Study Design and Setting
The ChronoWise observational study is a prospective single-centre cohort study designed to follow patients undergoing bariatric surgery at the Santo António Local Health Unit (ULSSA), a public hospital of Porto. This study will be carried out by the Institute of Public Health of the University of Porto (ISPUP) and ULSSA, with an expected total duration of 12 months. The duration of the study has been provisionally determined to span from December 2023 through December 2024.
Ethics
This research will be conducted according to the Ethical Principles for Medical Research Involving Human Subjects established by the Declaration of Helsinki. The study protocol was approved by the ULSSA and School of Medicine and Biomedical Sciences of University of Porto (ICBAS-UP) Ethics Committee (ref. 2023.325_CE) and by the Data Protection Office (DPO). All participants will be asked to provide written informed consent to participate in the study. Participants’ privacy and confidentiality will be ensured during all stages of data collection and processing. Identification data will be kept confidentially in files separated from the remaining data, linked by an identification ID.
Participants and recruitment
For study participation, a patient in the preoperative period of bariatric surgery must meet the following inclusion criteria: 1) age ≥18 years; 2) first-time bariatric surgery; 3) body mass index (BMI) ≥40.0 kg/m2 or BMI ≥35.0 kg/m2 with at least one health risk factor, particularly type 2 diabetes, dyslipidemia, obstructive sleep apnea, obesity hypoventilation syndrome, hypertension or osteoarthritis; 4) non-alcoholic, non-drug dependent and without a significant cognitive impairment or psychiatric disorder.
During the hospital stay prior to the surgery, each potential participant will be approached by a research investigator, who will provide information about the study’s objectives and methodology. A participant information sheet has also been developed and will be given to patients, inviting them to take part in the study and explaining the project. In cases of acceptance, participants will be asked to read and sign a duplicated Informed consent form. Additionally, consented participants will immediately attend a baseline evaluation. Those who decline to participate, under the condition of oral consent, will be requested to answer a small refusal questionnaire, encompassing questions regarding sociodemographics, chrono-related behaviours, consumption of fruits and vegetables and exercise practice.
Based on the weekly number of first bariatric procedures undertaken at ULSSA, the expected recruitment period for the study is six months. The recruitment phase and baseline data collection are ongoing and are anticipated to continue until June 2024. In case the sample size is not reached within this timeframe, additional screening time can be used to increase recruitment.
Data Collection
Data will be obtained at pre-surgery and on the 3rd and 6th month after surgery by trained interviewers following standard procedures. Baseline evaluation will be carried out by a face-to-face interview during the hospital stay. The follow-up evaluations will occur by telephone or video call interview according to the participant’s preference.
At all-time points, questionnaires will collect data on food intake, chrono-nutrition behaviours, sleep time behaviour, sleep quality, screen time, physical activity and exercise behaviours, anthropometric measurements, and biochemical parameters. The baseline preoperative questionnaire includes additional information on sociodemographics. Table 1 summarizes the details of the variables assessed at baseline and follow-up evaluations.
-------------------------------------------Please insert Table 1-------------------------------------------------------
- Food intake and chrono-nutrition
All participants will be required to report their dietary intake, which will be assessed through an adapted version of a validated food frequency questionnaire (FFQ)(29, 30) applied by a trained interviewer. The FFQ comprises a variety of food items and beverage categories, and a question regarding the intake of dietary supplementation will be added. For each item, the average consumption frequency, ranging from never/less than once a month to six or more times per day, will be gathered. Additionally, participants will also indicate whether their typical portion size is equal to, greater than or less than a standard average portion. At baseline and last follow-up assessments, the FFQ will cover the average dietary intake over the previous three months. In a different way, at the first post-operative evaluation, information on food intake will be gathered with reference to the eating habits of the past month. This is because patients must comply with specific dietary guidelines in the first seven weeks after undergoing bariatric treatment, starting subsequently a more non-restrictive diet. Energy and nutrient intake will be calculated by multiplying the frequency of consumption of each food by the selected portion obtaining daily total energy intake (TEI) and macronutrients (i.e., protein, fat, and carbohydrates) as the proportion of TEI, using the Portuguese Food Composition Table(31). Other indicators of diet quality, such as fruit and vegetables and energy-dense food daily consumption will be included as potential confounders.
The chrono-nutrition dimensions will be measured by the Chrononutrition Profile – Questionnaire(32), consisting of questions fitting behavioural patterns likely to influence the chrono-nutrition profile. For both work/school and free days, meal timing, eating window (duration of time between first and last eating event of the day), fasting during the night, as well as evening eating and nighttime snacking, will be obtained.
- Chronotype and Social Jetlag
The short version of the Munich Chronotype Questionnaire (MCTQ)(21, 22), validated for the Portuguese population(33), will be used to assess estimated sleep duration, chronotype and social jetlag over the last 6 weeks.
The MCTQ collects information on an individual’s sleep behaviours, distinguishing between workdays and free days. Considering that sleep on free days is less constrained by social factors, chronotyping using MCTQ is calculated based on the midpoint between sleep on- and offset on free days (midsleep point on free days) with a further correction for calculated sleep debt. The following formulas will be used:
Where MPSfreedays is the midsleep point on free days and SDfreedays and SDworkdays represent total sleep duration on free days and workdays, respectively.
Chronotype will be used as a continuous variable and further categorized according to the median or quartile distribution, as recommended by Roenneberg et al(34) and Reis et al(33).
Social jetlag, as a marker of circadian misalignment, will be calculated as the difference between the midsleep point on free days and the midsleep point on workdays(22).
The sleep quality will be assessed using the validated Portuguese version of Pittsburgh Sleep Quality Index (PSQI)(35), a self-rated questionnaire that gathers information regarding the previous month. It consists of 19 questions covering the seven following components: sleep quality, sleep onset latency, sleep duration, sleep efficiency, sleep disturbances, use of sleeping medications, and daytime dysfunction. Each component is scored on a 0 (no difficulty) to a 3 (severe difficulty) scale. The sum of the scores yields a global score, ranging from 0 to 21. A global score of >5 points indicate poor sleep quality(35).
- Physical activity and exercise behaviours
Physical activity level will be assessed using the Portuguese version of the International Physical Activity Questionnaire (IPAQ) – Short Form(36), which comprises the performance of activities over the past seven days. Information on the frequency (measured in days per week) and duration (time per day) is assessed regarding a set of activities, including walking, sitting, moderate-intensity and vigorous-intensity activities. The data collected will be computerized into metabolic equivalent (MET)-minutes per week, using established formulas, to determine energy expenditure, which is defined as the sum of walking, moderate, and vigorous MET-min per week scores(37). The total expenditure will be calculated by multiplying individual’s basal metabolic rate (estimated by the Harris-Benedict equations) by the physical activity level.
Additionally, questions concerning the practice of structured exercise will be included. Information on the type of exercise (e.g., Walking, Pilates, Yoga), duration, as well as the period of the day they usually practice it (i.e., morning, lunchtime, early afternoon, late afternoon, at night) will be collected.
- Sedentary activities and screen time
Behavioural aspects regarding sitting and screen time will be assessed. Participants will be asked to indicate the average time spent sitting (i.e., at a desk, reading, resting, watching television, using a mobile phone, or tablet, or listening to music) on both work/school and free days. In addition to recording the average daily screen time during the work/school and free days, participants will also detail the time of screen usage in the night period, particularly after dinner and before bedtime, and in bed with the lights on or off.
- Anthropometric and biochemical parameters
Anthropometric measurements, including weight and height, will be retrieved from the medical registries and self-reported. Body mass index (BMI) will be calculated. The post-operative percentage of excess weight loss (%EWL) will be estimated by dividing the weight loss (kg) by the excess weight (the initial weight minus weight at BMI=25.0 kg/m2) and multiplying by 100. Trajectories of weight loss over the follow-up time will be obtained and patients will be further classified as good or poor weigh loss responders according to the %EWL. Thus, patients with EWL≥50% at the last follow-up will be identified as having a good weight-loss response and the remaining as poor weight-loss responders(9).
Fasting venous blood samples will be collected and analysed at the ULSSA, following routine laboratory procedures, including glucose, insulin, glycated hemoglobin, triglycerides, and high-density lipoprotein (HDL-C), low-density lipoprotein (LDL-C), and very low-density lipoprotein (VLDL-C) cholesterol. The Homeostatic Model Assessment insulin resistance (HOMA-IR) will be calculated as ‘glucose (mg/dL) x insulin (µIU/mL)/405’.
Participants' diagnosed comorbidities, including hypertension, type 2 diabetes mellitus, dyslipidaemia, hepatic steatosis, and others, will be registered, along with details regarding the type of bariatric surgery and its date.
History of medication and supplementation intake considering the previous three months will be gathered.
Sample size calculation
The sample size was estimated through the following formula(38):
- π = proportion of interest.
- π0 = null hypothesis proportion.
- u = one-sided percentage point of the normal distribution corresponding to 100% the power, which is conventionally 80%.
- v = percentage of the normal distribution corresponding to the required (two-sided) significance level of 5% conventionally.
The observed proportion of late eaters among good weight-loss responders is 37% (р = 0.011) according to Ruiz-Lozano et al(26). Thus, we assumed that approximately 65% of good weight-loss responders will be early eaters. Based on these assumptions:
- π = proportion of early eaters among good weight-loss responders = 0.65.
- π0 = 0.5, assuming that there are no differences between the proportion of early and late eaters among good weight-loss responders.
- u = 0.84 (value of the standard normal distribution corresponding to the desired level of 80% of power).
- v = 1.96 (value of the standard normal distribution corresponding to the significance level [1.96 for a 2-sided test at the 0.05 level]).
The estimated sample size for this study was 85 participants. Taking into consideration a drop-out rate of 20% during the cohort, the sample size was adjusted by multiplying the obtained value by 100/ (100-20)(38). Thus, the minimum sample size required is 106 participants.
Statistical analysis plan
Descriptive statistics consisting of mean and standard deviation or median and interquartile range (IQR) will be described for continuous variables. Categorical variables will be reported as frequencies and percentages. The normality of data will be tested using the Kolmogorov–Smirnov test, visual inspection of box-plots or normal probability plots.
The student’s t-test for independent samples or the Mann–Whitney test will be performed to compare independent variables, according to the variable’s distribution. Chi-squared test or Fisher’s test will be used to compare the proportion variables. Comparisons throughout the follow-time will be assessed using one-way repeated measures ANOVA test or Friedman test for paired variables, as applicable. When a significant р-value is obtained, Bonferroni post hoc procedures will be used to evaluate pairwise differences.
The %EWL measured longitudinally at baseline and six months post-surgery will be analysed using linear mixed effects over the two post-surgical time points (3 months and 6 months), after controlling for the baseline BMI.
Mixed effects models will be used to evaluate the associations between exposures, such as chrono-nutrition, chronotype and social jetlag profile, and the weight trajectories at 3 and 6 months after undergoing bariatric surgery. To further evaluate the effect of chronotype and social jetlag on the association between chrono-nutrition and weight loss, interaction terms will be included in the models. Potential confounders such as sex, age, education, daily TEI, sleep duration and quality, physical activity, and other variables will be tested.
Data will be analysed with the SPSS IBM Statistical Software and р <0.05 will be considered statistically significant.