淫羊藿防治激素性股骨头坏死的作用机制研究

Mechanism of action of epimedium brevicornum in the prevention of steroid-induced osteonecrosis of femoral head

目的:探讨淫羊藿防治激素性股骨头坏死的作用机制。方法:将50只家兔随机分为5组,每组10只。B、C、D、E组家兔采用臀肌注射醋酸泼尼松龙的方法进行激素性股骨头坏死造模,A组家兔臀肌注射生理盐水。造模结束后,C、D、E组根据体质量分别按1.2 g·kg-1、0.6 g·kg-1和0.3 g·kg-1以淫羊藿提取液灌胃,每天1次,连续灌胃10周;A组和B组以10 mL生理盐水灌胃。分别于药物干预开始前(0周)及药物干预4周、6周、8周、10周后测定各组家兔血液中胆固醇和甘油三酯含量,同时在药物干预开始前(0周)及药物干预6周、10周后测定家兔股骨近端骨密度,并于药物干预10周后处死家兔,取双侧股骨头切片进行HE染色,观察骨小梁和骨陷窝的变化。结果:①血脂。药物干预开始后不同时间,家兔血胆固醇含量的差异总体上无统计学意义,即不存在时间效应(F=1.136,P=0.314);各组家兔血胆固醇含量的组间差异总体上有统计学意义,即存在分组效应(F=12.874,P=0.000),除0周和4周外,其余各时点B组均高于A、C、D、E组;时间因素和分组因素之间存在交互效应(F=4.910,P=0.001)。药物干预开始后不同时间,家兔血甘油三酯含量的差异有统计学意义,即存在时间效应(F=144.191,P=0.000);各组家兔血甘油三酯含量的组间差异总体上有统计学意义,即存在分组效应(F=12.358,P=0.000),B组和E组各时点血甘油三酯含量比较,差异均无统计学意义;除0周外,其余各时点B组血甘油三酯含量均高于A、C、D组;时间因素和分组因素之间存在交互效应(F=8.798,P=0.000)。②骨密度。药物干预开始后不同时间,家兔股骨近端骨密度的差异有统计学意义,即存在时间效应(F=289.909,P=0.000);除A组外,其余4组骨密度均呈下降趋势,B组下降趋势最明显;各组家兔骨密度的组间差异总体上有统计学意义,即存在分组效应(F=72.539,P=0.000),除0周外,其余各时点B组骨密度均低于A、C、D、E组;时间因素和分组因素之间存在交互效应(F=40.182,P=0.000)。③病理学检查结果。药物干预10周后,与其余4组相比,B组骨小梁广泛变细、断裂;5组家兔骨小梁面积分数和空骨陷窝率比较,组间差异均有统计学意义(F=62.032,P=0.000;F=20.515,P=0.000);B组骨小梁面积分数小于A、C、D、E组(P=0.000,P=0.000,P=0.000,P=0.000),空骨陷窝率大于A、C、D、E组(P=0.000,P=0.000,P=0.000,P=0.000)。结论:淫羊藿能够降低股骨头坏死家兔血液中胆固醇和甘油三酯的含量、延缓骨密度降低、降低空骨陷窝率、提高骨小梁面积分数,从而具有防治激素性股骨头坏死的作用。

Objective：To explore the mechanism of action of epimedium in the prevention of steroid-induced osteonecrosis of femoral head（SNFH）. Methods： Fifty rabbits were randomly divided into 5 groups, 10 cases in each group. The rabbit SNFH models were created in group B, C, D and E by intraglnteus injection of prednisolone acetate, while rabbits in group A were administrated with intragluteus injec- tion of normal saline. After that, rabbits in group C, D and E were intragastric administrated with aqueous extract of epimedium （ 1.2,0.6 and 0.3 g/kg,respectively） ,while the others in group A and group B were intragastric administrated with normal saline（ 10 ml） ,once per day for l0 consecutive weeks. Serum levels of cholesterol and triglyceride in rabbits were determined before drug intervention（0 week）and 4,6,8 and 10 weeks after the beginning of the drug intervention. The proximal femoral bone mineral density（BMD） were measured before drug intervention（0 week）and 6 and 10 weeks after the beginning of the drug intervention. At the end of the drug intervention, the rabbits were sacrificed and the bilateral femoral heads were sectioned and the change in bone trabecula and bone lacuna were observed though HE staining. Results：In general, there was no statistical difference in serum level of cholesterol （CHOL）among time points （F = 1. 136, P = 0.314） ,in other words,there was no time effect. There was statistical difference in serum CHOL level among groups in general（ F = 12. 874, P = 0. 000）, in other words, there was group effect. The serum CHOL level of group B was higher than that of group A, C, D and E except 0 and 4 weeks after drug intervention. There was interaction between time factor and grouping factor（ F = 4. 910 ,P = 0. 001 ）. There was statistical difference in serum level of triglyceride （TG） among time points （ F = 144. 191, P = 0. 000）, in other words, there was time effect. In general, there was statistical difference in serum TG level among groups （ F = 12. 358, P = 0. 000）, in other words, there was group effect. There was no statistical difference in serum TG level taken at different time points between group B and group E ; the group B sur- passed group A, C and D in serum TG level of at each time point except 0 week. There was interaction between time factor and grouping fac- tor（ F = 8. 798, P = 0. 000）. There was statistical difference in proximal femoral BMD among time points （ F = 289. 909, P = 0. 000 ）. There was a trend for decreased BMD in group B, C, D and E after drug intervention, and there was more significant decreased BMD in group B compared to other groups. There was statistical difference in BMD among different groups in general （ F = 72. 539, P = 0. 000）, in other words, there was group effect. The BMD of Group B was lower than that of group A, C, D and E at each time point except 0 week. There was interaction between time factor and grouping factor（ F = 40. 182, P = O. 000）. The bone trabecula became slim and broken extensively in group B compared with other groups 10 weeks after the beginning of drug intervention. There were statistical differences in empty bone lacu- na rate and trabecular bone area fraction among different groups （ F = 62. 032, P = 0. 000 ; F = 20. 515, P = 0. 000）, and the trabecular bone area fraction of group B was less than that of group A, C, D and E （ P = 0. 000, P = 0. 000, P = 0. 000, P = 0. 000）, while the empty bone la- cuna rate of group B was larger than that of group A, C, D and E （P = 0. 000, P = 0. 000, P = 0. 000,P = 0.000）. Conclusion ： The epimedi- um brevicoruum can decrease serum CHOL and TG contents, delay the decrease in BMD, decrease the empty bone lacuna rate and increase the trabecular bone area fraction in the rabbits with SNFH, so it can be used in the prevention and treatment of SNFH.