|Year : 2019 | Volume
| Issue : 1 | Page : 36-40
Epworth sleepiness score to predict sleep apnea in acute stroke: Do we need to delve deeper?
Rahul Tyagi, Pulikottil Wilson Vinny, Vivek Hande, Vijay Budhwar
Department of Medicine, INHS Asvini, Mumbai, Maharashtra, India
|Date of Submission||06-Sep-2018|
|Date of Acceptance||17-Dec-2018|
|Date of Web Publication||19-Jun-2019|
Surg Capt Pulikottil Wilson Vinny
Department of Medicine, INHS Asvini, Colaba, Mumbai - 400 005, Maharashtra
Source of Support: None, Conflict of Interest: None
Introduction: Globally, stroke is a leading cause of death and disability. Obstructive sleep apnea (OSA) is globally being recognized as an emerging public health problem and a risk factor for stroke. Epworth Sleepiness Score (ESS) is increasingly being used as a screening tool to determine the likelihood of OSA in a patient before polysomnography (PSG). However, ESS questionnaires in patients with acute stroke and their comparison with overnight PSG have not been studied. We conducted this study at a tertiary care respiratory center in Mumbai to determine the effectiveness of ESS in predicting the prevalence of sleep apnea in acute ischemic stroke. Materials and Methods: Twenty-eight patients of acute stroke were included in the study. These patients were subjected to ESS and technician attended overnight PSG. Apnea–Hypopnea Index, oxygen desaturation index, and minimum saturation were determined at the overnight PSG. Results: Majority of patients belonged to 61–80 years' age group (53.6%) and were predominantly male (89.3%). Only 5 patients (17.9%) showed an ESS of more than 8, whereas 23 patients (82.1%) showed evidence of OSA on PSG. Sensitivity of ESS in predicting OSA in patients with acute stroke was 17.3%, whereas specificity was 80%. Conclusion: The use of ESS in patients of acute stroke to screen for sleep-disordered breathing (SDB) should be avoided. A PSG may be considered in these patients as early diagnosis of SDB in these patients can help in improving recovery.
Keywords: Acute stroke, Epworth Sleepiness Score, obstructive sleep apnea, polysomnography
|How to cite this article:|
Tyagi R, Vinny PW, Hande V, Budhwar V. Epworth sleepiness score to predict sleep apnea in acute stroke: Do we need to delve deeper?. J Mar Med Soc 2019;21:36-40
| Introduction|| |
Globally, stroke is a leading cause of death and disability. In low- and middle-income countries, there has been a rising trend in stroke incidence, people affected are younger, and stroke mortality is more as compared to the developed world. In India, stroke prevalence has been recorded as high as 262/100,000 in rural and 424/100,000 in urban areas. Twenty-eight-day mortality at various centers varies from 29.8% to 42%, with significantly higher mortality in rural patients.
Obstructive sleep apnea (OSA) is globally being recognized as an emerging public health problem of great importance, and its association with various comorbidities including stroke is well documented. Sharma et al. in their study in urban Indians found that the prevalence of OSA varies from 2.4% to 4.96% in males and 1%–2% in females.
Epworth Sleepiness Score (ESS) is a validated questionnaire that uses eight items to determine the likelihood of sleeping in various situations. It has a specificity of 80% for predicting OSA. Data on sleep apnea in acute stroke are limited to few studies. In a study by Kim et al., the prevalence of sleep-disordered breathing (SDB) in patients with acute stroke, determined by the use of ESS, was approximately 48.8%. Camilo et al. found the prevalence of SDB in acute stroke, determined by overnight polysomnography (PSG), to be 76%. Bassetti and Aldrich also found similar results in their study on acute stroke patients. At most centers, ESS or other questionnaires are used to determine the likelihood of OSA in a patient before PSG. However, data are lacking on the effectiveness of these questionnaires (ESS) and its comparison with overnight PSG in patients of acute stroke. We conducted this study at a tertiary care center in Mumbai to determine the effectiveness of ESS in predicting the prevalence of sleep apnea in acute ischemic stroke.
| Materials and Methods|| |
The study was conducted between May 2016 and December 2016 at a tertiary care center in Mumbai. All consecutive patients with ischemic stroke presenting to the center were evaluated. Patients with the first episode of acute stroke presenting within 2 weeks of onset were included in the study. Patients with previous stroke, critical patients unfit to undergo PSG, patients with known OSA, patients on sedatives, and those unwilling to participate in the study were excluded from the study.
The ethical approval for the study was taken from the institutional review board. Forty-six patients with acute stroke presented to our center during the study period. Nine patients had a history of previous cerebrovascular accident, 5 patients were critical and were considered unfit to undergo PSG, 3 were known case of OSA, and 1 patient was on benzodiazepines. After excluding the above, 28 patients were included in the study. Written informed consent was obtained from the patient or his/her next of kin.
Assessment of severity of stroke was done using the National Institutes of Health Stroke Scale (NIHSS) score. NIHSS is a validated tool used to measure severity and effectiveness of treatment and predict outcomes in acute stroke. NIHSS scoring was done for all patients. All patients included in the study had NIHSS <15; patients were divided into NIHSS <5 (mild stroke) and NIHSS 6–15 (moderate stroke).
Assessment of excessive daytime sleepiness in patients with acute stroke was done using ESS. ESS was performed within 24 h of admission. An ESS score of more than 8 was considered as an indicator of poor sleep. All patients underwent in the laboratory, and technician attended overnight Type I PSG before being discharged. EEG, electrocardiography, chin and leg electromyography, electrooculography, nasal and oral airflow, chest and abdominal efforts, and pulse oximetry were included in the study. Standard definition for apnea, hypopnea, and oxygen desaturation index (ODI) was used. Manual scoring was done to confirm the recorded events before calculating various indices.
Apnea–Hypopnea Index (AHI) and ESS were considered as primary outcome variables.
The NIHSS, gender, age group, risk factors for stroke, total hypopneas, total obstructive apnea (OA), total central apnea (CA), ODI, and minimum saturation were considered as primary explanatory variables.
Descriptive analysis was carried out by mean and standard deviation for quantitative variables, frequency, and proportion for categorical variables.
The association between AHI and ESS was assessed by cross-tabulation and comparison of percentages. Chi-square test was used to test statistical significance.
Normality test for quantitative variables
A Shapiro–Wilk test (P > 0.05) and a visual inspection of their histograms, normal Q–Q plots, and box plots showed that the AHI and total hypopneas, total OA, total CA, ODI, and minimum saturation parameters were nonnormally distributed.
The comparison of total hypopneas, total OA, total CA, ODI, and minimum saturation across AHI categories was assessed by comparing the median values. Kruskal–Wallis test was used to assess statistical significance.
P < 0.05 was considered statistically significant. IBM SPSS Version 21.0. (IBM Corp., Armonk, NY, USA) was used for statistical analysis.
| Results|| |
A total of 28 people were included in the analysis. The descriptive analysis of demographic parameters in the study population is given in [Table 1].
|Table 1: Descriptive analysis of demographic parameter in the study population|
Click here to view
Majority of patients were in the age group of 61–80 years (n = 15, 53.6%). Hemiparesis was the most common neurological deficit noted (n = 14 [50%]), and hypertension was the most common risk factor for stroke noted among the study population (n = 13, 39.7%). More than 80% of the study population had ESS ≤8. All 28 patients underwent overnight attended Type I PSG. Twenty-three patients (82.1%) were noted to have an AHI >5 on PSG. Descriptive analysis of AHI and ESS in the study population is given in [Table 2].
|Table 2: Descriptive analysis of Apnea-Hypopnea Index and Epworth Sleepiness Score in the study population (n=28)|
Click here to view
In patients with ESS ≤8, 82.6% had AHI >5 and 80% of the patients with ESS >9 had AHI >5. AHI was similar in both the groups as evidenced by the difference proportion in the AHI across ESS groups [Table 3].
|Table 3: Comparison of Epworth Sleepiness Score with Apnea-Hypopnea Index (n=28)|
Click here to view
Among the 23 patients having AHI more than 5, 4 participants (17.4%) had ESS more than 9, whereas 19 (82.6%) participants had ESS ≤8. The difference proportion in the OSA across ESS was statistically not significant (P = 0.890). Hence, it can be concluded that almost comparable proportion of patients had ESS of 9 or more in both OSA and non-OSA groups [Table 4].
|Table 4: Comparison of obstructive sleep apnea with Epworth Sleepiness Score (n=28)|
Click here to view
The median values of total OA have shown an increasing trend with increasing AHI categories, and differences across AHI categories were statistically significant (P < 0.001). The median values of total CA were lowest in AHI <5 and gradually increased with increasing AHI category, but the differences in median total CA value across AHI categories were statistically not significant (P = 0.075). The median values of total hypopneas and ODI also had shown an increasing trend with increasing AHI category (P < 0.001). Minimum saturation value had shown no statistically significant differences across AHI categories (P = 0.685) [Table 5].
The sensitivity of ESS in diagnosing OSA was 17.4% (95% CI: 1.91%–32.9%) and specificity was 80.0% (95% CI: 44.94%–99.42%), false-positive rate was 20.0% (95% CI: 1.00%–55.1%), false-negative rate was 82.6% (95% CI: 67.11%–98.1%), positive predictive value was 80.0% (95% CI: 44.94%–96.62%), negative predictive value was 17.4% (95% CI: 1.91%–32.9%), and the total diagnostic accuracy was 28.6% (95% CI: 11.84%–45.3%)
|Table 6: Predictive validity of Epworth Sleepiness Score in predicting obstructive sleep apnea (n=28)|
Click here to view
| Discussion|| |
SDB is an umbrella term that includes various sleep-related problems such as OSA/hypopnea syndrome, central sleep apnea, and hypoventilation syndromes. PSG is the gold standard test for the diagnosis of OSA and for identifying other forms of SDB.
ESS uses eight questions answered by the patient to assess the tendency of sleeping or dozing during the day. In a review of various studies comparing ESS with PSG by Kapur et al., ESS had high false negatives limiting its utility for diagnosis of OSA. The same has been seen in our study also where ESS showed a false-negative rate of 82.6%. The task force also recommended that patients who are at high risk for nonobstructive SDB, including patients with stroke, should undergo PSG. The Indian consensus guidelines on OSA also recommend that ESS can be used in patients suspected of having OSA, but it is not mandatory. Despite this, the use of ESS for screening a patient suspected of OSA, before PSG is common in clinical practice. Kim et al. did a study to understand the relationship between sleep disturbance and functional status in mild stroke patients. In this study, 48.8% of the patients were found to have an ESS of more than 8 suggestive of sleep disturbance. However, the authors did not do a PSG to confirm SDB in these patients. Guimarães et al. studied the severity of sleepiness using ESS in 66 patients at baseline and after automatic positive airway pressure (APAP) treatment. They observed that there was no correlation between baseline ESS and AHI. They also determined that a retrospective baseline ESS after APAP treatment was significantly higher than the ESS recorded at baseline. They also found a considerable variability in individual sleepiness perception. In our study, ESS of more than 8 was seen only in 5 patients (17.8%). This is lower than the percentages seen in both these studies; however, no particular reason for this disparity could be determined. In our study, the sensitivity of ESS in determining OSA was found to be 17.3%. In a meta-analysis of seven studies done by Kapur et al., the sensitivity of ESS was found to be 0.27–0.52. The study population in our study was different as we were studying a specific cohort of acute ischemic stroke population.
Camilo et al. submitted 39 patients to full PSG on the first night after ischemic stroke. They found that the AHI ≥10/h was seen in 76% of patients. These patients had a median NIHSS of 11 (IR: 6–20). They observed that 48.7% of the patients had severe OSA and only two patients had predominantly CA In a meta-analysis of 29 studies, Johnson and Johnson determined that at a cutoff of AHI >5/h, the prevalence of SDB in stroke/TIA patients was 72%. In our study, OSA, as determined by AHI >5/h, was seen in 23 patients (82.1%). Although this may be considered higher than the 72% prevalence seen in the meta-analysis by Johnson and Johnson, the results are comparable to Camilo et al., as they used a higher AHI cutoff.
Increased incidence of ischemic stroke in patients with OSA can be multifactorial. Increased platelet aggregation, reduced fibrinolytic activity, and high plasma fibrinogen have all been documented in OSA. SDB is both an independent predictor of stroke and also a factor impairing recovery. Apneas and hypopneas can cause cerebral hypoxemia and impaired cerebral autoregulation leading to additive new cognitive defects, difficult rehabilitation, and increased neuronal loss in ischemic penumbra. OSA in a patient with stroke is associated with higher mortality with a correlation between increasing AHI and poststroke mortality. In a study done by Kaneko et al. on 61 patients admitted to the stroke rehabilitation unit, patients with sleep apnea had significantly worse functional capacity, both at admission and discharge, which was related to obstructive events but not central events. In our patients, OA was more than CA in all patients with AHI more than 5. Johnson and Johnson showed similar results demonstrating majority of SDB in stroke patients to be obstructive in nature.
SDB is a risk factor for stroke and also determines the outcomes in a patient of stroke, as has been determined by various studies mentioned earlier. ESS, although used as a screening test for OSA, has poor sensitivity in an acute stroke patient, which is unacceptably low as determined by our study. Furthermore, as the prevalence of SDB in stroke patients is high, there may be a point in considering PSG in all stroke patients, as evidenced by our study. Studies at a larger scale will further confirm our findings.
| Conclusion|| |
The use of ESS in patients of acute stroke to screen for SDB should be avoided. A PSG may be considered in these patients as early diagnosis of SDB in these patients may help in improving recovery. However, considering the small sample size and wide confidence intervals, there is a strong need to test the predictive validity with larger sample size, to generate better quality evidence.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Johnson W, Onuma O, Owolabi M, Sachdev S. Stroke: A global response is needed. Bull World Health Organ 2016;94:634.
Pandian JD, Sudhan P. Stroke epidemiology and stroke care services in India. J Stroke 2013;15:128-34.
Sharma SK, Katoch VM, Mohan A, Kadhiravan T, Elavarasi A, Ragesh R, et al.
Consensus and evidence-based Indian initiative on obstructive sleep apnea guidelines 2014 (first edition). Lung India 2015;32:422-34.
] [Full text]
Kim J, Kim Y, Yang KI, Kim DE, Kim SA. The relationship between sleep disturbance and functional status in mild stroke patients. Ann Rehabil Med 2015;39:545-52.
Camilo MR, Sander HH, Eckeli AL, Fernandes RM, Dos Santos-Pontelli TE, Leite JP, et al.
SOS score: An optimized score to screen acute stroke patients for obstructive sleep apnea. Sleep Med 2014;15:1021-4.
Bassetti C, Aldrich MS. Sleep apnea in acute cerebrovascular diseases: Final report on 128 patients. Sleep 1999;22:217-23.
Yao M, Hervé D, Allili N, Jouvent E, Duering M, Dichgans M, et al.
NIHSS scores in ischemic small vessel disease: A study in CADASIL. Cerebrovasc Dis 2012;34:419-23.
Kapur VK, Auckley DH, Chowdhuri S, Kuhlmann DC, Mehra R, Ramar K, et al.
Clinical practice guideline for diagnostic testing for adult obstructive sleep apnea: An American Academy of sleep medicine clinical practice guideline. J Clin Sleep Med 2017;13:479-504.
Guimarães C, Martins MV, Vaz Rodrigues L, Teixeira F, Moutinho Dos Santos J. Epworth sleepiness scale in obstructive sleep apnea syndrome – An underestimated subjective scale. Rev Port Pneumol 2012;18:267-71.
Johnson KG, Johnson DC. Frequency of sleep apnea in stroke and TIA patients: A meta-analysis. J Clin Sleep Med 2010;6:131-7.
Tofler GH, Brezinski D, Schafer AI, Czeisler CA, Rutherford JD, Willich SN, et al.
Concurrent morning increase in platelet aggregability and the risk of myocardial infarction and sudden cardiac death. N Engl J Med 1987;316:1514-8.
Andreotti F, Davies GJ, Hackett DR, Khan MI, De Bart AC, Aber VR, et al.
Major circadian fluctuations in fibrinolytic factors and possible relevance to time of onset of myocardial infarction, sudden cardiac death and stroke. Am J Cardiol 1988;62:635-7.
Loke YK, Brown JW, Kwok CS, Niruban A, Myint PK. Association of obstructive sleep apnea with risk of serious cardiovascular events: A systematic review and meta-analysis. Circ Cardiovasc Qual Outcomes 2012;5:720-8.
Wessendorf TE, Thilmann AF, Wang YM, Schreiber A, Konietzko N, Teschler H, et al.
Fibrinogen levels and obstructive sleep apnea in ischemic stroke. Am J Respir Crit Care Med 2000;162:2039-42.
Hermann DM, Bassetti CL. Role of sleep-disordered breathing and sleep-wake disturbances for stroke and stroke recovery. Neurology 2016;87:1407-16.
Das AM, Khan M. Obstructive sleep apnea and stroke. Expert Rev Cardiovasc Ther 2012;10:525-35.
Kaneko Y, Hajek VE, Zivanovic V, Raboud J, Bradley TD. Relationship of sleep apnea to functional capacity and length of hospitalization following stroke. Sleep 2003;26:293-7.
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6]