Computed tomographic analysis of the quality of trunk muscles in asymptomatic and symptomatic lumbar discectomy patients

Background: No consensus exists on how rehabilitation programs for lumbar discectomy patients with persistent complaints after surgery should be composed. A better understanding of normal and abnormal postoperative trunk muscle condition might help direct the treatment goals. Methods: A three-dimensional CT scan of the lumbar spine was obtained in 18 symptomatic and 18 asymptomatic patients who had undergone a lumbar discectomy 42 months to 83 months (median 63 months) previously. The psoas muscle (PS), the paraspinal muscle mass (PA) and the multifidus muscle (MF) were outlined at the L3, L4 and L5 level. Of these muscles, fat free Cross Sectional Area (CSA) and fat CSA were determined. CSA of the lumbar erector spinae (LES = longissimus thoracis + iliocostalis lumborum) was calculated by subtracting MF CSA from PA CSA. Mean muscle CSA of the left and right sides was calculated at each level. To normalize the data for interpersonal comparison, the mean CSA was divided by the CSA of the L3 vertebral body (mCSA = normalized fat-free muscle CSA; fCSA = normalized fat CSA). Differences in CSA between the pain group and the pain free group were examined using a General Linear Model (GLM). Three levels were examined to investigate the possible role of the level of operation. Results: In lumbar discectomy patients with pain, the mCSA of the MF was significantly smaller than in pain-free subjects (p = 0.009) independently of the level. The mCSA of the LES was significantly smaller in pain patients, but only on the L3 slice (p = 0.018). No significant difference in mCSA of the PS was found between pain patients and pain-free patients (p = 0.462). The fCSA of the MF (p = 0.186) and of the LES (p = 0.256) were not significantly different between both populations. However, the fCSA of the PS was significantly larger in pain patients than in pain-free patients. (p = 0.012). The level of operation was never a significant factor. Conclusions: CT comparison of MF, LES and PS muscle condition between lumbar discectomy patients without pain and patients with protracted postoperative pain showed a smaller fat-free muscle CSA of the MF at all levels examined, a smaller fat-free muscle CSA of the LES at the L3 level, and more fat in the PS in patients with pain. The level of operation was not found to be of importance. The present results suggest a general lumbar muscle dysfunction in the pain group, in particular of the deep stabilizing muscle system.