矿化胶原-聚甲基丙烯酸甲酯骨水泥经皮椎体后凸成形术治疗Ⅰ、Ⅱ期Kümmell病的临床研究

A clinical study of percutaneous kyphoplasty with mineralized collagen-polymethylmethacrylate bone cement for treatment of phase Ⅰ and Ⅱ Kümmell's diseases

目的:探讨矿化胶原-聚甲基丙烯酸甲酯(mineralized collagen-polymethylmethacrylate,MC-PMMA)骨水泥经皮椎体后凸成形术(percutaneous kyphoplasty,PKP)治疗Ⅰ、Ⅱ期Kümmell病的临床疗效和安全性。方法:收集2015年1月至2017年1月在宁波大学医学院附属医院住院治疗的Ⅰ、Ⅱ期Kümmell病患者的病例资料进行回顾性研究。符合要求的患者共23例,均行伤椎双侧PKP手术,术中使用MC-PMMA者9例(MC-PMMA组)、使用PMMA者14例(PMMA组)。比较2组患者的手术时间、术中出血量、住院时间、疼痛视觉模拟量表(visual analogue scale,VAS)评分、Oswestry功能障碍指数(Oswestry disability index,ODI)、伤椎前缘高度丢失率、伤椎后缘高度丢失率、脊柱后凸角矫正率、伤椎楔变角矫正率,以及治疗和随访期间的并发症发生情况。结果:(1)手术一般情况及并发症。2组患者均顺利完成手术。2组患者的手术时间、术中出血量及住院时间比较,组间差异均无统计学意义[(27.4±3.7)min,(27.0±4.5)min,t=0.248,P=0.807;(13.3±3.5)m L,(14.3±3.3)m L,t=-0.655,P=0.519;(8.1±1.1)d,(7.7±1.3)d,t=-0.780,P=0.444]。MC-PMMA组5例发生骨水泥渗漏,其中1例向椎体前缘渗漏、4例向椎间盘内渗漏;PMMA组7例发生骨水泥渗漏,其中1例向椎体前缘渗漏、5例向椎间盘内渗漏、1例向椎管内少量渗漏但无神经症状。MC-PMMA组2例发生椎体骨折,PMMA组3例发生椎体骨折,均经非手术治疗或PKP手术治疗后改善或治愈。2组骨水泥渗漏和术后新发骨折率比较,组间差异均无统计学意义(P=1.000;P=1.000)(2)疼痛VAS评分。时间因素和分组因素不存在交互效应(F=0.138,P=0.711)。2组患者疼痛VAS评分总体比较,差异无统计学意义,即不存在分组效应(F=0.095,P=0.760);手术前后不同时间点之间疼痛VAS评分的差异有统计学意义,即存在时间效应(F=302.813,P=0.000);2组疼痛VAS评分随时间均呈先降低后增加的趋势,且2组的变化趋势基本一致。(3)ODI。时间因素和分组因素不存在交互效应(F=0.299,P=0.590)。2组患者ODI总体比较,差异无统计学意义,即不存在分组效应(F=0.349,P=0.561);手术前后不同时间点之间ODI的差异有统计学意义,即存在时间效应(F=531.962,P=0.000);2组ODI随时间均呈先降低后增加的趋势,且2组的变化趋势基本一致。(4)伤椎前缘高度丢失率。时间因素与分组因素不存在交互效应(F=2.881,P=0.074);2组患者伤椎前缘高度丢失率总体比较,差异无统计学意义,即不存在分组效应(F=0.853,P=0.366);手术前后不同时间点之间伤椎前缘高度丢失率的差异有统计学意义,即存在时间效应(F=27.068,P=0.000);2组伤椎前缘高度丢失率随时间变化均呈先降低后增高的趋势,2组的变化趋势基本一致。(5)伤椎后缘高度丢失率。时间因素与分组因素不存在交互效应(F=2.488,P=0.102);2组患者伤椎后缘高度丢失率总体比较,差异无统计学意义,即不存在分组效应(F=3.871,P=0.062);手术前后不同时间点之间伤椎后缘高度丢失率的差异有统计学意义,即存在时间效应(F=20.016,P=0.000);2组伤椎后缘高度丢失率随时间变化均呈先降低后增高的趋势,2组的变化趋势基本一致。(6)脊柱后凸角矫正率。时间因素与分组因素存在交互效应(F=18.089,P=0.000);2组患者脊柱后凸角矫正率总体比较,差异无统计学意义,即不存在分组效应(F=0.001,P=0.973);术后不同时间点之间脊柱后凸角矫正率的差异有统计学意义,即存在时间效应(F=52.825,P=0.000);2组脊柱后凸角矫正率随时间变化均呈降低趋势,但2组的降低趋势不完全一致;术后即刻、术后3个月,2组脊柱后凸角矫正率的组间差异均无统计学意义(t=-1.867,P=0.076;t=-1.311,P=0.204);术后1年,PMMA组的脊柱后凸角矫正率明显低于MC-PMMA组(t=3.690,P=0.001)。(7)伤椎楔变角矫正率。时间因素与分组因素存在交互效应(F=10.315,P=0.000);2组患者伤椎楔变角矫正率总体比较,差异无统计学意义,即不存在分组效应(F=0.016,P=0.901);术后不同时间点之间伤椎楔变角矫正率的差异有统计学意义,即存在时间效应(F=49.888,P=0.000);2组伤椎楔变角矫正率随时间变化均呈降低趋势,2组的降低趋势基本一致。结论:应用MC-PMMA与应用PMMA行PKP治疗Ⅰ、Ⅱ期Kümmell病,均能迅速缓解疼痛症状、改善脊柱功能、恢复伤椎高度、纠正脊柱后凸畸形,而且具有较高的安全性;但后期均会出现一定程度的伤椎高度丢失和脊柱后凸矫正度丢失,应用MC-PMMA比应用PMMA能更好地维持脊柱后凸矫正度。

Objective： To explore the clinical curative effects and the safety of percutaneous kyphoplasty（ PKP） with mineralized collagen-polymethylmethacrylate（ MC-PMMA） bone cement for treatment of phaseⅠandⅡKümmell＇s diseases. Methods： The medical records of patients with phaseⅠandⅡKümmell＇s diseases who were treated in the Affiliated Hospital of Medical School of Ningbo University from January 2015 to January 2017 were collected and retrospectively studied. Twenty-three patients enrolled in the study were treated with bilateral PKP on injured vertebra using MC-PMMA（ 9） and PMMA（ 14） respectively. The operative time,intraoperative blood loss,hospital stay,pain visual analogue scale（ VAS） scores,Oswestry disability index（ ODI）,loss rate of injured vertebrae anterior border height,loss rate of injured vertebrae posterior border height,correction rate of spinal kyphotic angle,correction rate of angle of wedge shaped injured vertebrae and complications during treatment period and follow-up period were compared between the 2 groups. Results： The surgeries were finished successfully in all patients. There was no statistical difference in operative time,intraoperative blood loss and hospital stay between the2 groups（ 27. 4 ＋/-3. 7 vs 27. 0 ＋/-4. 5 min,t = 0. 248,P = 0. 807; 13. 3 ＋/-3. 5 vs 14. 3 ＋/-3. 3 m L,t =-0. 655,P = 0. 519; 8. 1 ＋/-1. 1 vs 7. 7 ＋/-1. 3 days,t =-0. 780,P = 0. 444）. The bone cements leaked out of anterior border of vertebral body（ 1） and leaked into intervertebral disc（ 4） in MC-PMMA group. The bone cements leaked out of anterior border of vertebral body（ 1） and leaked into intervertebral disc（ 5） and vertebral canal without nerve symptoms（ 1） in PMMA group. Meanwhile,the vertebral fractures were found in MC-PMMA group（ 2） and PMMA group（ 3）,and all patients were improved or cured after non-surgical treatment or PKP surgery. There was no statistical difference in the rates of bone cement leakage and postoperative new fracture between the 2 groups（ P = 1. 000; P = 1. 000）. There was no interaction between time factor and group factor in pain VAS scores（ F = 0. 138,P = 0. 711）. There was no statistical difference in pain VAS scores between the 2 groups in general,in other words,there was no group effect（ F = 0. 095,P = 0. 760）. There was statistical difference in the pain VAS scores between different timepoints before and after the surgery,in other words,there was time effect（ F = 302. 813,P = 0. 000）. The pain VAS scores presented a time-dependent trend of decreasing firstly and increasing subsequently in both of the 2 groups,and the 2 groups were basically consistent with each other in the variation tendency of pain VAS scores. There was no interaction between time factor and group factor in ODI（ F = 0. 299,P = 0. 590）. There was no statistical difference in ODI between the 2 groups in general,in other words,there was no group effect（ F = 0. 349,P = 0. 561）. There was statistical difference in ODI between different timepoints before and after the surgery,in other words,there was time effect（ F = 531. 962,P = 0. 000）. The ODI presented a time-dependent trend of decreasing firstly and increasing subsequently in both of the 2 groups,and the 2 groups were basically consistent with each other in the variation tendency of ODI. There was no interaction between time factor and group factor in the loss rate of injured vertebrae anterior border height（ F = 2. 881,P = 0. 074）. There was no statistical difference in the loss rate of injured vertebrae anterior border height between the 2 groups in general,in other words,there was no group effect（ F = 0. 853,P = 0. 366）. There was statistical difference in the loss rate of injured vertebrae anterior border height between different timepoints before and after the surgery,in other words,there was time effect（ F =27. 068,P = 0. 000）. The loss rate of injured vertebrae anterior border height presented a time-dependent trend of decreasing firstly and increasing subsequently in both of the 2 groups,and the 2 groups were basically consistent with each other in the variation tendency of loss rate of injured vertebrae anterior border height. There was no interaction between time factor and group factor in the loss rate of injured vertebrae posterior border height（ F = 2. 488,P = 0. 102）. There was no statistical difference in the loss rate of injured vertebrae posterior border height between the 2 groups in general,in other words,there was no group effect（ F = 3. 871,P = 0. 062）. There was statistical difference in the loss rate of injured vertebrae posterior border height between different timepoints before and after the surgery,in other words,there was time effect（ F = 20. 016,P = 0. 000）. The loss rate of injured vertebrae posterior border height presented a time-dependent trend of decreasing firstly and increasing subsequently in both of the 2 groups,and the 2 groups were basically consistent with each other in the variation tendency of loss rate of injured vertebrae posterior border height. There was interaction between time factor and group factor in the correction rate of spinal kyphotic angle（ F = 18. 089,P = 0. 000）. There was no statistical difference in the correction rate of spinal kyphotic angle between the 2 groups in general,in other words,there was no group effect（ F = 0. 001,P = 0. 973）. There was statistical difference in the correction rate of spinal kyphotic angle between different timepoints after the surgery,in other words,there was time effect（ F = 52. 825,P = 0. 000）. The correction rate of spinal kyphotic angle presented a time-dependent decreasing trend in both of the 2 groups,while the 2 groups were inconsistent with each other in the decreasing trend of correction rate of spinal kyphotic angle. There were no statistical difference in the correction rate of spinal kyphotic angle between the 2 groups immediately postoperatively and at 3 months after the surgery（ t =-1. 867,P = 0. 076; t =-1. 311,P = 0. 204）. The correction rate of spinal kyphotic angle was obviously lower in PMMA group compared to MC-PMMA group at 1 year after the surgery（ t = 3. 690,P = 0. 001）. There was interaction between time factor and group factor in the correction rate of angle of wedge shaped injured vertebrae（ F = 10. 315,P = 0. 000）. There was no statistical difference in the correction rate of angle of wedge shaped injured vertebrae between the 2 groups in general,in other words,there was no group effect（ F = 0. 016,P = 0. 901）. There was statistical difference in the correction rate of angle of wedge shaped injured vertebrae between different timepoints after the surgery,in other words,there was time effect（ F = 49. 888,P = 0. 000）. The correction rate of angle of wedge shaped injured vertebrae presented a time-dependent decreasing trend in both of the 2 groups,and the 2 groups were basically consistent with each other in the decreasing trend of correction rate of angle of wedge shaped injured vertebrae. Conclusion： PKP with MC-PMMA or PMMA can rapidly alleviate pain symptoms,improve spinal function,restore injured vertebrae height and correct kyphotic deformity in treatment of phaseⅠandⅡKümmell＇s diseases,and both of them have high safety. However,injured vertebra height and kyphotic correction rate can loss to a certain degree in later stage,and MC-PMMA surpasses PMMA in maintaining kyphotic correction rate.