Per-Protocol and Pre-Defined population analysis of the LINC study

Objective To perform two predefined sub-group analyses within the LINC study and evaluate if the results were supportive of the previous reported intention to treat (ITT) analysis. Methods Predefined subgroup analyses from the previously published LINC study were performed. The Per-Protocol population (PPP) included the randomized patients included in the ITT-population but excluding those with violated inclusion or exclusion criteria and those that did not get the actual treatment to which the patient was randomized. In the Pre-Defined population (PDP) analyses patients were also excluded if the dispatch time to ambulance arrival at the address exceeded 12min, there was a non-witnessed cardiac arrest, or if it was not possible to determine whether the arrest was witnessed or not, and those cases where LUCAS was not brought to the scene at the first instance. Results After exclusion from the 2589 patients within the ITT-population, the Per-Protocol analysis was performed in 2370 patients and the Pre-Defined analysis within 1133 patients. There was no significant difference in 4-h survival of patients between the mechanical-CPR group and the manual-CPR group in the Per-Protocol population; 279 of 1172 patients (23.8%) versus 281 of 1198 patients (23.5%) (risk difference −0.35%, 95% C.I. −3.1 to 3.8, p =0.85) or in the Pre-Defined population; 176 of 567 patients (31.0%) versus 192 of 566 patients (33.9%) (risk difference −2.88%, 95% C.I. −8.3 to 2.6, p =0.31). There was no difference in any of the second outcome variables analyzed in the Pre-Protocol or Pre-Defined populations. Conclusions The results from these predefined sub-group analyses of the LINC study population did not show any difference in 4h survival or in secondary outcome variables between patients treated with mechanical-CPR or manual-CPR. This is consistent with the previously published ITT analysis.