Corrigendum: Moderate-to-high risk of obstructive sleep apnea with excessive daytime sleepiness is associated with postoperative neurocognitive disorders: a prospective one-year follow-up cohort study
Year: 2024
Type: erratum
Abstract: Obstructive sleep apnea (OSA), characterized by chronic intermittent hypoxemia and sleep fragmentation, is primarily caused by episodes of complete or partial airway obstruction during sleep (Malhotra and White, 2002). OSA occurs in 10%–20% of the population, increases with age, and peaks after 60 (Bixler et al., 1998; Bixler et al., 2001; Peppard et al., 2013). Excessive daytime sleepiness (EDS) is the most common symptom in patients with OSA, and its clinical features are daytime drowsiness, reduced wakefulness, and decreased vigilance (Sauter et al., 2000). EDS has a prevalence of 16% to 22% in OSA patients and often means patients at high risk of OSA. Previous studies showed that OSA patients with EDS as a distinct clinical phenotype had lower oxygen saturation and more sleep disturbance than those without EDS (Zhou et al., 2016). Many studies recommended Epworth Sleepiness Scale (ESS) for measuring subjective EDS and multiple sleep latency test (MSLT) for objective EDS, respectively (Ren et al., 2016). Daytime naps recorded by actigraphy also reflect objective EDS, (Leng et al., 2018) which is a compensatory reaction to nocturnal sleep fragmentation, and it can also be a marker of EDS in older adults (Hsieh et al., 2016).Previous studies have shown that OSA is associated with impaired cognitive function, including attention, alertness, memory, and executive function (Andrade et al., 2018; Stepnowsky et al., 2019; Bubu et al., 2020; Léger and Stepnowsky, 2020; Liu et al., 2020). EDS may also contribute to cognitive impairment in OSA patients, especially attention and memory function because EDS is strictly related to sleep deprivation and fragmentation (Steiropoulos et al., 2019). Many studies reported that OSA patients with EDS have more significant cognitive impairments involving attention and vigilance, learning and memory, and executive function than those without EDS (Zhou et al., 2016). Several pathophysiological factors, such as intermittent hypoxia, systemic inflammation, and oxidative stress, may influence cognitive function in OSA patients. It has been speculated that the interaction between EDS and hypoxemia contributes to neurocognitive deficits (Roche, 2016).The prevalence of OSA in the surgical population is higher than in the normal population and varies widely among different surgery patients (Van Onselen et al., 2013; Zaremba et al., 2016). Multiple perioperative factors such as anesthetics and analgesics can worsen OSA. In addition, reorganization of sleep architecture with rapid eye movement sleep rebound attributed to postoperative disrupted, reduced, and poor-quality sleep leads to exacerbation of OSA and EDS (Finkel et al., 2009; Young et al., 2009). Add to them downstream reactions of surgery stress and anesthetics such as inflammatory responses, oxidative stress, metabolic disorders, and neuronal damage, the postoperative complications are more common in surgical populations with OSA than those without OSA (Memtsoudis et al., 2011; Roggenbach et al., 2014; Lam et al., 2017; Chan et al., 2019).In recent years, OSA as a substantial risk factor for postoperative delirium and cognitive decline has aroused scholars' attention. Postoperative neurocognitive disorder (PND) is a term that refers to cognitive impairment associated with anesthesia and surgery, including the acute event (postoperative delirium, POD) and cognitive decline diagnosed up to one year after surgery (postoperative cognitive dysfunction, POCD) (Evered et al., 2018). POD occurs in 13%-34% of surgical patients, and POCD occurs in 7%-50%, depending on surgery type and surgical populations (Postler et al., 2011; Norkienė et al., 2013; Krenk et al., 2014; Aceto et al., 2015; Goettel et al., 2017; Kotfis et al., 2018; Wang et al., 2018a; Chaiwat et al., 2019; Guo et al., 2020; Kang et al., 2020; Wang et al., 2020; Yang et al., 2020; Li et al., 2021). Previous studies showed that the prevalence of POD and POCD were 14% and 19% among elderly patients undergoing gastrointestinal surgery, respectively (Yang et al., 2020; Li et al., 2021). Enough evidence has shown that the development of PND is associated with inflammation, oxidative stress, metabolic disorders, increased blood–brain barrier permeability, and neuronal damage (Hudetz et al., 2011; Subramaniyan and Terrando, 2019; Lin et al., 2020). Patients who developed PND have an increased risk of postoperative complications and a worse prognosis (Abelha et al., 2013; Abawi et al., 2016; Lingehall et al., 2017; Ruggiero et al., 2017; Kotfis et al., 2018; Racine et al., 2018; Shi et al., 2019; Boone et al., 2020; Cai et al., 2020).So far, ten studies have explored the correlation between OSA and POD. Four studies confirmed that OSA increased the risk of POD, (Flink et al., 2012; Roggenbach et al., 2014; Nadler et al., 2017; Wu et al., 2022) while the remaining six did not identify their association (Gupta et al., 2001; Wang et al., 2018b; Strutz et al., 2019; Tafelmeier et al., 2019; King et al., 2020; Oldham et al., 2021). Only three human studies and three animal experiments have explored the correlation between OSA and POCD. Two human studies showed that the incidence of POCD before discharge in OSA patients was lower than that of non-OSA patients, (Wagner et al., 2018; Wagner et al., 2021) and one human study showed that OSA patients had a higher risk of POCD in the early postoperative period than patients without OSA (Wu et al., 2022). Three animal models also suggested that the incidence of POCD was higher in OSA mice than that in non-OSA mice (Dong et al., 2018; Zhang et al., 2019; Mei et al., 2020). However, previous studies assessed cognitive function within two days after extubating with a lack of long-term postoperative follow-up cognitive assessments at 1-month or 1-year after surgery. In addition, no study, to our knowledge, has shown the effect of moderate-to-high risk of OSA combined with EDS on PND within one year after surgery. Therefore, the first aim of our study was to explore the impact of moderate-to-high risk of OSA on the PND within one year postoperatively, and the second aim was to investigate the effect of moderate-to-high risk of OSA with EDS on PND within one year after surgery. We hypothesized that moderate-to-high risk of OSA with EDS was associated with the development of PND in older adults undergoing gastrointestinal surgery.Study designWith the ethical approval from the institutional review board of West China Hospital, this single-center prospective cohort study was conducted from June 4, 2019, to October 14, 2021, at a gastrointestinal surgical unit (≥80 beds, approximately 35 nurses, and 20 surgeons) of West China Hospital in Chengdu, China. All participants or their legal representatives provided written informed consent before the study. Patients did not receive financial compensation. Study methods and results are reported following the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement for prospective cohort studies (Vandenbroucke et al., 2007).ParticipantsWe screened older adults (≥65 years) who were scheduled for stomach and intestinal surgery between June 4, 2019, and October 14, 2020. Patients who finally underwent the gastrointestinal procedure with at least a 3-day hospital stay, without communication barriers, and had a suitable condition for evaluation were included. The exclusion criteria are as follows: 1) cognitive impairment with a Mini-Mental State Examination (MMSE) scores less than 20; 2) delirium (assessed by Confusion Assessment Method-Severity (CAM-S Short Form) at baseline; 3) a documented history of severe psychosis such as severe depression, severe anxiety, autism, and schizophrenia; 4) a terminal condition with a life expectancy of fewer than six months (metastatic cancer, multiple organ failure, or receiving chemotherapy); 5) exploratory laparotomy without tumor removal or transferring to other departments after the operation; and 6) alcohol abuse or dependence within the last three months. Patients were also excluded if they had missing baseline, procedural, or outcome data (Figure 1).Assessments The face-to-face evaluation was performed at the ward of the gastrointestinal surgery department on admission day by a researcher who received training in psychiatry and assessment methods. The baseline assessment included risks of OSA, subjective EDS, and objective napping. Then, the researcher and trained nurses assessed POD using the CAM-S twice daily (between 2–6 pm and 2–8 am) until postoperative day seven or discharge. Another trained researcher blinded to the POD results performed postoperative cognitive assessments at discharge using MMSE and Telephone Interview for Cognitive Status-40 (TICS-40). Postoperative cognitive function at one month and one year after surgery were also evaluated using TICS-40 by telephone. The criteria for loss to follow-up were as follows: 1) refused to participate; 2) cannot be contacted; and 3) died (Figure 1). Exposure The STOP-BANG questionnaire has good sensitivity and high diagnostic accuracy for detecting OSA in Asian populations (Chan et al., 2019). A STOP-BANG score of 3-8 indicates a moderate-to-high risk of OSA (OSA group) among the Chinese people, and a 0-2 score means a low risk of OSA (non-OSA group) (Chan et al., 2019). Subjective EDS was defined as an ESS score ≥11, and the Chinese version of ESS was a reliable and valid tool for measuring subjective EDS (Pak et al., 2017). Objective napping was measured using an Actiwatch-2 (Phillips Respironics Mini-Mitter), worn continuously on the non-dominant wrist for a minimum of two consecutive 24h periods before operation. Participants also completed sleep logs for the period they wore the actigraphy and reported information regarding times they napped. Data for devices were collected in 1-min epochs and scored using Actiware software version 6.0.9 (Phillips Respironics, Bend, OR). Daily napping duration was calculated by summing up the time of napping periods throughout the day and averaging across all recording days. A nap ≥1h/ day during the day was considered objective napping (Leng et al., 2018). For convenience, subjective EDS and objective napping will be referred to as “EDS” throughout the rest of the manuscript.Cognitive functionThe CAM-S provides a new delirium severity measure with strong psychometric properties and associations with important clinical outcomes. The Short Form includes the same four features as CAM: acute change or fluctuation, inattention, disorganized thinking, and altered level of consciousness. The diagnostic criteria of delirium are that the first and second criteria must be present, coming with the third or fourth criteria. The sum score of CAM-S Short Form ranges from 0 to et al., The Chinese version of the CAM-S good and in postoperative delirium among Chinese patients et al., 2019). MMSE is scored from 0 to with a of and a of The version of the MMSE has been in elderly a high sensitivity of and of et al., 2016). MMSE is the most widely in for cognitive in patients et al., 2021). The Chinese version of the TICS-40 was and scored on a of 0 to with a higher score cognitive function et al., 2021). A TICS-40 score was defined as cognitive and a score referred to to a previous study et al., 2021). for learning time with an that is using of MMSE and TICS-40 et al., 2015; et al., 2022). the and scores the of at postoperative evaluation to previous studies the et al., et al., 2009). also collected information on including age, and the of The Scale has an of in the Chinese et al., 2021). The score of is 0 means means and means et al., 2021). Form is a to the of the The Chinese version of performed and a score of is normal are at risk of and et al., et al., 2022). data and anesthesia or a of were from were the incidence of including POD during POCD at one and one year after surgery. on one of all postoperative days before discharge was as POD et al., 2014; et al., The term POCD in the from within after surgery to months (Evered and 2018). In 1-month has been as a time in which the acute effect of surgery and anesthesia has (Evered and 2018). POCD was developed early after surgery and cognitive function three months after surgery (Evered et al., 2018; and 2018). patients had cognitive decline up to year after this may a to (Evered et al., 2018). Therefore, early POCD was assessed at discharge and defined as at least one cognitive score and the TICS-40 or more to the baseline score following et al., et al., 2022). POCD at one month and one year were defined as a decline in 1-month and 1-year TICS-40 of or more to TICS-40 respectively et al., et al., et al., data were using for data and means with or with for between PND and using or test for and if the were not the test for The sum were also for For a multiple to and the effect of moderate-to-high risk of OSA, subjective EDS, and objective napping for to the effect of moderate-to-high risk of OSA and EDS, performed models that included to all of moderate-to-high risk of OSA and We first the moderate-to-high risk of OSA as a no moderate-to-high risk of OSA with no subjective EDS and no objective napping as a All participants in our study combined the was not included in the factors included age, and MMSE score in and factors included age, and level in and data were performed using version and All were and a was as of the study population on PND for the perioperative to the of PND are in A of participants years) were included in the POD and patients developed POD. older adults years) for POCD at had POCD. We finally included years) and years) patients for POCD at one month and one year The of patients who developed POCD at one month was and the of patients who POCD at one year were The postoperative cognitive assessment results are shown in was no significant among the PND POCD at POCD at one and POCD at one and in and surgical data for the scores and MMSE The scores were higher in the POD than in the In addition, MMSE scores were lower in the POCD to the at one between the moderate-to-high risk of OSA, subjective EDS, and objective between the moderate-to-high risk of OSA, subjective EDS, and objective napping are shown in The results showed that subjective EDS was associated with moderate-to-high risk of OSA and objective napping. However, as shown in patients with objective napping had a higher risk of OSA without of the moderate-to-high risk of OSA, subjective EDS, or objective napping on was and the results are shown in results in are shown in and results in are shown in for patients who reported subjective EDS had three times as to POCD at discharge with patients without subjective The participants who had objective napping also had three times as to POCD at discharge with those without objective napping. In addition, the incidence of POCD at one month after surgery increased by among patients with objective napping than those without napping. Objective napping also caused a in the risk of POCD at one year after surgery. However, patients with moderate-to-high risk of OSA had a higher incidence of POD POCD at discharge one month and one year after surgery with patients with low risk of OSA, which were not significant risk of OSA combined with subjective EDS or objective napping was associated with results of the are shown in results in are shown in and results in are shown in with moderate-to-high risk of OSA and subjective EDS showed an of for POCD at discharge with patients a moderate-to-high risk of OSA. Patients with moderate-to-high risk of OSA and subjective EDS also an of for risk of POCD at discharge with patients without moderate-to-high risk of OSA, subjective EDS, and objective with moderate-to-high risk of OSA and objective napping showed an of for POCD at an of for POCD at and an of for POCD at 1-year after surgery with patients with moderate-to-high risk of OSA patients with moderate-to-high risk of OSA and objective napping an of for POCD at an of for POCD at and an of for POCD at 1-year after surgery patients without moderate-to-high risk of OSA, subjective EDS, and objective our knowledge, this is the first prospective cohort study to explore the moderate-to-high risk of OSA and moderate-to-high risk of OSA combined with EDS are associated with We that patients with moderate-to-high risk of OSA had a higher incidence of PND than those with OSA, which were not significant However, moderate-to-high risk of OSA combined with subjective EDS increased the risk of POCD at discharge. moderate-to-high risk of OSA combined with objective napping increased the risk of POCD at one and one year after results did not a significant association between moderate-to-high risk of OSA and POD, which were with three studies that no association between high risk of OSA assessed by the STOP-BANG questionnaire and the incidence of POD et al., 2018b; King et al., 2020). However, were four cohort studies suggested that a higher will the risk of POD development (Flink et al., 2012; Roggenbach et al., 2014; Nadler et al., 2017; Wu et al., 2022). The of the results may be to the of the STOP-BANG questionnaire to the severity of OSA. Previous studies showed that the STOP-BANG had high of sensitivity and low of to of the population et al., 2018). Therefore, patients with a low risk of OSA were diagnosed with a moderate-to-high risk of OSA, which caused the in POD between OSA and non-OSA to be not the CAM-S by to POD will have low sensitivity et al., POD at other times may be for the evaluation Then, the incidence of POD in the OSA was in our study, which was lower than the and reported by previous (Flink et al., 2012; King et al., 2020). study may lack the to this in POD between the two to However, it is also to assessment of the STOP-BANG questionnaire the for POD risk regarding patients undergoing surgery who not which with high and time can the and of OSA. studies are to explore the association between OSA and POD and the of OSA and the accuracy of the STOP-BANG questionnaire for POD risk study assessed the effect of moderate-to-high risk of OSA on the risk of POCD at one and one year after surgery, and the effect of OSA on POCD during 1-year follow-up periods has no However, one clinical clinical data and that OSA diagnosed by may postoperative cognitive function before discharge (Wu et al., 2022). In to the of the STOP-BANG questionnaire to identify OSA the of to cognitive impairment may in the between the two for the of POCD are in the of multiple as a test (Evered et al., 2018). Another is that the incidence of POCD on the type of surgery and surgical Patients undergoing are more to PND than patients with gastrointestinal surgery, (Postler et al., 2011; Norkienė et al., 2013; Krenk et al., 2014; Aceto et al., 2015; Goettel et al., 2017; Kotfis et al., 2018; Wang et al., 2018a; Chaiwat et al., 2019; Guo et al., 2020; Kang et al., 2020; Wang et al., 2020; Yang et al., 2020; Li et al., 2021) and our study may more to in PND between the two previous models also confirmed that chronic intermittent worsen cognitive during the first four days after surgery (Dong et al., 2018; Zhang et al., 2019; Mei et al., 2020). The of animal models is that they a often level of risk and complete Therefore, the of OSA may an important in POCD in clinical and the between OSA and POCD within one year after surgery in a prospective cohort study with a and the for OSA and POCD be study that moderate-to-high risk of OSA with subjective EDS was associated with POCD at and moderate-to-high risk of OSA with objective napping was a for POCD at one and one year after surgery. The of the moderate-to-high risk of OSA with EDS may the incidence of PND is Only three animal experiments that POCD in to chronic intermittent be attributed to by and (Dong et al., 2018; Zhang et al., 2019; Mei et al., 2020). one study involving OSA patients suggested that OSA with objective EDS was the more severe phenotype of the disorder associated with et al., 2017). Another clinical study that EDS be a clinical marker to identify patients with severe OSA at risk of metabolic et al., 2016). A recent study that subjective EDS in OSA may be related to neuronal and in the et al., studies of also have in a that and damage are in patients with OSA and EDS et al., 2021). In OSA with EDS a severe phenotype and is as an important to outcomes. stress can and metabolic disorder and neuronal cognitive the for to attention to OSA in with EDS in elderly patients, which may impact the development of PND and clinical regarding the results of this study not the of OSA to the combined of the STOP-BANG questionnaire and subjective objective napping to Another important to is in our study, moderate-to-high risk of OSA with subjective EDS POCD at discharge. the moderate-to-high risk of OSA with objective naps POCD at one and one year after surgery. The be that subjective EDS reflect sleepiness is less to naps among older which to of napping study showed that older adults reported EDS and did not napping or they et al., 2020). studies have the effect of objective daytime naps and subjective EDS on cognitive et al., 2013; et al., The suggested that daytime a and objective measure of daytime patients with cognitive than subjective In addition, subjective EDS and objective napping may reflect two different The ESS the subjective of daytime sleepiness from impaired attention, et al., daytime nap is associated with an increased level of or et al., 2018). is by studies that not sleepiness is associated with in patients with OSA et al., 2017; et al., 2017). Therefore, objective to subjective EDS, is a of daytime impairment and PND is a to daytime naps in the evaluation of older adults with OSA before surgery. study has evaluated OSA using STOP-BANG not which may the severity of OSA, especially in the of high The association between OSA, especially severe OSA combined with EDS, and PND may be than our study actigraphy to objective daytime is considered the for the objective measure of daytime It is if actigraphy has the same effect as for measuring objective However, that is a and measure of EDS, is a to and methods of objective EDS to be in the evaluation of OSA patients. provided to actigraphy in older adults with OSA before surgery. did not a test considered the for PND to In addition, did not the Cognitive Assessment which has a higher sensitivity than MMSE among elderly patients, to cognitive at discharge. The of measuring the of patients with The association between OSA, EDS, or OSA with EDS and PND may while for learning of MMSE and patients baseline at one year that this was or that patients had cognitive at baseline to other or the incidence of POD in our study is which is lower than the of POD reported in previous study may lack the to this in POD between the two to factors, including postoperative anxiety, and may the association between sleep and long-term cognitive outcomes. to for that be between sleep and cognitive such will be important in to and for our study has of loss to follow-up our to all The of loss to follow-up at 1-month and 1-year after surgery were in OSA and non-OSA and was no in baseline data between patients to follow-up and those to our results are and In our study suggested that moderate-to-high risk of OSA combined with EDS a higher risk of POCD at one and one year after surgery. However, moderate-to-high risk of OSA not the development of PND within one year after surgery. that our study is the first to the of moderate-to-high risk of OSA and moderate-to-high risk of OSA with EDS on the development of more is to explore the effect of OSA and OSA with EDS on PND in prospective our results that of moderate-to-high risk of OSA and EDS are of clinical in the of PND in the and OSA and EDS be and to cognitive in elderly patients within one year after surgery.
Source: Frontiers in Neuroscience
Authors: Wenwen Wu, Lihui Pu, Xiuying Hu, Qian Chen, Guan Wang, Yanyan Wang
Language: English
Subfield: Physiology
Field: Medicine
Domain: Health Sciences
Open Access status: gold
Estimated APC paid: $3,295
OpenAlex ID: W4391784850