Investigators collected serum at admission and then at 1 and 3 hours to profile early circulating GDF-15 in patients presenting with NSTE-ACS. The study team reported a modest upward trajectory in GDF-15 over the initial 3-hour window. Short, early sampling captured a dynamic not apparent from a single admission value alone. The authors measured two other prognostic biomarkers—NT-proBNP and hs-cTnT—concurrently, permitting head-to-head temporal comparisons within the acute phase.
Data indicated that the 1-hour change in GDF-15 correlated with changes in hs-cTnT measured over the same interval. The authors used Spearman rank-correlation coefficients to assess pairwise associations and reported that contemporaneous troponin shifts aligned with early GDF-15 increases. Short sentences: correlation does not prove causation. Longer sentence: the team also applied linear regression models to identify variables associated with admission GDF-15 levels and with their temporal changes, although the abstract did not list the full set of covariates or model specifications.
The clinical outcome analysis focused on recurrent ischemic events over a 1-year follow-up. Patients whose GDF-15 rose at 1 hour compared with admission experienced a higher rate of recurrent ischemia during that year—reported rates were 1.9% versus 17.5% (P = 0.019). The authors then entered the 1-hour GDF-15 change into Cox proportional hazards models and found it remained an independent predictor after adjustment for established clinical variables and biomarkers; the adjusted hazard ratio reported was 1.71 (1.28 to 228.75), P = 0.03. Short, direct statement: these figures are reported in the preprint.
To evaluate whether serial GDF-15 adds to existing risk stratification, the study team calculated areas under the receiver operating characteristic curve for competing models. They reported that serial measurements provided additional prognostic information beyond traditional risk factors and the other biomarkers measured. The abstract indicates AUC comparisons were used to assess incremental value, but it did not provide the specific AUC values or the exact model compositions in the abstract text; readers should consult the full preprint for detailed performance metrics.
The authors described serial serum sampling at admission, 1 hour, and 3 hours. They measured GDF-15, NT-proBNP, and hs-cTnT; however, the abstract did not specify assay brands, limits of detection, nor laboratory processing details. For association testing the investigators used Spearman correlations and linear regression to link clinical and biomarker variables to GDF-15 levels and their short-term changes. For event prediction they applied Cox proportional hazards models and compared model discrimination via AUC. The abstract did not report sample size, inclusion and exclusion criteria, timing of follow-up visits beyond the 1-year event endpoint, or adjustment covariates in full; those methodological specifics are reported in the full text of the preprint.
The study protocol received approval from the Ethics Committee of Beijing Anzhen Hospital, Capital Medical University (2021-072), and the authors declared adherence to relevant ethical guidelines and consenting procedures. Funding support was acknowledged from the National Key Research and Development Program (2022YFC2009602). The authors stated that the data supporting their findings are available from the corresponding author upon reasonable request.
Short, candid caveat: this article is a medRxiv preprint and has not been peer-reviewed. The abstract reports statistically significant associations and model outputs, including an adjusted hazard ratio with a wide confidence interval upper bound; these estimates should be interpreted cautiously until peer review and full methodological detail are available. The abstract omitted assay and full model specifications and did not provide AUC point estimates in the summary, leaving unanswered questions about effect robustness, calibration, and net clinical reclassification. These items remain to be clarified by the authors in the complete manuscript or in subsequent peer-reviewed publication.