Circadian rhythms are regulated by the central clock located in the suprachiasmatic nucleus(Reference Chaput, Mc Hill and Cox1). The central clock interacts with peripheral clocks in various organs, including liver(Reference Reinke and Asher2), intestines(Reference Voigt, Forsyth and Keshavarzian3), pancreas(Reference Stenvers, Scher and Schrauwen4), adipose tissue(Reference Froy and Garaulet5) and cardiovascular system(Reference Morris, Purvis and Hu6). Together, they constitute the circadian timing system, which responds to external cues such as light and lifestyle factors like physical activity and meal timing. This synchronisation between the central and peripheral clocks influences the sleep-wake cycle, hormone release and organ activity at distinct times of the day(Reference Astiz, Heyde and Oster7).
Circadian misalignment occurs when there is a failure in the synchronisation between endogenous factors (central and peripheral clocks) and environmental signals, leading to alterations in the physiological circadian rhythm and the sleep-wake pattern(Reference Chaput, Mc Hill and Cox1). Social jet lag (SJL) addresses the difference between sleep midpoint during weekdays and weekends and has been a widely used measure of circadian misalignment(Reference Wittmann, Dinich and Merrow8,Reference Roenneberg, Allebrandt and Merrow9) . This measure has been associated with unhealthy habits, such as poor dietary patterns(Reference Zerón-Rugerio, Cambras and Izquierdo-Pulido10,Reference Mota, Silva and Balieiro11) , metabolic diseases(Reference Jang, Son and Lee12–Reference Sládek, Klusáček and Hamplová14) and poorer lifestyle habits(Reference Haynie, Lewin and Luk15–Reference Krueger, Stutz and Jankovic17). Similar to SJL, studies also show that a worse sleep pattern is associated with greater energy consumption, leading to weight gain – thus indicating a bidirectional relationship between sleep and obesity(Reference Imes, Bizhanova and Kline18).
Bariatric surgery is considered the most effective treatment for obesity, and a multidisciplinary team is very important to manage the weight loss(Reference Eisenberg, Shikora and Aarts19). Furthermore, bariatric surgery has shown that an improvement in sleep patterns is associated with better surgical outcomes(Reference Lodewijks, Schonck and Nienhuijs20). In a previous study with bariatric patients, we demonstrated that the group with greater SJL had lower weight loss, worse metabolic outcomes and poorer dietary pattern after 6 months of bariatric surgery compared with the group with less SJL(Reference Carvalho, Mota and Marot21). However, there is currently limited evidence regarding the influence of chronobiological issues – such as SJL and poor sleep patterns – on the weight loss process in patients undergoing bariatric surgery during first-year post-surgery. Given that bariatric surgery is widely performed worldwide and is regarded as a promising treatment for obesity(Reference Eisenberg, Shikora and Aarts19), albeit with associated challenges such as weight regain, understanding the variables that may impact patient outcomes is essential for optimising long-term clinical results.
Based on the points presented, it is plausible that improved sleep patterns and reduced circadian misalignment could be associated with more favourable surgical outcomes. This study aims to examine the relationship between circadian misalignment and sleep patterns and the progression of anthropometric, metabolic and dietary parameters during the first year following bariatric surgery. We hypothesised that individuals experiencing higher levels of SJL, poorer sleep quality and increased daytime sleepiness during the first year after bariatric surgery would exhibit lower weight loss, higher intake of calories and macronutrients, as well as worse metabolic outcomes.
Methods
Participants
This study received approval from the Research Ethics Committee of the Federal University of Uberlândia (66023717.8.0000.5152). The clinical trial number is NCT03485352 (URL: https://clinicaltrials.gov/ct2/show/NCT03485352). All participants were informed about the study objectives and provided signed informed consent. Further details regarding the methodology can be found in a previous study(Reference Carvalho, Mota and Marot21). This prospective cohort study included 122 patients (77 % women, median age 33 years) undergoing bariatric surgery (Roux-en-Y gastric bypass (79·5 %) or vertical gastrectomy (20·5 %)) at a private clinic in Uberlândia, Minas Gerais, between June 2017 and October 2018. Inclusion criteria were patients in the preoperative period of bariatric surgery, aged 18–59 years, with a BMI ≥ 35 kg/m2 associated with two comorbidities or BMI > 40 kg/m2 regardless of comorbidities(Reference Eisenberg, Shikora and Aarts19). The exclusion criterion was the performance of revisional surgery.
All participants were interviewed in-person for all study evaluations on the day of their routine clinic appointment by a trained nutritionist from our research group.
Social jet lag
SJL was calculated as the absolute difference between the midpoint of sleep on free days/weekends (mid-sleep on free days) and the midpoint of sleep on workdays/weekdays (mid-sleep on workdays)(Reference Wittmann, Dinich and Merrow8,Reference Roenneberg, Allebrandt and Merrow9) . This evaluation was conducted in-person at four assessment points (baseline, 3 months, 6 months and 1 year post-surgery) by a team experienced in sleep pattern research, using the following questions: ‘What time do you usually go to sleep during the week?’, ‘How long (min) do you stay up in bed before sleep onset (after turning off the lights) during the week?’, ‘What time do you usually wake up during the week?’, ‘What time do you usually go to sleep on weekends?’, ‘How long (minutes) do you stay up in bed before sleep onset (after turning off the lights) on weekends?’ and ‘What time do you usually wake up on weekends?’.
Sleep patterns
Sleep patterns were assessed by evaluating daytime sleepiness and sleep quality at four-time points (baseline, 3 months, 6 months and 1 year) during in-person appointments on the day of their routine clinic visit. Sleep quality was evaluated using the Pittsburgh Sleep Quality Index (PSQI) questionnaire, adapted for Portuguese(Reference Bertolazi, Fagondes and Hoff22). This questionnaire, which has been used in other Brazilian studies(Reference Bertolazi, Fagondes and Hoff22,Reference Mota, Waterhouse and De-Souza23) , comprises nineteen questions regarding sleep quality and disturbances over the past month, with scores above 5 indicating poor sleep quality. The Epworth Sleepiness Scale questionnaire assessed daytime sleepiness, with scores above 9 indicating high daytime sleepiness(Reference Bertolazi, Fagondes and Hoff24). All questionnaires were administered at four assessment time points: preoperative, 3 months, 6 months and 1 year after bariatric surgery.
Anthropometric variables
Weight and height were measured using standardised methods at all four follow-up appointments(25). BMI was calculated as kg/m2. In accordance with WHO guidelines (2000) for adults, a BMI of ≥ 25 kg/m2 was classified as overweight and > 29·9 kg/m2 as obesity(25).
Metabolic parameters
Fasting levels of glucose, insulin, total cholesterol (TC), HDL, LDL, VLDL and TAG, along with the HOMA-IR estimate, were established preoperatively, at 6 months and 1-year post-surgery. Blood samples were collected at a laboratory with rigorous adherence to high-quality standards, affiliated with the patient’s insurance, after a 12-hour fast. Blood samples were collected and processed on-site by a nurse under optimal conditions, adhering to widely established and standardised protocols. All analyses were conducted using methodologies that have been validated in the literature(Reference Friedwald, Levy and Fredrickson26,Reference Matthews, Hosker and Rudenski27) .
Dietary intake
Dietary intake was assessed by administering two 24-hour dietary recalls at each of the four assessment time points. All recalls were conducted by a trained nutritionist using the Multiple Pass Method(Reference Moshfegh, Rhodes and Baer28). At each time point, two recalls were completed: one during a weekday and the other on a weekend. The first 24-hour dietary recall was conducted in-person on the day of the routine clinic visit, while the second recall was conducted via telephone, ensuring the reliability of the method(Reference Moshfegh, Rhodes and Baer28). The Multiple Pass Method stipulates that, at the beginning of the assessment, a general listing of all foods consumed in the past 24 h is made, followed by additional details about each item, including preparation, ingredients and any potentially forgotten foods, as well as the time and location of meals. Subsequently, the quantities consumed are estimated, and the list is reviewed to ensure accuracy. A final check confirms that all foods and beverages have been included. Macronutrients, calories and fibre were evaluated using DietPro Clínico 5.0 software. A total of eight 24-hour dietary recalls were collected, and the averages for weekdays and weekends were calculated for each of the four assessment time points.
Statistical analysis
The Kolmogorov–Smirnov test was used to analyse the normality of the data. Student’s t test was employed for variables with a normal distribution, while the Mann–Whitney test was used for non-normally distributed variables. The Chi-square test was applied to analyse proportion variables.
Multiple linear regressions were used to assess the association between SJL exposure mean, sleep quality, daytime sleepiness level and weight loss, metabolic outcomes and dietary intake at 3 months, 6 months and 1 year after surgery.
Logistic regression was used to evaluate the likelihood of weight loss 1 year after surgery in groups more or less exposed to SJL, higher levels of daytime sleepiness and poorer sleep quality within 1 year. The mean for each exposure variable was estimated across the four assessment periods, and the sample was subsequently divided into two categorical groups according to the median of each parameter: more exposed (mean > median) and less exposed (mean < median).
Analyses of linear regression with metabolic outcomes were adjusted for potential confounders that, according to the literature, could influence the outcomes, including surgical technique, shift work, gender, age, family income, sleep medication use, diabetes diagnosis, exercise practice, BMI and daily caloric intake. Analyses of linear and logistic regression with anthropometric and dietary intake-related outcomes were adjusted for all variables mentioned above, except for BMI and caloric intake, respectively.
To address missing data, we used mean imputation, replacing missing values with the mean of observed values for each variable. This approach assumes that the data are Missing Completely at Random and was chosen to maintain the dataset’s overall structure and comparability across analyses. By imputing missing values, we ensured that the analyses remained robust while minimising potential biases introduced by data loss.
Statistical analysis was performed using SPSS version 20 (SPSS, Inc., Chicago, IL, USA), and P < 0·05 was considered statistically significant.
Results
Table 1 provides information on sociodemographic characteristics, surgical techniques, physical activity, sleep parameters (Epworth score, PSQI score and sleep duration) and SJL at baseline and at 3, 6 months and 1 year of treatment. We observed that the majority of participants were female (77 %), underwent Roux-en-Y gastric bypass surgery (79·5 %) and were married (66·4 %). Both sleep quality and daytime sleepiness significantly improved over the follow-up period after 1 year of bariatric surgery (P < 0·001).
Table 1. Socio-demographic characteristicals, surgical techniques, physical activity, sleep parameters and chronotype in 1-year follow-up
Values are presented as mean and standard deviation for normally distributed data or as median [interquartile range] for non-normally distributed data or as percentage. Participants in each evaluation moments: baseline (n 122), 3 months (n 117), 6 months (n 113) and 1 year (n 60). PSQI, Pittsburgh Sleep Quality Index.
Table 2 presents the results of the linear regression analysis assessing the association between the mean exposure to SJL, sleep quality and daytime sleepiness with weight loss in kilograms, percentage of weight loss and reduction in BMI at 3 months, 6 months and 1 year after bariatric surgery. We observed negative associations between SJL and weight loss (kg) (β = −0·14, P = 0·03; β = −0·24, P = 0·03), percentage of weight loss (β = −0·21, P = 0·02; β = −0·29, P = 0·03) and reduction in BMI (β = −0·18, P = 0·02; β = −0·25, P = 0·03) at 6 months and 1 year after bariatric surgery, respectively. Negative associations were also found between sleep quality and weight loss in kilograms (kg) (β = −0·22, P = 0·001; β = −0·33, P < 0·001; β = −0·33, P = 0·002), percentage of weight loss (β = −0·24, P = 0·005; β = −0·37, P < 0·001; β = −0·37, P = 0·004) and reduction in BMI (β = −0·26, P = 0·001; β = −0·34, P < 0·001; β = −0·42, P = 0·001) at 3 months, 6 months and 1 year after bariatric surgery, respectively. Additionally, a negative association was found between the average daytime sleepiness level and weight loss in kilograms (kg) (β = −0·28, P = 0·009), percentage of weight loss (β = −0·37, P = 0·002) and reduction in BMI (β = −0·35, P = 0·002) 1 year after bariatric surgery.
Table 2. Associations between changes in anthropometric parameters and mean exposure to social jet lag, sleep quality and daytime sleepiness at 3 months (n 117), 6 months (n 113) and 1-year follow-up (n 60)
Mean of social jet lag, sleep quality and daytime sleepiness were evaluated during three moments of evaluation for each variable (3 and 6 months and one year). Linear regression was adjusted for sex, age, family income, type II diabetes, surgical technique, shift work, physical activity, and energy intake and P < 0·05 was considered significant. Participants in each evaluation moments: baseline (n 122), 3 months (n 117), 6 months (n 113) and 1 year (n 60). PSQI, Pittsburgh Sleep Quality Index.
Table 3 shows the results of the logistic regression analysis, evaluating the association of SJL exposure, sleep quality and daytime sleepiness with weight loss 1 year after bariatric surgery. The group with greater exposure to SJL (P = 0·038; OR = 3·76 [1·07, 13·18]) and those with higher levels of daytime sleepiness (P = 0·04; OR = 3·13 [1·01, 9·65]) had an increased risk of experiencing less weight loss during the first year following bariatric surgery. No significant results were found for sleep quality (P = 0·10).
Table 3. OR evaluating the association for weight loss below the median after 1 year based on exposure to social jet lag, daytime sleepiness and sleep quality (less v. more exposed; below or above the group median) (OR and 95 % CI)
Values were calculated by logistic regression related median of weight loss percentage after 1 year (median 34·6 %) and presented as OR for weight loss according mean of each variable (SJL, PSQI and EPWORTH score) with 1 year.
We performed two categorical groups according to the mean of each parameter: more exposed (mean > median) and less exposed (mean < median).
P < 0·05 was considered significant.
Logistic regression analyses were adjusted for gender, age, surgical technique, physical activity, shift work and energy intake. PSQI, Pittsburgh Sleep Quality Index.
Table 4 presents the results of the linear regression analysis assessing the association between food consumption and mean exposure to SJL, daytime sleepiness, sleep quality, at 3 months, 6 months and 1 year after surgery. The results showed a positive association between the average SJL and calorie intake (β = 0·35, P = 0·01) and protein intake (β = 0·30, P = 0·03) 1 year after bariatric surgery. No significant results were found for the other analysed variables.
Table 4. Association between food consumption and mean of exposure social jet lag, daytime sleepiness and sleep quality after 3 months (n 117), 6 months (n 113) and 1-year (n 60) follow-up
Mean of social jet lag, daytime sleepiness and sleep quality was evaluated by three assessment periods (3 months, 6 months and 1 year). Linear regression was adjusted for sex, age, family income, type II diabetes, surgical technique, shift work, physical activity and BMI. Analysis of food consumption variables was evaluated according to 24-hour dietary recall questionnaire at the three assessment moments, and P < 0·05 was considered significant. Participants in each evaluation moments: Baseline (n 122), 3 months (n 117), 6 months (n 113) and 1 year (n 60). PSQI, Pittsburgh Sleep Quality Index.
Online Supplementary Table 1 shows the results of the linear regression analysis on the delta difference of metabolic parameters and the mean exposure to SJL, PSQI score and daytime sleepiness at 1 year after surgery. The results showed a negative association between the reduction in cholesterol (β = −0·55, P = 0·002) and TAG (β = −0·44, P = 0·02) and the mean of SJL. Negative associations were found between the reduction in insulin (β = −0·47, P = 0·02) and the mean PSQI score (a higher PSQI score show worse the sleep quality), as well as between the reduction in insulin (β = −0·57, P = 0·01), low-density cholesterol (β = 0·45, P = 0·006) and HOMA-IR (β = −0·53, P = 0·04) and the mean of daytime sleepiness (the higher score show greater the daytime sleepiness).
Online Supplementary Table 2 shows the results of linear regression analysis assessing the association between SJL exposure with the mean of sleep quality, daytime sleepiness and time of sleep at each moment (3 months, 6 months and 1 year). The results show a positive association of SJL with daytime sleepiness at 1 year after surgery and a negative association of SJL with sleep duration at 3 and 6 months after surgery.
We used G * Power to determine the sample size for the multiple linear regression analysis. The test was conducted post hoc. The parameters used for the calculation were as follows: Effect size (f2): 0·22, based on previous findings(Reference Carvalho, Mota and Marot21); alpha level (α): 0·05; sample size: 60; number of predictors: 1. The result showed a power (1-β) of 0·95.
Figure 1 shows the flowchart detailing the evolution of the number of patients during the 1-year follow-up study. A study cohort of 138 eligible patients was identified. However, eleven individuals opted not to participate, and five were scheduled for revisional surgery, meeting the exclusion criteria. This resulted in 122 participants undergoing baseline assessments. Subsequently, at the 3-month evaluation, five participants were missing follow-up compared with the baseline number, totalling 117 participants in this assessment. By the 6-month assessment, nine participants were missing follow-up and five participants missed the 3-month assessment but attended the 6-month assessment, bringing the total to 113 participants in this assessment. Finally, at the 1-year evaluation, sixty-two participants were missing follow-up, resulting in a total of sixty participants for this assessment. Participants who missed the 3-month, 6-month and 1-year routine consultations at the clinic did not complete the evaluations during these periods.
Figure 1. Flowchart of Participants in Baseline, 3-Month, 6-Month and 1-Year Evaluations after Bariatric Surgery. Note: Of the 138 eligible patients identified, 11 chose not to participate and 5 were excluded for revisional surgery, resulting in 122 participants for baseline assessments. At the 3-month follow-up, 5 participants were missing in relation to baseline (n 122), totaling 117 evaluations. At 6 months, nine participants were missing and five participants missed the 3-month assessment but attended the 6-month assessment, bringing the total to 113 participants in this assessment. By the 1-year follow-up, sixty-two participants were missing, resulting in sixty participants completing the evaluation. Participants who missed the 3-month (n 5), 6-month (n 9) and 1-year (n 62) routine consultations at the clinic did not complete the evaluations during these periods.
Figure 2 shows the effects of time on PSQI score, sleep duration, SJL and Epworth score during the first year after bariatric surgery. The results show a significant effect of time on PSQI (P < 0·001) and Epworth (P < 0·001) score during the first year following bariatric surgery.
Figure 2. Effects of time on PSQI score (a), sleep duration (b), Epworth score (c) and SJL (d) during the first year of bariatric surgery. Note: Values are presented as mean and standard error (Daytime Sleepiness and Sleep Duration) and as median and interquartile range (Social Jet Lag and Sleep Quality); P values were calculated by Generalised Estimation Equation (GEE) and P < 0·05 was considered significant. Analysis was adjusted for sex, age, family income, type II Diabetes, surgical technique, shift work, physical activity, energy intake and BMI.
The regression analyses conducted with the imputed data demonstrated that the results were consistent and reproducible with those obtained from the observed data (data not shown).
Discussion
This study assessed the impact of sleep patterns and SJL on anthropometric, metabolic and dietary outcomes during the first year following bariatric surgery. Our hypothesis was confirmed, as linear regression results demonstrated that greater exposure to SJL was associated with smaller weight loss and less reduction in BMI at both 6 months and 1-year post-surgery. Additionally, lower sleep quality was linked to decreased weight loss in kilograms and a reduction in the percentage of weight loss and BMI at 3, 6 and 12 months post-surgery. Elevated levels of daytime sleepiness were associated with reduced weight loss in kilograms, percentage of weight loss and BMI reduction 1-year post-surgery. In terms of dietary consumption, we observed a positive association between SJL and calorie as well as protein intake after one year of bariatric surgery.
Circadian misalignment appears to impact various metabolic processes, including alterations in gut microbiota(Reference Voigt, Forsyth and Green29), hunger and satiety hormones(Reference Senesi, Ferrulli and Luzi30,Reference Brum, Senger and Schnorr31) and suppression of melatonin secretion(Reference Vieira, Nehme and Marqueze32). Existing literature demonstrates the negative effects of SJL on obesity across different populations(Reference Mota, Silva and Balieiro33–Reference Suikki, Maukonen and Partonen35), as well as on obesity-related chronic diseases such as type II diabetes mellitus(Reference Mokhlesi, Temple and Tjaden36) and systemic arterial hypertension(Reference Bouman, Beulens and Groeneveld13). Furthermore, it has been associated with certain types of cancer, including colorectal(Reference Liu, Wang and Cheng37), hepatic(Reference Kettner, Voicu and Finegold38) and thyroid cancer(Reference Malaguarnera, Ledda and Filippello39). A systematic review with meta-analysis of sixty-eight studies confirmed that SJL is linked to a higher risk of increased BMI and waist circumference(Reference Bouman, Beulens and Groeneveld13). However, the relationship between obesity and SJL in individuals who have undergone bariatric surgery remains poorly understood. Our previous study(Reference Carvalho, Mota and Marot21) demonstrated a negative association between SJL and anthropometric, metabolic and dietary outcomes 6 months after bariatric surgery. In this study, we extend those findings over a longer follow-up period, revealing that individuals more exposed to SJL had less weight loss compared with those with lower exposure to SJL during the first-year post-surgery. This difference may become more evident and could be associated with obesity recurrence, a phenomenon frequently observed among individuals undergoing bariatric surgery(Reference El Ansari and Elhag40).
Our study establishes an association between SJL and sleep patterns, illustrating the intricate interconnection of these impairments(Reference Jin, Sutherland and Gislason41). Similar to SJL, impaired sleep patterns are also linked to unfavourable outcomes in bariatric surgery. We observed a negative statistical impact of sleep quality and daytime sleepiness on weight loss in bariatric patients during the first-year post-surgery. Importantly, sleep duration plays a pivotal role in evaluating sleep quality and has been identified as a crucial factor influencing body weight(Reference Creasy, Ostendorf and Blankenship42). Sleep restriction is associated with disruptions in the neuroendocrine mechanism that control appetite, leading to increased ghrelin levels and decreased leptin levels, thereby impacting energy consumption and, subsequently, body weight gain(Reference Spiegel, Tasali and Penev43,Reference Knutson and Van Cauter44) . A small study with fourteen participants who underwent bariatric surgery emphasised the significance of sleep patterns in long-term weight loss maintenance at the 6- and 9-year post-surgery marks(Reference Zuraikat, Thomas and Roeshot45). This study revealed an inverse relationship between sleep duration and BMI at the 6-year follow-up, implying that shorter sleep was associated with greater weight regain post-bariatric surgery. Inadequate sleep not only elevates the risk of obesity development but also influences the outcomes of weight loss interventions(Reference Antza, Kostopoulos and Mostafa46). Therefore, it is crucial to assess both the quantity and quality of sleep, emphasising sleep hygiene practices to optimise positive outcomes during the weight loss process in patients undergoing bariatric surgery(Reference Ackel-D’Elia, da Silva and Silva47,Reference Carneiro-Barrera, Díaz-Román and Guillén-Riquelme48) .
Our results indicate a negative association between SJL and both calorie and protein intake throughout the first-year post-bariatric surgery (Table 4). Consistent with our findings, previous studies in the literature involving other populations have demonstrated a negative impact of circadian misalignment on dietary patterns(Reference Zerón-Rugerio, Cambras and Izquierdo-Pulido10,Reference Suikki, Maukonen and Partonen35,Reference Mendoza49) . In our previous study with 792 individuals, a higher SJL was associated with increased intake of total calories, protein, total fat, saturated fat, cholesterol and servings of meat, eggs and sweets compared with those with SJL of up to 1 h(Reference Mota, Silva and Balieiro11). Individuals with SJL are more likely to experience sleep deprivation, particularly during the week, which may result in hormonal alterations related to hunger and satiety, leading to increased caloric intake and body weight gain. Several studies have also associated sleep deprivation with poorer dietary patterns(Reference Shaw, Dorrian and Coates50–Reference Hemmer, Mareschal and Dibner52). These findings highlight the importance of assessing sleep patterns during the nutritional monitoring of bariatric patients, particularly concerning the quantity and quality of food intake, as well as meal timing.
Based on all that has been discussed, it is possible to consider that the future of clinical practice for patients who have undergone bariatric surgery may include regular screening for sleep problems and the implementation of sleep hygiene measures. To minimise circadian misalignment and its significant negative impact, appropriate time-related interventions – such as optimising exposure to the light-dark cycle and aligning meal times – should be tested in the nutritional and clinical management of these patients. This proactive approach may help mitigate the effects of these factors on eating habits and overall health, ultimately enhancing the efficacy of bariatric treatments and supporting long-term success.
Concerning metabolic data (online Supplementary Table 1), our study revealed that greater SJL was associated with lower reduction in cholesterol and TAG levels during the first-year post-bariatric surgery. Additionally, higher daytime sleepiness was associated with a lower reduction in insulin, low-density lipoprotein, HOMA-IR, while poor sleep quality was associated with a lower reduction of insulin one year after surgery. Previous studies in the general population have also shown this negative effect of SJL on serum cholesterol levels(Reference Mota, Silva and Balieiro11,Reference Sládek, Klusáček and Hamplová14) and TAG, along with alterations in glycemic metabolism such as insulin, fasting glucose(Reference Mota, Silva and Balieiro53) and glycated haemoglobin(Reference Bouman, Beulens and Groeneveld13). A recent study demonstrated a correlation between sleep quality and glycated haemoglobin as well as HOMA-IR(Reference Tuna, Işık and Madenci54), while another associated daytime sleepiness with hypertriglyceridaemia and low high-density lipoprotein cholesterol(Reference Huang, Chen and Lin55). Our findings highlight the clinical significance of addressing SJL and sleep patterns in bariatric patients, as these factors negatively impact metabolic outcomes. The observed associations align with previous research linking sleep patterns and circadian misalignment to metabolic disturbances, emphasising the need for comprehensive management strategies that incorporate sleep quality improvements to optimise long-term metabolic health post-surgery. The literature on the influence of sleep patterns on metabolic outcomes in bariatric patients is still limited, highlighting the need for further research to better understand these findings and their potential association with weight regain in this population.
Despite the valuable insights gained from this study, it is essential to acknowledge its limitations. The use of questionnaires that depend on participants’ memory and subjective reporting introduces potential bias. Additionally, the small sample size, which only included patients undergoing two surgical techniques (bypass and vertical gastrectomy) in a private bariatric surgery service, limits the generalizability to patients undergoing other procedures at public health institutions. Regarding sleep evaluation, the absence of objective parameters such as polysomnography and actigraphy could impact the accuracy of sleep pattern measurements as sleep latency and the exact sleep time and our interpretation of results related to sleep quality and daytime sleepiness. Nevertheless, we employed validated questionnaires widely used in studies with similar objectives and methodologies.
Additionally, the loss to follow-up of 60 patients represents a significant factor that may have influenced the results, potentially introducing both selection bias and attrition bias. Despite the efforts made to reach these individuals, this loss underscores the considerable challenge of maintaining continuity in follow-up within this population, a challenge that may also arise in clinical practice. To address this limitation, we performed sensitivity analyses using imputed data for the missing cases, which confirmed that the results remained robust and consistent. It is also essential to highlight the strengths of our study. This 1-year follow-up research focuses on patients undergoing the post-bariatric surgery process, evaluating sleep and chrononutrition patterns – factors that have been relatively understudied in this context. By exploring these chronobiological elements, our study aims to shed light on their potential impact on treatment outcomes for obesity. If these findings are corroborated by future research, they could serve as crucial components in optimising obesity treatment strategies.
Conclusion
Our study highlights the negative associations of SJL, daytime sleepiness and poorer sleep quality on weight loss during the first year following bariatric surgery. Additionally, SJL was linked to higher calorie and protein intake. Longer-term studies utilising objective sleep data are needed to provide a more comprehensive understanding of these outcomes in bariatric patients.
Acknowledgements
We would like to thank all the patients and the LEV Clinic who participated in this research. We also thank the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq; CAC is CNPq fellow: #308466/2023-3).
This research didn´t receive any specific grant from any funding agency, commercial or non-profit sectors. All tests used in the research were already part of the patients’ routine clinic assessments as requested by the medical doctor at baseline, 6 months, and 1 year after surgery.
A. C. C. participated in the planning, data collect, interpretation of results, performed the statistical analysis and writing of the manuscript. L. P. M. participated in the interpretation of results and writing of the manuscript. L. A. M. participated in the interpretation of results and writing of the manuscript. J. A. G. S. participated in the interpretation of results and writing of the manuscript. A. C. T. A. participated in the interpretation of results and writing of the manuscript. C. T. C. A. participated in performed of data collect, interpretation of results and writing of the manuscript. M. C. M. participated in the interpretation of results and writing of the manuscript. C. A. C. participated in the planning, interpretation of results, support on the statistical analysis and writing of the manuscript.
A. C. C. works at LEV Clinic, is a PhD student, and is part of the Chrononutrition Research Group at the Federal University of Uberlândia. She does not declare any conflicts of interest. L. P. M. is part of the Chrononutrition Research Group at the Federal University of Uberlândia and does not declare any conflicts of interest. L. A. M. is a medical doctor at LEV Clinic and does not declare any conflicts of interest. J. A. G. S. is a medical doctor at LEV Clinic and does not declare any conflicts of interest. A. C. T. A. is a registered dietitian at LEV Clinic and does not declare any conflicts of interest. C. T. C. A. is an undergraduate student at University Center of Uberlândia and does not declare any conflicts of interest. M. C. M. is part of the Chrononutrition Research Group at the Federal University of Uberlândia and does not declare any conflicts of interest. C. A. C. is an Associated Professor and the Coordinator of the Chrononutrition Research Group at the Federal University of Uberlândia and does not declare any conflicts of interest.
Supplementary material
For supplementary material/s referred to in this article, please visit https://doi.org/10.1017/S0007114525000352
