentration. The sera were divided into a low-serum IgG stratum (sera 891, B026, 2987, A084) and a combined moderate- and high-serum IgG stratum (sera 1912, B020) and equality of variance for the IgG concentrations between sera was separately evaluated by IgG stratum. The estimated repeatability coefficient was separately calculated for STP and IgG concentration for serum with homogeneity of variance.13 A direct comparison between the tests for STP and IgG concentration could not be performed due to differences between the components being tested, therefore, CV was used to standardize the variance. The serum CV was averaged to calculate an average CV for STP. For IgG, separate average CVs were calculated for the low- and high-serum IgG strata.
To assess the secondary objective and evaluate the equality in CV by test, 5,000 replications of unrestricted random sampling (Proc SURVEYSELECT, SAS v9.4) were used to separately generate 5,000 datasets for STP and IgG concentration. A CV was generated for each test and serum within each replication, and these values were concatenated. Equal variance in the simulated CV was assessed between STP and IgG concentration for each serum using the Levene test for homogeneity with α = 0.10. Heterogeneity of variance was present between the tests by serum; therefore, the difference between the simulated CV for IgG concentration and STP for each serum was assessed using a Kruskal-Wallis test (Proc NPAR1WAY, SAS v9.4) with α = 0.05.
Results
Variance in refractometry
We included 13 observations of each serum in the analysis (Table 1). The Levene test analysis detected no difference in the variance produced by repeatedly measuring STP between the sera (P = .39, average mean = 53.8 g/L; variance = 5.3 [g/L]2, SD = 2.3 g/L). Serum total protein had an estimated repeatability coefficient of 6.5 g/L (95% CI: 1.2, 11.7 g/L) and an average CV of 4.2%.
Descriptive statistics for serum total protein (STP) concentration measured 13 times per sample with an optical refractometer vs IgG concentration measured 28 times per sample with radial immunodiffusion assay for banked sera samples from 6 neonatal calves (891, B026, 2987, A084, 1912, and B020) evaluated in a method-comparison study conducted between Dec. 19, 2020, and Nov. 10, 2021.
| Serum total protein | Immunoglobulin G | |||||||
|---|---|---|---|---|---|---|---|---|
| Serum ID | N | Mean (g/L) | SD | CV (%) | N | Mean (g/L) | SD | CV (%) |
| 891 | 13 | 40.0 | ± 0.4 | 1.09 | 28 | 6.2 | ± 1.0 | 16.43 |
| B026 | 13 | 42.0 | ± 2.0 | 4.50 | 28 | 9.4 | ± 1.1 | 11.68 |
| 2987 | 13 | 47.0 | ± 3.0 | 6.17 | 28 | 14.8 | ± 1.2 | 8.18 |
| A084 | 13 | 51.0 | ± 1.0 | 2.63 | 28 | 23.7 | ± 1.0 | 4.24 |
| 1912 | 13 | 68.0 | ± 7.0 | 10.09 | 28 | 54.1 | ± 8.0 | 14.45 |
| B020 | 13 | 75.0 | ± 0.6 | 0.85 | 28 | 62.0 | ± 10.0 | 16.46 |
CV = Coefficient of variation. SD = Standard deviation.
Variance in radial immunodiffusion assay
We included 28 observations of each serum in the analysis (Table 1). Sera 891, 2987, and B026 did not require dilution. Serum A084 required a 1:1 dilution for 1 observation. Serum 1912 required sixteen 1:1 dilutions and twelve 1:3 dilutions. Serum B020 required eight 1:1 dilutions and twenty 1:3 dilutions. The IgG concentrations were normally distributed for each serum. The standards used to determine the IgG concentrations of the serum had an average repeatability coefficient of 2.6 g/L (95% CI: 1.3, 4.0 g/L). The Levene test analysis determined that the variance produced by repeatedly measuring the IgG concentrations differed significantly between the sera (P < .0001). For the serum in the low-serum IgG stratum, the variance produced by repeatedly measuring the IgG concentration did not differ between the sera (P = .84; variance = 1.2 [g/L]2, SD = 1.1 g/L). The low-serum IgG stratum had an estimated repeatability coefficient of 3.0 g/L (95% CI: 2.8, 3.2 g/L) and an average CV of 10.1%. The Levene analysis did not detect a difference in the variance produced by repeatedly measuring the IgG concentrations between the sera of the moderate-serum IgG stratum and the high-serum IgG stratum (P = .31; variance = 84.1 [g/L]2, SD = 9.2 g/L). The high-serum IgG stratum had an estimated repeatability coefficient of 25.4 g/L (95% CI: 22.5, 28.7 g/L) and an average CV of 15.5%.
Coefficients of variation comparison
The number of observations for the bootstrap was 78 for STP and 168 for IgG. The minimum and maximum number of observations per serum in each replication was 1 and 27 for STP and 11 and 49 for IgG. The variance in the CV estimated from the random sampling using the Levene test found a significant difference between IgG concentration and STP (P < .001) for all sera except A084 (P = .31; Figure 1). For the sera with heterogenous variance, the estimated CV for IgG concentration frequently produced a larger variance compared with STP. The distribution of the estimated CVs was significantly different between STP and IgG concentration for each serum (P < .0001) with STP producing a lower CV compared to IgG concentration.
Distribution of 5,000 simulated coefficient of variation for serum total protein (STP) concentration measured 13 times per sample with an optical refractometer vs IgG concentration measured 28 times per sample with radial immunodiffusion assay for banked sera samples from 6 neonatal calves (891, B026, 2987, A084, 1912, and B020) evaluated in a method-comparison study conducted between Dec. 19, 2020, and Nov. 10, 2021. Six banked sera were selected by the number of 1:1 serum to phosphate buffered saline dilutions previously required for the precipitin ring to be within the range of the manufacturer’s standards: no dilution needed (891, B026, 2987, and A084), 1 dilution needed (1912), and 2 dilutions needed (B020). For each box-and-whisker plot, the solid line within the box represents the median; the lower and upper limits of the box represent the interquartile (25th and 75th percentiles) range, respectively; the whiskers delimit the range; and the x represent the mean. Within calf, data with different letter superscripts differ significantly (P < .001).
Citation: American Journal of Veterinary Research 84, 11; 10.2460/ajvr.23.05.0096
Discussion
The repeatability of STP values measured with an optical refractometer for neonatal calf serum had not been evaluated. Additionally, no studies have compared the repeatability of STP values measured using an optical refractometer and IgG concentrations measured using a RID. To evaluate these problems, STP values were repeatedly measured using an optical refractometer for 6 sera. The variance in STP produced by the repeated measures was compared to the variance produced by repeated measures of IgG concentrations measured using RID from the same 6 sera. The STP value varies with each repetition. The variance produced by repeated measures of the IgG concentration was greater than the variance produced by repeated measures of STP. This is the first investigation to compare the variance in STP from neonatal calves using an optical refractometer and IgG concentration using RID.
The limitations in the present study pertain to the tests used. Caution should be used when extrapolating the CV for the current refractometer to other refractometers. For example, we used an optical refractometer that may not represent the variance observed with a digital refractometer. Additionally, each refractometer manufacturer uses a proprietary equation to convert the refractive index to STP which may influence the observed variance.11 The variance observed in the commercial RID may be different from RID assays produced by other manufacturers. However, variance has been noted in other types of RID that may indicate an inherent imprecision in the test method.14–16 Also, caution should be used when relying upon a single measurement to determine the absorption of maternal immunity. Generally, the absorption of maternal immunity is evaluated once either using STP or RID. However, both STP values and IgG concentrations produce a variance. This is especially an issue when RID is used to quantify maternal antibodies in calves with a high IgG concentration and may contribute to misclassification of a calf’s transfer of maternal antibody status. Lastly, the present study could not explore whether dilution or changing the standard curve contributed to the large variance observed in the serum with high IgG concentrations. Previous studies have indicated that plate and date minimally contribute to observed variance, however, this could not be evaluated in the present study.10
Serum total protein produced different values when repeatedly measured using an optical refractometer. Refractometers use Snell’s law to measure the refraction of light as it passes between 2 media and reports a refractive index of the solution.17 It is difficult to interpret the meaning of a refractive index; therefore, to impart clinical relevance, a refractive index is converted into a meaningful value such as STP. Others have reported little variability in refractive index, however, we found that repeated measurements of STP produce a range of variances.18 Therefore, the observed imprecision may be due to the conversion of the refractive index to STP.
We found that IgG concentrations measured using RID produced greater variance for a given serum than STP values measured using an optical refractometer. Serum in the low-serum IgG stratum had a repeatability coefficient similar to that of the standards. Serum in the moderate- and high-serum IgG stratum had the greatest variance in IgG concentration. Serum with a precipitin ring diameter greater than the standard precipitin ring were diluted and re-evaluated which can contribute to the observed variance.9,10 Sera that require dilution are routinely evaluated on a different plate and day with a different standard curve as it is generally unknown which sera will require further dilutions. In our study, the main contributor to variance between dilution and changing standard curve was difficult to determine as the 1:1 and 1:3 dilutions were performed on separate days and plates and the IgG concentrations were quantified with separate standard curves. However, a previous study found that plate and date contributed minimally to the observed variance in RID assays.10 Regardless, proposed IgG threshold concentrations for quantifying adequate absorption of maternal immunity are between 5 and 24 g/L which are values that are within the range of the IgG standard provided with the RID assays.19,20 Serum that did not require dilution still produced a greater CV than STP evaluated in the same serum. A risk of misclassification exists for neonatal calves with IgG concentration near proposed threshold values for failed transfer of passive immunity. For sera that require dilution, it may be appropriate to state that the calf has adequate absorption of maternal immunity without quantifying an exact IgG concentration due to the imprecision potentially associated with dilution. Conversely, refractometers do not require serum to be diluted and are read on a standardized scale, limiting the influence of dilution and standard curve on the observed variance.
Radial immunodiffusion is sometimes used as the reference test to monitor the absorption of maternal antibodies in neonatal calves and is used to validate other diagnostic tests to monitor the absorption of maternal antibodies in neonatal calves. Therefore, imprecision in IgG concentrations measured with RID could lead to inaccuracies in the concentration of maternal antibodies or errors in the validation of other tests used to monitor the absorption of maternal antibodies in neonatal calves. Moreover, threshold values for classifying absorption of maternal antibodies are often extrapolated from RID to other diagnostic tests, such as STP. The imprecision in the reference test can translate into inaccurate threshold values for the indirect test. The variance of IgG concentrations measured using RID may contribute to the range of reported STP values, 48 g/L to 55 g/L, found within the literature that corresponds with an IgG concentration of 10 g/L.2,21,22 The variability produced by RID and optical refractometry makes it difficult to accurately determine the IgG concentrations of neonatal calves.
Previously, STP quantified with optical refractometers and IgG measured by RID have been associated with disease in pre-weaned calves.20,23–25 However, repeated measures of serum samples to estimate STP values measured with an optical refractometer and IgG concentrations measured with RID produced variable results. Estimating IgG concentration directly by RID or indirectly by STP lacks the precision that might affect the diagnostic interpretation of these tests regarding a calf’s absorption of maternal antibodies.
Acknowledgments
The authors would like to extend gratitude to Dr. Hannah Urig for being the masked reader for the optical refractometer. This study is a contribution of the Beef Cattle Population Health and Reproduction Program at Mississippi State University.
Disclosures
The authors declared no potential conflicts of interest concerning the research, authorship, and/or publication of this article.
Funding
This project was supported by the Mikell and Mary Cheek Hall Davis Endowment for Beef Cattle Health and Reproduction and the Mississippi State University College of Veterinary Medicine House Officer Grant Program.
References
- 1.↑
Stelwagen K, Carpenter E, Haigh B, Hodgkinson A, Wheeler TT. Immune components of bovine colostrum and milk. J Anim Sci. 2009;87(Suppl. 13):3–9. doi:10.2527/jas.2008-1377
- 2.↑
Buczinski S, Gicquel E, Fecteau G, Takwoingi Y, Chigerwe M, Vandeweerd JM. Systematic review and meta-analysis of diagnostic accuracy of serum refractometry and brix refractometry for the diagnosis of inadequate transfer of passive immunity in calves. J Vet Intern Med. 2018;32(1):474–483. doi:10.1111/jvim.14893
- 3.↑
Klaus GGB, Bennett A, Jones EW. A quantitative study of the transfer of colostral immunoglobulins to the newborn calf. J Immunol. 1969;16:293–299.
- 4.↑
McBeath DG, Penhale WJ, Logan EF. An examination of the influence of husbandry on the plasma immunoglobulin level of the newborn calf, using a rapid refractometer test for assessing immunoglobulin content. Vet Rec. 1971;88:266–270. doi:10.1136/vr.88.11.266
- 5.↑
Mohri M, Sharifi K, Eidi S. Hematology and serum biochemistry of Holstein dairy calves: age related changes and comparison with blood composition in adults. Res Vet Sci. 2007;83(1):30–39. doi:10.1016/j.rvsc.2006.10.017
- 6.↑
Butler JE. Bovine immunoglobulins: an augmented review. Vet Immunol Immunopathol. 1983;4:43–152. doi:10.1016/0165-2427(83)90056-9
- 7.↑
Marc S, Kirovski D, Mircu C, et al. Serum protein electrophoretic pattern in neonatal calves treated with clinoptilolite. Mol J Synth Chem Nat Prod Chem. 2018;23(6):1278. doi:10.3390/molecules23061278
- 8.↑
Thompson A, Smith DR. Failed transfer of passive immunity is a component cause of pre-weaning disease in beef and dairy calves: a systematic review and meta-analysis. Bov Pract. 2022;56(2):47–61. doi:10.21423/bovine-vol56no2p47-61
- 9.↑
Denholm K, Haggerty A, Mason C, Ellis K. Comparison of testing for failure of passive transfer in calf serum using four different testing methods. Vet J. 2022;281:105812. doi:10.1016/j.tvjl.2022.105812
- 10.↑
Thompson AC, Wills RW, Smith DR. Sources of variance in the results of a commercial bovine immunoglobulin G radial immunodiffusion assay. J Vet Diagn Invest. 2023;35(1):34–41. doi:10.1177/10406387221140047
- 11.↑
Vandeputte S, Detilleux J, Rollin F. Comparison of four refractometers for the investigation of the passive transfer in beef calves. J Vet Intern Med. 2011;25(6):1465–1469. doi:10.1111/j.1939-1676.2011.00816.x
- 12.↑
Faul F, Erdfelder E, Buchner A, Lang AG. Statistical power analyses using G*Power 3.1: tests for correlation and regression analyses. Behav Res Methods. 2009;41(4):1149–1160. doi:10.3758/BRM.41.4.1149
- 13.↑
Vaz S, Falkmer T, Passmore AE, Parsons R, Andreou P. The case for using the repeatability coefficient when calculating test–retest reliability. PLoS ONE. 2013;8(9):e73990. doi:10.1371/journal.pone.0073990
- 14.↑
Kalff MW. Quantitative determination of serum immunoglobulin levels by single radial immunodiffusion. Clin Biochem. 1970;3:91–104. doi:10.1016/S0009-9120(70)80011-X
- 15.
Mancini G, Carbonara AO, Heremans JF. Immunochemical quantitation of antigens by single radial immunodiffusion. Immunochemistry. 1965;2(3):235–254. doi:10.1016/0019-2791(65)90004-2
- 16.↑
Vergani C, Stabilini R, Acostoni A. Quantitative determination of serum immunoglobulins by single radial immunodiffusion on cellulose acetate. Immunochemistry. 1967;4:233–237. doi:10.1016/0019-2791(67)90184-X
- 17.↑
George JW. The usefulness and limitations of hand-held refractometers in veterinary laboratory medicine: an historical and technical review. Vet Clin Pathol. 2001;30(4):201–210. doi:10.1111/j.1939-165X.2001.tb00432.x
- 18.↑
Sutton RH. The refractometric determination of the total protein concentration in some animal plasmas. N Z Vet J. 1976;24(7):141–148. doi:10.1080/00480169.1976.34304
- 19.↑
Dewell RD, Hungerford LL, Keen JE, et al. Association of neonatal serum immunoglobulin G1 concentration with health and performance in beef calves. J Am Vet Med Assoc. 2006;228(6):914–921. doi:10.2460/javma.228.6.914
- 20.↑
Rea DE, Tyler JW, Hancock DD, et al. Prediction of calf mortality by use of tests for passive transfer of colostral immunoglobulin. J Am Vet Med Assoc. 1996;208(12):2047–2049.
- 21.↑
Calloway CD, Tyler JW, Tessman RK, Hostetler D, Holle J. Comparison of refractometers and test endpoints in the measurement of serum protein concentration to assess passive transfer status in calves. J Am Vet Med Assoc. 2002;221(11):1605–1608. doi:10.2460/javma.2002.221.1605
- 22.↑
Elsohaby I, Mcclure JT, Keefe GP. Evaluation of digital and optical refractometers for assessing failure of transfer of passive immunity in dairy calves. J Vet Intern Med. 2015;29(2):721–726. doi:10.1111/jvim.12560
- 23.↑
Donovan GA, Dohoo IR, Montgomery DM, Bennett FL. Associations between passive immunity and morbidity and mortality in dairy heifers in Florida, USA. Prev Vet Med. 1998;34(1):31–46. doi:10.1016/S0167-5877(97)00060-3
- 24.
Todd CG, McGee M, Tiernan K, et al. An observational study on passive immunity in Irish suckler beef and dairy calves: tests for failure of passive transfer of immunity and associations with health and performance. Prev Vet Med. 2018;159:182–195. doi:10.1016/j.prevetmed.2018.07.014
- 25.↑
Windeyer MC, Leslie KE, Godden SM, Hodgins DC, Lissemore KD, LeBlanc SJ. The effects of viral vaccination of dairy heifer calves on the incidence of respiratory disease, mortality, and growth. J Dairy Sci. 2012;95(11):6731–6739. doi:10.3168/jds.2012-5828
