February 23, 2023

2 min read

The Wellcome Trust funded this study. Finnegan reports being a co-investigator on a provisional U.K. patent. Please see the study for all other authors’ relevant financial disclosures.

We were unable to process your request. Please try again later. If you continue to have this issue please contact [email protected].

A model that factors in brain imaging markers of breathlessness expectation, questionnaires and physiology measures predicted changes in breathlessness during pulmonary rehabilitation, according to study results published in Thorax.

“Brain activity to breathlessness-related cues is a strong predictor of clinical improvement in breathlessness over pulmonary rehabilitation,” Sarah L. Finnegan, PhD, postdoctoral research scientist in the department of clinical neurosciences at University of Oxford, and colleagues wrote. “This implies that expectation is key in breathlessness perception. Manipulation of the brain’s expectation pathways (either pharmacological or non-pharmacological) therefore merits further testing in the treatment of chronic breathlessness.”

Infographic showing percentages of accuracy, sensitivity and specificity for predicting responsiveness to pulmonary rehabilitation in a model that combined brain and behavior measures.

Data were derived from Finnegan SL, et al. Thorax. 2022;doi:10.1136/thorax-2022-218754.

Chronic breathlessness in COPD is often treated with pulmonary rehabilitation, although 30% of patients derive no benefit from this practice, according to the researchers.

To better predict improvements in breathlessness during rehabilitation, Finnegan and colleagues analyzed data from a randomized double-blind controlled experimental medicine study of D-cycloserine — a partial agonist at the N-methyl-D-aspartate, or NMDA, brain receptor that influences brain expectation mechanisms — of 71 patients (median age, 71 years; 18 women) with mild to moderate COPD.

Researchers measured responsiveness to pulmonary rehabilitation based on a change in Dyspnea-12 (D12) score of at least three points. They also sought to determine whether D-cycloserine had an effect on expectation.

Researchers trained machine learning models with baseline variables including brain activity, self-report questionnaire responses and physiology measures of respiratory function and drug allocation to predict this change in D12 score. They then evaluated the accuracy, sensitivity and specificity of three models: a model with only brain imaging markers of breathlessness expectation, a model with only self-report questionnaires and physiology measures, and a combination model.

Researchers randomly assigned patients to receive 250 mg oral D-cycloserine (n = 37) or placebo (n = 34) half an hour before each of the first four pulmonary rehabilitation sessions, with data collected at the start of rehabilitation and then again at its completion 6 to 8 weeks later.

Of the total cohort, 41 patients had a minimal clinically relevant change in D12 score, including 24 patients receiving D-cycloserine and 17 patients receiving placebo.

In terms of which model performed the best, researchers noted that only those that contained brain imaging markers of breathlessness expectation were successful in predicting response to pulmonary rehabilitation based on D12 score.

Specifically, the combination model had 0.83 (95% CI, 0.75-0.9) accuracy, 0.88 sensitivity and 0.77 specificity in predicting responsiveness to pulmonary rehabilitation (one-tailed binomial test of model accuracy compared with the null information, P < .001). Researchers noted that whether patients received D-cycloserine was a significant feature of this model.

The brain-only model showed statistical significance overall (P = .02) with promising sensitivity (0.93) and accuracy (0.7; 95% CI, 58-81) but lower specificity (0.4), suggesting it could not sufficiently distinguish between responders and nonresponders.

Lastly, the model with questionnaires and clinical measures alone did not reach statistical significance, with poorer accuracy (0.66; 95% CI, 0.54-0.77), sensitivity (0.68) and specificity (0.2).

“While larger sample sizes are now required to translate these mechanistic models into clinical relevance, the data provides evidence that breathlessness expectation-related brain activity at baseline strongly influences how patients respond to treatment in a predictable manner,” Finnegan and colleagues wrote.

Source link