Study design and population

The CSAC is an ongoing multicentre prospective cohort study in China that was first established in West China Hospital in July 2020 and then gradually expanded to three additional medical centres (The First People’s Hospital of Longquanyi District since August 2021, West China Tianfu Hospital since May 2022, and The Second Hospital of Hebei Medical University since September 2022). The initial focus of the study was noncardiac surgeries, such as abdominal, thoracic or otorhinolaryngologic surgery; cardiac surgeries were included starting in July 2021.

Based on the age distribution of the population at risk of surgery with anaesthesia, as well as the reported incidence of surgical/anaesthesia complications [2, 12,13,14] and the major disease outcomes of interest (e.g., psychiatric disorders and neurodegenerative diseases) among the Chinese population [15], we aimed to recruit all patients aged 40–65 years who underwent elective surgeries with general anaesthesia in the included medical centres and who provided blood and hair samples. Therefore, the inclusion criteria were as follows: (1) patients aged between 40 and 65 years; (2) patients who planned to receive elective surgery; and (3) patients who agreed to receive general anaesthesia during the operation. The exclusion criteria included patients who were not residents of the city where the recruitment medical centre was located, patients who received day surgeries (the length of hospital admission < 2 days) or craniotomy, patients who had an education level less than primary school, and patients who were unable to understand the interview scales. The pilot study was conducted between January 1st and July 14th, 2020 in West China Hospital, during which we developed a customized pipeline for optimizing the standardized procedure of data and biosample collection and quality control.

Data collectionBaseline information

As shown in Fig. 1, data about sociodemographic factors, lifestyle factors, physical functioning, and preoperative psychological and cognitive conditions were collected through face-to-face interviews conducted by well-trained data collectors (training course checklist is shown in supplementary materials 1) at baseline (i.e., one day before surgery date) using touchscreen questionnaires implemented in a newly developed Cohort Data Collection and Management System (CD-CMS) (Version 1.0, Build 2021SR0484324. ©West China Hospital, Sichuan, China). Biological samples were collected on the day of surgery prior to anaesthesia.

Fig. 1

The data framework of the China Surgery and Anesthesia Cohort (CSAC)

Follow-ups

At least two phone numbers of the patient and/or their relatives were saved at initial recruitment for further contacts. Well-trained data collectors scheduled active follow-ups for each participant within 1 year (Fig. 1) using the abovementioned data collection system. Specifically, the face-to-face follow-ups on days 1, 3 and 7 after the operation mainly focused on assessing in-hospital anaesthesia and surgical-related complications, postoperative pain, and immediate changes in cognitive condition. The follow-ups after hospital discharge focused on measuring the trajectories of postoperative pain and cognitive and psychological conditions; follow-up assessments were conducted through telephone interviews or online questionnaires at 1 month, 3 months, 6 months and 1 year after the surgery. In practice, we could allow ± 1 week for these scheduled follow-ups, and unless the participants withdrew their informed consent or refused further contacts, we continued subsequent follow-ups even if the participants have missed the previous one(s). Additionally, to ensure high follow-up rates, data collectors reviewed the participants’ baseline records before calling them, in order to customize some specific questions and provide disease-related advice. We also offered to help participants with reaching their doctors/other specialists if needed.

Health consequences after 1 year are planned to be obtained by periodically linking the dataset to established administrative registers. Specifically, the linked data for diseases diagnosed 1 year after surgery will be obtained from the National Health Insurance claim databases where the medical diagnoses were documented according to the 10th version of the International Statistical Classification of Diseases (ICD-10). In addition, the underlying and contributing causes of death will be derived based on the death certificates reported to the regional mortality registries.

Other data resources

The CSAC study also included enriched clinical data through linkage with several electronic medical data resources (Fig. 1). For instance, details related to the surgery and general anaesthesia procedure were documented in the anaesthesia information management system (AIMS), and thus, our database has enriched variables on intraoperative events (e.g., hypotension, hypoxia, transfusion), dynamically measured vital signs, and medication and anaesthesia drug use. Information about disease diagnoses, the results of medical examination (laboratory tests and imaging examination), and medical care received in the hospital was obtained from the electronic medical record (EMR) system of each recruitment centre.

We summarized the variables available at each time point, as well as their measurement approaches and resources, in Supplementary Table 1.

Measurement of primary outcomes and conditions

Neuropsychological symptoms at each time point were assessed by several well-validated scales. Sleep quality was measured by the Pittsburgh Sleep Quality Index [PSQI] [16]. Stress was measured by the Impact of Event Scale-Revised version [IES-R] [17] and Posttraumatic Stress Disorder Checklist for DSM-5 [PCL-5] [18]. Anxiety was measured by the Generalized Anxiety Disorder 7-Item Scale [GAD-7] [19]. Depression was measured by the Patient Health Questionnaire [PHQ-9] [20]. Subjective cognitive function was measured by the Eight-item Informant Interview to Differentiate Aging and Dementia [AD-8] [21]. Short-term memory was measured with the three-word recall test [22]. Self-evaluated pain level was measured by the Brief Pain Inventory [BPI] [23]. We then determined the presence of those conditions using the cut-off points that have been validated in the Chinese population (Supplementary Table 2).

The medical records were manually reviewed by the trained data collector to determine the presence of postoperative complications of interest, including anaesthesia-related complications (i.e., postoperative nausea and vomiting, severe hypoxia, aspiration, cardiac arrest and intraoperative awareness) and surgery-related complications (i.e., myocardial infarction, pulmonary embolism [PE], stroke, acute kidney injury [AKI], acute respiratory distress syndrome [ARDS], anastomotic breakdown, postoperative infection, and cardiac arrest), in accordance with European Perioperative Clinical Outcome (EPCO) definitions [24].

On December 7, 2022, the Chinese government announced rollback of strict anti-COVID-19 measures [25]. We recontacted all participants to obtain data related to COVID-19 between January and April 2023, including vaccination history, infection status, diagnosis methods, symptoms (type, severity, and duration) and treatment (Supplementary Materials 2).

Biological sample collection and test

Peripheral blood (20 ml) and hair samples were collected from participants with their consent before the surgery in accordance with standard sample collection protocols (Supplementary materials 3). Genotyping data are derived from all collected blood samples. In July 2021 and March 2023, we successfully extracted DNA and performed GeneChip sequencing for 1,953 and 3,216 blood samples, respectively. DNA quality control and genotyping were performed at the WeGene Clinical Laboratory, Shenzhen. Genotyping was performed by the Illumina iScan System following the manufacturer’s instructions for the Magen-MD5111 kit. The initial quality control of GeneChip data was processed with Plink2 (pLink®), and we used Eagle2 for haplotype analysis and Minimac4 for genotype imputation, according to the Thousand Genomes Project phase III data.

In addition to GeneChip sequencing, we started a whole-genome sequencing (WGS) program in 2023, which planned to include samples from approximately 3,000 participants (528 samples have been sequenced in January 2023). The case-cohort design is being used as the sample selection strategy. We illustrated the conception of this study design, as well as the process and sample estimates in Supplementary Fig. 1 and Supplementary Table 3. Sequencing data were derived from a DNBSEQ-T7 next-generation sequencer (a target coverage depth of 30×). Genomic variants were called following the modified Genome Analysis Toolkit (GATK) best practice workflow in the Institute of Rare Disease at West China Hospital, Sichuan University [26]. After the standard quality control process, variants were aligned to GRCh38 and annotated using Variant Effect Predictor (VEP), resulting in more than 3 million polymorphic variants remaining for further analysis [27].

Quality control

Notably, the CSAC was characterized by multiple approaches for stringent quality control (Fig. 2). First, the CSAC has a professional data collection team of 20 full-time investigators, and each data collector and investigator was required to attend a training program (i.e., 14-day lecture and practice courses) and pass the designed examinations before starting their work (Supplementary materials 1). Additionally, we have a multiple-step protocol for assessing the quality of data, including automated logic checks and warnings implemented in CD-CMS, as well as manual audio-recording checks.

Fig. 2

The process of data collection and quality control of the China Surgery and Anesthesia Cohort (CSAC)

Ethical considerations

All the participants signed separate informed consent for data and biosample collection. The approval for the future use of derived data (including genetic data) extracted from those biosamples has also been specified in the later consent. We take all necessary steps to keep personal information confidential. The study was approved by the ethics committee of West China Hospital, Sichuan University (approval number 2020 − 469). The collection, storage, and use of biosamples were documented and approved by the Ministry of Science and Technology of the People’s Republic of China (approval number 2021-CJ0606).