This study uniquely evaluated sleep characteristics in relation to ideal CVH metrics, assessed using ideal cardiovascular health metrics, among HD patients. After adjusting for age, sex, job category, drinker, IPAQ, fluid overload, spkt/v, HDL, and potassium, this study demonstrated that short sleep duration(< 7h) and OSA risk were separated associated with poor cardiovascular health metrics (0–2 components), compared with ideal cardiovascular health metrics (5–7 components). Moreover, assessment of the combined effects of sleep duration and OSA risk on ideal CVH metrics in our study suggested existence of OSA risk disparities in the association between sleep duration and ideal CVH metrics in HD patients. To our knowledge, this is the first study to explore the separated and combined associations of sleep duration and OSA risk with ideal cardiovascular health metrics in HD patients.
Consistent with previous studies(4, 17), our study found that only 23.19% of HD patients met 5–7 ideal CVH metrics, less than a half of the general Chinese population. In fact, traditional risk factors for cardiovascular disease such as hypertension, diabetes, obesity and lack of exercise are more common in HD patients, while dialysis-related factors such as anemia, inflammation, oxidative stress and fluid overload also increase the incidence of CVD events(1). However, the results of our study showed that despite regular visits to hospitals, HD patients have a higher risk of cardiovascular disease than the general population and they are not well-controlled. Cardiovascular disease is the most common complication and the leading cause of death of HD patients, more attention should be paid to the prevention and control of CVD in HD patients.
Sleep complaints, including OSA risk and short sleep duration, are associated with higher risk of CVD events in several epidemiological studies(3, 7–9). In AHA GRFW SFRN population study, insomnia and OSA risk were associated with lower cardiovascular health scores(9). National Health and Nutrition Examination Survey (NHANES) found a significant inverted U-shaped relationship between sleep duration and number of ideal cardiovascular health metrics, which is similar to the Henan Rural Cohort(2, 8). To our knowledge, the association between sleep complaints and ideal cardiovascular health metrics (a measure of overall CVH) in HD patients has not been previously assessed. Notably, patients with end-stage renal disease (ESRD) are more likely to experience OSA (33%-56%)(5, 18),and short sleep duration (36.7%-67.1%)(6, 19).On the one hand, short sleep duration and OSA risk were closely associated with increase in the rate of kidney function decline in patients without hypertension or diabetes, On the other hand, dialysis-related fluid overload, inflammation, oxidative stress, circadian rhythm misalignment, and adverse emotions in HD patients are also the cause of disturbed sleep(9, 18, 20). In this study, after adjusting confounding factors such as fluid overload, inflammation and psychological state in Model 3 of Table 2, the relationship between short sleep duration and OSA risk and ideal CVH metrics is still significant. Nevertheless, the conclusiveness of our evidence may be limited by the small sample size of patients with long sleep duration (10.2%), more studies are needed to provide more reliable and robust evidence for the relationship between sleep complaints and ideal CVH metrics in HD patients in the future.
The management of traditional cardiovascular risk factors such as hypertension, diabetes and obesity in HD patients is crucial, hypertension and diabetes mellitus are common complications in patients with OSA, the recurrent episodes of characteristic intermittent hypoxia and sleep interruption will lead to sympathetic activation, oxidative stress and systemic inflammation, which are also their pathogenesis(21). In addition, obesity is an important indicator of OSA risk, and the relationship between OSA and traditional cardiovascular risk factors is clear.
However, the relationship between sleep duration and these risk factors is different. The U-shaped relationship between sleep duration and risk of type 2 diabetes is recognized. Compared with 7-hour sleep duration, the risk of type 2 diabetes increases by 9% per 1 hour decrease and 14% per 1 hour increase(22).But relationship between sleep duration and obesity and hypertension tends to be J- shaped, findings from China Health and Nutrition Survey(CHNS) suggested that short sleep duration is a risk factor for weight gain ≥ 5 kg in Chinese adults, increasing the probability of overweight/obesity(23).Besides, the relationship between short sleep duration and hypertension reported in epidemiological studies can be attributed to sympathetic nerve stimulation and circadian rhythm disorder(24).
Beyond their potential influence on clinical cardiovascular risk factors, sleep complaints may also be associated with other modifiable lifestyle behaviors, including diet, physical activity and smoking. Previous studies have reported that short sleepers and OSA patients are more likely to have irregular eating habits, such as fewer main meals, frequent intake of energy-dense snacks, or eating more animal viscera, fried foods, salted foods(25, 26). The influence of OSA and sleep duration on PA is less clear, and only one study found that OSA was associated with lower PA levels, while the quality of evidence for the effect of exercise on sleep intervention in patients with chronic kidney disease was low, suggesting a possible behavioral mechanism for the cardiovascular-impairing influence of sleep complaints(27, 28).
Data from the Henan Rural Cohort and NHANES also showed that people who slept less were more likely to smoke(2, 8). Among OSA patients, the proportion of frequent and occasional smokers was much higher than among non-OSA patients(26). There is weak evidence that insomnia exacerbates smoking and discourages quitting, in fact, genetic correlations and bidirectional, causal effects between sleep complaints and smoking indicating that sleep is a potential target for smoking treatment and prevention(29).
Considering the interactively confounding effects, we further explored the associations of the combination of OSA risk and sleep duration in relation to ideal CVH metrics in HD patients. Our study suggested that the association of sleep duration with ideal CVH metrics in HD patients varied with OSA risk. For patients without OSA risk, no category of sleep duration was found to be associated with ideal CVH metrics, and the relationship between short sleep duration and ideal CVH metrics was statistically significant only in patients at risk for OSA. The combined effect of sleep duration and OSA risk suggests the importance of ensuring adequate quality and amounts of sleep. In HD patients without OSA risk, the conflicting evidence for the correlation of sleep duration and ideal CVH metrics may be due to the failure of some studies to take the effect of risk for OSA into consideration(2, 8).
Strength and limitations
This study is the first to demonstrate in HD patients that OSA risk and short sleep duration are significantly associated with poor CVH, respectively, and that the co-existence of OSA risk and short sleep has a greater impact on poor CVH. Our assessment of clinical and lifestyle risk factors for CVD is very strict. The validity and reliability of the questionnaire are verified by standardized and widely used questionnaires(11, 15). Standardized anthropometric and blood pressure measurements are carried out by trained research staff and blood indicators are assessed using standardized protocols(10).Furthermore, we fully considered the relationship between OSA risk, sleep duration and ideal CVH metrics, the results of this study could suggest the important role of sleep in primary prevention of CVD in HD patients, and provide reference for clinical intervention.
This study also has some limitations. First, information about and OSA risk and sleep duration were obtained from self-reported questionnaires rather than objectively assessed. The gold standard for sleep assessment is polysomnography, but it is not feasible to obtain objective sleep data in large population studies. STOP questionnaire is a valid and effective tool for OSA risk screening in different medical population, and self-reported sleep duration as the most commonly used method is moderately consistent with PSG-measured assessments(15, 30). Second, our study eventually included a medium number of participants, and only 10.2% of them were long sleepers, which may limit the statistical effect of regression analysis. Finally, our study is a cross-sectional study, indicating that OSA risk and sleep duration are significantly correlated with ideal CVH metrics, but lack of temporality limits the inference of causality.