台灣中老年族群發炎指標與第二型糖尿病危險之關係： 以SEBAS資料庫族群之橫斷性研究

Abstract

目的：分析發炎生物指標，如白血球總數及其分類、白細胞介素-6、C反應蛋白與台灣族群第二型糖尿病盛行率的關聯，找出在第二型糖尿病風險評估中最具實用性與參考價值的發炎生物指標。

方法：使用台灣老人健康之社會因素與生物指標研究 (Social Environment and Biomarkers of Aging Study, SEBAS) 資料進行橫斷性研究。第二型糖尿病依據血液檢查結果、用藥情況、以及自述資料定義。分析採用羅吉斯回歸模型估算與第二型糖尿病的勝算比，並運用樣條模型 (restricted cube spline models) 分析其關係型態。次群體分析用以探討潛在風險因子的效應修飾，並用重分類改善指標淨值 (net reclassification improvement, NRI) 與整合性鑑別度改善指數 (integrated discrimination improvement, IDI) 評估各模型的辨別能力。

結果：在 1,402 位參與者中，較高的白血球總數與第二型糖尿病盛行率風險顯著相關 (勝算比 = 2.56，95%信賴區間：1.72-3.81；趨勢檢定：P < 0.001)，其次為嗜中性球、淋巴球、白細胞介素-6及C反應蛋白。僅有白細胞介素-6與第二型糖尿病盛行率之間呈非線性關係，為飽和曲線的型態。此外，次群體分析顯示，白血球數與第二型糖尿病的關係在非吸菸者中較強 (交互作用P值 = 0.010)，而C反應蛋白的關聯則在有規律運動者中較為明顯 (交互作用P值 = 0.024)。最後，綜合曲線下面積 (AUC)、重分類改善指標淨值 (NRI) 與整合性鑑別度改善指數 (IDI)，白血球數在各發炎指標中展現出最佳的辨別能力。

結論：在台灣族群中，較高數值的白血球、嗜中性球、淋巴球、白細胞介素-6、C反應蛋白與第二型糖尿病的盛行率具有顯著關聯。其中，白血球數展現出最強的辨別能力，有助於臨床醫師識別高風險案例並及早介入。

Objective: To identify the most practical and informative inflammatory biomarker associated with type 2 diabetes risk in the Taiwanese population, we examined the relationships between various biomarkers, including WBC count and its differentials, IL-6, and hs-CRP, and the prevalence of type 2 diabetes.

Methods: We conducted a cross-sectional study using data from the Social Environment and Biomarkers of Aging Study (SEBAS). Type 2 diabetes was defined based on results from blood tests, medication use, and self-reported information. Odds ratios for type 2 diabetes were calculated using logistic regression models, and restricted cubic spline models were applied to understand the relationship pattern. The subgroup analyses examined effect modification by potential risk factors, and discrimination was compared using the net reclassification improvement (NRI) and integrated discrimination improvement (IDI) approaches.

Results: Among 1,402 participants, higher WBC count was significantly associated with type 2 diabetes risk (odds ratio = 2.56, 95% confidence interval: 1.72-3.81; P for trend < 0.001), followed by neutrophils, lymphocytes, IL-6, and hs-CRP. Only IL-6 exhibited a nonlinear relationship with a saturation curve regarding the prevalence of type 2 diabetes. Furthermore, subgroup analyses revealed that the association between WBC count and type 2 diabetes was stronger among nonsmokers (P for interaction = 0.010), while the association for hs-CRP was more pronounced among individuals with regular exercise habits (P for interaction = 0.024). Lastly, WBC count demonstrated the best discriminatory performance based on AUC, NRI, and IDI.

Conclusion: Elevated levels of WBC, neutrophils, lymphocytes, IL-6, and hs-CRP were significantly associated with the prevalence of type 2 diabetes in the Taiwanese population. Among these biomarkers, WBC demonstrated the strongest discriminatory ability, potentially assisting clinicians in identifying high-risk individuals and initiating early interventions.