If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Address correspondence to: Dr Tracy A. Smith, PhD, BSc, MBBS, Department of Respiratory and Sleep Medicine, PO Box 533, Wentworthville, 2145, Australia.
UNSW Sydney, Faculty of Medicine, St Vincent's Clinical School, Sydney, AustraliaDepartment of Respiratory and Sleep Medicine, Westmead Hospital, Sydney, Australia
Noninvasive ventilation (NIV) is commonly used to manage acute respiratory failure due to decompensated cardiorespiratory disease. We describe symptom burden in this population.
Measures
Fifty consecutive, consenting, English-speaking, cognitively intact patients, admitted to wards other than the intensive care unit in a tertiary teaching hospital and treated with NIV for hypercapnic respiratory failure, were recruited. The 14-item Condensed Memorial Symptom Assessment Scale was used to assess physical and psychological symptoms within 36 hours of commencing NIV. Breathlessness (using Borg score), pain location and intensity using a numerical rating scale, and four symptoms potentially prevalent in patients undergoing NIV (cough, sputum, gastric bloating, and dry eyes) were also assessed.
Outcomes
Patients reported a median of 10 symptoms (IQR 9–13). A median of five symptoms (IQR 3–7) were rated as severe. Breathlessness was the most prevalent and most distressing symptom, with participants reporting a mean maximum Borg score of 7.55 over the 24 hours before admission. Dry mouth, lack of energy, cough, sputum, difficulty sleeping, and psychological symptoms were prevalent. Pain, when reported, was of moderate intensity and contributed to distress.
Conclusions/Lessons learned
This study describes the patient-reported symptoms occurring during an episode of acute respiratory failure. Understanding the symptom profile of patients in this setting may allow clinicians to target symptom relief while simultaneously managing respiratory failure, enhancing care.
NIV improves gas exchange and can “buy time,” whereas other treatments, such as bronchodilators, corticosteroids, diuretics, and/or antibiotics, as appropriate to the underlying illness, reverse acute physiological disturbances.
Although life-saving measures are crucial, symptom assessment and management are also important.
Symptom burden is inherently subjective and may be assessed by validated, patient-reported outcome measures. Understanding the epidemiology of symptoms in specific populations serves to enable development of strategies to ameliorate distress and reduce suffering.
Understanding breathlessness: cross-sectional comparison of symptom burden and palliative care needs in chronic obstructive pulmonary disease and cancer.
Worsening symptoms are a common trigger for hospital presentation and are intrinsic to the diagnosis of exacerbations of chronic obstructive pulmonary disease (COPD)
The prevalence and associated distress of physical and psychological symptoms in patients with advanced heart failure Attending a South African medical center.
The Severe Respiratory Insufficiency Questionnaire scored best in the assessment of health-related quality of life in chronic obstructive pulmonary disease.
However, the overall symptom burden in patients with respiratory failure requiring treatment with NIV has not been reported to our knowledge.
The aim of this study was to describe the symptom burden of patients admitted to hospital with acute hypercapnic respiratory failure who were treated with NIV.
Methods
We describe a prospective, cross-sectional study of patients with hypercapnic respiratory failure (confirmed on arterial blood gas sampling) requiring NIV. This study was undertaken in a tertiary teaching hospital in Australia. Data were collected between January 2010 and December 2011. Ethical approval for the study was obtained from the South Eastern Sydney Illawarra Area Health Service Human Research Ethic Committee–Central Network (approval number 09/STG/128). Written informed consent was obtained from all participants. The Strengthening The Reporting of OBservational studies in Epidemiology guidelines were used to guide development of this article.
English-speaking patients were offered participation in the study in a time frame of 12–36 hours after starting NIV. Participants were excluded if they had a documented history of dementia, were unable to understand the consent process, declined consent, or were unable to communicate. In the study institution, NIV was administered in the emergency department, high dependency unit, and the respiratory ward. Participants treated in the intensive care unit (ICU) were excluded. Fifty patients completed the study. Details of enrollment and exclusions are outlined in Figure 1. Patients most often declined participation because they felt too unwell or were uninterested.
Fig. 1Details of screening and reasons for exclusion.
The CMSAS was administered verbally to patients as the population was frail and some were wearing NIV masks at the time making reading difficult. The patients were asked if the 11 physical symptoms listed on the CMSAS had been present in the last 72 hours. If the symptom was present, they were asked how much the symptom bothered them—“not at all,” “somewhat,” “quite a bit,” or “very much” as per the CMSAS. Participants were then asked about the presence of three psychological symptoms and how frequently they had experienced these over the last 72 hours. The scoring process for this instrument is outlined in Table 1. As suggested by Portenoy et al.,
physical symptoms rated as “quite a bit” or “very” bothersome and psychological symptoms present “frequently” or “almost constantly” were rated as “severe.” We also asked patients to rate the bothersomeness of four additional symptoms potentially prevalent in patients treated with NIV, namely cough, sputum production, dry eyes, and gastric bloating using the same process as other physical symptoms on the CMSAS and the same cut points to rate severity. These additional symptoms were selected by an expert panel consisting of medical, nursing, and physiotherapy staff familiar with NIV.
Table 1Scores for Patient Symptom Descriptors for the Condensed Memorial Symptom Assessment Scale (CMSAS)
The CMSAS summary score is the mean of all 14 symptom scores, whereas the physical and psychological subscale scores are the mean scores of the physical and psychological scores, respectively.
Breathlessness and pain were further assessed. Patients were asked to rate their breathlessness using the modified Borg scale (0–10, higher score = worse breathlessness).
Patients were asked to report their breathlessness at the time of assessment, and their highest and lowest score over the preceding 24 hours. Patients were asked if they had chest pain, nasal bridge pain (i.e., where the NIV interface often causes irritation), or pain elsewhere. Patients responding positively were asked to rate the severity of any reported pain site on a 0–10 numerical rating scale. In addition, the bothersomeness of pain at each pain site was collected using the same scoring system as the CMSAS.
Other data collected included the patient's usual level of breathlessness before admission, using the modified Medical Research Council scale
Clinical data, including arterial blood gas measures and respiratory rate, were collected either before or within 30 minutes of NIV initiation. Hospital length of stay was collected from hospital records. The cause of patients' respiratory failure and total time spent on NIV were assessed by the first author at the end of hospitalization using all available clinical information in patients' medical record.
Descriptive statistics of clinical and demographic data were calculated. Correlation between CMSAS summary score using either Spearman (ordinal variables) or Pearson (interval or ratio variables) with demographic and clinical variables was performed. GraphPad Prism, version 6.07, (GraphPad Software, 2016) was used for statistical calculations.
Results
The characteristics of the population are outlined in Table 2. The mean age of participants was 72.0 ± 13.1 (mean ± SD) years, and 32/50 were female. Participants were assessed a mean of 23.3 ± 8.69 hours after starting NIV. At presentation, the study population had moderate-to-severe respiratory failure (mean pH 7.29 ± 0.08; mean PaCO2 58.3 ± 12.7 mm Hg) and moderate illness severity (mean APACHE II score 17.6 ± 5.93). No patients were intubated or transferred to the ICU during the study; five were intolerant of NIV but were included as they tolerated at least one hour of NIV; and three of the cohort died during hospitalization in the time period after initial assessment.
Fraction of inspired oxygen (FiO2) was highly variable. At the time the ABG was taken, the FiO2 was not recorded for two patients; 22% were breathing room air; 40% were receiving an FiO2 of greater than 40%, with the remainder receiving an FiO2 between 21% and 40%.
85.2 (48.14)
Pre-NIV SaO2, mean (SD)
89.44 (11.15)
APACHE II Score, mean (SD)
17.6 (5.93)
Time between starting NIV and symptom assessment (hours), mean (SD)
23.3 (8.69)
Length of hospital stay (days), mean (SD)
11.3 (9.4)
AECOPD = acute exacerbation of chronic obstructive pulmonary disease; APO = acute pulmonary edema; SDB = sleep disordered breathing; mMRC = modified Medical Research Council dyspnea score; NYHA = New York Heart Association; NIV = noninvasive ventilation; APACHE II = Acute Physiology and Chronic Health Evaluation II.
a Most often a family member.
b Fraction of inspired oxygen (FiO2) was highly variable. At the time the ABG was taken, the FiO2 was not recorded for two patients; 22% were breathing room air; 40% were receiving an FiO2 of greater than 40%, with the remainder receiving an FiO2 between 21% and 40%.
Figure 2 outlines symptom prevalence. Breathlessness, dry mouth, lack of energy, difficulty sleeping, cough, and sputum were the most prevalent symptoms. Psychological symptoms were highly prevalent with half of the cohort reporting the presence of worrying, feeling sad, and feeling nervous. The mean CMSAS summary score for the population was 1.60 (SD .66), the mean CMSAS physical symptom subscale score was 1.62 (SD .64), and the CMSAS psychological symptom subscale score was 1.57 (SD 1.23). The Cronbach alpha for the CMSAS summary score was 0.72, for the CMSAS physical subscale was 0.65, and for the psychological subscale was 0.74.
Fig. 2Symptom prevalence. CMSAS = Condensed Memorial Symptom Assessment Scale.
Table 3 details the number of symptoms reported and the number of severe symptoms. Participants reported a median of 10 symptoms (IQR 9–13) of the 18 symptoms assessed, of these a median of 5 (IQR 3–7) were reported as severe.
Table 3Number of Symptoms and Number of Severe Symptoms Reported
Figure 3 details the bothersomeness of physical symptoms and frequency of psychological symptoms. Symptoms with the highest mean scores (in rank order) were breathlessness, lack of energy, pain, dry mouth, difficulty sleeping, and worrying. More detail regarding mean scores for individual symptoms appears in the Supplementary Material (Table S1).
Fig. 3Symptom bothersomeness or frequency. CMSAS = Condensed Memorial Symptom Assessment Scale; psych. = psychological.
The mean Borg score at the time of assessment was 2.63 (SD 1.90). The mean worst (highest) Borg score over the previous 24 hours was 7.55 (SD 3.31), and the mean best (lowest) was 2.33 (SD 1.86). The overall prevalence of pain at any site was 58% (see Fig. 2). Detailed pain results are outlined in Table 4. Mean pain intensity ratings suggest pain was of moderate intensity, and mean bothersomeness ratings (ranging from 2.4–3.2; max score 4) suggest that pain was a significant concern to patients.
The total number of patients able to complete reporting of detailed pain data is 46—four patients declined to answer this part of the assessment as they were feeling too fatigued. NB: patients may report pain in more than one location.
a The total number of patients able to complete reporting of detailed pain data is 46—four patients declined to answer this part of the assessment as they were feeling too fatigued. NB: patients may report pain in more than one location.
b Severity is rated on a 0–10 numerical rating scale. Mean score for patients reporting pain.
c Bothersomeness rated as per the CMSAS, range 0–4. (0 = not present; 4 = very bothersome).
There was no statistically significant correlation between the CMSAS summary score and a range of clinical or demographic variables including age, APACHE II score, breathlessness measures, length of hospital stay, Charlson score, or Karnofsky score as detailed in the Supplementary Material (Table S2).
Discussion
The wide array of symptoms described by patients with hypercapnic respiratory failure highlights the importance of symptom assessment. Breathlessness was, unsurprisingly, highly prevalent (although not universal) and was often reported as severe and distressing. Dry mouth, perhaps related to anticholinergic medication, diuretics, and/or NIV itself, has a high prevalence and bothersomeness, suggesting the need for ameliorating strategies. Lack of energy, while clinically explicable, was reported as bothersome and may be linked to other prevalent, bothersome symptoms, including poor sleep and feeling drowsy. Pain was both prevalent and bothersome. More than half the cohort reported constipation. Psychological symptoms were highly prevalent with half of the cohort reporting all three psychological symptoms assessed.
To our knowledge, this is the first study to assess a wide breadth of symptoms using a validated symptom assessment instrument in inpatients with respiratory failure requiring treatment with NIV, although another study has described the related concept of HRQoL in this population.
The Severe Respiratory Insufficiency Questionnaire scored best in the assessment of health-related quality of life in chronic obstructive pulmonary disease.
Symptoms reported in this study are similar to other studies conducted in patients with COPD or heart failure in both the inpatient and outpatient setting.
Understanding breathlessness: cross-sectional comparison of symptom burden and palliative care needs in chronic obstructive pulmonary disease and cancer.
The prevalence and associated distress of physical and psychological symptoms in patients with advanced heart failure Attending a South African medical center.
Measuring the symptom experience of seriously ill cancer and noncancer hospitalized patients near the end of life with the memorial symptom assessment scale.
As expected, breathlessness was found to be a high burden symptom with a mean maximum Borg score of 7.55. Although clinicians routinely ask about breathlessness, it is not commonly quantified objectively in the inpatient setting. Structured and comprehensive assessment of breathlessness may lead to improved clinical management.
Improvements in mask technology may account for some of this difference. Given the interaction between pain, breathlessness, and quality of life, treatment interventions targeting one symptom might lead to improvement in overall in quality of life and diminish patient distress.
Given the stress associated with hospitalization, the high prevalence reported in this study is unsurprising. Understanding the epidemiology of these phenomena is a starting point for developing effective management strategies.
Difficulty sleeping was reported by 70% of patients in this study. A recent qualitative study reported patients perceive the ability to sleep once NIV is initiated as, of itself, therapeutic.
It remains unclear which instrument should be used for routine assessment. Although the COPD Assessment Test has been predominantly used in outpatient settings, it has also been used to assess hospital inpatients
Items in the COPD Assessment Test would capture some prevalent symptoms (e.g. cough, mucus/sputum and sleep), but some high burden symptoms (particularly dry mouth and psychological symptoms) would be missed. The data contained in this study, if replicated, might inform the development of a symptom assessment instrument for inpatients. Given the importance of the inpatient care, both in terms of patient outcomes and health economics, instruments specifically assessing this may serve to inform, develop, and assess effective models of care.
The broad range of symptoms reported in this study suggests that clinicians who deliver NIV should be skilled in detecting and managing symptoms concurrently with delivering NIV. The range and intensity of symptoms reported here suggests that patients may have complex symptom needs and expert input from palliative care teams may be additive. In end-stage renal failure, palliative care clinicians are increasingly being embedded within the renal care teams.
This study raises a question of whether similar strategies might be effective for patients with respiratory failure.
Optimal symptom assessment and management, including both physical and psychological symptoms, has many implications that go beyond the individual patient experience. For example, these may influence discharge readiness and readmission rates, which may have a significant patient and health economic impact. Understanding longitudinal changes in symptoms, from admission through to postdischarge follow-up, may allow clinicians to provide reassurance that some symptoms resolve with time, while prioritizing management of symptoms which appear most persistent and/or troublesome.
To our knowledge, this study is the first to look at the broad symptom experience of a population treated with NIV from a subjective perspective. We assessed patients within 36 hours of hospitalization giving an insight into symptomatology close to the beginning of admission, but the study does not provide longitudinal data, and this would be an important area for follow-up research. The choice of the CMSAS to assess symptom burden deserves some comment. This tool has only been validated in a cancer population
Two potentially relevant modifications of the original Memorial Symptom Assessment Scale (from which the CMSAS is derived) have been undertaken—one in a population with heart failure and the other in a population with COPD—with both reporting good validity and reliability.
Both these instruments were thought to be too long for patients with acute respiratory failure to complete. In developing the Memorial Symptom Assessment Scale–Short form, Chang et al. demonstrated that correlation between symptom scores and HRQoL remained strong when only the bothersomeness of physical symptoms and the frequency of psychological symptoms are assessed.
Thus, to minimize respondent burden in this very ill inpatient population, the CMSAS was chosen.
There are other symptom assessment tools available, which are short and could be potentially useful in this context. For example, the Edmonton Symptom Assessment Scale assesses the intensity of 8–10 symptoms depending on the version used
assesses a combination of intensity, frequency, and distress but does not include psychological symptoms that were thought to be important in this population. Based on Chang et al.'s work, we conclude that bothersomeness is more informative than intensity.
As we were interested in a broad range of symptoms, tools that focus predominantly on respiratory symptoms, such as the Chronic Bronchitis Symptom Assessment scale
were thought to be too narrow in scope. We should acknowledge differences between our administration of the CMSAS and the method reported. First, we administered this instrument verbally for this very ill population. Second, the wording of the physical symptom question in our study varied slightly from the original CMSAS instrument. The original CMSAS asks “How much did the (symptom) bother or distress you?”
and in this study, patients were simply asked, “How much did the (symptom) bother you?” In addition, the time frame about which the symptom enquiry was “in the last 72 hours” was different from the original instrument which is “in the last 7 days.”
The clinical heterogeneity of this population reflects the real-world experience of patients with respiratory failure. Generalizability may be limited by the large number of potential participants excluded and the inclusion of only English-speaking participants who underwent arterial blood gas sampling before initiating NIV. The symptoms of those who do not speak English and/or did not undergo arterial blood gas assessment before commencing NIV may be different, and exploring these would be important future work. In addition, the examined cohort included a disproportionately high number of women, again limiting generalizability. No correlation between symptom burden and other clinical demographics was found, perhaps partially explained by the relatively small number of participants. We did not specifically assess “anxiety,” although we did assess “feeling nervous” and “worrying” and found they were common. Future studies might explore differences in these emotional states from the patient perspective. The Cronbach alpha for the CMSAS summary score and CMSAS psychological score are acceptable (0.72 and 0.74, respectively). The relatively low Cronbach alpha for the CMSAS physical subscale (0.65) may suggest that the bothersomeness of some physical symptoms are relatively independent of each other, perhaps due to our combination of high and low prevalence symptoms. Future studies enrolling a larger cohort would be necessary to undertake valid statistical analysis to establish this.
Conclusions
These patients with acute respiratory failure requiring NIV reported a high prevalence of a wide range of physical and psychological symptoms, reporting a median of 10 symptoms, five of which were severe. The breadth and prevalence of symptoms reported suggests an opportunity to develop treatment strategies that address the total patient experience, laying the foundation for developing clinically useful tools for quantifying symptom burden in inpatients with respiratory failure. Such work is crucial to developing a truly person-centered approach to addressing the needs of people treated with NIV.
Disclosures and Acknowledgments
The authors thank the staff of St George Hospital, Sydney, particularly the Department of Respiratory Medicine where Dr. Smith was an Honorary Fellow during this project. Special thanks to Ms. Mary Dunford, Dr. Elizabeth Clark, and Dr. Steven Lindstrom. Thanks also to Dr. Trish Davidson, Dr. Katy Clarke, and Dr. Meera Agar who assisted with developing the research protocol and Mr. Michael Piza for assistance with preparation of the initial statistical plan.
More importantly, the authors thank the participants for their willingness to participate. Without you, this research would not be possible.
Professor Ingham's research work on this paper was undertaken, in part, with funding support from the Cancer Institute New South Wales Academic Chairs Program. The views expressed herein are those of the authors and are not necessarily those of the Cancer Institute NSW. There are no other conflicts of interest to declare.
Data statement: Original data are available from the authors on application.
Supplementary Material
Mean symptom score for people with hypercapnic respiratory failure treated with NIV, all symptoms assessed using the Condensed Memorial Symptom Assessment Scale.
Table S1Individual Symptom Mean Scores
Mean
SD
Bothersomeness of CMSAS physical symptoms
SOB
3.25
1.246
Lack of energy
2.38
1.441
Dry mouth
2.37
1.262
Pain
1.84
1.699
Difficulty sleeping
1.82
1.567
Feeling drowsy
1.33
1.358
Difficulty concentrating
1.25
1.492
Constipation
1.20
1.316
Lack of appetite
1.04
1.224
Nausea
0.99
1.39
Weight loss
0.42
.828
Frequency of psychological symptoms
Worrying
1.96
1.607
Feeling sad
1.57
1.528
Feeling nervous
1.27
1.381
Bothersomeness of other symptoms potentially prevalent in patients treated with NIV
pH and PaCO2—as assessed on arterial blood gas analysis either before initiation of NIV or within 30 minutes of NIV initiation.
−0.06
0.66
Borg score at assessment
0.092
0.53
Best Borg
0.05
0.71
Worst Borg
0.10
0.51
Total hours on NIV
−0.081
0.58
Initial respiratory rate
−0.131
0.53
APACHE II
−0.087
0.55
NYHA
−0.078
0.59
mMRC
−0.093
0.52
Charlson
0.038
0.79
Karnofsky
0.227
0.11
Length of hospitalization (days)
0.095
0.529
CMSAS-SUM = Condensed Memorial Symptom Assessment Scale–Summary score; NIV = noninvasive ventilation; APACHE = Acute Physiology and Chronic Health Evaluation; NYHA = New York Heart Association breathlessness score; mMRC = modified Medical Research Council breathlessness score.
a pH and PaCO2—as assessed on arterial blood gas analysis either before initiation of NIV or within 30 minutes of NIV initiation.
Understanding breathlessness: cross-sectional comparison of symptom burden and palliative care needs in chronic obstructive pulmonary disease and cancer.
The prevalence and associated distress of physical and psychological symptoms in patients with advanced heart failure Attending a South African medical center.
The Severe Respiratory Insufficiency Questionnaire scored best in the assessment of health-related quality of life in chronic obstructive pulmonary disease.
Measuring the symptom experience of seriously ill cancer and noncancer hospitalized patients near the end of life with the memorial symptom assessment scale.
31052-2&id=gr1.jpg){kind=link}
31052-2&id=gr2.jpg){kind=link}
31052-2&id=gr3.jpg){kind=link}