LabMed, Vol. 2, Pages 8: Evaluating Interleukin-6, Tumour Necrosis Factor Alpha, and Myeloperoxidase as Biomarkers in Severe Osteoarthritis Patients: A Biostatistical Perspective
LabMed doi: 10.3390/labmed2020008
Authors:
Laura Jane Coleman
John L. Byrne
Stuart Edwards
Rosemary O’Hara
Objective: This study employed advanced biostatistical methods to investigate Interleukin-6 (IL-6), Tumour Necrosis Factor Alpha (TNF-α), and Myeloperoxidase (MPO) levels in serum and plasma samples from patients with severe osteoarthritis (OA) compared to volunteers. The primary aim was to evaluate the diagnostic potential of these biomarkers and address statistical challenges, including non-normal data distribution and non-aged-matched groups. Design: Using Enzyme-Linked Immunosorbent Assays (ELISAs), IL-6, TNF-α, and MPO concentrations were analysed in 58 OA patients and 28 volunteers. Statistical analyses included Shapiro–Wilk tests to assess normality, a Mann–Whitney U (MWU) test to compare biomarker levels, and sensitivity analyses using Rank-based ANCOVA, and regression models were used to address non-normal data distributions and to validate the findings under adjustments for age and gender. Levene’s test was used to evaluate the homogeneity of variables. Results: Serum TNF-α and plasma MPO were significantly higher in OA patients than in volunteers (p < 0.05), while IL-6 levels were non-significant (p = 0.160). MWU tests confirmed significant differences for TNF-α (p = 0.045) and MPO (p = 0.0001). Sensitivity analysis using Rank-based ANCOVA and regression models confirmed the robustness of these biomarkers, with TNF-α (p = 0.037) and MPO (p = 0.0099) retaining statistical significance after adjusting for covariates. IL-6 remained non-significant across all analyses. Conclusions: TNF-α and MPO emerged as statistically robust biomarkers for severe OA, with the serum samples better reflecting inflammation than plasma. These findings underscore the importance of using advanced biostatistical methods such as Rank-based ANCOVA and regression to validate biomarkers, particularly in heterogenous datasets. Future research should incorporate larger, more diverse cohorts and detailed demographic profiling to explore the early diagnostic potential of these biomarkers and further understand OA progression.
Source link
Laura Jane Coleman www.mdpi.com
