How small modifications in laboratory workflow of blood cultures can have a significant impact on time to results.

Rapid identification and antimicrobial susceptibility testing of micro-organisms causing bloodstream infections is crucial in the management of septic patients. In this study, we compared a period of twice-daily and a period of thrice-daily reading of subculture agar plates. In 2016, 10,644 positive blood cultures bottles (bioMérieux) from 2608 patients were analyzed at UZ Leuven. Identification and antimicrobial susceptibility testing were performed by MALDI-TOF MS (Bruker Daltonics) and Vitek 2 (bioMérieux) respectively. In period 1 (January to June), subculture plates were read at 8:30 a.m. and 2:00 p.m. during the weekdays. In period 2 (August until December), reading was performed at 8:30 a.m., 2:00 p.m., and 5:00 p.m. Time to identification and time to AST results after positivity were compared. In period 1, median time to identification of all organisms was 22.8 h compared to 20.2 h in period 2 (p < 0.01). Moreover, micro-organisms were identified before 12 h in 9% of samples in period 2, a significant increase compared to 2% in period 1 (p < 0.01). In period 2, AST results were known within 36 h in 39% of samples, compared to 31% in period 1 (p < 0.01). Optimization of the reading frequency of subcultures of blood cultures significantly decreases time to results. Further optimization can be done by introducing lab automation. We will use the data of this study as a baseline to analyze the impact of introducing WASPLab (Copan Diagnostics) automation on time to results.

On the use of historical control data in pre-clinical safety studies.

A number of methods to formally incorporate historical control information in pre-clinical safety evaluation studies have been proposed in literature. However, it remains unclear when one should use historical data. Focusing on the logistic-normal model, we investigate situations where historical studies may prove to be useful. Aspects of estimation (precision and bias) and testing (power) for treatment effect are investigated under different conditions such as the number of historical control studies, the degree of homogeneity amongst them, the level of treatment effect and different control rates. The possibility to use a selected subset of historical control studies is also explored.

Further prospective evidence of a circadian variation in the frequency of call for sudden cardiac death. Belgian Cardiopulmonary Cerebral Resuscitation Study Group.

OBJECTIVE: To determine whether in a larger data base call for sudden cardiac death exhibits a specific circadian rhythm similar to that recently demonstrated by Levine et al. DESIGN AND SETTING: The time of the day of calls received for out-of-hospital cardiac arrests (OOHCA) prospectively registered between 1983 and '90 by 7 major Belgian pre-hospital EMS-MICU services. Chrono-biologic assessment was made by two-harmonic linear regression analysis of the data tabulated by hour of the day. The hourly distribution of calls for OOHCAs was subjected to Fourier transformation resulting in a periodogram. PATIENTS: 3471 OOHCAs with presumed cardiac etiology and age of more than 18 years versus 2007 inpatients registered in the same period. MEASUREMENTS AND RESULTS: Significant and remarkably similar circadian patterns were found (R-square = 0.84) for the cardiac origin OOHCAs and the ventricular fibrillation OOHCAs. There is a low incidence during the night and an increased incidence from 6 a.m. until noon with an additional early afternoon-peak. The data were always better fitted when applying sinusoids with periods of 8 and 24 h instead of 12 and 24 h. Our observed circadian distribution resembles the reported circadian variation of ischaemic episodes, ventricular tachycardia and acute myocardial infarction in the awake hours. The time distribution of OOHCA (cardiac origin) differs significantly from OOHCA (non-cardiac origin) and from in-hospital cardiac arrests. The in-hospital CA pattern shows less deviation. The age dependent variation in the incidence of cardiac origin OOHCAs, was not obvious for the ventricular fibrillation subgroup. CONCLUSION: Knowledge about the cyclical nature of incidence of cardiac arrests is useful to improve intersystem comparisons and make sound decisions about prophylaxis, treatment and allocation of resources.

Does age have an effect on the time of occurrence of cardiac arrest of presumed cardiac etiology? Belgian Cardiopulmonary-Cerebral Resuscitation Study Group.

To assess possible age-related differences in the time of occurrence (hour of the day, day of the week, month of the year) of cardiopulmonary collapses of presumed cardiac etiology, 3,305 out-of-hospital patients registered by the Belgian Cardiopulmonary-Cerebral Resuscitation Study Group between 1983 and 1990 were studied. Occurrence of call-time was separately tabulated for patients less than versus above 60 years of age and tested for differences with the Kolmogorov-Smirnov two-sample test. Only the circadian patterns of all out-of-hospital cardiac arrests of presumed cardiac etiology showed a significant difference between the two age-groups; the prominent morning peak (6 a.m.-10 a.m.) in incidence was less pronounced in the younger age-group. The overall incidence is lower in the summer and during mid-week. These trends are similar for both age-groups.