Effects of daytime hypercapnia on logical memory and working memory in patients with obstructive sleep apnea hypopnea syndrome

Abstract

Objective: To investigate the effects of daytime hypercapnia on logical memory and working memory in patients with obstructive sleep apnea hypopnea syndrome (OSAHS). Methods: This prospective study recruited patients complaining of snoring and diagnosed with OSAHS at the Sleep Center of the Second Affiliated Hospital of Soochow University from January to November 2020. Patients were assessed clinically and scored for their memory function. All patients underwent daytime transcutaneous carbon dioxide (PtcCO2) test, and overnight polysomnography (PSG). Logical memory was scored using the Logical Memory Test (LMT), while working memory was evaluated by Digit Span Test (DST) and Cambridge Neuropsychological Test Automated Battery (CANTAB) which included Pattern Recognition Memory (PRM), Spatial Span (SSP), and Spatial Working Memory (SWM). Patients were divided into the normocapnic group and the hypercapnic group using the daytime PtcCO2 test. The clinical and PSG parameters and the memory test scores between the two groups were compared. Binary logistic stepwise regression was conducted to identify risk factors of memory impairment in OSAHS patients. Results: Among the 123 enrolled OSAHS patients, 79 were normocapnic and 44 were hypercapnic. There was no significant difference in the general clinical parameters between the two groups. The snoring history in years in the hypercapnic group was longer than that in the normocapnic group (P<0.05). Compared with the normocapnic group, the apnea-hyponea index (AHI), oxygen desaturation index (ODI) and percentage of total sleep time with oxygen saturation level<90% (TS90) of the hypercapnic group were higher (all P<0.05), while other PSG parameters exhibited no statistically significant differences. There was no statistically significant difference in the immediate logical memory and PRM immediate accuracy rate between the two groups, while the delayed logical memory, verbal and spatial working memory, and executive function were worse in the hypercapnic group, as shown by lower total LMT scores, lower DST, lower SSP scores (all P<0.05), and higher between errors and strategy scores (P<0.01) of SWM in the hypercapnic group. Binary logistic stepwise regression showed that PtcCO2 ≥45 mmHg (1 mmHg=0.133 kPa, OR=3.055, 95%CI 1.359-6.868, P=0.007) and higher body mass index (BMI) (OR=1.132, 95%CI 1.005-1.275, P=0.041) were risk factors for poor performance in Digit Span Backwards Test. Therefore, PtcCO2 ≥45 mmHg was an independent risk factors for poor performance in delayed LMT, SSP, and between errors and strategy scores in SWM (OR=3.109, 3.941, 3.238 and 2.785, respectively, all P<0.05). Conclusion: Hypercapnia had negative impacts on logical memory and working memory of OSAHS patients, especially on the delayed logical memory, verbal working memory and spatial working memory impairment.