Bayesian estimation of the receiver operating characteristic curve for a diagnostic test with a limit of detection in the absence of a gold standard.

The receiver operating characteristic (ROC) curve is commonly used for evaluating the discriminatory ability of a biomarker. Measurements for a diagnostic test may be subject to an analytic limit of detection leading to immeasurable or unreportable test results. Ignoring the scores that are beyond the limit of detection of a test leads to a biased assessment of its discriminatory ability, as reflected by indices such as the associated area under the curve (AUC). We propose a Bayesian approach for the estimation of the ROC curve and its AUC for a test with a limit of detection in the absence of gold standard based on assumptions of normally and gamma-distributed data. The methods are evaluated in simulation studies, and a truncated gamma model with a point mass is used to evaluate quantitative real-time polymerase chain reaction data for bovine Johne's disease (paratuberculosis). Simulations indicated that estimates of diagnostic accuracy and AUC were good even for relatively small sample sizes (n=200). Exceptions were when there was a high per cent of unquantifiable results (60 per cent) or when AUC was < or =0.6, which indicated a marked overlap between the outcomes in infected and non-infected populations.

Acute phase proteins in the diagnosis of bovine subclinical mastitis.

BACKGROUND: The California mastitis test (CMT) and somatic cell count (SCC) are commonly used for diagnosis of subclinical mastitis in cattle. Acute phase proteins (APPs), as alternative biomarkers of mastitis, may increase in concentration in the absence of macroscopic changes in the milk, or may precede the onset of clinical signs. OBJECTIVE: The aim of this study was to compare the accuracy of APPs measured in milk and in serum with bacterial culture for the diagnosis of bovine subclinical mastitis. METHODS: One hundred and seventy-five Holstein cows were randomly selected from 7 dairy farms. Quarter milk and serum samples were taken from all cows. Milk samples were analyzed using a CMT and SCC, and for haptoglobin (MHp) and amyloid A (MAA) concentrations, and were also submitted for bacterial culture. Serum samples obtained concurrently were analyzed for haptoglobin (SHp) and amyloid A (SAA). Two-sample Wilcoxon (Mann-Whitney) test was used to compare SCC, MAA, MHp, SAA, and SHp concentrations between culture-positive and culture-negative animals. Receiver-operating characteristic analysis was used to assess the performance of each test using bacterial culture as the reference method. RESULTS: MAA concentration was the most accurate of the 5 tests, with a sensitivity of 90.6% and specificity of 98.3% at concentrations >16.4 mg/L. MAA and MHp had significantly larger areas under the curve than the respective serum proteins, SAA and SHp. CONCLUSIONS: The results suggest that measuring haptoglobin and amyloid A in milk is more accurate than serum analysis for the diagnosis of subclinical mastitis in Holstein cows.

Determination of the time required for appropriate chemical de-epithelialization of an ileal segment for cystoplasty: an animal model.

UNLABELLED: Another group of authors from Iran attempted experimentally to determine the required time for the appropriate enzymatic treatment of the ideal segment to complete de-epithelialization, thus reducing its absorptive function. They found that 25 min of enzymatic treatment of the ideal segment was adequate for this, and that it was recommended from their experimental study for cystoplasty. OBJECTIVES: To determine the time required for the appropriate enzymatic treatment of an ileal segment to de-epithelialize its mucosa and to reduce its absorptive function for cystoplasty in dogs. MATERIALS AND METHODS: Twenty-one adult female Persian mixed-breed dogs were divided into seven equal groups: group 1 (negative control group) had no ileocystoplasty; group 2 (positive control group) had a routine ileocystoplasty with no enzymatic treatment of the ileal segment; and groups 3-7 had an ileocystoplasty with 5, 10, 15, 20 or 25 min, respectively, of enzymatic treatment of the ileal segment with collagenase and trypsin. The seven groups were then compared for haematological, biochemical and histological changes, and glucose reabsorption assessed using a glucose-absorption test. RESULTS: No dogs showed any signs of metabolic disturbances, biochemical and haematological changes. There were significant differences in blood glucose level (BGL) for the groups at different times after the glucose-absorption test, but a pair-wise comparison showed significant differences in BGL between group 1 and the other groups, except group 7. The degree of histopathological change was associated with the duration of enzymatic treatment, in that changes were more prominent in group 7. There was no shrinkage or collagen deposition. CONCLUSIONS: In these conditions, 25 min of enzymatic treatment of the ileal segment is sufficient to remove the absorptive function of the augmented bladder, and is recommended for cystoplasty in dogs.

The sensitivities and specificities of total plasma protein and plasma fibrinogen for the diagnosis of traumatic reticuloperitonitis in cattle.

We performed a prospective clinical study to select cut-off points for total plasma protein (TPP) and plasma fibrinogen (PF) to differentiate between traumatic reticuloperitonitis (TRP) and other gastrointestinal diseases with similar clinical signs, and to estimate the dependence and accuracy of TPP and PF when used in series or in parallel. TPP and PF were estimated preoperatively by refractometry in 93 cattle with TRP and 65 cattle with gastrointestinal diseases such as vagus indigestion (n = 16), liver abscesses (n = 17), omasal impaction (n = 19) and Johne's disease (n = 13). Three different cut-off points were selected by two-graph receiver-operating characteristic (TG-ROC) analysis for TPP and PF. Conditional covariances were calculated as a measure of dependence between sensitivities and specificities of TPP and PF. The cut-off points of 7.22, 7.78 and 8.82g/dl for TPP and 622, 691 and 766mg/dl for PF were suggested by TG-ROC based on different requirements of test performances. There was moderate negative dependence between sensitivities of TPP and PF at the 8.82g/dl and 766mg/dl cut-off points, and mild negative dependence between their specificities at the 7.78g/dl and 691mg/dl cut-off points, respectively. Acceptable accuracy (98 or 86% specificity with 62 or 88% sensitivity, respectively) was obtained with serial interpretation of the tests.