Multidimensional analysis of the clinical spectrum and symptom burden of unexplained myofascial pain

This prospective cross-sectional study demonstrates that symptom burden, particularly physical function and pain catastrophizing, explains the majority of variability in pain intensity and interference among adults with unexplained neck/shoulder myofascial pain, suggesting that multidimensional network analysis can improve clinical risk stratification and shift treatment focus from regional pain contributors to the whole patient.

The Gist

The Big Picture: Why Does Pain Feel Different for Everyone?

Imagine you have a headache. For one person, it's a dull throb that makes them want to nap. For another, it's a blinding lightning bolt that makes them nauseous and anxious. Even though the "headache" is the same thing, the experience is totally different.

This study is about Myofascial Pain (MP). Think of this as those annoying, deep "knots" in your muscles (like in your neck or shoulders) that doctors can sometimes feel when they press on them.

For a long time, doctors have treated these knots like a simple mechanical problem: "Find the knot, press it, fix it." But this paper argues that this approach is like trying to fix a car engine by only looking at the spark plugs, while ignoring the fact that the driver is stressed, the fuel is bad, and the tires are flat.

The researchers wanted to know: Is the "knot" the whole story, or is the whole person's body and mind involved?

---

The Experiment: Three Groups of People

The researchers recruited 82 adults and split them into three groups based on a physical exam:

1. The "Active" Group: These people had knots that hurt all the time, even when they weren't touching them. They were in the most pain.
2. The "Latent" Group: These people had knots, but they only hurt when a doctor pressed hard on them. They didn't hurt on their own.
3. The "Normal" Group: These people had no pain and no obvious knots. (Surprise! Even this group had some "knots" when the doctors pressed, but they didn't feel pain from them).

The Twist: The researchers didn't just look at the muscles. They asked everyone about their sleep, their anxiety, their depression, how much they "catastrophize" (worry that the pain will never end), and how flexible their joints were.

---

The Findings: The "Symptom Burden" is the Real Boss

Here is the big discovery, explained with an analogy:

The "House" Analogy:
Imagine your pain is a house on fire.

• The Old Way: Doctors only looked at the flames (the muscle knot). They thought if they put out the flame, the house was safe.
• The New Way: This study found that the fire is actually being fed by a whole ecosystem: the wind (anxiety), the dry wood (poor sleep), and the lack of a sprinkler system (low physical function).

What the Data Showed:

1. It's Not Just the Knot: The presence of a muscle knot didn't perfectly predict how much pain a person felt. Some people with "knots" felt fine; others felt terrible.
2. The "Symptom Burden" Explains Everything: The researchers found that if you look at the whole person (their sleep, their mood, their physical strength, and their worry levels), you can explain 75% to 92% of why their pain is bad.
   • In the "Active" group (the worst pain), Pain Catastrophizing (the fear that pain is the end of the world) was the biggest driver.
   • In the "Normal" group (no pain), Physical Function (how well they could move) was the biggest factor.
3. The "Network" of Pain: The researchers used a special computer map (Network Analysis) to see how these factors connected.
   • In the "Active" group: It was a tangled web. Anxiety, depression, poor sleep, and bad physical function were all holding hands and making the pain worse. It was like a domino effect where one bad thing triggered the next.
   • In the "Latent" group: The web was much simpler. It was mostly just about anxiety and depression.
   • In the "Normal" group: There was almost no web at all. The factors weren't connected.

---

Why This Matters: A New Way to Treat Pain

The Old Approach:
"Your neck hurts? Let's massage the knot."

• Result: Sometimes it helps, sometimes it doesn't. It's a hit-or-miss game.

The New Approach (Based on this paper):
"Your neck hurts. Let's look at the whole system."

• If your pain is driven by anxiety and sleep, we need to fix your sleep and calm your mind, not just massage your neck.
• If your pain is driven by low physical strength, we need to build your body up.
• If your pain is driven by worrying too much, we need to teach you how to stop the "catastrophizing" thoughts.

The Takeaway

This paper is a call to stop treating pain like a broken part in a machine. Instead, we should treat it like a complex garden.

If your garden is full of weeds (pain), you can't just pull one weed (the knot). You have to check the soil (sleep), the water (stress), and the sunlight (mood). If you fix the whole ecosystem, the weeds are much more likely to go away.

In short: To fix chronic pain, we need to stop looking just at the "knot" and start treating the whole person.

Technical Summary

1. Problem Statement

Chronic musculoskeletal pain (cMSKP), particularly myofascial pain (MP), is a leading cause of global disability. However, the field faces significant challenges:

• Diagnostic Ambiguity: There is no definitive test for MP. Diagnosis relies on clinical assessments of myofascial trigger points (MTrPs), which suffer from low to moderate reproducibility and reliability.
• Regional vs. Systemic Focus: Traditional approaches focus on regional anatomy (e.g., neck/shoulder) and local physical findings (MTrPs), failing to capture the complex biopsychosocial nature of the patient's lived experience.
• Unexplained Variability: There is high variability in pain intensity and interference among patients with similar physical findings. It remains unclear whether MP is a distinct regional syndrome or a component of a broader non-specific cMSKP spectrum influenced by central and peripheral mechanisms.
• Lack of Holistic Models: Current treatment paradigms often fail because they do not account for the interplay between physiological, functional, and psychosocial determinants.

2. Methodology

Study Design:

• Type: Prospective cross-sectional study.
• Setting: George Mason University and surrounding healthcare practices (Virginia/DC/Baltimore area).
• Participants: 96 adults recruited (Jan 2023–Dec 2024). 82 had complete data for analysis.
  • Inclusion: Adults (≥18) with persistent/episodic neck/shoulder pain (≥3 months) or pain-free controls (no pain in past 5 years).
  • Exclusion: Fibromyalgia, radiculopathy, recent surgery, medication changes, or fractures.
• Group Classification: Based on physical examination (Travell and Simon's criteria), participants were classified into three whole-person phenotypes:
  1. Active MP: Spontaneous pain without provocation ($n=23$).
  2. Latent MP: Pain reproduced only upon provocation/palpation ($n=38$).
  3. Normal: Asymptomatic with no provokable pain ($n=21$).

Outcome Measures:

• Primary Dependent Variable: Pain, Enjoyment of Life, and General Activity (PEG) score.
• Predictors (Symptom Burden):
  • Psychosocial: Pain Catastrophizing Scale (PCS), PROMIS Physical Function (PF), Sleep Disturbance, GAD-2 (Anxiety), PHQ-2 (Depression).
  • Physical/Clinical: Modified Beighton-Brighton score (Hypermobility), Pressure Pain Threshold (PPT), Quantitative Sensory Testing (QST) for wind-up.
• Statistical Analysis:
  • Linear Regression: To determine if symptom burden explains the variance in PEG. Models were adjusted for sex, age, and pain group.
  • Network Analysis: Graphical LASSO (Least Absolute Shrinkage and Selection Operator) was used to construct partial correlation networks. This identified conditional dependencies (edges) between variables (nodes) to map the structure of symptom interactions within each group.

3. Key Contributions

• Paradigm Shift: Moves the focus from regional MTrP identification to a whole-person biopsychosocial model. The study demonstrates that regional physical findings alone cannot explain the variability in patient-reported pain outcomes.
• Phenotype Differentiation: Provides evidence that "Active," "Latent," and "Normal" groups exhibit distinct network topologies and predictor profiles, suggesting they are mechanistically different despite overlapping physical findings.
• Methodological Innovation: Successfully applies Network Analysis to MP, revealing unique symptom clusters and interconnections that traditional regression cannot detect.
• High Predictive Power: Demonstrates that a multidimensional set of patient-reported outcomes and objective measures can explain a vast majority (75–92%) of the variance in pain intensity and interference.

4. Key Results

Demographics & Overlap:

• The sample had a mean age of 32.2 years (57% female).
• Crucial Finding: There was significant overlap in PEG scores between the "Latent" and "Normal" groups. Many subjects classified as "Normal" by physical exam criteria reported significant symptom burdens, indicating that physical exams alone are insufficient for stratification.

Regression Analysis (Predicting PEG):

• Overall Sample: The model explained 75% of the variance in PEG.
  • Key Predictors: Physical Function (Partial $R^2$ = 25.6%) and Pain Catastrophizing (PCS) (Partial $R^2$ = 16.7%).
• Active Group: Explained 85% of variance.
  • Dominant Predictor: PCS (Partial $R^2$ = 37.5%).
• Normal Group: Explained 92% of variance.
  • Dominant Predictors: Physical Function (70.7%), PCS (41.9%), Hypermobility (35.6%), and GAD-2 (34.4%).
• Latent Group: Explained only 46% of variance.
  • Dominant Predictor: PHQ-2 (Depression) (Partial $R^2$ = 20.2%). This group showed the weakest correlation between predictors and pain outcomes.

Network Analysis:

• Active Group: Showed a complex, highly connected network (median 14 edges).
  • Key Links: Physical Function was negatively correlated with Hypermobility and PEG. PCS was positively correlated with PEG and Anxiety (GAD-2). This suggests a "high burden" phenotype where physical, emotional, and sensitization factors are tightly interlinked.
• Latent Group: Showed a sparse network (median 4 edges).
  • Key Links: Weak connections between Physical Function and PCS; PHQ-2 and GAD-2 remained linked.
• Normal Group: No reliable non-zero edges were found, indicating a lack of strong conditional associations among variables in this group.

5. Significance and Implications

• Clinical Risk Stratification: Network analysis offers a new tool for stratifying patients. The "Active" profile, characterized by a dense network of interconnected symptoms (catastrophizing, anxiety, poor function, hypermobility), likely requires earlier, more intensive, multidisciplinary intervention compared to the "Latent" profile.
• Treatment Guidance: The findings argue against siloed, regional treatments (e.g., only treating the trigger point). Instead, interventions should target central influencers identified in the network (e.g., addressing pain catastrophizing or physical function) to disrupt the cycle of pain.
• Research Direction: The study supports the hypothesis that MP is an emergent property of a complex system involving central sensitization and the "myofascial unit." Future research should focus on longitudinal studies to determine causality and directionality within these networks.
• Diagnostic Refinement: The high explanatory power of biopsychosocial measures suggests that diagnostic criteria for MP should evolve to include whole-person symptom burden rather than relying solely on the unreliable identification of local trigger points.

Conclusion:
The study concludes that the variability in unexplained neck/shoulder myofascial pain is best explained by a multidimensional symptom burden. Network analysis reveals that active MP is a distinct, highly interconnected biopsychosocial state, necessitating a shift from regional pain management to holistic, mechanism-informed clinical guidance.