Does frailty and sarcopenia affect outcomes in spinal oncology?

What is frailty? A lot of the time, our approach is ‘I know it when I see it.’ Yet it’s not always easy to define. Some patients may be old, but not frail, and it’s important that we are able to recognize those frail patients at risk of adverse outcomes. Frailty is an age-related syndrome. It reflects a decreased resilience, in the face of stress. Essentially what's happening is when you are frail, you don't bounce back. Let's say you have an illness or an injury. It decreases your function. Normally you would go back to your pre-level of injury function if you're not frail, but when you are frail, you never go all the way back. So, you can go from an independent state to a dependent state.

The challenges of measuring frailty and sarcopenia

Frailty is associated with higher age, but it's not equal to age. It's an unsuccessful model of aging. Which begs the question: how do we assess frailty? Frailty has been defined using two models: the first, a phenotypic model, and the second, a cumulative deficit model. Both models have been validated and used in large studies, and both of them have been correlated with death.

Under the Fried phenotypic model, we are looking at the physical characteristics of frailty. The criteria are weight loss, weakness, exhaustion, slowness, and no physical activity level. When you meet three of these criteria, you're considered frail.

On the other hand, the cumulative deficit model is the theory of a lifelong accumulation of deficit. This means, you can accumulate deficit in your life, but at some point, you'll reach a maximum, after which you will become frail. When we say deficit, we are talking about disease symptoms or signs, disability, laboratory value, or radiologic finding. Then you add the deficits. You take the number of your deficit, and you divide them by the total number of deficit considered, and you get a score going from zero to one. The frailer you are, the higher the score is going to be.

Subsequently was developed the modified frailty index, which has been used and validated from large NSQIP database. Essentially, it's only 11 of those deficits that are mapped to this modified frailty index. It’s important to understand that frailty is not a comorbidity score, but we do have to acknowledge that most people with disability are frail.

Those are the two main theories, but there’s a wide variety of frailty indices, which will take their rationale from the cumulative model or the phenotypic model. It's important also to examine the content of the frailty index you're looking at because some claim to be frailty index, but in fact are just risk prediction tools.

Sarcopenia and frailty are related, but sarcopenia will focus on the muscle loss and strength. How do you measure a sarcopenia? You can either go with a functional capacity—get the patient walking—or you could look at muscle mass. In general surgery, they usually look at the total abdominal musculature of the abdomen on a CT scan, while some people have moved on to only look at the psoas.

When you look at the psoas, the standard approach is to do a measure of the cross-sectional area of both psoas. You need to normalize it for height because this would change your psoas. Another way is to look at the total psoas area (right + left) over the vertebra body area at L3 or L4. This gives you a ratio that correlates with the normalized total psoas area but is simpler to measure.

Frailty and sarcopenia as a predictor of surgical complications

So, what has been done so far with frailty, sarcopenia, and spine surgery? In the degenerative population using the MFI, frailty was a predictor of major complications, re operation, discharge position, and mortality. This was achieved using the NSQIP database. We also published a systematic review on the impact of frailty and sarcopenia in post operative outcome in adult spine surgery. Across the board, frailty was a predictor of major and minor morbidity mortality, length of stay and discharge to a higher level of care. Interestingly, it's important to note that most studies that have been published regarding frailty and spine use this modified frailty index.

What about sarcopenia in the degenerative population? Well, there are contradictory results. In our degenerative lumbar case series, sarcopenia did not predict adverse events. Another study, published around the same time, showed that sarcopenia was a predictor of morbidity and mortality after surgery.

Why is this important in spinal oncology? We know that adverse events are quite frequent in oncology. Many papers have been published on adverse events in metastasis, and the adverse events are quite high. This is a vulnerable population. Even in the primary bone tumor population, which is expected to be healthier, the adverse event rate is still very high, going up to 79%. Unfortunately, there’s not a whole lot of literature today regarding frailty and sarcopenia in spinal oncology. In the metastatic spine population, there is one study which used the MSTFI. This is one of the types of index I mentioned previously, which is not really a frailty index. Regardless, they found that this index was predicting mortality, adverse event and length of stay using the MFI, in the metastatic spine population. It was not correlating with mortality at 30 days in one year. Recent validation study did not report the MSTFI to be a good predictor in the metastatic spine population.

When we used the MFI and MSTFI in our metastatic spine population, MFI was not a predictor of adverse event or mortality. The MSTFI was not a predictor of adverse events but was a predictor of mortality at three months. However, I'd like to point out that when you look at the MFI score of our population it was quite low, it's hard to believe that it would happen in a patient that is metastatic. Therese are patients who has been through chemo and radiation therapy, they’re really sick but score as non-frail on the MFI. This is a problem with the tool, not about being frail, and I think the result of our study reflects that.

We have also looked at this in our primary bone tumor population, and MFI was not a predictor of an adverse event. Operative duration was the predictor of length of stay, complications, and re-operation. The same group that created the MSTFI acknowledged that their tool was including items not related to frailty, and they have modified their scoring to use it in the primary bone tumor population. They showed that it was correlating with more adverse events and mortality. Further validation is needed.

Now if we look at sarcopenia in our metastatic spine population, a lower ratio of the total psoas/ vertebral body (the lowest quartile) was associated with increased mortality and also increased adverse events. In our population, the first quartile ratio associated with the bad outcome was 0.7298, but we cannot consider this a cutoff. This is a value that might guide us for now, but we need more studies about it. There's also a group in Michigan that has been really prolific looking at sarcopenia. They took a different approach to us. We did it at L3, they did it at L4; we used CT scan, they used MRI. However, the results are similar, meaning that patients with the lowest quartile were having increased mortality. This was a nonsurgical cord, but it was a metastatic spine patient.

More research is needed

So where do we go from now on? For frailty, we know that frailty does predict mortality and length of stay, and this in multiple spine population. But is the MFI the right tool? I think in the spine oncology population, it does not apply due to the fact that it doesn't capture really how those patients are. Moving forward, we should probably focus more on phenotypic definitions, such as those defined by the fried frailty phenotype—which is kind of cumbersome—or just go with the simple clinical frailty scale. Another alternative would be to develop a metastatic spine frailty index that would capture the unique characteristic of this vulnerable population.

For sarcopenia, we identified a predictor of adverse events and mortality, but we need to investigate further to figure out which modality is the best imaging method. Maybe the value or the cutoff point for predictive mortality or adverse events might be different. Again, which levels should we use? This has never been used in the primary bone tumor population.

What we do know, is that frailty and sarcopenia represent a state of reduced resilience to physiological stress. This may occur out of proportion to chronological age. We know that in spine, frailty has been established as a predictor for the first event. In oncology, the role of frailty and sarcopenia warrants further investigation.

***

You can help us move this research forward by sharing your views and practices regarding frailty and spine metastatic disease. The AO Spine Knowledge Forum Tumor aims to analyze the perceptions, opinions, and practices of spine surgeons about frailty regarding spine metastatic disease and identify what spine surgeons consider to be relevant preoperative factors and postoperative outcomes.

Access the survey here.

The results may be used to build a new preoperative assessment tool which could help in evaluating metastatic spine tumor patients’ frailty and candidacy for relevant spine surgeries. The results will be published in a peer-reviewed journal and appear on the AO Spine website.