Pedicle screw fixation and posterior fusion for lumbar degenerative diseases: effects on individual paraspinal muscles and lower back pain; a single-center, prospective study

It is widely known that patients undergoing lumbar surgery are at increased risk of developing atrophy of the lumbar extensors [13]. Yong et al., who conducted an animal study to compare a fusion group and a control group, reported that there was a significant decrease in the root mean square (RMS) and median frequency (MF)—which served as indicators of the activity of paraspinal muscles—in the fusion group at the 6-month follow-up when compared with the preoperative values. These authors also showed that histologically there was also a significant decrease in the volume of muscle fibers, which served as an indicator of amyotrophy [14]. A more recent human study observed significantly increased denervation at the 6-month follow-up [15]. Our study showed similar results in that both multifidus and erector spinae muscles were still significantly denervated at the 6-month follow-up (compared with their preoperative status). In terms of the changes in adjacent levels, Yong et al. reported that there was a significant increase in the RMS and MF of paraspinal muscles at the adjacent cranial and caudal levels when compared with those that had been subjected to surgery [14]. The current study, however, demonstrated that there was a significant difference in the degree of change in multifidus and erector spinae muscles between the adjacent cranial and caudal levels. We think it may result from the different subjects being studied (adult New Zealand white rabbits versus adult humans), the degree of damage to the paraspinal muscles, and the fusion level.

For the purpose of avoiding a bias when evaluating the degree of denervation in the paraspinal muscles on EMG, we limited the spinal levels to L3–5. There were significant degrees of denervation in the paraspinal muscles at the surgical level and its adjacent levels at 3 months. This might be due to the retraction, compression, and intraoperative damage to the posterior primary ramus and disuse muscle atrophy due to fusion and orthoses.

Waschke et al. demonstrated there was slightly decreased denervation at the 12-month follow-up in adult humans [15]. In the current study, we found that, in group I, there was more rapid reinnervation in the upper spinal segments that at other levels. This might be due to technical problems such as less traction, less severe direct damage to paraspinal muscles during dissection, and a smaller area of the multifidus being removed. We also assumed that a greater amount of paraspinal muscles’ dissection in the cranial and lateral directions and a more immobilized spinal segment might lead to slower reinnervation in group II than in group I.

In group II, although there was no significant reinnervation of the denervated multifidus and erector spinae muscles until the 18-month follow-up, the upper segment of the denervated erector spinae was significantly reinnervated at 18 months. Also, the other muscles of each segment had a tendency to be reinnervated during serial follow-up evaluations. These results suggest that the postoperative denervation in paraspinal muscles resulting from one- and two-level pedicle screw fixation and posterior fusion might not be a permanent phenomenon. Further studies are therefore warranted to confirm the significant reinnervation of paraspinal muscles following multiple levels of pedicle screw fixation and fusion.

In the current study, we assumed that the degrees of reinnervation in denervated erector spinae and multifidus muscles at the 12- and 18-month follow-up evaluations provide evidence that muscle function is likely to recover to the preoperative level even in patients undergoing an instrumentation and fusion procedure.

As described by Macintosh and Bogduc, there is limited evidence to explain the correlation between denervated paraspinal muscles and failed back surgery syndrome (FBSS) in patients undergoing posterior lumbar surgery [16]. In contrast, there are many reasons for an unsatisfactory clinical outcome after posterior instrumentation and fusion of the lumbar spine. Among them, dysfunction of the paraspinal muscles due to atrophy is one of the most discussed issues [17]. Sihvonen et al. reported that the degree of denervation in paraspinal muscles was significantly greater in patients with FBSS than in those with satisfactory surgical outcomes [6]. Ranaten et al. also suggested that inactivation of and damage to axons might be associated with atrophy of type 2 muscle fibers and unsatisfactory clinical outcomes [18]. Wilbourne and Aminoff demonstrated that denervation of paraspinal muscles leads to significant changes in spinal biomechanics, which could cause LBP [19]. In terms of the clinical consequence of denervation of paraspinal muscles, the current study showed results similar to those reported for previous studies.

Moreover, there was significant alleviation of LBP at the 12-month follow-up in group I and at the 18-month follow-up in group II. Thus, the LBP diminishes prior to reinnervation in denervated paraspinal muscles.