Is Neuromonitoring on OLIFs Warrented?

Click here for original paper: Safety of Lateral Interbody Fusion Surgery without Intraoperative Monitoring

Background

A direct lateral lumbar approach results in psoas and lumbar plexus injury 19%-63% of the time. The mini-open OLIF accesses the disc obliquely, anterior to the psoas muscle. Unlike the lateral lumbar approach, the OLIF does not need to pass through the psoas muscle. With this approach the incidence of lumbar plexus or psoas injury is reduced to 4% -14%. Because of this reduction in injury, the authors wanted to investigate if IONM is necessary for OLIF procedures.

Methods

This was a retrospective review. The first half of the study’s 57 patients had monitoring (IONM [EMG] Group)].  The second half of the study’s 72 patients did not (Non-IONM Group). They used Medtronic's surgeon-directed equipment and did do not specify modalities, techniques or muscles recorded. They do state the neurophysiological responses were interpreted by a "qualified" individual.

For up to 3 months post-surgery, patients were assessed by a blinded clinical coordinator. If their deficits resolved within the first 30 days after surgery they were labeled transient. And if they lasted more than 30 days they were labeled persistent.

Results

IONM patient demographics having OLIF

Patient demographics

There was a statistically significant difference between the two groups pre-operatively; the IONM group suffered preoperative symptoms longer than the non-IONM group. The IONM group also took 20.4 minutes more setup time. And patients from both groups spent on average 12-14 days in the hospital post-op.

Lumbar plexus or psoas muscle injuries occurred in 15.7% (10.5% transient and 5.3% persistent) of the IONM Group and 20.8% (16.7% transient and 4.2% persistent) in the Non-IONM Group. The difference between the transient injuries was statistically significant.

OLIF Lumbar Plexus Injuries

Conclusions:

The authors conclude that IONM is not essential for OLIF surgery as there is no statistical difference between the two groups when evaluating persistent injuries.

Comments

Articles like this show us why we need to read more than the title.

What IONM?

The largest gaping hole in this study is that they never define IONM. Which of the modalities did they monitor? What parameters? What muscles? All of these will significantly affect the sensitivity of IONM. They mention using a surgeon-directed Medtronic unit so the assumption is that they monitored EMG. EMG is infamous for being unreliable or insensitive to nerve stretch and gradual compression.  These are both mechanisms of lumbar plexus injury in these surgeries.

No mention or discussion is made as to the possible value of multimodality IONM. Is monitoring the saphenous nerve SSEPs possible and of value? What about motor evoked potentials? A recent paper by Riley et al (2018) nicely demonstrates the benefit of motor evoked potentials in lateral approach surgeries.

IONM Group was Biased

Statistically the two groups were not equal and any subsequent comparison of them is corrupted. The IONM Group had experienced preoperative symptoms for a longer period of time than the non-IONM group. This is a classic case of selection error. One can assume this meant that the IONM patients’ nerves were exposed to greater pathology making both their monitoring and recovery possibly problematic.

Furthermore, the study did not randomize the patients and the first half of the patients had IONM and the second half did not. Therefore, the results may be further contaminated by systematic changes over time and practice effects (i.e. the more you do something the better you get at it).

Left Out the Main Finding

In both the results and discussion, the authors fail to even mention that transient injuries were statistically significantly lower in the EMG group.

OLIF statistical outcome

Analysis of OLIF complications

Also, despite claiming to look at the utility of IONM to reduce nerve damage, they lumped in non-physiological complications (e.g. swelling, infections) into a single "procedure-related" category when examining incidents. In the adjacent figure you can see the rates of complications they compare are higher than simply the physiological complications.

Future Experiments

It would be great to redo this experiment properly.  For OLIF surgeries randomize your two patient groups, and for your IONM group do EMG, saphenous SSEPs and TcMEP monitoring and compare outcomes.

Take Away

IONM question

Lee et al.’s retrospective study used a biased patient sample and found EMG monitoring on a surgeon-directed system resulted in significantly fewer transient injuries.

Questions

On lateral approach cases what IONM do you monitor or would recommend monitoring? Let us know in the comments below.