Does reduced cough capacity in minor thoracic injury is associated with atelectasis development?


 Context: Minor thoracic injury causes non-negligible pain that could reduce the cough capacity and can cause infectious problems and atelectasis. Objectives: To describe the association between atelectasis and cough capacity, and to assess the concordance of cough capacity perceptions between health professionals and the patient. Design:The data were collected from 2006 to 2012 in 4 Canadian emergency departments (ED). Participants: Patients with a chief complaint of minor thoracic injury, ≥ 16 years old, discharged home from the ED after an evaluation by the attending physician, were included. They have three visits, at the ED, 7-day and 14-day follow-up. The initial ED visit included medical evaluation, data collection and mandatory chest and rib radiography. Main outcome measures: The presence of atelectasis was noted at the initial ED and subsequent visits.Participants cough capacity was noted by a physician, a nurse and the patients himself at subsequentvisits and classified as good, diminished, weak /absent.Results: Among the 1474 patients, 8.89% (95% CI: 7.55 - 10.47), 7.33% (95% CI: 6.04 – 8.89) and 4.63% (95% CI: 3.51 – 6.09) had atelectasis at the initial visit, 7-day and 14-day follow-up visit respectively. Except for patients with weak or absent cough capacity at the 7-day visit, which had a 2.89 (95%CI: 2.05 – 4.05) folds atelectasis proportion relative to those with a good cough capacity, they were no associations between cough capacity and atelectasis. The weighted kappa coefficient suggests a moderate to substantial agreement between the cough capacity measured by patients and nurses (0.52 to 0.65).Conclusion: There was not strong evidence of an association between atelectasis and cough capacity and the best agreement between cough capacity perception was between nurses and patients.


Introduction
Minor thoracic injury's(mTI) are de ned by the presence of chest abrasion, chest contusion or minor rib fractures (1,2). These injuries account for approximately 15% of all injuries treated in the emergency department, many of which will be treated on an outpatient basis (3). They have also been associated with a high risk of morbidity and mortality and complications such as delayed pneumothorax, hemothorax and pneumonia have been reported (1,4,5). It is also known that patients with mTIs and rib fractures can experience non-negligible pain (1,2,6,7), which can lead to reduced pulmonary function, decreased mucous clearance and decreased cough capacity (4,(8)(9)(10)(11)(12). Coughing is a normal mechanism of the body that allows the airway to remain free of foreign objects and to expel secretions produced by the lungs (10). When impaired, cough can cause infections and atelectasis (10,11,13). Atelectasis is a collapse of a major lung segment and can occur when the residual functional capacity is lower than the volume at which dependent airways close (14,15). Ineffective clearance of secretions is one of the predisposing factors of this pathology (14). Low tidal volumes also encourage the progressive lung collapse (14,15). Therefore, the pain syndrome caused by mTI can induce low cough capacity and Atelectasis. Thus, if it were possible to effectively assess the cough capacity of these patients in the emergency department (ED), this might help practitioners prevent atelectasis. To our knowledge, there is no study evaluating cough capacity and its relationship with atelectasis development in patients with mTI at the ED. Therefore, the primary objective of this study was to describe the association between cough capacity and the incidence of atelectasis in patients with a chief complaint of mTI. The secondary objective was to assess the concordance of cough capacity perception between health professionals and the patients themselves.

Methods
Setting: This is a secondary analysis of a prospective multicenter observational cohort study that took place between November 2006 and May 2012 in four Canadian university a liated emergency departments (ED),two of which were level I trauma centers, one level II trauma center and one academic non-trauma center (16).

Participants and recruitment:
Discharged patients aged ≥ 16 who presented to the ED with a chief complaint of mTI were included. Patients were excluded if they had: a con rmed hemothorax, pneumothorax, ail chest, lung contusion or any other important thoracic or internal abdominal injury upon initial visit. Patients who were unable to attend follow-up visits, unable to consent, and those who presented to the ED more than 3 days after the initial trauma were excluded as well.

Procedure and measures:
After informed written consent was obtained, an index medical evaluation was performed by the attending physician or delegate. Relevant sociodemographic and clinical data, such as age, sex, mechanism of injury, dyspnea, COPD, asthma, oxygen saturation and smoking status were collected and patients underwent a chest and rib radiography.
Patients returned for 7-and 14-day in person medical evaluations, during which they underwent followup chest radiographies to evaluate the presence of atelectasis and other complications such as pneumothorax or hemothorax. These evaluations also included pain assessment and clinical examinations performed by a physician and the data collected was recorded on a standardized data collection sheet.
Radiologists who were blinded to the study's objectives assessed each chest x-rays and determined whether a patient had developed atelectasis using the standard radiological criteria for its diagnosis. Cough capacity was assessed with a multiple-choice question (QCM) by the physician, a nurse and the patient himself at 7 and 14 days. The QCM was described as good/normal decreased, low/absent. Patients also self-assessed their chest pain using the visual Visual Analog Scale (VAS) pain (0: no pain. 10: extreme pain) at each visit.

Statistical analysis:
Multiple imputation with chained equation was used for imputing missing values of patient characteristics at the initial visit and cough capacity (17).We used censoring weights (IPCW)to account for the loss to follow-up. The aim was to weight each subject by their inverse probability of having each measurement observed (i.e. that they have not yet dropped out) (18).
These weights were then used in a generalized estimating equation GEE accounting for data clustering to compute relative proportion of atelectasis within each cough capacity level at each visit adjusted for age, sex, dyspnea and smoking history (19). To assess the agreement between the cough capacity perception according to the patient or health professionals, we used the weighted kappa coe cients (20). The following standards were used as strength of agreement, a kappa coe cient ≤0=poor, 0.01-0.20=slight, 0.21-0.40=fair, 0.41-0..60=moderate, 0.61-0.80=substantial, and 0.81-1=almost perfect (21).
All analyses were strati ed according to the presence or absence of atelectasis at the initial visit and statistical tests were adjusted for multiple comparisons. Rubin's rule was used to combined results across imputed datasets (22).

Ethics approval:
This study was approved by the research ethics committee of each institution.

Results
A total of 1474 patients were enrolled. Among those,14% (n=213) of patients did not attend their 7-followup appointment and 23% (n=341) at 14days.The mean age of participants was 54.48 years, male represent two third of the study population and fall from height was the most common mechanism of injury. All characteristics of participants are presented in Table 1.Among the 1474 patients,8.9% (95% CI: 7.55-10.47) had atelectasis at the initial visit. Accounting for losses to follow-up with IPCW, we estimated that the proportion of atelectasis was 7.3% (95% CI:6.04 -8.89)and 4.6%(95% CI:3.51 -6.09), respectively at 7 and 14 days. Table 2 presents association between cough capacity and atelectasis development. Compared with patients with good self-reported cough capacity, those with low or decreased capacity seemed to have a higher proportion of atelectasis. This pattern was higher at the 7-day follow-up for patients without atelectasis at the initial visit, relative proportions of 2.89 (95% CI 1.93 -4.33).
The cough capacity improves between the 7-and 14-day follow-up for patients with and without atelectasis. That improvement was mostly due to the fact patients with low or diminished cough capacity at 7 day had a good cough capacity at the 14-day visit (Table 3).We also observed a decreased in the means of pain at each follow-up (Table 4).
The weighted kappa coe cient suggests a moderate to substantial agreement between the cough capacity measured by patients and nurses (0.52 to 0.65); and a fair agreement between patients and physicians (0.33 to 0.42) ( Table 5).

Discussion
In this descriptive study, we did not nd a strong evidence of an association between atelectasis and cough capacity, except at the 7-day follow-up for patients without atelectasis at the initial visit. However, the 95% con dence intervals were wide, and we lacked precision to make any conclusion. In addition, we observed a decreased of the overall atelectasis proportions through the study follow-up, an improvement in cough capacity between the 7-and 14-day follow-up and a decrease in pain level. We also observed that the best concordance among cough capacity perception measures is between nurses and patients.
Our results are in accordance with the physiological phenomena that a decrease of pain can increase the ability to cough and decrease the risk of atelectasis (11)(12)(13).The large uncertainty regarding the association between cough capacity and atelectasis might be due to a lack of power, given that analyses were performed within strata of the presence of atelectasis at the initial visit. Several studies reported a moderate or good concordance among nurses and patients for the evaluation of different health questionnaires (23,24). Our results are in concordance with the literature While the QCM are quick and easy to use in medicine, the literature questions their validity since the answers given seem to be correlated to the level of prior knowledge(25).. Then QCM is an subjective evaluation and the gold standard to evaluate the cough capacity is the cough frequency assessment(26). This measure is often used in patients with neurodegenerative diseases, but they are time and resources consuming and are not practical in an emergency department setting. Moreover, one study showed a mild to moderate relationship between objective cough frequency and subjective measures of cough(27).It is therefore important to develop a valid, user-friendly and reliable patient questionnaire like Patient reported outcome measures (PROM) to evaluate the cough at the ED(28). PROMs allow patients to measure their functional well-being and health status. They are standardized and validated questionnaires. PROMs are recommended in the overall evaluation of the patient because they can help us understand if the services provided and the therapies really help the health and the quality of life of the patient(29, 30). Therefore, to increase the reproducibility of study evaluating cough capacity perception by the patient, PROM measures should be prioritized.

Limit and strength:
The association between cough capacity and atelectasis was a secondary analysis from a cohort study designed to build a clinical decision rule for delayed hemothorax after mTI (16). In addition, the measurement of atelectasis and coughing ability were done simultaneously, which may limit the interpretation of the association observed in this study.
In addition, losses to follow-up and missing data could have worsened the lack of power and introduced bias in the estimation of atelectasis proportions. However, using multiple imputations and IPCW may have helped mitigate those issues(17, 31).

Conclusion
There was not strong association between atelectasis and cough capacity, and a higher agreement between patients and nurse regarding cough capacity perceptions compared to physicians. Further studies could focus on developing a PROM questionnaire targeting patients with mTI as it is a common emergency department complaint, in order to improve medical management and patient quality of life.