Journal Club: “Reach” and Its Impact on Large Populations

Every Thursday afternoon we have journal club. This is an opportunity for the residents to sit down with a faculty member, present a research paper, discuss, and apply the findings to patient care.  A few weeks ago, a resident presented an article on the use of sublingual cyclobenzaprine (Flexeril) for the treatment of fibromyalgia.

As of today, fibromyalgia is a syndrome, not necessarily a disease. There is a muddled distinction, but essentially a syndrome, in most circumstances, does not have a known physiologic mechanism leading to progressive dysfunction.  The symptoms exist, a group of patients have it, and we don’t know why. The goal is to alleviate whatever ails the patient with pre-approved tools from other known diseases/syndromes.

Cyclobenzaprine hails from a group of medications known as muscle relaxers.  These drugs work on various parts of the neuro-muscular junction to reduce skeletal muscle contractility and, in theory, alleviate muscle related pain.  Fibromyalgia translates loosely into, fibrous pain.  So, muscles, fascia, connective tissue, etc, anything that forms from fibers.

Now, one does not tamper with the neuro-muscular junction without some downside. If a drug’s intention is to turn your skeletal muscle system into Jell-O, you’re not exactly going to be able to dunk a basketball.  Most likely, you’re going to sink into the couch and fall asleep. Naturally patients feel less pain, because they spend fewer hours awake and their bodies are forced to move in slow motion.

The article presented was a randomized control trial comparing cyclobenzaprine sublingual 5mg nightly to placebo in patients with fibromyalgia.  The authors used the classic definition of fibromyalgia which included assessment of tender points and a fibromyalgia scale that aided in the diagnosis.

The authors successfully recruited 250 patients for each arm of the study. After 14 weeks, the experimental group had a reduction of 0.4 points on a daily pain scale compared to placebo. There were also associated improvements in FIQR scores (Fibromyalgia impact questionnaire), and PROMIS scores (measure of well-being).

There was a 60% adverse event rate in the experimental group and a whopping 46% adverse event rate in the placebo group (mind you that is a sugar pill). The adverse event (AE) rate in the placebo group should provide some insight into the nature of this population. The most common AE in the experimental group was oral hypoesthesia (numbness and tingling of the mouth).

The authors concluded that sublingual cyclobenzaprine improves fibromyalgia related symptoms with a tolerable safety profile.  Which is true, statistically.

There are so many issues I have with this study, but for the sake of this piece I’m just going to focus on potential reach, or in business terms scale. What I mean by reach is how far this intervention will go within the general population. 

Fibromyalgia is a bugaboo for physicians.  It lacks any objective test to diagnose or track severity. There is no specific or sensitive physical exam finding, other than a high percentage of tender points along different muscle groups. The only reliable tool to determine treatment effectiveness are fibromyalgia symptom surveys.  These surveys are not practical in everyday use and were designed for research purposes, leaving the practitioner with subjective information provided during the office interview as grounds for decision making.

Fibromyalgia waxes and wanes but there is no progression to something sinister. It won’t kill you. It may go away. It may stick around. Who knows?

The objective during these office visits is to figure out something relatively safe, that can help with your symptoms, then move to the next patient.

Thai chi, therapeutic exercise, pain psychologists, are all proven therapies for fibromyalgia, but they are hard. They require action, intention, resources, and time before you get results. Given these barriers, most patients do not attempt, or abandon these interventions before they become effective.

I mention these characteristics for the simple fact that a quick “fix” to an issue like this will catch like wildfire. It’s desired by patients and practitioners. An electronic script is signed. It takes less than a minute to fill out. The doctor feels good because they offered some solution; the patient feels good because something happened that might help; the only burden is going to the pharmacy to pick up a pill to be taken at night. Everyone wins, nobody is better off.

This study looked at 500 highly specific individuals.  The classic fibromyalgia patient.  Let’s estimate the fibromyalgia population in the U. S. to be around 4,000,000.  That’s 8000x the number in the study. Let’s be kind at say 3,000,000 see a doctor on a regular basis complaining of symptoms (it’s probably a higher percentage). Given the low barrier for use, the only thing holding back cyclobenzaprine is awareness. Once a guideline gets published, or it makes its way onto an UpToDate treatment algorithm, this gap is likely to diminish quickly.

Knowing, for a fact, that cyclobenzaprine is not a placebo, and it is going to have some effect on patients, it’s important to weigh the benefit and risk in the context of the syndrome.

So, the syndrome limits quality of life, but not disabling or deadly. The benefit to the intervention is a few points on a pain scale. The reported downside, is tolerable. Some mouth numbness and slurred speech, but did the researchers know what to look for? Was their sample large enough to look for what really matters?

On a scale of this magnitude, what you didn’t look for, may be huge.  The goal of any patient engaging with the healthcare system should be to avoid certain ruin.  If the disease is to result in certain ruin (cancer, heart attack, stroke) the decision is easy, you take your chance with intervening on that process. If the disease is not to result in certain ruin, your decision is more difficult, downside becomes important.

Let’s say out of these 3,000,000 patients 33% of them are above the age of 65 (1,000,000). Of these 15% have osteoporosis (150,000).  Cyclobenzaprine disrupts the neuro-muscular junction affecting gait. Of those 150,000, the midnight fall rate on cyclobenzaprine is 15% per annum (20,000 falls per annum). Of those falls 10% result in a hip fracture.  That’s 2000 additional hip fractures.

This is just one scenario…its sinister. Its clandestine. Its plausible.

Let’s look at a different one…3,000,000 patients, 33% above 65 (1,000,000), majority of these are women (grandmas). Half of them fall asleep at their granddaughter’s dance recital (~400,000). They return to their doctor complaining of fatigue. All of them get at least some work-up (CBC, TSH, Sed rate, CMP,)Medicare $$$$). Half of them get a polysomnogram (sleep study $$$$) to screen for sleep apnea.  Since they’re on cyclobenzaprine most of these are positive (~150,000). Some are false positives, some are true positives, either way, it results in 150,000 new CPAP orders.

I could do this all day.

The point is, this will all be identified post hoc; that is after inertia has set in and cyclobenzaprine is making its way into medicine cabinets.  This happens for several reasons: 1) The researchers don’t have the resources or ability to run a trial at scale; 2) They don’t know what to look for; 3) There is no reliable way to track temporal, causative adverse events post hoc, and 4) They cannot differentiate signal from noise in low frequency events. Was it the cyclobenzaprine, or was it fate?

But as you can see, as reach increases, even something that occurs 1% of the time, will happen many times in a large population.

This is not some massive conspiracy theory to keep the hamster on the illness wheel (nobody wants you to be sick, despite what influencers on tiktok tell you). In large populations these are natural forces, they happen slowly, they are clandestine, they are noisy, and they are everywhere.

All for the sake of altruism, and a few points on a pain scale.