Abstract
Dear Editor,
We thank Drs. Aydın et al 1 for their interest in our study 2 and for raising several methodological points that merit a response.
Regarding the potential confounding effect of age within the fibrosis-4 (FIB-4) formula, this concern is important and was addressed through several pre-specified analyses. Variance inflation factor (VIF) assessment demonstrated that while age itself showed high internal collinearity (VIF = 27.28), reflecting its mathematical incorporation into FIB-4, the index itself maintained acceptable collinearity (VIF = 2.79), and the direct correlation between age and FIB-4 was only modest (r = .115, P = .012). A sensitivity analysis excluding age from the multivariable Cox model confirmed that FIB-4 retained independent prognostic significance (hazard ratio [HR] 1.45, 95% confidence interval [CI]: 1.30-1.62, P < .001). Prior cardiovascular studies have similarly demonstrated that FIB-4 retains independent prognostic value after adjustment for age and established risk factors,3,4 and a large-scale study involving >80 000 patients with non-alcoholic fatty liver disease (NAFLD) confirmed FIB-4 as the strongest independent predictor of major adverse cardiovascular events (MACE) after comprehensive multivariable adjustment. 5
Concerning NAFLD diagnostic methodology, we acknowledge that magnetic resonance imaging (MRI)-based hepatic fat quantification represents the reference standard. However, our study 2 employed a dual-modality approach combining the hepatic steatosis index (HSI) with blinded abdominal ultrasonography graded by experienced radiologists using a validated 0-3 scale. Ultrasonography is recommended as the first-line imaging method for hepatic steatosis screening by current clinical practice guidelines and demonstrates high diagnostic accuracy for moderate-to-severe steatosis, which constitutes the clinically relevant category for cardiovascular risk stratification. 6 Patients with NAFLD confirmed by both modalities showed 21.5% MACE incidence versus 8.3% in those without NAFLD by either method (adjusted HR 3.15, 95% CI: 1.52-6.53, P = .002), supporting the robustness of this dual approach. Routine MRI evaluation in 472 consecutive myocardial infarction with non-obstructive coronary arteries (MINOCA) patients during acute hospitalization is not feasible in practice; the combination of HSI and ultrasonography therefore represents a pragmatic and reproducible clinical alternative.
With respect to MINOCA etiological subclassification, we agree that the pathophysiological heterogeneity of MINOCA—encompassing coronary vasospasm, microvascular dysfunction, spontaneous coronary artery dissection, and thromboembolism—poses an inherent challenge for any single biomarker study. Coronary microvascular dysfunction (CMD) has been identified as a hallmark across all MINOCA subtypes, 7 and the mechanistic pathways linking FIB-4 to adverse outcomes—systemic inflammation, insulin resistance, oxidative stress, and endothelial dysfunction—have been consistently implicated in MINOCA pathophysiology across etiological subtypes. 8
The commentators’ suggestion 1 that FIB-4 functions best as part of a composite rather than as a standalone tool is fully aligned with our conclusions. While FIB-4 alone achieved an area under the curve (AUC) of 0.710, when incorporated into a comprehensive integrated model alongside conventional clinical parameters—including age, ejection fraction, renal function, cardiac troponin, and guideline-directed therapy—discrimination improved substantially to AUC 0.834, with net reclassification improvement (NRI) of 23.3% (P = .002) and integrated discrimination improvement (IDI) of 4.8% (P = .006). A recently published independent cohort of 321 MINOCA patients followed for a median of 26.6 months similarly confirmed FIB-4 as an independent MACE predictor, 9 lending further external support to our findings. Systematic investigation of whether combining FIB-4 with novel inflammatory indices such as the pan-immune-inflammation value (PIV) or systemic immune-inflammation index (SII) provides additional reclassification benefit in MINOCA populations is an important direction for future research.
