Wilkins Jeffrey, Ghosh Palash, Vivar Juan, Chakraborty Bibhas, Ghosh Sujoy
1Biomedical Biotechnology Research Institute, North Carolina Central University, 1801 Fayetteville Street, Durham, NC 27707 USA.
2Centre for Quantitative Medicine, Duke-NUS Medical School, 8 College Road, Singapore, 169857 Singapore.
BMC Obes. 2018 Aug 6;5:21. doi: 10.1186/s40608-018-0196-2. eCollection 2018.
Obesity is positively associated with low-level chronic inflammation, and negatively associated with several indices of health-related quality of life (HRQOL). It is however not clear if obesity-associated inflammation is partly responsible for the observed negative associations between obesity and HRQOL, and also whether systemic inflammation independently affects HRQOL. We conducted an exploratory analysis to investigate the relationships between obesity, systemic inflammation and indices of HRQOL, using NHANES survey data.
Data for the variables of interest were available for 6325 adults (aged 20-75 years, BMI > 18.5 kg/m). Demographic, body mass index (BMI), C-reactive protein (CRP), inflammatory disease status, medication use, smoking, and HRQOL data were obtained from NHANES (2005-2008) and analyzed using sampling-weighted generalized linear models. Data was subjected to multiple imputation in order to mitigate information loss from survey non-response. Both main effects and interaction effects were analyzed to evaluate possible mediation or moderation effects. Model robustness was ascertained via sensitivity analysis. Averaged results from the imputed datasets were reported in as odds ratios (OR) and confidence intervals (CI).
Obesity was positively associated with poor physical healthy days (OR: 1.59, 95% CI: 1.15-2.21) in unadjusted models. 'Elevated' and 'clinically raised' levels of the inflammation marker CRP were also positively associated with poor physical healthy days (OR = 1.61, 95% CI: 1.23-2.12, and OR = 2.45, 95% CI: 1.84-3.26, respectively); additionally, 'clinically raised' CRP was positively associated with mental unhealthy days (OR = 1.66, 95% CI: 1.26-2.19). The association between obesity and physical HRQOL was rendered non-significant in models including CRP. Association between 'elevated' and 'clinically raised' CRP and physical unhealthy days remained significant even after adjustment for obesity or inflammation-modulating covariates (OR = 1.36, 95% CI: 1.02-1.82, and OR = 1.75, 95% CI: 1.21-2.54, respectively).
Systemic inflammation appears to mediate the association between obesity and physical unhealthy days. Clinically raised inflammation is an independent determinant of physical and mental unhealthy days. Importantly, elevated (but sub-clinical) inflammation is also negatively associated with physical healthy days, and may warrant more attention from a population health perspective than currently appreciated.
肥胖与低水平慢性炎症呈正相关，与健康相关生活质量（HRQOL）的多个指标呈负相关。然而，尚不清楚肥胖相关炎症是否部分导致了肥胖与HRQOL之间观察到的负相关，以及全身炎症是否独立影响HRQOL。我们利用美国国家健康与营养检查调查（NHANES）数据进行了一项探索性分析，以研究肥胖、全身炎症与HRQOL指标之间的关系。
6325名成年人（年龄20 - 75岁，BMI > 18.5 kg/m）的相关变量数据可用。从NHANES（2005 - 2008）获取人口统计学、体重指数（BMI）、C反应蛋白（CRP）、炎症性疾病状态、用药情况、吸烟及HRQOL数据，并使用抽样加权广义线性模型进行分析。对数据进行多重填补以减轻调查无应答导致的信息损失。分析主效应和交互效应以评估可能的中介或调节效应。通过敏感性分析确定模型稳健性。报告插补数据集的平均结果为比值比（OR）和置信区间（CI）。
在未调整模型中，肥胖与身体不健康天数呈正相关（OR：1.59，95% CI：1.15 - 2.21）。炎症标志物CRP的“升高”和“临床升高”水平也与身体不健康天数呈正相关（OR分别为1.61，95% CI：1.23 - 2.12，以及OR = 2.45，95% CI：1.84 - 3.26）；此外，“临床升高”的CRP与精神不健康天数呈正相关（OR = 1.66，95% CI：1.26 - 2.19）。在纳入CRP的模型中，肥胖与身体HRQOL之间的关联变得不显著。即使在调整肥胖或炎症调节协变量后，“升高”和“临床升高”的CRP与身体不健康天数之间的关联仍然显著（OR分别为1.36，95% CI：1.02 - 1.82，以及OR = 1.75，95% CI：1.21 - 2.54）。
全身炎症似乎介导了肥胖与身体不健康天数之间的关联。临床升高的炎症是身体和精神不健康天数的独立决定因素。重要的是，升高（但亚临床）的炎症也与身体健康天数呈负相关，从人群健康角度来看，可能值得比目前更多的关注。
