Given the narrow safe working zones described above, adequate patient and fluoroscopic positioning are critical to ensuring the proper placement of retractors during the lateral approach. Poor patient positioning and a lack of understanding how to find the true lateral or anterior/posterior (AP) imaging will lead to the placement of instrumentation through critical structures. The patient is placed in the lateral decubitus position with an axillary roll. The arms are well padded with the top arm supported by either a pillow or arm sling. The head is supported with padding and the leg on top is placed on a pillow and flexed. Flexing the top leg allows relaxation of the ipsilateral iliopsoas and plexus thus reducing traction on them during the procedure. The iliac crest is placed just below the break in the table, and the torso is lowered to open the space between the ribs and iliac crest on the approach side. The bed is then placed in reverse Trendelenburg to bring the spine parallel to the floor. Finally, the patient is secured in this position with tape across the chest and hips.

AP fluoroscopic imaging is obtained to ensure a true lateral position. On AP the spinous processes at the levels of interest should be perfectly centered between their respective pedicles. If there is a large discrepancy, the patient is repositioned accordingly. If there is a minor discrepancy, the bed can be rotated to obtain a true AP image. Having the patient in a perfectly lateral position allows for proper placement of the instrumentation in the safe working zones described above. It also ensures that the discectomy and cage placement are performed safely. A lateral image is obtained, the borders of the disc space are marked on the skin, and the incision is planned (single- or two-incision technique). The flank incision should always be in a horizontal direction to avoid injury to the plexus branches traversing the lateral abdominal musculature and to respect the Langer lines.

When performing LLIF through an MIS approach, the surgeon must also be aware of the anatomy of the psoas muscle.

Though injury to the plexus is most frequently encountered while dissecting through the psoas muscle, the nerves traveling outside of the psoas in the retroperitoneal space are also at risk of injury during the abdominal dissection. Dakwar et al. performed cadaveric studies looking at the trajectories of the major motor and sensory branches of the lumbar plexus outside of the psoas muscle and within the retroperitoneum and abdominal wall [28]. They found that four plexus branches were at risk of injury during dissection through the abdominal wall and retroperitoneal space: subcostal, iliohypogastric, ilioinguinal, and lateral femoral cutaneous nerves. Injury to these nerves can occur during the flank or lateral incision and can lead to abdominal wall paresis and sensory deficits in the corresponding dermatome.

In order to avoid injury to these nerves, once the external oblique muscle fascia is sharply opened, blunt muscle dissection should be performed until the retroperitoneal cavity is identified. Bovie cauterization and bipolar must be limited, and any electrocautery should be limited to bipolar electrocoagulation. If a nerve is discovered during dissection, it can be carefully dissected and mobilized. Once the retroperitoneal space is encountered, blunt dissection of the peritoneal contents from posterior to anterior is performed until the transverse process and psoas muscle are encountered [28]. If using finger dissection, one must be careful not to avulse any presumed fibrous bands as they could be nerves.

Triggered electromyography (t-EMG) has been described to detect motor branches of the plexus during dissection through the psoas muscle [6, 35]. Using threshold stimulation, one can potentially determine the proximity of the dilators to adjacent nerves. Clinically normal nerves elicit an EMG response with stimulation ranging from 1 to 5 mA, with a mean of about 2 MA [36, 37]. The closer the stimulator is to the nerve, the lower the current needed for stimulation. Thus, thresholds of 5 mA or less indicate that the stimulator is possibly in direct contact with the nerve. Thresholds between 5 and 10 mA are generally considered to be a gray zone where caution must be elicited, while thresholds greater than 10 mA are considered safe.

Some dilators have directional stimulation that allows the operator to test where the lowest threshold (and thus the nerve) is located relative to the dilators. If a threshold less than 10 mA is encountered, the dilators can be repositioned away from the stimulated nerves. Tohmeh et al. prospectively studied the EMG threshold values of 102 consecutive patients undergoing LLIF. They found that 55.7 % of the cases had alert-level EMG feedback (thresholds less than 10 mA) with initial dilator placement, which lead to repositioning. Transient motor deficits occurred in three patients (2.3 %) with all resolving by 6 months. Of note, two of the three patients had spontaneous EMG (S-EMG) activity in the affected myotome during the surgery while trial spacers were being inserted. Despite these promising findings, t-EMG is limited to motor nerves only leaving the sensory nerves vulnerable. Thus, Tohmeh et al. found that 17.6 % of their patients experienced thigh sensory deficits immediately postoperatively. However, t-EMG does not warn of stretch injury, and false negatives may occur from shunting of the stimulation through the blades [38].

When further visualization of the disc space after inserting the working retractor into the psoas muscle, it is important to only open the anterior blade. Opening the posterior blade can lead to stretch injury of the plexus and/or can crush the plexus between the blades of the retractor and the transverse processes. Also the length of surgery has been found to correlate with increased extremity symptoms, presumably from longer periods of psoas and plexus retraction [8]. Thus, once the retractor is opened, the disc space must be prepared in a timely fashion.

As illustrated above, good t-EMG may significantly lower the risk of injury to the motor nerves but it does not limit sensory nerve injury. To this end a different technique is described whereby instead of serially dilating through the psoas muscle, the working retractor is guided onto the surface of the psoas muscle without entering it [38]. Once the working retractor is in proper position as determined by fluoroscopy, the psoas muscle is bluntly dissected in a longitudinal fashion exposing the disc space (Fig. 12.4). Any nerves will likely be clearly visualized or at least amenable to “mapping” of the surface of the psoas muscle to detect the proximity of any lumbosacral plexus nerves.

During dissection, t-EMG can be used to distinguish motor from sensory nerves. These exposed nerves are then dissected away from the exposed disc space so that the interbody work can safely be performed. Of note, the retractor is not advanced through the psoas muscle limiting retraction injury on both the muscle and nerves. Acosta et al. described their experience utilizing this technique on 15 consecutive patients. No patients were noted to have postoperative weakness, numbness, or pain immediately or at last follow-up; however, the study was admittedly of small sample size [38].

The oblique approach is a variant of the lateral approach discussed in great detail in the following chapter. It is mentioned here since its main function is limiting psoas muscle and plexus injury. During this approach, the anterior aspect of the psoas muscle is exposed, dissection is performed between the anterior border of the psoas muscle and the sympathetic trunk, and the entire psoas muscle is retracted posteriorly while the sympathetic trunk is retracted anteriorly [39, 40]. This has the advantages of being further away from the plexus than the standard lateral approach. Also, no dissection through the psoas muscle is needed limiting any injury to it. However, this approach has its own set of challenges including: manipulations of the sympathetic trunk, rocking of the inserter posteriorly to obtain correct positioning of the cage, and closer proximity to the great vessels.