Inflammatory burden index: associations between osteoarthritis and all-cause mortality among individuals with osteoarthritis

Abstract

Background: The newly described inflammatory burden index (IBI) reflects a patient's inflammatory burden. This study aimed to estimate the association between IBI, osteoarthritis (OA), and all-cause mortality in patients with OA.

Methods: We extracted the data of adults from the National Health and Nutrition Examination Survey database between 1999 and 2018. After using appropriate survey weights to correct for sample bias, we conducted multivariate logistic regression analyses to explore the association between IBI and OA across three models: in the unadjusted model, partially adjusted model (adjusting age, sex, race, education level, marital status, PIR, BMI, smoking status, drinking status, stroke, CVD, DM, and hypertension) and fully adjusted model (which included additional variables: HBA1C, ALT, AST, BUN, TC, and HDL). And the odds ratios (OR) and 95% confidence intervals (CI) were calculated. Similarly, using comparable survey weights and covariates adjustments, we employed Cox proportional hazards regression analysis to investigate the association between IBI and all-cause mortality in the other 3 models. The Cox proportional hazards regression models were fitted to calculate the hazard ratios (HR) and 95% CI of the association between IBI and all-cause mortality. A restricted cubic spline (RCS) was used to explore the nonlinear relationships between association effects. Subgroup analysis was performed to validate the reliability of their effects.

Results: In total, 22,343 eligible participants were included. Multiple logistic regression models revealed that participants with the highest IBI had 2.54 times (95%CI, 2.23, 2.90)) higher risk of OA than those with the lowest IBI in Model 1, whereas the OR was 1.21 (95%CI, 1.03, 1.42) in Model 2 and 1.23 (95%CI,1.05, 1.45) in Model 3. Multiple Cox regression models showed participants with the highest IBI had 186% (95%CI, 1.50, 2.31) times risk of developing all-cause death than those with the lowest IBI in Model 1. This trend remained stable in Models 2 (HR,1.54; 95%CI,1.22, 1.95) and 3 (HR, 1.41; 95%CI, 1.10, 1.80). The RCS revealed a significant positive association between IBI and OA risk. With respect to the association between IBI and all-cause mortality, a slight decrease in mortality was observed from the lowest quartile to the second quartile of IBI, and the mortality risk increased with increasing IBI. Subgroup analyses showed that age, cardiovascular disease, and hypertension were pivotal in the association of IBI with all-cause mortality, whereas the association of IBI with OA remained stable after stratification by other factors such as sex, race, education level, marital, smoking, and drinking status, hypertension, and most serological indices.

Conclusions: This study provides evidence of a positive association between IBI, OA, and all-cause mortality. IBI may be a promising signature for assessing the inflammatory burden in patients with OA, which, in turn, is conducive to precise references for high-risk population recognition, anti-inflammatory guidance, and reducing mortality intervention.

Keywords: All-cause mortality; CDC; Inflammatory burden index; NHANES; Osteoarthritis; Positive association.

Fig. 1

Flowchart depicting the subject selection process for this study

Fig. 2

DAG of IBI and OA with Detailed Factors. This DAG illustrated the potential causal relationships and confounding factors associated with the IBI and OA. Each node represents a variable that may influence IBI and OA. Arrows indicate the direction of influence or causal relationship between variables. DAG, Directed Acyclic Graph; IBI, Inflammatory Burden Index; OA, Osteoarthritis; EDUCATION, Education Level; FAMILY PIR, Family Income to Poverty Ratio; BMI, Body Mass Index; CVD, Cardiovascular disease; SMOKE, Smoking; DM, Diabetes Mellitus; HBA1C, Hemoglobin A1c; ALT, Alanine Aminotransferase; AST, Aspartate Aminotransferase; BUN, Blood Urea Nitrogen; TC, Total Cholesterol; HDL, High-Density Lipoprotein. The variable, IBI, is the primary exposure variable. The variables, OA and MORTALITY, are the outcome

Fig. 3

Forest plots from multiple imputations (10 times). (A) Forest plot of odds ratios (OR). (B) Forest plot of hazard ratios (HR). CI, confidence interval. ^*P < 0.05

Fig. 4

Restricted cubic spline plot illustrating the non-linear association between continuous inflammatory burden index (IBI) and the risk of osteoarthritis (OA) in the entire study population. The solid line represents the hazard ratio, while the dotted line denotes the 95% confidence interval. Fig. S1 shows the risk for OA according to IBI quartiles in the entire study population

Fig. 5

Restricted cubic spline plot demonstrating the non-linear relationship between continuous IBI and the risk of all-cause mortality among individuals with OA. The solid line signifies the hazard ratio, and the dotted line indicates the 95% CI. Fig. S2 shows the risk for all-cause mortality according to IBI quartiles in the OA population

Fig. 6

Forest plot of subgroup analyses exploring the association between IBI and OA risk in the entire study population. The 95% CI are presented for each subgroup

Fig. 7

Forest plot of subgroup analyses investigating the relationship between IBI and all-cause mortality risk in the OA population. The 95% CI are displayed for each subgroup