Introduction

We know that factors of the biopsychosocial model almost always influence pain. Yet surgical treatments mostly focus on correcting or repairing damaged structures, such as rotator cuff tears. Since the outcomes after rotator cuff repair show high variability – with some people largely improving while others remain with pain and deficits – selecting individuals who might benefit from rotator cuff surgery is paramount. As there is no relationship between tear size and pain intensity (Dunn et al., 2014) (Pietroski et al. 2022), more factors are in play. This study wanted to examine factors beyond the biomedical features for making a solid rotator cuff repair prognosis. It considered psychosocial factors, sleep, and central pain processing to make a prognosis about recovering from pain and improving function and quality of life.

Methods

This observational, longitudinal study included patients undergoing first-time rotator cuff repair at a Swiss hospital. Patients older than 18 were included if they were scheduled for their first rotator cuff surgery.

A wide array of candidate prognostic factors were included. A stepwise reduction after interim analyses took place to retain the significant factors in the model. As such the study analyzed multiple modifiable and non-modifiable factors, including but not limited to:

• Psychosocial factors: Pain catastrophizing, perceived stress, and patient expectations.
• Sleep-related variables: Sleep quality and sleep efficiency.
• Central pain processing: Central Sensitization Inventory (CSI) scores, cold hyperalgesia, and pressure pain thresholds.

The primary outcome was the Western Ontario Rotator Cuff Index (WORC) for pain, function, disability, quality of life, and emotional health. Secondary Outcomes included the Subjective Shoulder Value (SSV) for shoulder function and the EuroQol EQ-5D-5L for overall quality of life.

The study aimed to determine how psychosocial factors, sleep-related indices, and proxies of central pain processing, along with age, sex, and body mass index, influenced postoperative outcomes at 12 weeks and 12 months. The preoperative measures were taken at 1-21 days before surgery.

Results

142 participants were included in the study. Their baseline characteristics are displayed in the table here under. An equal proportion of people had isolated supraspinatus tendon or combined supraspinatus and infraspinatus tendon tears. A minority (11%) had multi-tendon tears affecting the supraspinatus, infraspinatus, and subscapularis tendons. The same proportion (10%) of combined supraspinatus and subscapularis tendon tears was present at baseline. Isolated subscapularis tears occurred the least commonly (6%). Almost half of the participants had sustained a traumatic tear (56%) while the other half had nontraumatic tears (44%).

The model identified five significant prognostic factors influencing recovery. To establish a rotator cuff repair prognosis for the primary outcome WORC at 1 year following surgery, the following variables should be considered :

• Expectations of symptom reduction at 1 year: Higher expectations for improvement were linked with better outcomes in WORC scores over time.
• Neuropathic Pain (DN4): Higher scores on the Douleur Neuropathique 4 (DN4) scale were associated with worse WORC scores.

Three significant prognostic factors remained for the secondary outcomes:

• Injury Perception (IPQ-R Consequence Subscale): Patients who perceived their injury to have a significant negative impact on their daily life had worse shoulder function (SSV) scores.
• Sleep Quality: Better sleep quality was linked to improved overall quality of life (EQ-5D-5L) and emotional function.
• Sleep Efficiency: Higher sleep efficiency, measured as the ratio of time asleep versus time in bed, predicted better quality of life outcomes.

Questions and thoughts

When establishing a rotator cuff repair prognosis, the expectations about surgery to relieve complaints were linked to a better outcome on symptoms, shoulder function, and related quality of life, as reflected by better WORC outcomes. This is understandable from a more biomedical view of the patient. When a patient expects surgery to repair the tendon and “fix” their shoulder issue, they will likely improve significantly after surgery. On the other hand, patients who have negative expectations, or who feel their shoulder problem is completely irreparable, will have less improvement in the WORC outcomes of shoulder symptoms, function, and quality of life. In these patients, good counseling should be prioritized to alter their expectations, but when no shift in expectations can be reached, the question arises whether surgery is the right treatment for them.

The presence of pain catastrophizing, central sensitization based on the CSI, and the pain distribution/surface of the pain were prognostic factors that were retained in the model based on statistical significance but their confidence intervals indicated nonsignificance by spanning zero. The authors indicate that these factors may potentially hinder positive expectations and hence the trajectory of the WORC outcomes.

Neuropathic pain features influence the course of the improvements in the primary outcome WORC. Previously, a cutoff score of 4 was determined to indicate the presence of neuropathic pain features and studies confirmed a doubled risk of prolonged acute postoperative pain in case patients with these characteristics were operated on.

Talk nerdy to me

Establishing a rotator cuff repair prognosis can be difficult. Using the framework this study shared, we can improve our prognostic reasoning. In establishing a prognosis, this study emphasized the importance of early detection of psychosocial and sleep-related disorders.

It was valuable that the study started with a large set of candidate prognostic factors rather than a small subset. Often, in prognostic research, a fixed set of variables is studied or an approach of forward selection is used. The stepwise reduction as was used in the current study is less prone to bias. A limitation lies in the fact that the study did not account for non-modifiable characteristics such as tear size, which can affect outcomes. Further, self-reported sleep quality and efficiency may be biased.

The pragmatic approach made the integration into real-world settings easier by using a longitudinal design, which makes the results applicable to surgeons and physiotherapists. Furthermore, selection bias was avoided since the data from consecutive patient consultations were used. Another good aspect of this study was the use of 3 different models to study the primary outcome WORC and the secondary outcomes SSV and EQ-5D-5L separately.

Take home messages

To make an accurate rotator cuff repair prognosis at one year, expectations, neuropathic pain characteristics, injury perception, and sleep should be questioned. These factors are modifiable and we can use the results to improve preoperative assessments by identifying patients at risk of poor recovery due to psychosocial and sleep problems. Interventions like cognitive-behavioral therapy (CBT) for pain catastrophizing and sleep hygiene education may enhance outcomes. Furthermore, educating patients about central sensitization and managing neuropathic pain features before surgery may assist in setting realistic expectations and possibly boost postoperative recovery.

Reference

Schwank A, Struyf F, Struyf T, Mertens M, Gisi D, Benninger E, Meeus M. Psychosocial Factors, Sleep, and Central Pain Processing for Making a Prognosis About Recovery of Pain, Function, and Quality of Life After Rotator Cuff Repair: An Exploratory Longitudinal Study. J Orthop Sports Phys Ther. 2024 Aug;54(8):530-540. doi: 10.2519/jospt.2024.12398. PMID: 39096935.