Academic Editor: Rafael Franco
Background: Both obstructive sleep apnea-hypopnea syndrome (OSAHS) and
obesity are related to cognitive deficits, but the interaction effects of OSAHS
and abdominal obesity on cognitive function are unclear. Thus, we performed this
study to investigate this issue. Methods: We recruited subjects who
received polysomnography test, anthropometric measurements and cognitive function
assessment and/or blood protein test. Correlations between apnea-hypopnea index
(AHI) and cognitive function were assessed. Analysis of covariance was used to
compare the differences in cognitive function between groups and detect the
interactions of OSAHS and obesity on cognitive function. Multiple linear
regression models were used to determine the associations between OSAHS and
cognitive function. Results: In total, 196 subjects with Montreal
Cognitive Assessment (MoCA), 161 subjects with Symbol Digit Modalities Test
(SDMT) and Trail making test, and 44 subjects with blood protein test were
enrolled. Significant negative correlations between AHI and visuo-spatial and
executive, language, delayed recall and total score of MoCA were observed. After
adjusting for multiple confounding factors, subjects with severe OSAHS had
significant lower delayed recall score and total score of MoCA, SDMT index, and
Obstructive sleep apnea-hypopnea syndrome (OSAHS) is a highly prevalent disease. It is been reported that approximately 25% of men and 13% of women worldwide suffer from OSAHS . Recent evidence showed an significant positive association between OSAHS and cognitive deficits . Compared with patients without OSAHS, those with OSAHS exhibited significant impaired cognitive function, including impaired vigilance and executive function, declined memory and motor coordination deficits, and the degree of cognitive deficits elevated with the severity of OSAHS . A recent meta-analysis showed that the risk of cognitive deficits in patients with OSAHS increased by 26% . The cognitive deficits associated with OSAHS affect the quality of life of patients and lead to frequent occurrence of malignant accidents at work and traffic, which severely damage national health . Thus, more attention should be attached to cognitive deficits in patients with OSAHS.
Although a robust association between OSAHS and cognitive deficits is established by previous studies, Kilpinen et al.  pointed out that future studies were needed to further clarify the joint effects of OSAHS and other comorbidities including obesity on cognitive function. Obesity often coexists with OSAHS. Obesity can not only increase the risk of OSAHS, but also have a negative impact on cognitive function . Studies have shown that obesity itself can lead to cognitive dysfunction [7, 8]. A variety of obesity indexes, including body mass index (BMI), Waist to hip ratio (WHR) and waist circumference (WC) have been shown to hold a negative impact on learning, memory and language ability . Meanwhile, in children and adolescents, obesity and OSAHS have been shown to exert a synergistic effect on cognitive deficits [10, 11, 12]. However, the joint effects of OSAHS and obesity on cognitive function in adults are unclear.
A study found that in young people, compared with mild OSAHS and the control
group, moderate to severe OSAHS is associated with increased total tau levels
. Another study found that in the middle-aged, the blood A
Thus, we performed this cross-sectional study to further verify the relationships between OSAHS and cognitive function, as well as the interactions of OSAHS and obesity on cognitive function. Except for that, we also explored the relationships between OSAHS and cognitive-related protein to provide more evidence.
The subjects were the patients admitted to the First Affiliated Hospital of Nanchang University and underwent polysomnography (PSG) between February 2019 and September 2020 due to snoring symptoms. All subjects received a questionnaire survey of basic information and past medical history before PSG. After completing these survey, the Epworth Sleepiness Scale (ESS)  was used to assess subjective sleepiness. This study was approved by the ethics committee of the First Affiliated Hospital of Nanchang University and was carried out following the Declaration of Helsinki. Subjects were informed of the general content of the study before taking the test, followed the principle of voluntariness, and signed the informed consent.
Inclusion criteria: (1) aged no less than 18 years old; (2) education years: no less than 5 years; (3) never underwent related surgery, ventilator, or related treatment (including various types of weight-loss measures, OSAHS treatment surgery, continuous positive airway pressure, etc.). Exclusion criteria: (1) serious systemic diseases (including heart failure, severe cerebral infarction, myocardial infarction, etc.); (2) a history of Mental or neurological diseases, or a history of psychotropic drug dependence; (3) taking drugs that affect nerve function or drinking alcoholic beverages within 24 hours before receiving relevant examinations; (4) a history of other sleep-related diseases; (5) unable to cooperate with the research due to various reasons; (6) missing or incomplete data.
Height, weight, neck circumference (NC), WC, hip circumference (HC), and BMI are
measured and assigned readings by the same professional researchers using the
same measuring equipment (tape measure, automatic height and weight instrument),
based on the WHO standard method. The criteria for determining abdominal obesity
of Chinese adults: male WC
Subjects are not allowed to drink tea, alcohol, or coffee within 24 hours before monitoring and must not have a history of taking sedatives and hypnotic drugs shortly. All subjects used a PSG based on a non-interference single laboratory to monitor and scored respiratory events. Simultaneously recorded electroencephalogrphy, electromyogram, electrooculogram, electrocardiogram, nasal cavity and oral airflow, Chest and abdominal cavity respiratory activity, snoring, pulse oximetry, and other data. All tests started before 10:00 at night and ended after 6:00 AM the next day.
After the monitoring, the professional and technical personnel will interpret
and review according to the American Academy of Sleep Medicine (AASM) criteria
and related event interpretation standards . The apnea-hypopnea index (AHI)
refers to the average number of apneas and hypopneas per hour during sleep [17, 18]. Lowest oxygen saturation (LSpO2) refers to the lowest value of blood oxygen
during sleep, in % [17, 18]. Oxygen-desaturation index (ODI) refers to the
number of times that blood oxygen drops by 4% or more from baseline per hour
during sleep [17, 18]. According to the standards established by the AASM, AHI
All MoCA scale evaluations are conducted under the same standard guidance by the professionally trained personnel. After completing the evaluation, the professional analyzed each field’s scores and total scores according to the same standards.
(1) Chinese-Beijing Version of Montreal Cognitive Assessment (MoCA) : MoCA is used to initially and quickly detect the overall cognitive level of patients, including 8 items, namely visuo-spatial and executive, naming, attention, language, abstraction, delayed recall, orientation and total score.
(2) Symbol Digit Modalities Test (SDMT) : SDMT is a scale used to assess attention and processing speed. The subjects were asked to fill in the symbols corresponding to the numbers in the form within 90 seconds. The number of correctly completed within 90 seconds (correct number) is recorded as an analysis indicator.
(3) Trail making test-Part A (TMT) : TMT is also a scale used to assess attention and processing speed. Record the time (in seconds) spent to complete all connections, the number of errors, and the number of pen strokes. The shorter the time, the better the performance.
After receiving the PSG monitoring, we draw 5 mL of fasting blood from the
subjects participating in the blood test. Use a centrifuge to separate the serum
within 60 minutes at 3000 rpm and 10 minutes. After separation, use a cryotube to
store and immediately freeze at –80
Data are presented as means
In total, 196 subjects who completed (MoCA) test and 161 subjects who completed the SDMT and TMT were enrolled.
For subjects who completed MoCA, they were divided into subjects with severe
OSAHS (n = 119) and subjects with non-severe
OSAHS group (n = 77). Compared to subjects with non-severe OSAHS, those with
severe OSAHS were more obese (exhibited by higher BMI, NC, WC and HC), and had
significant higher ESS and more severe OSAHS (exhibited by higher AHI and ODI,
and lower LSPO2) (all p
|Variable||Overall||Non-Severe OSAHS||Severe OSAHS||p
|(n = 196)||group (n = 77)||group (n = 119)|
||26.45 (24.65, 29.02)||25.71 (23.88, 27.60)||26.81 (24.90, 29.76)||0.003||-|
|Age, years||41.00 (34.00, 52.00)||40.00 (32.00, 52.00)||43.00 (35.00, 52.00)||0.164||-|
|NC, cm||40.00 (38.00, 42.00)||38.50 (37.00, 41.00)||41.00 (39.00, 43.50)||-|
|WC, cm||96.00 (90.88, 101.62)||94.00 (88.50, 97.00)||97.80 (92.25, 104.00)||-|
|HC, cm||101.00 (97.00, 105.00)||100.00 (96.00, 103.50)||102.50 (98.00, 106.00)||0.025||-|
|ESS||10.00 (6.00, 14.00)||8.00 (4.00, 13.00)||12.00 (8.00, 15.00)||-|
|Education level, years||11.00 (8.00, 15.00)||11.00 (8.00, 15.00)||11.00 (8.00, 14.00)||0.609||-|
|Male, N (%)||173 (88.3)||65 (84.4)||108 (90.8)||0.263||-|
|Abdominal Obesity, N (%)||156 (79.6)||56 (72.7)||100 (84.0)||0.082||-|
|Hypertension, N (%)||55 (28.1)||23 (29.9)||32 (26.9)||0.771||-|
|Diabetes, N (%)||7 (3.6)||3 (3.9)||4 (3.4)||1||-|
|Smoking, N (%)||93 (47.4)||32 (41.6)||61 (51.3)||0.237||-|
|Drinking, N (%)||96 (49.0)||32 (41.6)||64 (53.8)||0.127||-|
|AHI||38.10 (19.42, 63.12)||12.70 (4.50, 21.30)||60.60 (44.60, 71.45)||-|
|LSPO2||78.00 (65.75, 86.00)||87.00 (83.00, 90.00)||70.00 (61.00, 78.50)||-|
|ODI||35.95 (14.88, 67.60)||9.80 (3.00, 18.10)||62.80 (39.00, 74.65)||-|
|Visuo-spatial and executive||4.00 (3.00, 5.00)||4.00 (4.00, 5.00)||4.00 (3.00, 5.00)||0.119||0.265|
|Naming||3.00 (3.00, 3.00)||3.00 (3.00, 3.00)||3.00 (3.00, 3.00)||0.383||0.355|
|Attention||6.00 (6.00, 6.00)||6.00 (6.00, 6.00)||6.00 (6.00, 6.00)||0.464||0.298|
|Language||2.00 (2.00, 3.00)||2.00 (2.00, 3.00)||2.00 (2.00, 2.00)||0.081||0.506|
|Abstraction||2.00 (1.00, 2.00)||2.00 (1.00, 2.00)||2.00 (1.00, 2.00)||0.201||0.278|
|Delayed recall||2.00 (1.75, 4.00)||3.00 (2.00, 4.00)||2.00 (1.00, 3.00)||0.007||0.012|
|Orientation||6.00 (6.00, 6.00)||6.00 (6.00, 6.00)||6.00 (6.00, 6.00)||0.844||0.965|
|Total score||25.00 (23.00, 27.00)||26.00 (24.00, 27.00)||25.00 (22.00, 26.00)||0.002||0.002|
For subjects who completed SDMT and TMT, they were also divided into subjects
with severe OSAHS (n = 98) and subjects with non-severe OSAHS group (n = 63).
Compared to subjects with non-severe OSAHS, those with severe OSAHS were more
obese, and had significant higher ESS, more alcohol consumption and more severe
OSAHS (all p
|Variable||Overall||Non-Severe OSAHS||Severe OSAHS||p
|(n = 161)||group (n = 63)||group (n = 98)|
||26.06 (24.22, 28.96)||25.31 (23.61, 27.52)||26.45 (24.78, 30.06)||0.01||-|
|Age, years||43.00 (34.00, 51.00)||40.00 (30.50, 51.00)||44.50 (34.25, 51.00)||0.16||-|
|NC, cm||40.00 (38.00, 42.00)||38.00 (37.00, 40.25)||41.00 (39.00, 43.00)||-|
|WC, cm||96.00 (90.00, 101.00)||93.00 (87.50, 97.00)||96.50 (92.62, 104.00)||0.001||-|
|HC, cm||101.00 (96.00, 105.00)||101.00 (94.75, 104.00)||101.25 (97.25, 106.00)||0.066||-|
|ESS||10.00 (7.00, 14.00)||9.00 (4.00, 13.00)||11.00 (8.00, 15.00)||0.001||-|
|Education level, years||8.00 (8.00, 14.00)||8.00 (8.00, 14.00)||11.00 (8.00, 14.00)||0.384||-|
|Male, N (%)||141 (87.6)||53 (84.1)||88 (89.8)||0.412||-|
|Abdominal Obesity, N (%)||126 (78.3)||45 (71.4)||81 (82.7)||0.136||-|
|Hypertension, N (%)||31 (19.3)||11 (17.5)||20 (20.4)||0.796||-|
|Diabetes, N (%)||2 (1.2)||1 (1.6)||1 (1.0)||1||-|
|Smoking, N (%)||86 (53.4)||29 (46.0)||57 (58.2)||0.179||-|
|Drinking, N (%)||70 (43.5)||20 (31.7)||50 (51.0)||0.025||-|
|AHI||39.70 (19.70, 63.20)||12.20 (3.65, 21.35)||59.25 (45.02, 74.12)||-|
|LSPO2||78.00 (66.00, 87.00)||88.00 (83.00, 90.00)||69.00 (61.00, 76.00)||-|
|ODI||37.40 (11.40, 66.00)||7.90 (2.10, 16.80)||56.10 (41.02, 79.70)||-|
|Attention and processing speed|
|SDMT||46.00 (34.00, 57.00)||47.00 (36.00, 59.00)||45.00 (34.00, 55.75)||0.154||0.019|
|TMT||43.00 (32.00, 56.00)||43.00 (32.00, 58.00)||43.00 (34.25, 55.75)||0.981||0.48|
For subjects who completed A
|Variable||Overall||Non-Severe OSAHS||Severe OSAHS||p
|(n = 44)||group (n = 25)||group (n = 19)|
||27.39 (24.47, 30.57)||25.95 (23.71, 29.41)||27.74 (26.06, 31.12)||0.09||-|
|Age, years||42.50 (32.25, 52.25)||46.00 (38.00, 57.00)||36.00 (30.50, 51.00)||0.148||-|
|NC, cm||39.75 (36.88, 42.12)||38.00 (35.00, 40.00)||41.00 (39.50, 43.50)||0.004||-|
|WC, cm||94.50 (87.75, 101.00)||92.00 (87.00, 96.00)||100.00 (93.00, 104.50)||0.063||-|
|HC, cm||102.00 (94.38, 107.00)||99.00 (92.00, 106.00)||104.00 (99.50, 108.50)||0.061||-|
|ESS||11.00 (7.75, 13.25)||12.00 (7.00, 13.00)||9.00 (8.00, 14.00)||0.695||-|
|Education level, years||8.00 (8.00, 11.00)||8.00 (5.00, 11.00)||11.00 (8.00, 11.50)||0.066||-|
|Male, N (%)||33 (75.0)||16 (64.0)||17 (89.5)||0.114||-|
|Abdominal Obesity, N (%)||32 (72.7)||17 (68.0)||15 (78.9)||0.641||-|
|Hypertension, N (%)||7 (15.9)||3 (12.0)||4 (21.1)||0.691||-|
|Diabetes, N (%)||0 (0.0)||0 (0.0)||0 (0.0)||1||-|
|Smoking, N (%)||20 (45.5)||11 (44.0)||9 (47.4)||-||-|
|Drinking, N (%)||18 (40.9)||12 (48.0)||6 (31.6)||0.431||-|
|AHI||21.05 (4.50, 53.08)||5.80 (1.60, 17.00)||58.80 (48.55, 73.75)||-|
|LSPO2||83.50 (65.00, 87.50)||87.00 (86.00, 90.00)||61.00 (55.00, 71.00)||-|
|ODI||20.55 (4.38, 56.60)||4.60 (0.80, 12.40)||65.30 (43.80, 82.00)||-|
|Attention and processing speed|
||242.39 (159.80, 334.49)||278.22 (151.54, 339.04)||215.48 (164.53, 293.94)||0.32||0.011|
||185.93 (125.02, 243.24)||189.51 (126.22, 241.82)||169.17 (116.00, 228.21)||0.53||0.176|
|tau||27.78 (16.21, 42.73)||30.50 (16.31, 41.40)||27.64 (16.41, 45.79)||0.972||0.547|
To explore the relationships between severe OSAHS and cognitive functions, we
first ran a correlation analysis between AHI and MoCA. Significant negative
correlations between visuo-spatial and executive, language, delayed recall and
total score of MoCA and AHI have been observed (p
The correlation between AHI and each score of the MoCA scale.
The correlation between AHI and SDMT and TMT-A indexes. (A) SDMT. (B) TMT-A.
The correlation between A
After adjusting for gender, age, years of education, ESS score, abdominal
obesity, smoking and alcohol consumption, compared to subjects with non-severe
OSAHS, those with severe OSAHS had significant lower delayed recall score and
total score of MoCA (p = 0.012, and p = 0.002, respectively)
(Table 1 and Fig. 4). And subjects with severe OSAHS also had significant lower
SDMT index (p = 0.019) (Table 2 and Fig. 5) and lower A
Differences in each score of MoCA scale between severe OSAHS and non-severe OSAHS groups (That is non-to-moderate OSAHS in the figure). (A) Visuo-spatial and executive. (B) Naming. (C) Attention. (D) Language. (E) Abstraction. (F) Delayed recall. (G) Orientation. (H) Total score.
Differences in them between severe OSAHS group and non-severe OSAHS group (That is non-to-moderate OSAHS in the figure). (A) SDMT. (B) TMT-A.
Differences in A
For those significant different cognitive function assessment between the groups
including delayed recall, MoCA total score, SDMT index and A
|Severe OSAHS||Abdominal obesity||Gender||Age||ESS||Smoking||Alcohol intake||Education|
|MoCA total score||Beta||–0.180||–0.050||–0.073||–0.261||–0.044||–0.047||–0.040||0.454|
|MoCA, Montreal Cognitive Assessment; SDMT, Symbol Digit Modalities Test; OSAHS, Obstructive sleep apnea-hypopnea syndrome; ESS, Epworth Sleepiness Scale.|
Subjects were divided into four groups according to abdominal obesity and severe
OSAHS. Strong main effects of severe OSAHS were detected on delayed recall (F =
|Non-abdominal obesity||Abdominal obesity||p
||Main effect, p
|Non-Severe OSAHS||Severe OSAHS||Non-Severe OSAHS||Severe OSAHS||Abdominal obesity||Severe OSAHS|
|(n = 21)||(n = 19)||(n = 56)||(n = 100)|
|Visuo-spatial and executive||4.62
|Total MoCA score||26.71
|Non-abdominal obesity||Abdominal obesity||p
||Main effect, p
|Non-severe OSAHS||Severe OSAHS||Non-severe OSAHS||Severe OSAHS||Abdominal obesity||Severe OSAHS|
|(n = 18)||(n = 17)||(n = 45)||(n = 81)|
An interaction of severe OSAHS and abdominal obesity was found for language (F =
Interaction plot between severe OSA and abdominal obesity on language function.
In this study, we found that significant negative correlations between AHI and
visuo-spatial and executive, language, delayed recall and total score of MoCA.
Subjects with severe OSAHS had significant lower delayed recall score and total
score of MoCA, SDMT index, and A
We have found that severe OSAHS increased the risk of cognitive deficits, evidenced by lower cognitive function assessments in subject with severe OSAHS and negative associations between severe OSAHS and cognitive function assessments, which are consistent with previous results. Systematic meta-analysis generally suggests that OSAHS can cause delayed long-term visual and language memory, impaired attention/alertness, declined visuospatial/structural ability and executive function deficits . It is generally believed that language ability and psychomotor function are not affected by OSAHS, when OSAHS has ambiguous effects on working memory, short-term memory and overall cognitive function . Not only that, the correlation of OSAHS with cognitive functioning (attention, memory, thinking, speech) observed in Berdina et al. ’s study also supports our results.
At present, it is mainly believed that OSAHS especially impaired cognitive function through intermittent hypoxia and disrupted sleep architecture. Long-term intermittent hypoxia can cause cerebral vasoconstriction and changes in cerebral hemodynamics, which slows down the blood flow rate, further making the brain in a state of low perfusion for a long time, and ultimately aggravating neuronal hypoxia . The cerebral cortex, hippocampus, and cerebellum are particularly sensitive to hypoxia and easy to be damaged to cause cognitive deficits . Meanwhile, the inflammatory response and oxidative stress response triggered by intermittent hypoxia [26, 27] will also lead to cognitive deficits, showing short-term decreased attention and vigilance, long-term memory, and decreased executive function . Besides, sleep structure disturbances, manifesting as sleep fragmentation and total sleep time reduction caused by repeated arousals, prolonged light sleep, and shortened deep sleep, are also be shown to aggravate cognitive deficits . Sleep fragmentation damages the fine structures of the brain and destroy the integrity of axons extensively . Long-term lack of sleep causes permanent neuronal degeneration, especially in memory-related brain regions such as the hippocampus, reduces the activation of the frontal and parietal networks and changes the internal functions of the thalamus .
Previous studies have found that the lack of sleep, disturbed sleep and
intermittent hypoxia of OSAHS elevated the level of A
Obesity has been identified as a risk factor for OSAHS [36, 37]. Weight gain reduces the size of the upper respiratory tract, thereby increasing the incidence of apnea events . And the physiological component of obesity, including blood sugar control, insulin action, and leptin signaling, cause a greater reduction in the muscle tone of the pharyngeal dilator and increase the chance of obstruction during sleep . Meanwhile, obesity is shown to have a negative impact on cognitive function [6, 7, 8, 9]. In our study, we also found that abdominal obesity had strong main effects on visuo-spatial and executive ability and language ability (Table 5).
Although numerous studies have confirmed the negative effects of obesity and OSAHS on cognitive function alone, the joint effect of the two on cognitive function is still unclear. Thus, we tried to explore the interaction of abdominal obesity and severe OSAHS on cognitive function. After adjusting for confounding factors, the interaction of severe OSAHS and abdominal obesity on language score of MoCA was found. In a further simple effect analysis, in non-abdominal obesity and non-severe OSAHS levels, severe OSAHS and abdominal obesity had a significant negative impact on language scores. On the levels of abdominal obesity and severe OSAHS, respectively, severe OSAHS and abdominal obesity had no significant effect on language scores. This result suggests that severe OSAHS and abdominal obesity may have an antagonistic impact on language function, which is confusing and need to be further studied.
There are some limitations in this research. First of all, the cognitive domains evaluated in this study are limited, including only the metrics that cover cognitive domains and individual attention and information processing speed in MoCA. Secondly, although this study has a larger sample size than many previous single-center studies, it still failed to get more positive results, suggesting that such studies require larger sample size data to support. Therefore, the results of this study still need to be confirmed in large samples and in other populations in the future. Third, although we have adjusted some factors that may affect cognitive function, we cannot adjust the interference of other confounding factors that cannot be measured, such as APOE4 genotype and tester’s skill level. Fourth, we have not been able to find a suitable mechanism to explain the interaction of abdominal obesity and severe OSAHS on language ability. Despite this, the sleep data and the relatively large sample size increased the credibility of our results.
We found that OSAHS increased the risk of cognitive deficits. And interaction effect of severe OSAHS and abdominal obesity was detected on language. More large-sample-size studies are required to further clarify the interaction of OSAHS and comorbidities on cognitive deficits.
ZYZ and YYX designed the research; XF prepared the figures and drafted the manuscript; YYX, YPX and XF analyzed the data; XF contributed analytic tools and finalized the manuscript; XF, YZ, LLZ and JQL participated in the writing of the manuscript. All authors have read and approved the final manuscript.
This study was approved by the ethics committee of the First Affiliated Hospital of Nanchang University (Approval No. 2019- 051) and was carried out following the Declaration of Helsinki. Subjects were informed of the general content of the study before taking the test.
The authors sincerely thank all the participants and the supporting from the Research Fund Project of Jiangxi Provincial Department of Education (180017) and the Jiangxi Natural Science Foundation Project (Grant No. 20202BABL216028).
The authors declare no conflict of interest.