Thalidomide for the Treatment of Cough in Idiopathic Pulmonary Fibrosis: A Randomized Trial

Dear Editor:We read with interest the article by Horton and colleagues on thalidomide for IPF-related cough.(1) Besides dyspnea and fatigue, cough is a hallmark symptom—for a significant minority of IPF patients, cough is most debilitating. IPF patients describe an ever-present “nagging” sensation but “never feeling relieved after coughing.”(2) In our unpublished work, IPF patients describe cough when they “talk too long,” breathe “hard” and/or “fast” or when their oxygen saturation is low. Cough can occur in frightening paroxysms, cause significant oxygen desaturation and leave patients “wiped out.” The potential for cough to profoundly affect IPF patients’ lives makes these investigators’ results all the more impressive. Their statistical models would suggest effect sizes > 0.6—and > 0.9 for the primary outcome. While recognizing the goal was to test thalidomide and not necessarily to elucidate how cough impacts IPF patients, their data could prove useful there. On the face of it, many items from the CQLQ appear relevant to IPF patients, but many do not. Might the authors provide more detail on scores from specific CQLQ domains? Did one or more of the six CQLQ domains drive the change scores in response to thalidomide? We are also curious about the wording for the cough VAS? Did it ask about cough “frequency”? “Affects on daily life?” “Impact on quality of life?” Data for the SGRQ are interesting. In most IPF trials that included SGRQ, the Activity domain—which assesses physical activities that either cause or are limited by dyspnea(2)—is most impaired. Possibly because this cohort consisted largely of self-referred subjects, presumably because their coughs were so distressing; here, the Symptoms domain—items on cough, phlegm, wheezing and others—was most impaired. It may prove informative to drill into the Symptoms domain to see where thalidomide had its effects. The absence of change in Activity domain would suggest thalidomide had no effect on dyspnea. The baseline score from the Impacts domain—assesses effects of respiratory disease on daily life, including psycho-social functioning—suggests greater impairment than in other IPF treatment trials. Presumably, when responding to Impacts items, subjects focused on the effects of cough.Whatever the case, the results of this study are exciting. They pave the way for investigation into the mechanistic pathways of IPF-related cough, into discerning more precisely how cough impacts IPF patients’ lives and they provide the impetus for large-scale, systematic assessments of therapies for this vexing symptom.

References

1. Horton MR, Santopietro V, Mathew L, Horton KM, Polito AJ, Liu MC, et al. Thalidomide for the treatment of cough in idiopathic pulmonary fibrosis: a randomized trial. Ann Intern Med. 2012;157(6):398-406.

2. Jones PW, Quirk FH, Baveystock CM. The St George's Respiratory Questionnaire. Respir Med. 1991;85 Suppl B:25-31; discussion 3-7.

We appreciate Dr. Swigris and colleagues’ interest in our article and would like to respond to their questions regarding the Cough Quality of Life Questionnaire (CQLQ)(1), the visual analog scale (VAS) of cough, and the St. George’s Respiratory Questionnaire (SGRQ)(2). The CQLQ is a 28-item instrument developed and validated in a cohort of patients with idiopathic cough, and as such contains some questions that are not relevant to patients with IPF. However, the CQLQ contains many questions that are applicable to virtually anyone with a severe cough. There are six subscales of the CQLQ; Physical Complaints, Psychosocial Issues, Functional Abilities, Emotional Well-Being, Extreme Physical Complaints, and Personal Safety Fears. In our study four of the subscales improved significantly after treatment with thalidomide. The two subscales that did not improve significantly were the Emotional Well-Being and Personal Safety Fears subscales, which both contain statements that are not pertinent to patients with IPF, such as “I want to be reassured that I do not have anything seriously the matter with me.” The cough VAS we used was meant to be broad and was worded “Considering all the ways your cough affects you, on the average, how have you been doing in the past week?” The anchors on the VAS were “No Cough” and “Worst Cough.” Therefore it does little to assess cough frequency or its impact on QOL. The SGRQ is respiratory specific QOL instrument developed for use in asthma and COPD, but it has been used in a wide variety of lung diseases. Many of the items in the SGRQ are dichotomous and therefore are therefore not particularly sensitive to subtle changes in symptoms. Additionally, our study was not designed to look at individual questions from the SGRQ. With these limitations in mind, there were several questions that showed statistically significant improvement in dyspnea after thalidomide treatment. The proportion of people who stated that sitting or lying still made them feel breathless decreased from 10% to 5%, (p=0.002). The proportion who felt breathless when bending over decreased from 68% to 42%, (p=0.012). Unfortunately, looking into responses on the CQLQ and SGRQ in more detail does little to further our understanding of the mechanistic pathways of IPD-related cough. Nevertheless, we share Dr. Swigris’ excitement about the ability of thalidomide to reduce IPF-related cough and hope to see a subsequent, larger clinical trial of thalidomide for IPF.

Reference List

(1) French CT, Irwin RS, Fletcher KE, Adams TM. Evaluation of a cough-specific quality-of-life questionnaire. Chest 2002; 121(4):1123-1131.

(2) Jones PW, Quirk FH, Baveystock CM. The St George's Respiratory Questionnaire. Respir Med 1991; 85 Suppl B:25-31.