Study Population and Data Collection
This was a follow-up study. We conducted convenience sampling among the patients with gastric cancer at a tertiary hospital in China, from March 2021 to September 2021.
The inclusion criteria included patients who: (1) diagnosed with gastric cancer by endoscopy or pathology; (2) planned to receive radical surgery for the first time; (3) aged ≥60 years.The exclusion criteria included patients who: (1) with severe physical disability or cognitive impairment (determined by a score ≥ 8 on the Short Portable Mental Status Questionnaire 15); (2) received preoperative radiotherapy or chemotherapy; (3) with other sites of malignant tumors; (4) with severe heart, liver, or renal insufficiency.
The enrolled patients would be collected baseline information and measured for frailty by researchers within 24 hours after admission. The postoperative total complications and length of stay (LOS) were captured from the electronic medical records while the information of disability and quality of life (QOL) was obtained by researchers from the telephone at 30 days after discharge.
Ethical Considerations
All patients signed informed consent after being informed that their clinical information would be used anonymously for research. The study was approved by the Ethics Committee of our university in China (Number: 2020-273). The study was verified by the Chinese Clinical Trial Registry (Number: ChiCTR2100052769).
Measurements
Baseline Information
Baseline data were collected on demographic characteristics (age, gender, BMI, marital status), and disease-related information (cancer stage, and type of surgery).
Frailty Instrument
Frailty was investigated with part B of the TFI which include 15 frailty deficits mapping into three domains: physical, psychological, and social frailty 10. The 8 components of physical frailty are poor physical health, unexplained weight loss, difficulty in walking, difficulty in maintaining balance, poor hearing, poor vision, lack of strength in the hands, and physical tiredness. The 4 items of psychological domain are problems with memory, feeling down, feeling nervous or anxious and unable to cope with problems. The social frailty consists of living alone, lack of social relations (loneliness) and lack of social support. The sore of the TFI range from 0-15, and the defined cutoff points are 5, 3, 2, and 2 for total, physical, psychological, and social frailty, respectively 16. The TFI has been extensively culturally adapted and validated in China 17.
Outcome Measures
We assessed postoperative complications and LOS as in-hospital outcomes. Postoperative complications were graded by Clavien-Dindo classification and total complications were considered as Clavien-Dindo grade ≥ 2 according to previous studies 18. LOS was defined from the data of admission to date of discharge.
The out-of-hospital outcomes included disability and QOL at 30 days of discharge. Disability was assessed by Elderly Disability Assessment Scale (EDAS) 19, which developed based on the ICF theoretical framework and the survey data of the Chinese elderly population. The EDAS comprises 28 items focusing on 7 domains: mental function, organ function, communication, activity, self-care ability, family life, economic and social life. The scores of EDAS range from 28-196, with lower scores representing more severe disability. The EDAS has been validated in China and Cronbach's α for total scale was 0.966 20. QOL was measured using the validated Chinese version of the gastric cancer-specific module of the European Organization for Research and Treatment of Cancer (EORTC QLQ-STO22) 21. The Cronbach’s α coefficient was above 0.7 for the total scale. The EORTC QLQ-STO22 including 22 questions evaluating 9 symptom scales: dysphagia, eating restriction, pain, reflux, anxiety, dry mouth, body image, hair loss and taste problem. Each scale is represented by a score ranging from 0 to 100 with higher score indicating poorer QOL.
Statistical Analysis
Descriptive data that in normal distribution were presented as mean ± standard deviation (SD) for continuous variables, otherwise, they were presented as the median to enable numerical comparisons. Frequencies and percentages were used as appropriate for categorical variables. The sample size was estimated based on the PASS version 15.0 (PASS 15 Power Analysis and Sample Size Software, NCSS, LLC). One previous study showed that the prevalence of frailty in preoperative patients with gastric cancer was 17.71% 22. Considering the missing rate of 10%, the minimum sample in our study was 267 patients. We used the non-parametric Kruskal–Wallis test to analyze differences between LOS, QQL, and disability since the data didn’t follow a normal distribution. The Bonferroni multiple comparison test was employed for post hoc analyses and adjusted for P-value 23. As for the categorical variables, the chi-square test was performed with P < 0.05 considering statistically significant. While employed post hoc analyses, the partitions of c2 method was used to estimate the differences between two of the groups, with level of significance changed into α < 0.002 (to reduce class I errors) 24. Statistical analysis was performed using SPSS (IBM SPSS Statistics for Windows, Version 26.0. Armonk, NY: IBM Corp).