Prevalence and Risk Factors of Obstructive Sleep Apnea in Hypertensive Emergency.

INTRODUCTION: Obstructive sleep apnea (OSA) is a common factor associated with hypertensive crises. There is limited evidence of prevalence and risk factors of OSA in hypertensive emergency.
METHODS: This study recruited adult patients who diagnosed as hypertensive emergency and tested for OSA. The study period was between July 2019 and January 2020. The patients were categorized as OSA and non-OSA groups by the evidence from polysomnography. Prevalence and risk factors for OSA were executed.
RESULTS: During the study, there were 52 eligible patients. Of those, 30 patients (57.69%) were diagnosed with OSA. The stepwise logistic regression analysis for predicting OSA had two remaining factors: body mass index and diastolic blood pressure. Only body mass index was independently associated with OSA with an adjusted odds ratio of 1.166 (95% confidence interval of 1.033, 1.316). The body mass index of 25.02 kg/m2 gave sensitivity and specificity of 80.00% and 59.09%, respectively. The area under the receiver operating characteristic curve was 70.98%.
CONCLUSION: OSA had high incidence rate in patients with hypertensive emergency. High body mass index was a predictor for OSA associated with hypertensive emergency. Copyright:

INTRODUCTION

Hypertension and obstructive sleep apnea (OSA) are common diseases in clinical practice. The prevalence of hypertension in the Chinese community was high as 64.05% in 2014.[1] While, the prevalence of OSA was found up to 49% of middle-aged men in the United States and Europe.[2] Both diseases are closely related in terms of cause and prevalence. Since 2003, the Joint National Committee on hypertension control (JNC 7) has defined OSA as a secondary cause of hypertension.[3] OSA may be found 30%–50% of hypertensive patients and may be high as 71% in resistant hypertension, 85% in those who have at least one symptom of OSA, and 90% in hypertensive in the young.[4567]

Hypertensive crisis particularly hypertensive emergency is an emergency condition with high morbidity and mortality.[8] A 12-month mortality was 38.9% in hypertensive emergency patients. A previous study found that OSA was common in hypertensive crises: up to 70.8%.[9] The study did not report on risk factors of OSA in hypertensive crises. In addition, the prevalence was not specific to hypertensive emergency which is more critical condition. This study, therefore, aimed to study the risk factors of OSA only in hypertensive emergency condition.

METHODS

This was a retrospective study and recruited adult patients who diagnosed as hypertensive emergency and admitted to Srinagarind Hospital, a university hospital of Khon Kaen University, Khon Kaen, Thailand. Those who are pregnant or unstable to perform polysomnography were excluded. The diagnosis of hypertensive emergency defined as systolic blood pressure of 180 mmHg or over and/or diastolic blood pressure of 100 mmHg or over plus acute target organ damage.[9] The study period was between July 2019 and January 2020.

The eligible patients were evaluated by the Alice PDx device.[10] OSA was diagnosed if an apnea-hypopnea index (AHI) was five times/hour or over and classified as mild, moderate, and severe with the AHI of 5–14, 15–29, 30, or over, respectively. Baseline characteristics, comorbid diseases, blood pressure, and laboratory tests were studied. The patients were categorized as OSA and non-OSA groups by the evidence from polysomnography.

Statistical analyses

Sample size calculation. Required sample size was calculated using one sample formula with an approximate prevalence of 70%.[9] With a confidence of 95% and an error of 0.12, the required sample size was 57.

The prevalence of OSA in hypertensive emergency patients was calculated. Descriptive statistics were used to evaluate differences between those with and without OSA. Predictors for OSA associated with hypertensive emergency were executed by the stepwise logistic regression analysis. An independent factor was defined by a P < 0.05 in the final logistic regression model. The final model was tested for goodness of fit by the Hosmer–Lemeshow method. The independent predictor for OSA was chosen for the best diagnostic properties if it was numerical variable. Sensitivity, specificity, and area under a receiver operating characteristic (ROC) curve were reported. The statistical analyses were performed by the STATA software (College Station, Texas, USA).

RESULTS

During the study, there were 52 eligible patients. Of those, 30 patients (57.69%) were diagnosed with OSA. The severity of OSA was categorized as follows: mild (12 patients, 40.00%), moderate (13 patients, 43.33%), and severe (5 patients, 16.67%). There were two significant factors between both groups: body mass index and serum blood urea nitrogen level. The OSA group had significantly higher body mass index (23.97 vs. 28.60 kg/m2; P = 0.010) and lower serum blood urea nitrogen level (23.65 vs. 34.5 mg/dL; P 0.039) than the non-OSA group [Tables 1 and 2].

Table 1

Baseline characters of patients with hypertensive emergency categorized by the presence of obstructive sleep apnea

	Non-OSA (n=22)	OSA (n=30)	P
Age (years)	64.5 (23-86)	63.0 (24-81)	0.704
Body weight (kg)	66.50 (36-92)	77.75 (50-140)	0.004
Height (cm)	157 (53-180)	160 (140-176)	0.306
BMI (kg/m2)	23.97 (16.58-38.41)	28.60 (19.69-72.25)	0.010
Male, n (%)	11 (50.00)	14 (46.67)	0.999
Female, n (%)	11 (50)	16 (53.33)	0.999
Hypertension	22 (100)	28 (93.33)	0.502
Diabetes mellitus	15 (68.18)	22 (73.33)	0.762
Coronary artery disease	2 (9.09)	1 (3.33)	0.567
Heart failure	7 (31.82)	14 (46.67)	0.392
Chronic kidney disease	15 (68.18)	18 (60.00)	0.576
Stroke	3 (13.64)	7 (23.33)	0.488
Atrial fibrillation	4 (18.18)	7 (23.33)	0.741

Data presented as median. BMI: Body mass index, OSA: Obstructive sleep apnea

Table 2

Laboratory results of patients with hypertensive emergency categorized by the presence of obstructive sleep apnea

	Non-OSA (n=22)	OSA (n=30)	P
AHI (times/h)	<5	16.55 (5.3-56.6)
OSA severity, n	0	30
Mild, n (%)	0	12 (40.00)
Moderate, n (%)	0	13 (43.33)
Severe, n (%)	0	5 (16.67)
SBP (mmHg)	191 (183-200)	191 (186-200)	0.766
DBP (mmHg)	106 (98-115)	103 (96-111)	0.316
HbA1C (%)	6.3 (0-13.6)	6.6 (0-11.2)	0.528
SpO2 (%)	97 (83-100)	96 (63-100)	0.353
Hemoglobin (g/dL)	10.65 (6.8-14)	11.75 (6.8-15.8)	0.258
Hematocrit (%)	32.95 (21.3-40.7)	36.2 (20.3-51.5)	0.326
White blood cell (/µL)	7815 (3450-11910)	6895 (3120-16340)	0.359
Neutrophil (%)	63.0 (55.8-71.3)	66.3 (54.3-73.2)	0.697
Lymphocyte (%)	26.7 (20.9-31.7)	24.9 (16.9-29.8)	0.617
Monocyte (%)	5.5 (4.8-6.7)	6.6 (5.0-7.7)	0.266
Eosinophil (%)	3.3 (1.9-5.0)	3.0 (1.2-5.5)	0.970
Basophil (%)	0.7 (0.4-1.0)	0.6 (0.3-0.8)	0.160
Platelet (/µL)	362,000 (42,000-319,000)	233,000 (8000-378,000)	0.135
BUN (mg/dL)	34.5 (10.5-79.7)	23.65 (6.9-64.8)	0.039
Creatinine (mg/dL)	2.78 (0.64-13.39)	1.49 (0.74-11.56)	0.069
GFR (ml/min/1.73 m2)	21.42 (3.4-122.1)	44.35 (4.31-125.72)	0.056

AHI: Apnea-hypopnea index, SBP: Systolic blood pressure, DBP: Diastolic blood pressure, BUN: Blood urea nitrogen, GFR: Glomerular filtration rate, OSA: Obstructive sleep apnea, HbA1C: Glycated hemoglobin A1c

The stepwise logistic regression analysis for predicting OSA had two remaining factors: body mass index and diastolic blood pressure [Table 3]. Only body mass index was independently associated with OSA with the adjusted odds ratio of 1.166 (95% confidence interval of 1.033, 1.316). The Hosmer–Lemeshow Chi-square of the final model was 7.21 (P = 0.513). The body mass index of 25.02 kg/m2 gave sensitivity and specificity of 80% and 59.09%, respectively. The area under the ROC curve was 70.98% [Figure 1].

Table 3

Factors predictive of hypertensive emergency caused by obstructive sleep apnea

Factors	95% CI
	Unadjusted odds ratio	Adjusted odds ratio
BMI	1.132 (1.019-1.258)	1.166 (1.033-1.316)
DBP	0.981 (0.947-1.017)	0.965 (0.926-1.006)

BMI: Body mass index, DBP: Diastolic blood pressure, CI: Confidence interval

Figure 1

A receiver operating characteristic curve of body mass index and hypertensive emergency caused by obstructive sleep apnea

DISCUSSION

The incidence of OSA in hypertensive emergency was over 50% (57.69%) which was lower than the previous study (70.8%). However, the incidence rate of OSA in this study is more specific to only hypertensive emergency; not hypertensive crises as in the previous study. The proportions of OSA severity were quite comparable in moderate severity (43.33% vs. 40.35%), while in this study found more mild OSA (40.0% vs. 25.43%) but fewer severe OSA (16.67% vs. 34.21%) than the previous study. Once again, the results of this study were more specific to hypertensive emergency. There are several mechanisms to explain severe hypertensive emergency from OSA. First, blood pressure may be as high as 240/130 mmHg after each apnea episode.[1112] In each night, OSA patients may have over several hundred of apneic events depending on the AHI. Second, a meta-analysis found that OSA increased angiotensin II and aldosterone levels significantly compared with controls by 3.39 ng/L and 1.35 ng/mL, respectively.[13] Least but not last, OSA had proportions of a reverse dipper than control on both systolic (73.5% vs. 26.5%) and diastolic (69.6% vs. 30.4%) blood pressures.[14]

Obesity is a major risk factor for OSA. Those with body mass index over 30 kg/m2 had OSA is 63%.[15] A study also found a close relationship between body mass index, OSA, and hypertension.[16] The body mass index had the adjusted coefficient of 0.46 with the blood pressure. These data support the positive correlation of body mass index and the occurrence of hypertensive emergency with a 16% risk per 1 kg/m2 increase in body mass index [Table 3]. Those with a body mass index of 25 kg/m2 increased the risk of OSA with hypertensive emergency (80% sensitivity). The cut point was similar to the obesity criterion for the Asian population.[17]

There are some limitations in this study. First, the cut point of body mass index may not be applicable to other ethnicities. However, the body mass index of 30 kg/m2 may be justified for the non-Asian populations. Second, there was no long-term outcome of the patients in this study. Further studies are needed to evaluate if continuous positive airway pressure treatment for OSA may improve the hypertensive emergency outcome.[18] Finally, some other aspects of OSA or metabolic diseases were not studied.[192021222324]

CONCLUSION

OSA had a high incidence rate in patients with hypertensive emergency. High body mass index was a predictor for OSA associated with hypertensive emergency.

Research quality and ethics statement

This study was approved by the Institutional Review Board / Ethics Committee approval number [HE611131]. The authors followed applicable EQUATOR Network (http://www.equator-network.org/) guidelines during the conduct of this research project.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.