Introduction

Spinal surgery is offered to those who, despite conservative treatment, do not achieve meaningful improvements. With those surgeries, the aim is to address the underlying spinal pathology. Examining which factors may predict functional recovery after spinal surgery for chronic low back pain is crucial given the fact that the number of surgeries performed is increasing and their effects are often reported to be suboptimal. We all have encountered a patient who’s had spinal surgery but hasn’t recovered well, right? To avoid this, it would be interesting to see who will most likely benefit from these surgical interventions. Therefore, this study aimed to identify factors predicting functional recovery after spinal surgery for chronic low back pain from the existing literature.

Methods

This systematic review examined preoperative factors predicting functional recovery after spinal surgery in adults with chronic low back pain. The participants were scheduled for primary lumbar or lumbosacral surgery. Chronic low back pain was defined as back pain lasting or recurring for at least 3 months. Lumbar radicular pain could be included as well and was defined as pain radiating to the leg due to nerve root compression. The most common pathologies included disc herniation, degenerative disc disease, and spondylosis. The surgeries performed were spinal fusion and discectomy.

Return to work served as a proxy for functional recovery and was assessed at least 3 months after surgery. Eligible study designs were randomized and nonrandomized clinical trials and prospective or retrospective cohort studies, case-control studies, and registry-based studies.

Results

In total, 8 reports of six studies were available for inclusion. Sociodemographic factors predicting functional recovery after spinal surgery supported by moderate-quality evidence were older age, duration of sick leave, and having legal representation. This means the older the patient is when he undergoes spinal surgery, the less likely he will return to work. Age was defined as a categorical predictor: being >48 or >50 years. The duration of sick leave had a significantly small effect on return to work. This means, the longer the duration of the sick leave, the less likely the patient will return to work. The same was true for those people having legal representation, where the effect on return to work was also negative but small.
A small association between income and return to work was found in 2 low-risk bias studies, where participants with higher weekly wages and household income had higher odds of RTW after surgery. Yet this effect was not supported by the meta-analysis as the pooled adjusted OR was not significant. The meta-analysis had high heterogeneity (84%), which may be a part of the reason for the non-significance. This should be analyzed further.

Psychological factors supported by moderate-quality evidence were having a psychiatric comorbidity and depression. Participants with any psychiatric comorbidity (including affective disorders and schizophrenia) had lower odds of return to work after surgery in the adjusted analysis. Participants with a clinical diagnosis of depression had lower odds of return to work than those without. Unfortunately, the forest plot of these psychological factors was not presented.

Questions and thoughts

The surgery should be able to correct the underlying spinal pathology. In practice, this should be confirmed. In reality, patients often attribute their complaints to a disc herniation they had years ago. But as we know that herniations may resolve spontaneously in some, there wouldn’t be an indication for surgery. Yet still, some patients want to proceed to surgery. When you encounter a patient who’s had a disc herniation in the past, but they present to you with resorption of leg pain and without neurological findings, you can question whether the persisting back pain is still attributable to a herniated disc. Probably not. Spinal surgery won’t be very effective in this case. There, you have an important role in the education of this patient. Explain that most likely no nerve root is pinched or no disc has slipped. In this study, lumbar radicular pain was included as well and defined as “pain radiating to the leg due to nerve root compression”. Sadly, this way, the thought that radiating pain down the leg is caused by a compression of the nerve root is reinforced. Yet, we know that inflammation of the nerve root or around the nerve can also contribute to radicular leg pain. This information may already be reassuring to many patients affected. Instead, try not to create a nocebo effect when you are communicating with someone who sees you for their radiating leg pain!

Return to work was the primary outcome of this study and this should reflect full functional recovery. Indeed, when one is able to take up his duty again, it may seem that he has had a full recovery. However, return to work may have different definitions; return to unrestricted work, return for at least 6 months, return to work with adaptations, part-time return to work,… There appears to be still quite some heterogeneity in the return to work outcome measure, therefore.

Talk nerdy to me

The review was prospectively registered and reported according to the PRISMA guidelines. Relevant databases and reference lists of relevant systematic reviews were searched to retrieve as many eligible articles as possible. The methodology used also aimed to be unrestrictive. There was no defined minimum set of comparator prognostic factors. Prognostic factors in the field of radiography or genetics or factors collected intra- or postoperatively were beyond the scope of this research. There was no requirement to be working before surgery. To me, this doesn’t seem like much of a problem. The study characteristics section revealed that all participants were working prior to surgery, except in one study, where only 14% were unemployed before their surgery.

The follow-up had to have a duration of at least 3 months. The length of the follow-up in the included studies ranged between 6 and 36 months. Yet, most of the studies included follow-up periods of 24 or 36 months. Therefore, these results can be interpreted as mid-term outcomes assessed several years after the spinal surgery was performed.

Five out of the included studies were prospective in nature, three included retrospective cohorts. This means that in nearly one-third of included studies data was collected from events in the past. This may have some limitations as recall bias for example may have had a spell on the results. For the factors examined in a meta-analysis, heterogeneity was low, except for the factor of income. But the meta-analysis revealed that the effect of income was not significant. Maybe, the high heterogeneity in the two pooled studies could have influenced the significance.

Take home messages

In general, the study included few articles, so the evidence base is limited. Therefore I would not overemphasize the effects supported by low or very low evidence until these are confirmed by future studies. The factors predicting functional recovery after spinal surgery that are supported by moderate evidence (age, duration of sick leave, legal representation, psychological comorbidity, and depression) may be guiding your prognostication. You understand that these factors are not modifiable, so they will be purely informative in determining your prognosis, but they will not be part of your treatment strategy.

Reference

Halicka M, Duarte R, Catherall S, Maden M, Coetsee M, Wilby M, Brown C. Systematic Review and Meta-Analysis of Predictors of Return to Work After Spinal Surgery for Chronic Low Back and Leg Pain. J Pain. 2022 Aug;23(8):1318-1342. doi: 10.1016/j.jpain.2022.02.003. Epub 2022 Feb 18. PMID: 35189352.