Assessing the impact of bariatric surgery on cardiovascular risk in type 2 diabetes: a systematic review and meta-analysis using OMICs data Authors: Balqees Almazrouei, Mira Mousa, Nour Al Dain Marzouka, Juan S Barajas-Gamboa, Carlos Abril, Habiba Al Safar Published: 2025 ABSTRACT AbstractType 2 diabetes (T2D) is a complex, multifactorial metabolic disorder, and while bariatric surgery has emerged as a promising intervention for obesity-related T2D with significant metabolic benefits, its long-term durability and potential for remission vary among patients. This systematic review and meta-analysis explore how omics modalities—such as genomics, epigenomics, transcriptomics, metabolomics, proteomics, and gut microbiome—can reveal potential biomarkers linked to cardiovascular disease (CVD) of T2D patients who undergo bariatric surgery. Following PRISMA 2020 guidelines, a systematic search in PubMed identified 49 eligible studies. The meta-analysis of eight proteomic biomarkers, showed significant post-surgery improvements in total cholesterol (mean difference (MD) 0.44 (95% CI: 0.06–0.82), p = 0.02), triglycerides (MD 1.00 (0.77–1.24), p < 0.00001), LDL cholesterol (MD 0.27 (0.02, 0.52), p = 0.03), HDL cholesterol (MD -0.22 (-0.30, -0.15), p < 0.00001), hsCRP (MD 0.64 (0.44, 0.84), p < 0.00001), C peptide levels (MD 1.29 (0.96, 1.61), p < 0.0001), and IL-6 (MD 1.84 (0.85, 2.84), p = 0.0003). These findings highlight the value of integrated omics in developing personalized diagnostics, predicting disease risks, and designing targeted therapies. The present study is the first systematic review presenting the omics disciplines that offer a comprehensive view of the effectiveness of bariatric surgery in managing T2D, subsequently reducing the risk of CVD.Supplementary InformationThe online version contains supplementary material available at 10.1186/s13098-025-02002-1.Keywords: Type 2 diabetes, Bariatric surgery, Omics, Cardiovascular disease, Precision medicine, Risk predictor Introduction Type 2 diabetes (T2D) is a chronic, metabolic disorder that is characterized by a complex set of genetic, metabolic, and environmental factors. The International Diabetes Federation (IDF) reports that nearly half of all diabetes mellitus, (mainly T2D) patients, totaling 240 million, are undiagnosed. In 2021, there were 537 million diagnosed cases, with projections suggesting an increase to 643 million by 2030 and 783 million by 2045 [1]. While insulin resistance (IR) and β-cell dysfunction are recognized as key pathogenic players, obesity is one of the predominant risk factors in the development of T2D and its associated vascular complications [2]. Weight loss has been demonstrated to ameliorate the metabolic complications inherent to T2D. Personalized management of T2D demands the development of a clinical strategy individualized to the patient’s specific requirements, complemented by an optimal weight-loss approach tailored to each individual [3]. However, achieving persistent weight loss and complete or partial remission of T2D can be challenging in certain cases. Bariatric surgery has emerged as a robust alternative for patients with T2D concomitant with obesity, leading to weight loss, glycemic control, and, often, T2D remission [4–7]. The primary mechanism underpinning these outcomes is the anatomical alteration of the gastrointestinal tract, which results in modifications of gut hormones, bile acids, and the microbiome. Roux-en-Y gastric bypass (RYGB), sleeve gastrectomy (SG), biliopancreatic diversion with duodenal switch (BPD/DS), and one anastomosis gastric bypass (OAGB) [8] may modify the anatomy and influence the entero-insular axis, enhancing insulin sensitivity and modulating postprandial glycemic excursions [9, 10]. Both RYGB and SG are considered the gold standard bariatric interventions due to their durable weight loss outcomes and associated metabolic benefits [11–14]. Despite evidence that surgical treatments can promote weight loss and help manage T2D, the long-term durability of these benefits remains uncertain. In a subset of patients, bariatric surgery may not lead to sustained improvement, as some individuals experience weight regain and a subsequent decline in glycemic control. [15]. While bariatric surgery offers promising results, comprehensive patient evaluation and post-operative management are imperative to optimize metabolic outcomes, cardiovascular disease (CVD) risk and minimize complications, especially in patients with T2D. [16]. Leroy Hood and other pioneers have championed systems medicine, an approach often referred to as P4 medicine due to its focus on prediction, prevention, personalization, and participation [17]. This paradigm aims to reduce mortality and morbidity from chronic diseases by implementing a deep knowledge of disease mechanisms and systems of the human body, and evidence-based personalized interventions. In this context, the development of a comprehensive approach to support P4 has fostered new diverse fields of research, commonly referred to as “OMICS”. Omics modalities including genomics, transcriptomics, proteomics, metabolomics, and microbiome with deep phenotypic data, are currently utilized to understand biological processes, disease phenotypes, biomarker discovery, and prognostic assessments (Fig. 1). Utilizing diverse omics data facilitates the identification of new molecular pathways and biomarkers for complex diseases, subsequently en0061bling the assessment of an individual’s health and offering early indicators of potential disease progression [18]. Conceptual illustration of multi-OMICs integrating genomics, epigenomics, transcriptomics, microbiome, metabolomics, and proteomics to study disease. Genomics mainly studies DNA sequence and associated heritable biochemical modifications through rapid elucidation of an entire genome. Epigenomics studies factors affecting gene expression. Transcriptomics studies the detection and quantification of RNA molecules. Metabolomics captures comprehensive profiling of metabolite concentration in biological systems to investigate the chemical processes involved in metabolism. Proteomics analyze all expressed proteins in biological systems and their function. These disciplines’ interaction leads to a phenotype change defined as any observable characteristic The field of omics has been implemented to identify factors associated with the progression of T2D and its related conditions [19–23], as well as to improve our understanding of the pathophysiological significance of β-cell dysfunction and its role in metabolic disorders [24]. Currently, omics technologies can be utilized to outline the main outcomes of bariatric surgeries, investigate T2D biomarkers, and lay the groundwork for innovative T2D treatment approaches. This approach helps in determining which patients stand to benefit most from various therapeutic interventions, including bariatric surgery. Yet, limited pioneering research has explored multiple omics layers to define the impact of bariatric procedures on T2D and CVD outcomes. This review aimed to systematically evaluate clinical studies that applied at least one omics layer to investigate patients with T2D undergoing bariatric surgery. Given the significant gap in current literature, we focused on identifying potential biomarkers associated with CVD and highlighted their relevance to future precision medicine strategies. To the best of our knowledge, this is the first systematic review to integrate multiple omics disciplines, offering a comprehensive perspective on the effectiveness of bariatric surgery in managing T2D and reducing CVD risk. Methods Selection strategy This systematic review was registered in PROSPERO (ID: CRD420251090077). A search was conducted in April 2024 on the PubMed database for clinical studies that utilized omics methods to assess the effect of bariatric surgery on CVD in patients with T2D. The search was executed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA 2020) guideline [25]. PICO method [26] was used to formulate the research question (population, intervention, control, and outcomes); Population of interest was patients with T2D with an age ≥ 18 years old, Intervention was bariatric surgery, Controls were patients with T2D at baseline pre-surgery, and outcomes were defined as OMICS integration to evaluate the impact of bariatric surgeries on T2D remission, inflammation and CVD changes between baseline and follow-up. Search terms included a combination of the following keywords and MeSH terms: “Bariatric surgery”, “T2D”, “OMICS”, and “CVD”. The detailed search strategy is described in the Supplementary file. Data extraction and selection process The inclusion process contained studies published in peer-reviewed journals written only in the English language. Review articles, conference abstracts, meta-analysis, case reports and comments were excluded from the search. Rayyan Software [27] was used to conduct the selection process. The selection was first implemented on the basis of evaluating the title and abstract where studies that do not match the selection criteria were excluded. Next, the remaining studies were selected after evaluating the full text by two reviewers. The following study characteristics were extracted from the studies: publication year, study design, type of bariatric intervention, study cohort, sample type, analytical technique, number of patients, gender, mean age and Body mass index (BMI), major inclusion and exclusion criteria, follow-up period, omics approach, and outcome parameters. The extracted data were then documented in an Excel spreadsheet for analysis and reference. Quality assessment The quality assessment for each study included in the systematic review was performed using the National Institutes of Health (NIH) quality assessment tool by two independent reviewers [28]. Disagreement was resolved by a third reviewer. The classifications of each study were defined as “good”, “fair”, or “poor” quality according to the cutoff points defined by the NIH tool. A summary of the quality assessment is presented in the Supplementary file. Forty studies were “good” quality, nine studies of “fair” quality, and none of “poor” quality. Statistical analysis A meta-analysis was performed using the summarized data of the most frequently studied biomarkers. Biomarkers examined in three or more studies were analyzed using Review Manager 5.4 software (RevMan). Mean differences and their 95% confidence intervals (CI) were calculated for continuous outcomes, and statistical heterogeneity was assessed using a random-effects model. Heterogeneity among studies was assessed using the Q statistical (Cochran’s Q) test and I2 statistic. I2 of 50% or above was designated significant heterogeneity and 30–49% moderate heterogeneity. Forest plots demonstrate the pooled prevalence estimates of the meta-analysis results. P < 0.05 was used as the threshold for statistical significance. Results Screening and article selection The search strategy reported a total of 6,103 studies, and 895 duplicate entries were identified. After a review of the titles and abstracts, 5,019 studies were excluded based on the inclusion criteria. The remaining 189 studies, a full-text review was conducted for eligibility. Out of these, 140 studies were excluded; 116 due to unrelated outcomes of interest and 24 because they focused on an unrelated patient population. A total of 49 studies were included in the review, comprising 35 prospective cohort studies (of which 7 were randomized controlled trials) and 7 retrospective case–control studies. The inclusion process is illustrated in the PRISMA flow diagram shown in Fig. 2a. Studies were grouped based on omics approaches. Out of the 49 studies, three focused on the genomics approach, three addressed epigenomics, and five explored the transcriptomics approach. Thirty-six studies focused on proteomics, six on metabolomics, and three on the gut microbiome. Six of the studies reported multiple omics approaches (Fig. 2b) [29–34]. Of the 49 studies included, 48 were rated as good and 1 as fair based on the NIH Quality Assessment Tool (Supplementary Table 2). A summary of all the included studies can be found in Table 1. PRISMA flowchart and upset plot of included studies.(a) PRISMA Flowchart illustrating the selection process. (* Note: Six studies included multiple approaches).(b) Upset plot summarizing shared OMICs approaches within the included studies. Overlapping OMICs approaches are indicated by intersections in the dot-and-line chart. The upper bar chart illustrates the number of studies per set intersection and the lower left bar chart shows the number of studies addressing each approach Table 1OMICS studies examining bariatric surgery effects on inflammation and cardiac health in T2D patientsStudyStudy designInterventionCohortNo. T2D subjectsMean age (Years)Mean BMI (kg/m2)Inclusion criteriaFollow-up (Years)Sample typeAnalytical techniqueBiomarkers and outcomes post-surgeryBenenati et al. [35]; 2022Retrospective cohort studyRYGB, OAGB, SG and GBUnit of Endocrinology, University of Siena, Italy3243 ± 1145.3 ± 7.0BMI 30–35 kg/m2 and T2D with poor control despite optimal medical therapy1Peripheral blood leukocytesGenotypingNo association between Thr92Ala polymorphism and the presence of T2DSchnor N. et al. [36]; 2016Prospective cohort studyRYGBBrazilian76 (100% female)Not reportedNot reportedFemale aged 20–50 years1Whole bloodGenotypingThe prevalence of T2D among Brazilian women with obesity was associated with the genetic variants rs3813929 (p = 0.001) in the 5-HT2C gene and rs1800849 (p = 0.019) in the UCP3 geneGatta-Cherifi B. et al.[37]; 2023Prospective cohort studySGEuropean12464NA1Peripheral bloodMiseq sequencingA novel loss-of-function variant in MRAP2 was found in a T2D patient with severe obesityZerrweck C. et al. [38]; 2016Prospective cohort studyGBMexican12Not reported38.5 ± 2.93T2D patients with severe obesity0.5Subcutaneous adipose tissue biopsyDNA methylationA total of 1152 CpG sites differentially methylatedVohl M-C et al. [33]; 2015Prospective cohort studyBPD/DSQuebec Heart and Lung Institute; Canada20 (100% female)28.9 ± 3.745.0 ± 7.2Women with severe obesity and T2D4–22Whole blood, plasma, and buffy coatELISA, Microarray, Bisulfite conversion and quantitative DNA methylationDifferential methylation analysis identified 15,343 genes demonstrating at least one site (p < 1.43 × 10–7). FKBP1A and ZNF323 genes showed the highest correlations with insulin and HOMA-IR (p < 0.0001)Benton M. et al. [30]; 2015Prospective cohort studyGBWakefield Hospital, New Zealand15 (100% female)44 ± 1047.6 ± 11.3Women with obesity1.5 ± 0.5Subcutaneous and omental adipose tissue and whole bloodDNA methylation and mRNA qPCRA greater proportion of CpGs are hypermethylated before weight loss. Robust correlations were identified between fasting glucose and HDAC4, SLC37A3 and DENND1C in subcutaneous adiposeAlmby K. et al. [29]; 2021Randomized controlled trialRYGBDepartment of Endocrinology and Diabetes at Uppsala University Hospital; Sweden13 (76.9% female)51 ± 1036.8 ± 3.9T2D patients aged 18–60 years, with a BMI of 30–45 kg/m2 and diabetes duration < 10 years and treated with < three antidiabetic agents but not insulin2Adipose tissuemRNA qPCRAdipogenesis genes (PPARG, CEBPA, CEBPB, BMP4, FABP4, and FAS) (p < 0.05), glucose transport (GLUT1, GLUT4, AKT1, and IRS1 (p < 0.1)), and fatty acid oxidation (CPT1B) (p < 0.001) were increased. While IL6 (p < 0.1) and IL18 (p < 0.01) were decreasedRega-Kaun et al. [39]; 2020Prospective cohort studyRYGBCaucasian; Austria17Not reportedNot reportedNot reported1Plasma and serummiRNA qPCRmiR-122, miR-130, and miR-132 were significantly decreased, while miR-375 was increased (p < 0.001 for all miRNAs)Fachim et al. [40]; 2020Prospective cohort studyRYGB, and SGSalford Royal Hospital; United Kingdom5Not reportedNot reportedPatients with impaired glucose regulation aged 18–80 years0.5Adipose tissuemRNA qPCRCaveolin-1 was discreetly increased (p = 0.084)Lee SJ et al. [44]; 2021Prospective cohort studyRYGB, and SGKorean Obesity Surgical Treatment Study18 (27.8% female)43.8 ± 1337.8 ± 5.9BMI ≥ 35 kg/m2 or 30.0–34.9 kg/m2 and obesity-related comorbidities0.5SerumUPLC/TQ-MSIncreased levels of TMAO (p = 0.043) and decreased levels of its precursor choline (p = 0.091) after RYGB; while no change after GBHärm A. et al. [31]; 2020Prospective cohort studyRYGBOulu University Hospital; Finland16 (63% female)49.8 ± 8.141.8 ± 4.4Adults aged 18–65 years undergoing bariatric surgery0.5Feces and serumNot reportedFecal short-chain fatty acids (SCFAs), especially acetate (p < 0.002) and butyrate (p < 0.03) levels, were significantly lower after surgeryNarath S. et al. [45]; 2016Prospective cohort studyRYGBMedical University of Graz Austria and Interdisciplinary Obesity Center in St. Gallen Switzerland2446.8 ± 11.343.9 ± 5.4Adults aged > 18 yers1SerumLC-HRMSA significant decline in Sarcosine (p = 0.031), pyroglutamic acid (p = 0.044), alanine (p = 0.005) and leucyl-proline (p = 0.049) in patients with diabetes remission compared to patients without diabetes remissionArora T. et al. [46]; 2015Prospective cohort studyRYGBNot reported14 (71.4% female)48.45 ± 2.8749.1 ± 2.74Patients undergoing RYGB42 daysPlasmaGC/TOF–MST2D in remission exhibited higher pre-surgery levels of tricarboxylic acid cycle intermediates and TG with long-chain fatty acids compared to those not in remissionThaker V. et al. [47]; 2024Prospective cohort studyRYGB and GB10 US hospitals93 (69.9% female)51.5 ± 9.8346.63 ± 7.43T2D adults aged ≥ 18 years undergoing RYGB or GB2PlasmaLC/MSChanges in BCAAs and branched chain keto acids were positively linked with the change in A1c (false discovery rate q < 0.001)Zhao S. et al. [34]; 2024Prospective cohort studyRYGBOulu University Hospital; Finland16 (63% female)49.8 ± 8.141.8 ± 4.4Adults aged 18–65 years undergoing bariatric surgery0.5Faces and serumELISA and NMRTG, cholesterol, Apo-B-100, lactate, and glycoprotein acetyls were decreased (p < 0.0031). The concentrations of BCAA (leucine, isoleucine, and valine) and aromatic amino acids (phenylalanine and tyrosine) decreased (p < 0.0031) by 17% for tyrosine and 23% for leucineSong Z et al. [54]; 2022Prospective cohort studyRYGBJinshazhou Hospital of Guangzhou University of Traditional Chinese Medicine3Not reported40.41 ± 5.65Adults with obesity aged between 18–70 years who are candidates for bariatric surgeryNot reportedOmental adipose tissue and serumLC–MS/MS and ELISAAIP1 was increased (p < 0.05) and TNF-α was decreased (p < 0.05) post-RYGBAl Shawaf E et al. [51]; 2020Prospective cohort studySGDasman Diabetes Institute; Kuwait11Not reported41 ± 3.35Normal weight and patients with T2D and severe obesity1Plasma and serumImmunoassay analyzer and ELISAThe Atherogenic Index of Plasma AIP (log TG/HDL) (p < 0.01) and hsCRP (p < 0.01) were significantly decreased after SG. No significant change was observed in Ox-LDLAdam S et al. [49]; 2019Prospective cohort studyRYGBSalford Royal Hospital; England20Not reportedNot reportedNot reported1Plasma and serumImmunoturbidimetric assays and ELISAhsCRP, IL-6, TG (p < 0.001) and small, dense SD-LDL ApoB (p < 0.001) decreased, while HDL cholesterol increased (p < 0.001) significantly. No significant change was observed in total ApoBBonaventura A et al. [52]; 2019Prospective cohort studyRYGB and BPDSurgical Department of Genoa University; Italy31 (16.13% female)56 ± 7.7734.9 ± 9.96T2D patients aged between 35–70 years, with diabetes duration > 3 years, and A1c ≥ 7.5%5SerumELISALow NLR is associated with long-term T2D remission and CRP levels decreased after surgeryWei J-H et al. [53]; 2022Prospective cohort studyOAGBMin-Sheng General Hospital; Taiwan40 (57.5% female)47.8 ± 11.339.3 ± 8.5Adults with obesity aged between 18–65 years with a BMI of ≥ 27.5 kg/m2 and A1c > 7% and who failed the medical control for obesity with acceptable operative risk1SerumELISAFGF-19 and ox-LDL decreased significantly (p < 0.001) and Corin increased significantly (p = 0.027). Changes in FGF-21 and sRAGE were not significantGhanim H. et al. [55]; 2018Prospective cohort studyRYGBNew York15 (66.67% female)53.8 ± 32.952.1 ± 18.6T2D patients with severe obesity with a BMI of ≥ 40 kg/m20.5PlasmaELISANeprilysin was decreased post-RYGB (p < 0.05). LDL and TG were decreased while was HDL increased (p < 0.05). No significant change was observed in total cholesterol levelHo JH. et al. [62]; 2020Prospective cohort studyRYGB, SG and omega loop bypassSalford Royal NHS Foundation Trust tertiary weight management centre; United Kingdom3751.4 ± 10.549.7 ± 10Patients with severe obesity undergoing bariatric surgery1Plasmachemiluminescent immunoassays and ELISALp(a) levels were increased while OxPL-apoB (p < 0.05) levels were decreased. No significant change was observed in Ox-LDLLee P. et al. [56]; 2018Retrospective cohort studySG and GB (RYGB and mini GB)Ethnic composition consists of Chinese, Malay, and Indian from recruited from a hospital in Singapore145 (57.9% female)47.6 ± 9.140.0 ± 7.6T2D patients who underwent either primary SG or GB and who have at least 1 year of follow-up1Not reportedNot reportedHDL levels increased (p < 0.001), while TG decreased (in SG (p = 0.003) and in GB (p < 0.001). LDL increased (p = 0.004) after SG but was unchanged after GB (p = 0.239)Huang Y. et al. [57]; 2022Prospective cohort studySGTaipei Medical University Hospital, Taiwan95 (53.7% female)41.8 ± 8.7842.5 ± 7Undergoing the primary SG surgery1SerumNot reportedAccording to the Framingham 10-year CHD risk model, the overall CVD risk decreased from 32.4 to 17.2% in the men and from 15.8 to 5.8% in the female, respectively, after SG (p < 0.001). TG decreased (p < 0.05) and HDL increased (p < 0.05)Billington E. et al. [63]; 2019Randomized controlled trialSGNorth Shore Hospital; New Zealand22 (41% female)49.3 ± 8.7241.6 ± 7.92Adults aged 22‐55 years undergoing bariatric surgery with at least 6 months duration of T2D with a BMI of 35‐65 kg/m2 for at least 5 years1PlasmaELISAFGF23 (p = 0.005), PTH (p = 0.009) and leptin (p = 0.01) declined following SGHärm A. et al. [31]; 2021Prospective cohort studyRYGBOulu University Hospital; Finland16 (63% female)49.8 ± 8.141.8 ± 4.4Adults aged 18–65 years undergoing bariatric surgery0.5Faces and serumNot reportedTG decreased (p = 0.026) and fecal calprotectin levels (p < 0.001), active LPS concentration (p = 0.001), ASCA (p = 0.003), and MG-H1 (p < 0.001) increased significantly, while the SCFA acetate (p = 0.04) was significantly lowerIqbal Z. et al. [50]; 2022Prospective cohort studyRYGBSpecialized and complex obesity/bariatric service in Manchester; United Kingdom24Not reported52.29 ± 8.86Not reported1Serum and plasmaELISA and immunoturbidimetric assaysThe levels of IL-6 (p < 0.0001), glycApoB (p < 0.05), HOMA-IR (p < 0.001), CRP (p < 0.001) decreased significantly, while HDL (p < 0.001) and total cholesterol (p < 0.38) increasedLau E. et al. [32]; 2021Randomized controlled trialRYGBCentro Hospitalar São João (CHSJ); Portugal8 (50% female)53.4 ± 8.7833.6 ± 1.84T2D adults aged between 20–65 years with a BMI ≥ 30 kg/m2 and T2D duration > 3 months, overnight fasting C-peptide > 0.7 ng/ml and negative antiGAD autoantibody1SerumEnzymatic colorimetric test, HPLC and electrochemiluminescence immunoassayC-peptide, and HOMA-IR were significantly decreased (p = 0.020 and p = 0.027, respectively), while HDL (p = 0.003) increased significantlyFaramia J. et al. [64]; 2021Prospective cohort studyRYGB, SG and BPD/DSIUCPQ; Canada20 per surgery type (70% female per surgery type)49.5 ± 6.4944.6 ± 5.46Adults aged ≥ 18 years with aBMI > 40 kg/m2 or > 35 kg/m2 with a comorbidity1PlasmaELISAPlasma IGFBP-2 increased significantly (p < 0.0001) reaching higher levels after BPD-DS than after RYGB and SGShah R. et al. [58]; 2019Retrospective cohort studySG and GBBariatric surgery and Framingham Heart Studies; United StatesBariatric cohort: 17Framingham Heart cohort:192 (42% female)Bariatric cohort: Data not reported,Framingham Heart cohort: 56.3 ± 11.6Bariatric cohort: data not reported. Framingham Heart cohort: 32.4 ± 6.3Adults from Bariatric surgery and Framingham Heart cohortsBariatric cohort: 1Framingham Heart cohort:6.3PlasmaProximity Extension Assay and ELISAMore than 180 proteins were analyzed, and only IGFBP-2 protein was associated inversely with diabetes risk and increased after surgery ((OR = 0.63, 95% CI 0.49–0.79, p = 0.0001)Jahansouz C. et al. [65]; 2019Randomized controlled trialRYGBUniversity of Minnesota, Columbia University in New York, and Mayo Clinic in Rochester34 (65% female)51 ± 936.1 ± 2.7T2D adults aged > 30 years, with a BMI of 30.0–39.9 kg/m2, undergoing treatment for T2D for at least 6 months, A1c ≥ 8.0%, and C-peptide > 1.0 ng/ml 90 min after a liquid mixed meal1SerumELISACirculating FABP4 decreased by 42% after surgery (p = 0.002)Seki Y. et al. [59]; 2018Retrospective cohort studyLSG-DJBJapanese72 (48.6% female)46.8 ± 9.031.7 ± 2.0T2D patients with Mild obesity with a BMI < 35 kg/m2 underwent LSG-DJB B from September 2007 to March 2015, and the patients who presented for at least 1 year of follow-up1Not reportedNot reportedFasting C-peptide, systolic BP, diastolic BP, LDL-C, and TGs decreased (p < 0.001); while HDL increased significantly from the baseline (p < 0.001)Demyanets S. et al. [66]; 2020Prospective cohort studyRYGBAustrians25Not reportedNot reportedAdults aged ≥ 18 years with aBMI > 40 kg/m2 or > 35 kg/m2 with secondary disease and the failure or futility of a structured conservative program1SerumELISAThe pro-inflammatory marker sST2 decreased by 43% after surgery (p = 0.0016) and showed positive correlation with IL-33 (r = 0.44, p = 0.05)Almby K. et al. [29]; 2021Randomized controlled trialRYGBDepartment of Endocrinology and Diabetes at Uppsala University Hospital; Sweden13 (76.9% female)51 ± 1036.8 ± 3.9T2D patients aged 18–60 years, with a BMI of 30–45 kg/m2 and diabetes duration < 10 years and treated with < three antidiabetic agents but not insulin2Not reportedEnzymatic immunoassaysHDL increased (p < 0.01), while FFA, TGs and CRP reduced significantly (p < 0.001, p < 0.01, p < 0.05; respectively)Istomin N. et al. [60]; 2022Prospective cohort studyRYGBOulu University Hospital; Finland16 (62.5% female)49.8 ± 8.141.8 ± 4.4Adults aged between 18–65 undergoing RYGB0.5Plasma and fecesChemiluminescence immunoassayTotal fecal IgA was increased (p = 0.020), while total IgM and IgG were not affected by the surgeryEspinosa O. et al. [61]; 2018Prospective cohort studyRYGBMexicans23 (87% female)44.4 ± 7.733.1 ± 1.1Mexican patients with mild obesity (BMI 30–34.9 kg/m2) aged 18–60 years with T2D who underwent laparoscopic gastric bypass between January 2014 and January 20151.5Not reportedNot reportedCardiovascular risk was improved significantly (p = 0.03) which was calculated by dividing total cholesterol by HDL. C peptide was decreased significantly (p < 0.001)Munkonda et al. [87]; 2012Prospective cohort studyBPD/DSBariatric surgery clinic of the Institut universitaire de cardiologie et de pneumologie de Que´bec (IUCPQ); Canada22 (100% female)42.0 ± 10.649.8 ± 7.9Women, aged > 18 years, with BM ≥ 40 or BMI ≥ 35 kg/m2 with associated comorbidities1PlasmaTurbidimetric inhibition immunoassay and ELISA,Acylation stimulating protein was decreased (p < 0.001), as well as glucose, insulin, HOMA-IR, TG, cholesterol, LDL, HDL and hsCRP levelsKim M. et al. [88]; 2013Prospective cohort studyRYGBYeouido St. Mary’s Hospital57 (66.7% female)43.5 ± 10.134.2 ± 4.1T2D patients aged 19–64 years, with a BMI > 30 kg/m21SerumELISALevels of BMP-4 (p = 0.024), PAI-1, hsCRP, and FFAs were significantly decreasedZhao X. et al. [89]; 2017Prospective cohort studyRYGBJiahe Surgical Hospita; China20 (65% female)42.70 ± 12.6034.20 ± 6.22Chinese patients with T2D and obesity undergoing RYGB surgery1.5SerumImmunoassay analyzerThe 10-year cardiovascular risk of coronary heart disease decreased from 13.05% to 3.81% (p = 0.001) and the 10-year risk of stroke decreased from 19.66% to 14.22% (p = 0.002)Shimizu H. et al. [90]; 2017Prospective cohort studySGTokyo Metropolitan Tama Medical Center; Japan10 (50% female)48.8 ± 2.740.9 ± 3.2BMI > 35 kg/m20.5SerumELISAhsCRP and plasminogen activator inhibitor-1 were decreased at 6 months. No significant changes were observed in IL-6, IL-8, TNF-α, serum amyloid A protein, and monocyte chemotactic protein-1Liang Z. et al. [91]; 2013Randomized controlled trialsRYGBSouthwest Hospital of Third Military Medical University; China31 (29% female)50.81 ± 5.4430.48 ± 0.94T2D patients aged 30–60 years with BMI > 28 kg/m21Plasma and serumELISASurgery group had the greatest degree of improvement (p < 0.05) compared to the usual care group in the inflammation index (hsCRP, TNF-a, and high molecular weight adiponectin)Cazzo E. et al. [92]; 2014Retrospective cohort studyRYGBHospital das Clínicas da Unicamp; Brazil96 (23% female)46 ± 10.844.3 ± 8.74Subjects with obesity who underwent RYGB1Not reportedNot reportedSignificant reductions (p < 0.0001) in fasting glucose, fasting insulin, A1c, LDL, and TG levels, along with an increase in HDL levels (p < 0.0001)Kota S. et al. [93]; 2012Prospective cohort studyII-DSGKirloskar Hospital; India17 (29.4% female)29.2 ± 7.550.7 ± 8.1T2D of 5 years duration aged 25–70 years with BMI ≥ 20 kg/m2 and stimulated C-peptide level > 1 ng/ml1.5Serum and plasmaChemiluminescence, Immunoturbidimetry assay and HPLCSignificant decrease in glycemic, lipid parameters, and microalbuminuria (p < 0.05)Boza C. et al. [94]; 2011Retrospective cohort studyRYGBPontificia Universidad Católica; Chile30 (56.6% female)48 ± 933.5 ± 1.2T2D patients with a BMI 30–35 kg/m22Serum and plasmaChemiluminescent immunoassaySignificant reductions (p < 0.05) in fasting glucose, fasting insulin, TG and A1c, along with an increase in HDL levels. No significant change in LDL, and TC levelsCourcoulas A. et al. [95]; 2024Randomized controlled trialRYGB, SG and GBCleveland Clinic, Joslin Diabetes Center, University of Pittsburgh, and University of Washington; US16649.0 ± 9.036.6 ± 3.6T2D adults aged between 18–65 years with a BMI 27–45 kg/m212Not reportedNot reportedPatients who underwent bariatric surgery had better glycemic control, fewer diabetic medications, and greater rates of remission compared to those who received medical/lifestyle interventionsPané A. et al. [96]; 2024Prospective cohort studyRYGB and SGSpain18 (75% female)51.0 ± 5.443.2 ± 4.3Adults aged between 18–70 years with a BMI > 40 or > 35 kg/m2 with obesity-related comorbidities1PlasmaNot reportedSignificant decrease (p < 0.001) in fasting glucose, TG, hs-CRP and A1cMartínez-Montoro J et al. [97]; 2023Prospective cohort studySG and BPDVirgen de la Victoria University Hospital; Spain98 (73.5% female)48.08 ± 8.8749.53 ± 7.18T2D adults aged ≥ 18 years with a BMI ≥ 35 kg/m21SerumImmunoassaySignificant differences in T2D remission were found between patients with TG ≥ 150 mg/dl and < 150 mg/dl (p = 0.002)Happonen N. et al. [98]; 2024Prospective cohort studyRYGBOulu University Hospital; Finland16 (62.5% female)49.8 ± 8.141.8 ± 4.4A medical indication for RYGB with an age of 18–65 years0.5Plasma and facesChemiluminescent immunoassayImprovement in the systemic inflammation state evidenced by decreased levels of total plasma IgG, better hypertension and hyperglycemia, and their correlations with lower levels of certain plasma antibodiesQi QYD et al. [99]; 2023Randomized controlled trialGBAustralia21 (81% female)5329.5T2D adults aged 18–65 years with a BMI 25–30 kg/m210Not reportedNot reportedHDL increased at 10 years (p = 0.030), but there was no significant change in blood pressure or other lipid measuresChakaroun R. et al. [67]; 2021Prospective cohort studyNot reportedUniversity of Leipzig Medical Center; Germany24 (70.83% female)50.1 7.9449.0 ± 6.68BMI ≥ 40 kg/m2, or ≥ 35 kg/m2 with at least one obesity comorbidity1Whole EDTA blood16S rRNA sequencingAt the phylum level:­ (Proteobacteria, Firmicutes, and Patescibacteria)¯ (Cyanobacteria and Actinobacteria)At the genus level:¯ (Sphingomonas, Pseudomonas, Bacillaceae, and Burkholderiaceae)­ (CM1G08, Acinetobacter, and Acidibacter)Lau E. et al. [32]; 2021Randomized controlled trialRYGBCentro Hospitalar São João (CHSJ); Portugal8 (50% female)53.4 ± 8.7833.6 ± 1.84T2D adults aged between 20–65 years with a BMI ≥ 30 kg/m2 and T2D duration > 3 months, overnight fasting C-peptide > 0.7 ng/ml and negative antiGAD autoantibody1Stool16S rRNA sequencingAt the phylum level:­ Proteobacteria and rPF;No significant difference was observed in rFBAt the genus level:­ prokaryotic genera richness (p = 0.03) and ­ (Klebsiella, Gammaproteobacteria, Enterobacter, unclassified Gammaproteobacteria, and unclassified Veillonellaceae)¯ (Ruminococcus, unclassifed Lachnospiraceae family and Faecalibacterium)Assal K. et al. [68]; 2020Prospective cohort studyRYGBColoproctology and Digestive Surgery Service of Hospital das Clínicas of the University of São Paulo, School of Medicine (HC-FMUSP); Brazil2545.80 ± 7.9546.40 ± 5.48Aged 18–60 years, BMI ≥ 35 kg/m2, confirmed diagnosis of T2D with a fasting glucose ≥ 126 mg/dL, glycated hemoglobin (A1c) ≥ 7%, and current use of oral anti-diabetic drugs1Stool16S rRNA sequencingAt the phylum level:­ Proteobacteria and ¯ rFB (rFB = 1.53 vs. 3.27, p < 0.046)At the genera level:­ (Veillonella and Streptococcus) and ¯ (Flavonifractor, Blautia, and Butyricicoccus) (p < 0.005)Only the Oribacterium↓ between the 3- and 12-month time points postoperatively (p < 0.003)aHR: adjusted hazard ratio; AIP: atherogenic index of plasma; AIP1: Abstract Actin-interacting protein 1; AMI: acute myocardial infarction; ApoB: Apolipoprotein B100; ASCA: anti-saccharomyces IgA antibodies; BCAAs: Branched-chain amino acids; BPD/DS: Biliopancreatic diversion with duodenal switch; CHD: coronary heart disease; CI, Cumulative incidence; CRP: C-reactive protein; ELISA: enzyme-linked immunosorbent assay; FABP4: Fatty acid binding protein 4; FFA: Free fatty acid; FGF: fasting serum fibroblast growth factor; GB: gastric bypass; GC/TOF–MS: gas chromatograph combined with time-of-flight mass spectrometer; HDL: high density lipoprotein; HF: heart failure; HOMA –IR: Homeostatic model assessment of insulin resistance; HPLC: High Performance Liquid Chromatography; hsCRP: Highly sensitive C-reactive protein; IGFBP-2: insulin-like growth factor binding protein-2; IGFBP: Insulin-like growth factor-binding protein; II-DSG: Ileal interposition with diverted sleeve gastrectomy; LC-HRMS: liquid chromatography coupled to high resolution mass spectrometry; IL: interleukin; IUCPQ: Institut Universitaire de Cardiologie et de Pneumologie de Québec; LDL: low-density lipoprotein, Lp(a) lipoprotein(a); LPS: active lipopolysaccharide; LSG-DJB: Laparoscopic sleeve gastrectomy with duodenojejunal bypass; MG-H1: methylglyoxal-hydro-imidazolone; NLR: Neutrophil-to-lymphocyte ratio; NMR: Nuclear magnetic resonance; OAGB: one-anastomosis gastric bypass; ox-LDL: oxidized low-density lipoprotein; OxPL: oxidised phospholipids; PAD: peripheral artery disease; qPCR: quantitative polymerase chain reaction; rFB, Firmicutes/Bacteroidetes ratio; rPF, Proteobacteria/Firmicutes ratio; RYGB, Roux-en-Y Gastric bypass; SCFAs: short-chain fatty acids; SG, sleeve gastrectomy; SNP: single nucleotide polymorphism; sRAGE: soluble receptor for advanced glycation end-products; sST2: soluble suppression of tumorigenicity-2; TG: triglycerides; TMAO: Trimethylamine N-oxide; TNF-α: Tumor necrosis factor; UPLC/TQ-MS: ultra-high-performance liquid chromatography/triple quadrupole mass spectrometry OMICS studies examining bariatric surgery effects on inflammation and cardiac health in T2D patients Adults aged ≥ 18 years with a BMI > 40 kg/m2 or > 35 kg/m2 with a comorbidity Bariatric cohort: 17 Framingham Heart cohort: 192 (42% female) Bariatric cohort: Data not reported, Framingham Heart cohort: 56.3 ± 11.6 Bariatric cohort: 1 Framingham Heart cohort: 6.3 Adults aged ≥ 18 years with a BMI > 40 kg/m2 or > 35 kg/m2 with secondary disease and the failure or futility of a structured conservative program At the phylum level: ­ (Proteobacteria, Firmicutes, and Patescibacteria) ¯ (Cyanobacteria and Actinobacteria) At the genus level: ¯ (Sphingomonas, Pseudomonas, Bacillaceae, and Burkholderiaceae) ­ (CM1G08, Acinetobacter, and Acidibacter) At the phylum level: ­ Proteobacteria and rPF; No significant difference was observed in rFB At the genus level: ­ prokaryotic genera richness (p = 0.03) and ­ (Klebsiella, Gammaproteobacteria, Enterobacter, unclassified Gammaproteobacteria, and unclassified Veillonellaceae) ¯ (Ruminococcus, unclassifed Lachnospiraceae family and Faecalibacterium) At the phylum level: ­ Proteobacteria and ¯ rFB (rFB = 1.53 vs. 3.27, p < 0.046) At the genera level: ­ (Veillonella and Streptococcus) and ¯ (Flavonifractor, Blautia, and Butyricicoccus) (p < 0.005) Only the Oribacterium ↓ between the 3- and 12-month time points postoperatively (p < 0.003) aHR: adjusted hazard ratio; AIP: atherogenic index of plasma; AIP1: Abstract Actin-interacting protein 1; AMI: acute myocardial infarction; ApoB: Apolipoprotein B100; ASCA: anti-saccharomyces IgA antibodies; BCAAs: Branched-chain amino acids; BPD/DS: Biliopancreatic diversion with duodenal switch; CHD: coronary heart disease; CI, Cumulative incidence; CRP: C-reactive protein; ELISA: enzyme-linked immunosorbent assay; FABP4: Fatty acid binding protein 4; FFA: Free fatty acid; FGF: fasting serum fibroblast growth factor; GB: gastric bypass; GC/TOF–MS: gas chromatograph combined with time-of-flight mass spectrometer; HDL: high density lipoprotein; HF: heart failure; HOMA –IR: Homeostatic model assessment of insulin resistance; HPLC: High Performance Liquid Chromatography; hsCRP: Highly sensitive C-reactive protein; IGFBP-2: insulin-like growth factor binding protein-2; IGFBP: Insulin-like growth factor-binding protein; II-DSG: Ileal interposition with diverted sleeve gastrectomy; LC-HRMS: liquid chromatography coupled to high resolution mass spectrometry; IL: interleukin; IUCPQ: Institut Universitaire de Cardiologie et de Pneumologie de Québec; LDL: low-density lipoprotein, Lp(a) lipoprotein(a); LPS: active lipopolysaccharide; LSG-DJB: Laparoscopic sleeve gastrectomy with duodenojejunal bypass; MG-H1: methylglyoxal-hydro-imidazolone; NLR: Neutrophil-to-lymphocyte ratio; NMR: Nuclear magnetic resonance; OAGB: one-anastomosis gastric bypass; ox-LDL: oxidized low-density lipoprotein; OxPL: oxidised phospholipids; PAD: peripheral artery disease; qPCR: quantitative polymerase chain reaction; rFB, Firmicutes/Bacteroidetes ratio; rPF, Proteobacteria/Firmicutes ratio; RYGB, Roux-en-Y Gastric bypass; SCFAs: short-chain fatty acids; SG, sleeve gastrectomy; SNP: single nucleotide polymorphism; sRAGE: soluble receptor for advanced glycation end-products; sST2: soluble suppression of tumorigenicity-2; TG: triglycerides; TMAO: Trimethylamine N-oxide; TNF-α: Tumor necrosis factor; UPLC/TQ-MS: ultra-high-performance liquid chromatography/triple quadrupole mass spectrometry Genomics Identifying genetic factors related to T2D may help to guide the establishment of novel interventions for patients and minimize the risk of CVD. It could provide weight management strategies pre- and post-bariatric surgery and influence CVD outcomes. Of the 49 studies included in this review, three addressed a genomic approach [35–37]. Benenati et al. [35] investigated the impact of the DIO2 p.Thr92Ala substitution in 32 T2D subjects which caused by a single nucleotide polymorphism (SNP) (rs225014) on the response to bariatric surgery in patients with T2D at 12 months follow-up. The study found that there was no association between the p.Thr92Ala polymorphism and the presence of T2D (p = 0.8), hypertension, dyslipidemia or metabolic syndrome regardless of the different DIO2 gene expression forms (Thr/Thr, Thr/Ala, Ala/Ala). The p.Thr92Ala polymorphism is a common SNP in the DIO2 gene, resulting in a threonine-to-alanine substitution at codon 92. Another study by Schnor et al. [36] showed that rs3813929 (p = 0.001) and rs1800849 (p = 0.019) polymorphisms in the 5-HT2C and UCP3 genes are associated with the prevalence of T2D among 76 Brazilian women with obesity who were candidates for bariatric surgery. Additionally, Gatta-Cherifi B. et al. [37] reported a novel loss-of-function variant in MRAP2 found in one T2D patient with severe obesity who had a successful outcome following bariatric surgery. Currently, insufficient data at present on predicting patients’ responses to bariatric surgery suggest that these genetic variations cannot yet be employed in clinical practice. Moreover, the genomics approach illustrates only one essential element of biological variation, which must be considered alongside novel biomarkers emerging from proteomics, metabolomics, and microbiomes approaches. Epigenomics In the context of bariatric surgery, understanding the epigenetic alterations that occur before and after the procedure can provide valuable insights into the mechanisms driving T2D remission. This knowledge can be leveraged to identify patients who are more likely to experience positive epigenetic changes following surgery, enabling further personalization of treatment plans. This review includes three studies that investigated the epigenetic impact of bariatric surgery on T2D, all of which indicated significant DNA methylation changes in adipose tissue and blood post-surgery [30, 33, 38]. In Mexican populations, 1152 CpG sites showed differential methylation in abdominal subcutaneous adipose tissue in patients with T2D and obesity compared to patients with obesity but without T2D [38]. Enrichment analysis indicated that genes associated with differentially methylated CpG sites are implicated in metabolic pathways, with DNA methylation remodeling correlating with improved metabolic status. Furthermore, whole blood DNA methylation analyses in women with obesity revealed variations in methylated genes at the gene body region after surgery, linked to diabetic and inflammatory functions, including ABCF1, FKBP1A, IFIT2, and HTT, indicating durable metabolic improvements post-surgery [33]. Another study found that a larger proportion of CpGs are hypermethylated prior bariatric surgery in 15 T2D women, and increased methylation is observed in the 3′ untranslated regions and gene bodies compared to the promoter regions [30]. In addition, they reported that bariatric surgery altered the methylation of PRDM16 within adipose tissue. These findings suggest that the benefits of bariatric surgery for patients with T2D can be partially explained by epigenetic remodeling. While DNA methylation is the most extensively studied epigenetic regulatory mechanism, other epigenetic mechanisms, such as microRNAs and long non-coding RNAs, may also contribute to the regulation of T2D, CVD, and body weight. Additionally, further research is needed to understand the relationship between genetic and epigenetic factors in bariatric surgery outcomes. Transcriptomics Five transcriptomic studies have investigated the impact of bariatric surgery on T2D [29, 30, 33, 39, 40]. Almby K. et al. [29] studied the mRNA expression of different genes in patients with T2D undergoing bariatric surgery. They reported an increase in the expression of genes involved in adipogenesis (PPARG, CEBPA, CEBPB, BMP4, FABP4, and FAS), glucose transport (GLUT1, GLUT4, AKT1, and IRS1), and fatty acid oxidation (CPT1B) 104 weeks after RYGB. In addition, the levels of pro/anti-inflammatory cytokines (IL6 and IL18) were shown to be decreased post-surgery. These changes suggest enhanced lipid oxidation after surgery and highlight the importance of systemic inflammation and adipose tissue function in preventing microvascular and cardiovascular comorbidities. Rega-Kaun et al. [39] reported a significant change in miRNA profiles after bariatric surgery in patients with T2D. Specifically, they observed a significant decrease in miR-122, miR-130, and miR-132 levels one-year post-surgery, while miR-375 was increased (p < 0.001 for all miRNAs). These miRNAs were previously found to be associated with insulin resistance and β-cell dysfunction in people with obesity [41, 42]. On the other hand, Fachim et al. [40] found that the gene expression of Caveolin-1 in fat tissue increased slightly, but not significantly (p = 0.084), in five patients with T2D six months after bariatric surgery. Caveolin-1 is known for its role in stimulating the production of major pro-inflammatory cytokines [43]. Vohl M-C et al. [33] identified 15,343 genes from differential methylation analysis demonstrating at least one site with significant correlation (p < 1.43 × 10–7). The FKBP1A and ZNF323 genes showed the highest correlations with insulin and HOMA-IR. In addition, Benton M. et al. [30] examined the mRNA expression of five genes that are associated with obesity and T2D with differential promoter methylation (ACACA, CETP, CTGF, S100A8, S100A9 and PLIN4) in 15 T2D women. Findings reveal that the mRNA expression of CTGF (p = 0.002), S100A8 (p = 0.0005), S100A9 (p = 0.0009) and PLIN4 (p = 0.0002) was upregulated, while ACACA (p = 0.001) and CETP (p = 0.00008) downregulated. The previous studies employed traditional transcriptomic techniques, such as Quantitative PCR method (qPCR) and microarray. However, single-cell RNA sequencing technology has emerged to overcome many limitations of these traditional techniques, especially in detecting low-expressed and novel transcripts. Despite the promise of the transcriptomic approach for understanding T2D, there are still insufficient data highlighting major modifications induced by bariatric surgery in patients with T2D. Metabolomics Detecting an individual’s metabolomics profile has a key role in predicting metabolic outcomes of disease states and early changes in particular biochemical pathways. Six prospective studies have studied metabolites profile to evaluate the impact of bariatric surgery on CVD risk in patients with T2D [31, 34, 44–47]. Lee SJ et al. [44] reported elevated levels of the microbiota-derived metabolite, trimethylamine N-oxide (TMAO) (p = 0.043) and a trend toward decreased levels of its precursor choline (p = 0.091) six months after RYGB in seven patients with T2D. However, no change in TMAO metabolite was observed in 11 patients with T2D who underwent SG. Further, Härm A. et al. [31] found that fecal short-chain fatty acids (SCFAs), especially acetate (p < 0.002) and butyrate (p < 0.03), were significantly lower after RYGB. These SCFAs are recognized for their anti-inflammatory features, which therefore, could be key contributors to preventing gastric inflammation [48]. Narath S. et al. [45] reported significant declines in sarcosine (p = 0.031), pyroglutamic acid (p = 0.044), alanine (p = 0.005) and leucyl-proline (p = 0.049) in patients with diabetes remission compared to those without remission after bariatric surgery. In addition, Arora T. et al. [46] showed that patients with T2D remission exhibited higher pre-surgery levels of tricarboxylic acid cycle intermediates and triglycerides with long-chain fatty acids compared to those not in remission. Thaker V. et al. [47] found that changes in branched-chain amino acids (BCAAs) and branched-chain keto acids after surgery were positively linked with changes in A1c (false discovery rate q < 0.001). Moreover, Zhao S. et al. [34] reported that BCAAs were the most affected metabolites six months after RYGB, with concentrations of aromatic amino acids and BCAAs decreasing by 17% for tyrosine and 23% for leucine. Bariatric surgery has beneficial impacts on attenuating risks of CVD; however, additional long-term studies are needed to better understand the role of different metabolites in predicting the outcomes of bariatric surgery in patients with T2D. Proteomics The search strategy identified 36 studies exploring proteomics technology to uncover a variety of proteins that are associated with CVD risk in patients with T2D undergoing bariatric surgery. Inflammatory profiles play a crucial role in the metabolic disturbances of T2D, with inflammatory adipocytokines exacerbating diabetic vascular complications and coagulation responses. Key mediators of diabetogenic inflammation include Interleukin-6 (IL)−6 [49, 50], C-Reactive Protein (CRP) [50], CRP-high-sensitivity CRP (hs-CRP) [32, 49, 51–53], Tumor necrosis factor-alpha (TNF)-α [54], and lipid profiles [29, 31, 51, 53, 55–61] all implicated in insulin resistance and modulated by bariatric surgery. Other markers such as IL-8, Adiponectin, Resistin and Leptin also exhibit significant changes post-surgery. However, inconsistent data on short- and long-term inflammatory marker levels post-procedure highlights variability in surgical interventions. Nevertheless, a consistent decline in CRP levels across studies supports the fact that substantial attenuation of inflammation occurs post-bariatric surgery. Additional CVD-related biomarkers include B-type natriuretic peptides (BNP) [55], oxidized low-density lipoprotein (oxLDL) [62], total apolipoprotein B (ApoB) [49, 50], fibroblast growth factors (FGF) 19, 21 [53] and 23 [63], insulin-like growth factor-binding protein-2 (IGFBP-2) [64], fatty acid binding protein 4 (FABP4) [65], and soluble suppression of tumorigenicity-2 (ST2) [66]. These molecules hold promise as potential biomarkers for predicting CVD risk in patients with T2D and for the monitoring of patient responses to bariatric surgery.Fig 3 demonstrates the Forest plots of mean differences and confidence intervals for most redundant inflammation and CVD biomarkers (Total cholesterol, Triglycerides, LDL, HDL, hsCRP, C peptide, ApoB, and IL-6) across the studies that met the inclusion criteria for the meta-analysis. Total cholesterol levels were reported in 22 studies. Results showed that bariatric surgery significantly decreased total cholesterol levels (mean difference (MD) 0.44, 95% CI 0.06 to 0.82, p = 0.02, I2 = 94%) (Fig. 3a). Triglycerides and LDL cholesterol were reported in 24 studies. Their levels also decreased post-surgery (MD 1.00, 95% CI 0.77 to 1.24, p < 0.00001, I2 = 88% for triglycerides; MD 0.27, 95% CI 0.02 to 0.52, p = 0.03, I2 = 91% for LDL) (Fig. 3b and 3c). Conversely, HDL cholesterol increased (MD −0.22, 95% CI −0.30 to −0.15, p < 0.00001, I2 = 84%) (Fig. 3d), and hsCRP levels decreased (MD 0.64, 95% CI 0.44 to 0.84, p < 0.00001, I2 = 93%) (Fig. 3e) post-surgery. C peptide levels also significantly decreased (MD 1.29, 95% CI 0.96 to 1.61, p < 0.00001, I2 = 62%) (Fig. 3f), while ApoB levels increased but did not reach significance (MD −0.05, 95% CI −0.15 to 0.05, p = 0.29, I2 = 48%) (Fig. 3g). IL-6 was reported in three studies, with a I2 value of 33%. Its levels were decreased post-surgery (MD 1.84, 95% CI 0.85 to 2.84, p = 0.0003, I2 = 33%) (Fig. 3h). These findings underscore the favorable impact of bariatric surgery on lipid profiles and inflammatory markers crucial for cardiovascular health in patients with T2D. However, further research is needed to elucidate the mechanisms through which bariatric surgery mitigates inflammation and improves cardiometabolic outcomes in these individuals. Forest plots of mean differences and confidence intervals for inflammation biomarkers pre- and post-surgery. Analysis was performed at several postoperative time points using Random effects model of mean differences to assess the continuous data and statistical heterogeneity Microbiome Three studies have investigated the impact of bariatric surgery on gut microbiota in patients with T2D [32, 67, 68]. Overall, these studies have shown that bariatric surgery leads to alterations in the gut microbiome and improvements in dysbiosis compared to baseline. Specifically, changes in the relative abundance of various bacterial phyla, orders, families, genera, and species have been observed post-surgery. At the phylum level, all three studies reported an increased relative abundance of Proteobacteria and a decrease in Firmicutes [32, 67, 68]. However, at the genus level, all studies reported contradictory results after bariatric surgery. Chakaroun R. et al. [67] observed reductions in Sphingomonas, Pseudomonas, Bacillaceae, and Burkholderiaceae, along with an increase in CM1G08, Acinetobacter, and Acidibacter, which are negatively associated with metabolic disease and inflammation markers, whereas positively associated with HDL cholesterol and blood bacterial quantity. In contrast, Lau E. et al. [32], reported an increase in Klebsiella, Gammaproteobacteria, Enterobacter, unclassified Gammaproteobacteria, and unclassified Veillonellaceae, while a decrease in Ruminococcus, unclassifed Lachnospiraceae family and Faecalibacterium. Assal K. et al. [68], found an increase in Veillonella and Streptococcus and a reduction in Flavonifractor, Blautia, Butyricicoccus, and Oribacterium. These discrepancies likely stem from individual variations in how the gut microbiome responds to bariatric surgery. Therefore, further research is essential to comprehensively understand the mechanisms underlying these microbiota changes and their implications for improving metabolic outcomes in patients with T2D undergoing bariatric surgery. Clarifying these mechanisms could establish a foundation for targeted microbiome interventions to enhance the health benefits of surgical weight loss strategies in diabetes management. Discussion Postoperative outcomes of bariatric surgery show considerable heterogeneity among individuals, with patients achieving different responses. This systematic review, for the first time, identifies 49 studies from the existing literature that integrate omics datasets to provide a comprehensive analysis of the molecular and genomic determinants influencing individualized therapeutic responses to bariatric surgical interventions. Genetic and epigenetic factors may influence how patients with T2D respond to bariatric surgery. Several SNPs, such as those in the MC4R, FTO, UCP2, ACSL5, 5-HT2C, UCP3, and MRAP2 genes, have shown associations with obesity, energy balance, and surgical outcomes [36, 37, 69–74]. However, the DIO2 p.Thr92Ala polymorphism remain inconclusive [35]. While a global meta-analysis suggests it may be linked to poorer glycemic control and increased insulin resistance in patients with T2D [75]. Similarly, epigenetic signatures can also predict responses to bariatric surgery. Various epigenetic mechanisms can alter gene expression in response to environmental factors. Several studies have found significant DNA methylation changes in adipose tissue and blood following surgery [30, 33, 38]. In addition, the impact of bariatric surgery on specific gene promoter methylation has been reported in several candidate gene studies [76–78], although, none of these studies provided clear data specifically on patients with T2D undergoing bariatric surgery. It is crucial to note that the majority of epigenetic studies have been conducted with small sample sizes, therefore, the results should be interpreted carefully. The lack of sufficient genomics and epigenomics data on patients with T2D to fully explain the benefits of bariatric surgery presents a significant challenge. Overcoming this obstacle will require replication studies, as well as large-scale and longitudinal studies to pave the way for more effective and personalized healthcare solutions. Transcriptomics is another key component of the omics approach to understanding gene expression changes in patients with T2D before and after surgery. Studies have identified altered expression in genes linked to obesity, T2D comorbidities, cardiovascular disease, adipogenesis, glucose regulation, inflammation, and fatty acid oxidation [29, 30, 33, 40, 79]. Bariatric surgery also affects miRNA levels related to pancreatic function and β-cell stress [39]. However, most studies rely on qPCR at single time points and lack RNA sequencing, limiting discovery of novel gene transcripts and dynamic changes over time. Moreover, transcriptomic data alone do not reveal functional effects, highlighting the need for complementary methods like proteomics or metabolomics for deeper biological insight. Metabolomics can deliver a comprehensive view of these changes by spotting alterations in metabolite profiles coupled with improved insulin sensitivity, lipid metabolism, and inflammation; key factors in T2D and CVD. Bariatric surgery significantly alters amino acids, especially BCAAs, and reduces microbiota-derived metabolites like TMAO and SCFAs such as acetate and butyrate. [31, 34, 44, 47]. High TMAO levels are associated with T2D, obesity, and increased CVD risk, while SCFAs have anti-inflammatory properties that may help prevent gastric inflammation [48, 80–83]. Other important metabolites like free fatty acids (FFAs) and lipids also need further study. However, variability in metabolite profiles due to genetics, lifestyle, and gut microbiome differences complicates identifying consistent metabolic patterns across patients. The review meta-analyzed 36 proteomics studies and found that bariatric surgery improves metabolic health, reduces inflammation, and lowers cardiovascular risk in patients with T2D. Post-surgery, lipid profiles (total cholesterol, LDL, HDL, triglycerides) and inflammatory markers (IL-6, TNF-α, hs-CRP) improved, along with cardiac biomarkers like BNP, NT-proBNP, and adiponectin. Meta-analysis confirmed significant benefits, but high heterogeneity (> 50%) was observed, likely due to differences in study design and assay techniques. Most studies used targeted immunoassays (e.g., ELISA), while others applied broader approaches such as the Proximity Extension Assay (PEA) or LC–MS, with LC–MS offering greater potential for discovery-based analysis [54, 58]. Microbiome dysbiosis is increasingly linked to T2D and its complications [84, 85]. Three studies using 16S rRNA sequencing found consistent phylum-level shifts post-bariatric surgery, including increased Proteobacteria and decreased Firmicutes, likely due to dietary and nutrient changes post-surgery [32, 67, 68]. However, findings at the genus level varied across studies, likely due to inherent inter-individual variability in microbial compositions. These variabilities create a challenge to identifying consistent microbial signatures associated with T2D, CVD, or bariatric surgery outcomes. While 16S rRNA provides limited resolution, advanced techniques like shotgun metagenomics offer deeper insight into microbial function and composition [86]. Such tools may uncover therapeutic targets, with microbiome modulation emerging as a promising adjunct to improve metabolic outcomes in T2D management. Across the 49 studies, only six investigated multiple omics layers simultaneously [29–34]. Almby K. et al. [29] covered both transcriptomics and proteomics. Härm A. et al. [31] and Zhao S. et al. [34] covered both metabolomics and proteomics layers. Lau E. et al. [32] explored both gut microbiome and proteomics. Benton M. et al. [30] examined epigenomics and transcriptomics, while Vohl M-C et al. [33] integrated epigenomics, transcriptomics, and proteomics. With the ongoing advancements in systems biology, diabetes research is expected to become more systematic and in-depth. Multi-omics signatures offer a comprehensive view of the complex interplay of biological processes implicated in T2D remission and CVD risk reduction post-surgery. They promise the development of personalized diagnostics, predictive disease risk assessments, and targeted therapies that are not only more effective but also less likely to cause adverse effects in patients with T2D. Despite these promising efforts, the critical appraisal of included studies reveals several methodological limitations. Follow-up durations were highly variable, ranging from a few weeks to several years, which complicates direct comparisons of post-surgical outcomes. Furthermore, inconsistencies in outcome definitions, such as T2D remission or CVD improvement were common ranging from short-term glycemic control to longer-term medication independence, or from surrogate biomarker changes to hard cardiovascular outcomes. This heterogeneity undermines the ability to draw unified conclusions across studies by limiting comparability, constraining meta-analytic power, and complicating translation into clinical practice. A notable limitation is the heavy reliance on surrogate endpoints rather than definitive clinical outcomes, which, while providing mechanistic insights, may not fully predict long-term patient benefit. Future research would benefit greatly from the adoption of harmonized, standardized, and clinically relevant endpoints. This would enhance reproducibility, strengthen evidence synthesis, and ultimately improve the applicability of findings to patient care. A significant imbalance in the application of omics technologies was also observed. Most studies focused heavily on proteomics, with relatively fewer examining genomics, epigenomics, metabolomics, transcriptomics, or microbiome data. This overrepresentation narrows the potential for discovering broader molecular pathways involved in surgical response. While proteomics has yielded valuable insights, expanding research into these underrepresented omics layers offers significant promise. Metabolomics can capture dynamic metabolic adaptations that underlie both glycemic improvements and cardiovascular risk reduction after surgery. While the gut microbiome represents a critical modifiable factor influencing host metabolism, energy balance, and surgical outcomes. Large, longitudinal cohorts that integrate these modalities alongside proteomics and other omics layers will be essential to achieve a more comprehensive systems-level understanding of T2D remission and cardiometabolic health after surgery. Additionally, selection and publication biases may limit generalizability. Beyond the statistical heterogeneity reported, additional variability likely arises from differences in assay platforms, cohort ethnicity, baseline T2D severity, and surgical techniques. Addressing these factors through standardized protocols and harmonized study designs will be essential to improve comparability and enhance the reliability of future findings. It is crucial to be aware that the study of bariatric surgery outcomes in patients with T2D via omics technology poses some challenges. Clinical studies involving patients with T2D undergoing bariatric surgery often have limited sample sizes due to the invasive nature of these procedures, ethical considerations, and resource constraints. Small sample sizes may reduce the statistical power and limit the extension of the findings into a broader context. In addition, ethnic diversity could influence individual responses to bariatric surgery, as genetic and epigenetic factors affecting T2D and CVD may vary among different populations. The clinical implementation of omics technologies also faces logistical and financial barriers. Robust infrastructure is required, including access to high-throughput platforms, bioinformatics expertise, and secure data storage. Multi-omics profiling, particularly in proteomics, metabolomics, and microbiome analysis, can be prohibitively expensive, especially in resource-limited settings. In addition, a lack of standardization in assay methods, data analysis pipelines, and result interpretation can introduce technical variability in biomarker quantification. Which, therefore, hinders the scalability of omics-based approaches across healthcare systems. Future research should explore the interplay between bariatric surgery and omics profiles in larger, ethnically diverse cohorts to better understand how these factors influence outcomes. Longitudinal studies with extended follow-up periods are necessary to capture the temporal dynamics of molecular and clinical changes following surgery. Omics-derived biomarkers must also be validated in independent populations to assess their reproducibility and predictive value for clinical outcomes such as T2D remission and cardiovascular risk reduction. Emerging omics biomarkers have the potential to enhance personalized decision-making in bariatric surgery. In the future, clinicians could use these biomarkers to stratify patients according to their predicted likelihood of T2D remission, cardiovascular risk reduction, or weight-loss response, enabling tailored surgical planning and individualized preoperative counseling. For instance, specific metabolomic or microbial signatures may help identify patients who would benefit most from a particular surgical procedure or who require closer postoperative monitoring. Although routine clinical implementation is not yet established, integrating omics data into multidisciplinary care pathways could support precision medicine approaches, optimize patient outcomes, and guide postoperative follow-up strategies as the evidence base grows. Successful regulatory approval and clinical adoption of OMICS technologies will require demonstrating added value beyond current standards of care. In addition, practical barriers such as high assay costs, infrastructure requirements, and variable regulatory frameworks, particularly in low- and middle-resource settings, pose significant challenges to translation. To overcome these limitations, future efforts should focus on developing cost-effective assays, establishing clinically standardized workflows, and fostering collaborative networks between researchers, clinicians, and regulatory bodies. Such coordinated initiatives will be essential to enable the integration of OMICS into personalized, evidence-based care across diverse healthcare environments. Conclusion This systematic review highlighted the pivotal role of omics disciplines in unraveling the impact of bariatric surgery on managing T2D and mitigating the risk of CVD. By exploring the realms of genomics, epigenomics, transcriptomics, metabolomics, proteomics, and additional omics disciplines, we ascertain profound understanding of the intricate molecular and physiological pathways underpinning a diverse spectrum of pathological conditions. This multifaceted data serves as a foundational pillar for precision medicine, facilitating the formulation of individualized diagnostic tools, predicting disease susceptibilities, and developing tailored therapeutic interventions. Integrating multi-dimensional omics data into clinical practice represents a significant advancement toward personalized medicine, enhancing therapeutic efficacy while reducing the risk of adverse events for patients with T2D. Supplementary Information Supplementary file 1. Acknowledgements The authors acknowledge the support of Salsabeel Junaidi in screening eligible studies for the review. Author contributions BA: Writing – original draft. MM: Writing – review & editing. NM: Writing – review & editing. J.S.B.-G: Writing – review & editing. C.A: Writing – review & editing. H.A: Writing – review & editing. Funding This work was supported by The Abu Dhabi Department of Education and Knowledge (ADEK) (grant number 8434000344). Data availability No datasets were generated or analysed during the current study. Declarations Ethics approval and consent to participate Not applicable. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary file 1. Data Availability Statement No datasets were generated or analysed during the current study.