医学
甲基强的松龙
不利影响
观察研究
入射(几何)
脊髓损伤
随机对照试验
临床试验
脊髓
内科学
麻醉
加药
精神科
光学
物理
作者
Zhongyu Liu,Yang Yang,Lei He,Mao Pang,Chunxiao Luo,Bin Liu,Limin Rong
出处
期刊:Neurology
[Ovid Technologies (Wolters Kluwer)]
日期:2019-07-30
卷期号:93 (9)
被引量:154
标识
DOI:10.1212/wnl.0000000000007998
摘要
Due to the continuing debates on the utility of high-dose methylprednisolone (MP) early after acute spinal cord injury (ASCI), we aimed to evaluate the therapeutic and adverse effects of high-dose MP according to the second National Acute Spinal Cord Injury Study (NASCIS-2) dosing protocol in comparison to no steroids in patients with ASCI by performing a meta-analysis on the basis of the current available clinical trials.We searched PubMed and Cochrane Library (to May 22, 2018) for studies comparing neurologic recoveries, adverse events, and in-hospital costs between ASCI patients who underwent high-dose MP treatment or not. Data were synthesized with corresponding statistical models according to the degree of heterogeneity.We enrolled 16 studies (1,863 participants) including 3 randomized controlled trials (RCTs) and 13 observational studies. Pooled results indicated that MP was not associated with an increase in motor score improvement (RCTs: p = 0.84; observational studies: p = 0.44) and incidence of recovery by at least one grade on the American Spinal Injury Association Impairment Scale or Frankel (p = 0.53). Meanwhile, MP did not lead to better sensory recovery (p = 0.07). However, MP was associated with a significantly higher incidence of gastrointestinal hemorrhage (p = 0.04) and respiratory tract infection (p = 0.01). The difference in the overall in-hospital costs between MP and control groups was not statistically significant (p = 0.78).Based on the current evidence, high-dose MP treatment, in comparison to controls, does not contribute to better neurologic recoveries but may increase the risk of adverse events in patients with ASCI. Therefore, we recommend against routine use of high-dose MP early after ASCI.
科研通智能强力驱动
Strongly Powered by AbleSci AI