Citations

BACKGROUND A rise or fall of cardiac troponin is a prerequisite for the diagnosis of acute myocardial infarction. Defining significant changes requires knowledge of both biological and analytical variation. The short-term biological variation of cardiac troponin in healthy individuals is 3%–48%. However, healthy individuals may not be representative for patients in whom cardiac troponin measurement is often of clinical importance. Therefore, we studied the individual variation of cardiac troponin in patients with symptoms of stable coronary artery disease. METHODS Twenty-four patients scheduled for elective coronary angiography were included. Blood samples were drawn once at enrollment and serially at six 4-h intervals on the day before coronary angiography. Cardiac troponin was measured with hs-cTn assays from Abbott Laboratories (premarket cTnI assay) and Roche Diagnostics (Elecsys® cTnT assay with two different lots). RESULTS The short-term individual variation in cardiac troponin I (cTnI) was 14%, the reference change value (RCV) 49%, and RCV-log-normal (rise/fall) 54%/−35%. The corresponding values for cTnT were 7%, 23%, and 26%/−21%. The long-term variation for cTnI was 24%, RCV 69%, and RCV-log-normal (rise/fall) 97%/−49%. The corresponding values for cTnT were 11%, 32%, and 37%/−27%. CONCLUSIONS The short-term individual variation of cardiac troponin in patients with symptoms of stable coronary artery disease is similar to the biological variation previously demonstrated in healthy individuals. Our results suggest that a change in cardiac troponin concentrations of >50% can be used in attempting to diagnose acute myocardial injury. To detect significant long-term changes in cardiac troponin concentrations, larger changes will be required.

BACKGROUND There is little information available on cardiac troponin concentrations in healthy young children. METHODS Using a precommercial high-sensitivity assay from Abbott Diagnostics, we measured cardiac troponin I (cTnI) in longitudinal blood samples collected at ages 8, 10, and 12 years from a cohort of healthy, community-dwelling children. The 99th percentile values were calculated and estimates of the long-term biological variation were made. RESULTS cTnI concentrations were above the limit of detection in 87%, 90%, and 98% of the children at ages 8, 10, and 12 years. The 99th percentiles were lower compared to a healthy adult population in both male and female children at all ages studied. At the 3 periods of study assessment, different children had cTnI concentrations above the 99th percentile. The calculated 99th percentile varied markedly depending upon whether the lowest or highest cTnI measurement for an individual child was included in the calculation. Biological variation varied markedly between 0% and 136%, the index of individuality was low at 0.36, and the reference change value was an increase of 147% or a decrease of 59%. CONCLUSIONS In this longitudinal study of cTnI concentrations in healthy children as determined by a high-sensitivity assay, different children had concentrations of cTnI above the 99th percentile at the 3 episodes of assessment. These results suggest that in children the 99th percentile may not be a reliable index of silent cardiac disease, but rather may be indicating low-grade intercurrent illness.

In a 1964 study of the pharmacokinetic determinants of penicillin cure of gonococcal urethritis, 45 male prisoner volunteers were experimentally infected with strains of Neisseria gonorrhoeae having known in vitro penicillin susceptibility. After developing urethritis, subjects received intramuscular penicillin G and had serum samples obtained serially to determine penicillin concentration. Using a multiple regression technique, we studied patient-associated parameters and parameters of the serum penicillin curves to determine the best predictors of treatment results. Cure was best predicted by the time the serum penicillin concentration remained above three to four times the penicillin minimum inhibitory concentration of the infecting strain (probability of correct classification, >0.80). Those cured had serum penicillin concentrations which remained in this range for means of 7 to 10 h. Our findings confirm principles of antimicrobial therapy derived from animal models and may have application in studying therapy of gonorrhea and other infectious diseases.