Quantifying breathlessness following lung resection – concordance between scoring tools

Abstract

Introduction Dyspnoea is the subjective feeling of breathing discomfort and is the most important but least understood symptom in cardiorespiratory disease. Patient and public involvement from our research group confirms this finding. The Medical Research Council Dyspnoea scale (MRC scale) is used to measure dyspnoea based on the work load required to provoke breathlessness. It is validated to predict outcome in patients with chronic obstructive pulmonary disease (COPD) however broad grading categories mean clinically important changes may be missed. The minimally clinically important difference (MCID) for the MRC scale is 1. The University of California and San Diego Shortness of Breath Questionnaire (UCSD-SOBQ) is another validated tool in respiratory conditions to measure dyspnoea with a maximum score of 120 and a MCID of 5. To date, no validated scoring tool (nor MCID) exists to quantify dyspnoea in the lung cancer population despite strong evidence this population have a high incidence of breathlessness both before and after surgery.1 The aim of this analysis is to compare tools used to measure breathlessness following lung resection for cancer and assess the concordance between these. Methods This study is a secondary analysis of PROFILES study (NCT03888937) exploring the prediction of dyspnoea following lung resection. With informed consent and ethics approval we prospectively recruited 107 patients undergoing lung resection in a single centre. The MRC scale and UCSD-SOBQ were completed by each patient pre-operatively and 3 months post-operatively. Data are displayed as median (IQR). A four-quadrant plot was created and direction of change analysis performed for concordance. Wilcoxon signed rank test was used to compare scores pre-operatively vs 3 months post-operatively. Results At 3 months 75 patients returned questionnaires for comparison of pre and post-operative breathlessness (return rate 77%). Median MRC dyspnoea was reduced post-op (2(2-3) compared to pre-op (2(1-2), p=0.01) as was median UCSDSOBQ post-op(24(11,56) compared to pre-op (8(2,24), p=0.01. Concordance for detecting a MCID change in dyspnoea score was 59%. Sixty-one% of patients reported a MCID change in MRC score post-op vs 81% reporting a MCID change in UCSDSOBQ, p=0.02. Discussion This study demonstrates an increase in dyspnoea following lung resection, irrespective of scoring tool. When using questionnaires to measure dyspnoea it is important to ascertain the MCID. No validated questionnaire exists for the lung cancer population. Validated self-assessment tools to measure dyspnoea in other populations have moderate concordance when describing a change in dyspnoea level following lung resection surgery. Future studies should focus on; validation of a tool to measure dyspnoea, reliable methods to predict dyspnoea and a greater understanding of the patho-physiological mechanisms involved.