HMG-CoA還原酶抑制劑對高血脂症患者的瞼板腺形態之影響

Abstract

目的：本研究旨在了解服用常規口服降血脂藥物 (HMG-CoA還原酶抑制劑 [3-hydroxy-3-methyl-glutaryl-coenzyme A reductase inhibitors, statins]) 對瞼板腺形態和眼表生理之影響，及探討高血脂症與瞼板腺炎的關係。 方法：研究團隊在2017至2018年期間於台大醫院收集共159位由內科及家醫科門診轉介至眼科門診的高血脂症病人作追蹤樣本。病人被分為兩組：於收案日起已常規服用降血脂藥物的病人為藥物治療組，以及只以改善生活形態的方式控制血脂的病人為非藥物治療組。眼睛診斷方面，瞼板腺形態是透過瞼板腺觀察儀 (meibography) 作瞼板腺分級 (meiboscore)，而瞼板腺炎的嚴重程度則透過眼瞼邊緣異常等級(lid margin abnormality score)及瞼板腺分泌品質(meibum quality) 決定，其他測量包括淚液分泌功能和淚膜破裂時間等。病人同時會接受血脂及血糖的常規檢查。目前73人已完成長達六個月或以上的追蹤觀察，而上述的診斷項目均於病人回診時再次執行。 結果：藥物治療組的總膽固醇及低密度脂蛋白的平均值顯著性低於非治療組 (P值分別為0.012及0.016）。兩組之間的乾眼症及瞼板腺的其他參數沒有顯著差異。使用逐步多元回歸分析發現，三酸甘油脂、總膽固醇、低密度脂蛋白、高密度脂蛋白、糖化血色素及飯後血糖等，可作為預測上眼瞼瞼板腺分級的參數 (P值0.0004)；此外，年齡及糖化血色素/飯後血糖，可分別作為下眼瞼(p值0.01)及總瞼板腺分級(P值0.0004) 的預測參數。三酸甘油脂、總膽固醇、低密度脂蛋白及高密度脂蛋白則可作為瞼板腺分泌品質的顯著預測變量 (P = 0.043)。六個月追蹤後發現無論有無使用藥物治療，高血脂症患者的瞼板腺擴張及擁塞，會隨著高血脂的進步而改善。在六個月後，組間比較發現非藥物治療組的空腹血糖比藥物治療組明顯下降 （P值0.038）。組內比較則發現藥物治療組在觀察前後，上眼瞼的瞼板腺分級、瞼板腺分泌品質、瞼板腺阻塞、及皮膚黏膜交界處位移等指標均有顯著惡化（P值0.019, < 0.0001, 0.034及0.024）；非藥物治療組的總膽固醇和空腹血糖於觀察六個月前後有顯著進步（P值0.012及0.018），瞼板腺分級、瞼板腺分泌品質及眼瞼邊緣異常等等指標，在觀察期間並沒有顯著變化 (P > 0.05)。 結論：無論有無使用降血脂藥物治療，瞼板腺擁塞會隨著高血脂的控制而改善。即使血脂濃度及血糖不變，長期的高血脂及糖尿病與瞼板腺萎縮、阻塞及分泌品質等指標惡化有相關。高血脂及血糖代謝皆有可能影響瞼板腺的形態變化。

Purpose: To evaluate the possible association of dyslipidemia with meibomian gland (MG) morphological changes by standardized meibography. Methods: Patients with dyslipidemia with or without HMG-CoA reductase inhibitor (statin) treatment were recruited. MG morphology described by meiboscore, lid margin abnormality, meibum quality, and dry eye parameters were compared between the two groups. The longitudinal MG morphological changes in dyslipidemia patients under treatment with statins over the study period were also characterized. Results: A total of 159 participants were enrolled in our present study in the cross-sectional period, and 73 participants completed the longitudinal part of study over 6 months. Total cholesterol (TC) and low density lipoprotein (LDL) were significantly lower in statin group compared to non-statin group (P = 0.012 and 0.016 respectively). No statistically significant intergroup difference was found in terms of meibomian gland and dry eye parameters among patients with or without statin use. Stepwise multiple regression showed that serum triglycerides (TG), TC, LDL, high density lipoprotein (HDL), HbA1c and postprandial glucose are statistically significant predictors of meiboscore of upper eyelid (P = 0.0004). Age was a significant predictor of meiboscore of lower eyelid (P = 0.01) while HbA1c and postprandial glucose were significant predictors of total meiboscore (P = 0.0004). Moreover, meibum quality could be described by the regression model with predictors including TG, TC, LDL and HDL (P = 0.043). Longitudinal observation among patients with dyslipidemia revealed subtle improvement of ductal dilatation and glandular congestion, corresponding to improvement in lipid profile either with statin usage or diet control. There were statistically significant changes in meiboscore of upper eyelid (P = 0.019), meibum quality (P < 0.0001), MG orifice plugging (P = 0.034) and displacement of mucocutaneous junction (P = 0.024) in the statin group after 6 months. No significant difference was observed in terms of serum lipid profile or glucose level, subjective or objective dry eye parameters. On the other hand, there were statistically significant differences of total cholesterol and fasting glucose level in the without-statin group (P = 0.012 and 0.018 respsectively). However, no changes in meiboscores, meibum quality, lid margin abnormality scores, subjective and objective dry eye parameters were noted in the without-statin group (P > 0.05) after 6 months. Conclusion: Improvement in MG congestion was observed in patients with improved lipid profile, regardless of statin usage. Both dyslipidemia and diabetes might interact in changes of MG morphology. Meibomian gland atrophy, orifice plugging and meibum quality might worsen among patients with hyperlipidemia and diabetes in longterm even under statin usage.