Evaluation of Five Assays for the Diagnosis of Failure of Passive Transfer of Immunity in Foals

Most commercially available assays for diagnosis of failure of passive transfer of immunity (FPT) are appropriate for use as initial screening tests. However, many of the assays evaluated lack specificity, and their use as definitive tests would result in unnecessary therapy of foals with adequate immunoglobulin G (IgG) concentrations.

1. Introduction

Sepsis is the leading cause of morbidity and mortality in newborn foals. Several studies have documented a positive correlation between failure of passive transfer of immunity (FPT) and bacterial sepsis in foals [1-5]. FPT is typically defined as serum immunoglobulin G (IgG) concentrations of <400 mg/dl 18 - 24 h after birth, and partial FPT is defined as serum IgG concentrations between 400 and 800 mg/dl 18 - 24 h after birth [6]. The prevalence of FPT in foals has ranged from 3% to 20% [7,8]. The objective of this study was to assess the performance of five commercially available assays for the diagnosis of FPT and partial FPT in foals.

2. Materials and Methods

One hundred blood samples were collected from two sources. Blood samples were taken from foals presented to the University of Florida Veterinary Medical Center at the time of admission (n = 65) and from clinically normal foals at various times post-partum (n = 35). IgG concentration in serum was assessed using zinc sulfate turbidity (Equi Z [a]), glutaraldehyde coagulation (Gammacheck E [b]), semiquantitative immunoassays (Midland 4 and 8 Quick Test Kits [c]; Snap [d]), and a quantitative immunoassay (DVM Stat [e]). Two single radial immunodiffusion (RID) assays were used as reference methods [f,g]. Each assay was performed as described by the manufacturer. Samples were coded in such a way that the individual performing the assays was unaware of the source of the serum sample (ill versus healthy foal) and RID results. Sensitivity, specificity, and accuracy were calculated and compared between assays.

3. Results

The prevalence of FPT at serum concentrations of IgG < 400 mg/dl and IgG < 800 mg/dl was 27% and 42%, respectively. For the detection of IgG < 400 mg/dl, sensitivity of the DVM Stat test (100%) was not significantly different from that of the Midland 4, Equi Z, or Snap tests (88.9%). Specificity of the DVM Stat (96.0%) and Snap (95.8%) tests was significantly higher than that of the Equi Z (79.4%) and the Midland 4 (78.1%) tests. Similarly, accuracy of the DVM Stat (97.0%) and Snap (93.4%) tests was significantly higher than that of the Equi Z (82.0%) and the Midland 4 (81.0%) tests.

For the detection of IgG < 800 mg/dl, sensitivity of the DVM Stat (97.6%), the Gammacheck E (92.9%), the Equi Z (81.0%), and the Snap (81.0%) tests were significantly higher than that of Midland 8 test (52.4%). Specificity of the Midland 8 (100%), the Snap (94.7%), and the DVM Stat (82.8%) tests was significantly higher than that of the Equi Z (56.9%) and Gammacheck E (58.6%) tests. Accuracy of the DVM Stat (89.0%) and Snap (88.9%) tests was significantly higher than that of the Equi Z (67.0%) and Gammacheck E (73.0%) tests.

4. Discussion

When used for the purpose of screening, a test with a high sensitivity must be selected [9]. This ensures a high predictive value of a negative test and therefore, identification of the vast majority of foals with FPT. To confirm the disease, the samples with positive results should be retested with a different confirmatory test [9]. In this second test, a high specificity and positive predictive value are required. At a cut-off of 400 mg/dl, sensitivity was similar for each assay evaluated, resulting in negative predictive values in excess of 95%. As a result, all the assays evaluated would be suitable screening tests. However, specificity of the Equi Z or Midland 4 tests was significantly low, resulting in much lower positive predictive values. Positive results with the Midland 4 or Equi Z tests should not be interpreted as an indication for treatment, and results should be confirmed with a more specific assay.

At a cut-off of 800 mg/dl, sensitivity was greater with the DVM stat (97.6%) and Gammacheck E (92.9%) tests than with the Equi Z (81.0%) and Snap (81.0%) tests. Any of these tests would be adequate as a screening test. In contrast, the sensitivity of the Midland 8 kit was too low to justify its use as a screening test. The specificity of the DVM Stat (82.8%), Equi Z (56.9%), and Gammacheck E (58.6%) tests was significantly lower than that of the Snap (94.7%) or Midland 8 (100%) tests. As a result, a positive test with the DVM Stat, Gammacheck E, or Equi Z tests should not be interpreted as an indication for treatment, and results should be confirmed with a more specific test.

The authors thank the Florida Thoroughbred Breeders' and Owners' Association for their support of the equine research breeding herd as well as VMRD, Plasvacc USA, Midland BioProducts, Idexx Laboratories, and VDx for providing the assays.

Footnotes

[a] Equi Z Equine FPT Test Kit, VMRD Inc., Pullman, WA 99163.
[b] Gamma-Check-E, Plasvacc USA Inc., Templeton, CA 93465.
[c] Midland Plasma Foal IgG Quick Test Kits, Midland BioProducts Corporation, Boone, IA 50036.
[d] Snap Foal IgG, Idexx Laboratories, Westbrook, ME 04092.
[e] DVM Stat, VDx Inc., Belgium, WI 53004.
[f] Equine IgG RID Kit, VMRD Inc., Pullman, WA 99163.
[g] Equine IgG Test Kit, Kent Laboratories/Triple J Farms, Bellingham, WA 98226.