Response to Newman et al. (March 2006): factitious increase in thyrotropin in a neonate caused by a maternally transmitted interfering substance.

with CAD or MyD, cTnI was detectable in 69.2% of patients and undetectable in the remainder. In the control group (Con), cTnI was detectable in 26.2% and undetectable in 73.8% of the patients. Of the detectable values, 87.5% were 0.07 g/L (the 99th percentile value) in the CAD and MyD groups; the remainder were between 0.07 and 0.1 g/L (8.3%) or 0.1 g/L (4.2%). In the control group, all detectable values were 0.07 g/L. The ROC curve analysis suggested that a cTnI cutoff concentration of 0.02 g/L best distinguished persons with ischemic (CAD) or nonischemic (MyD) myocardial impairment (area under the curve, 0.74; 95% confidence interval, 0.66–0.83) from those with milder disease. Regression analysis showed that the only variable that correlated with cTnI was left ventricular ejection fraction (r 0.097; r 0.311; P 0.001). In all 12 patients with an ejection fraction 50%, cTnI concentrations were detectable [median (75th–25th percentiles), 0.045 (0.0275–0.06) g/L]. Our data suggest a difference in cTnI concentrations between groups of cardiovascular outpatients with more severe disease compared with less severely affected patients. These differences are all below the 99th percentile of a reference interval. The mechanisms responsible for measurable cTnI values of this type are unclear. The group data are intriguing and suggest that when better assays permit evaluation of low values, there appears to be additional information that can be used to evaluate patients.