Skip to main content
Download PDF

Application of fourier transform infrared vibrational spectroscopy in identifying early biochemical changes in lipid profiles of individuals undergoing Roux-en-y gastric bypass

BMC Surgery volume 25, Article number: 33 (2025) Cite this article

Abstract

Background

Fourier transform infrared spectroscopy (FTIR) is an analytical technique increasingly applied in biological analysis. This study investigates the application of FTIR to identify early biochemical changes, particularly in lipid profiles, in individuals undergoing Roux-en-Y gastric bypass (RYGB).

Methods

An observational study involving patients from a university hospital’s Bariatric and Metabolic Surgery Program, with evaluations performed before (T0) and two months after (T1) RYGB. Biochemical parameters, anthropometric data, and body composition were assessed. FTIR spectra were pre-processed and analyzed using Principal Component Analysis and Partial Least Squares Discriminant Analysis. The normality of the data was evaluated using the Kolmogorov-Smirnov test, followed by paired T-tests or Wilcoxon tests as appropriate. Spearman correlation analysis of spectral information with biochemical parameters was also performed. A significance level of p < 0.05 was set for all tests. The university hospital’s Research Ethics Committee approved the study (protocol CAAE 59075722.7.0000.5071).

Results

The study evaluated 29 individuals (86.2% female) with a mean age of 41.2 ± 7.8 years. Significant differences were observed in anthropometric parameters and body composition (p < 0.001). Additionally, early improvements in the lipid profile were noted, with significant decreases in triglycerides, total cholesterol, and LDL cholesterol (p < 0.05) just two months post-surgery. FTIR identified correlations between biochemical parameters and specific spectral regions at T0 and T1. Notably, serum triglycerides showed a significant correlation with the lipid-specific spectral region (1796–1685 cm− 1) at both time points (p < 0.05).

Conclusion

FTIR can effectively monitor biochemical changes in RYGB surgery patients. The spectral range associated with lipid functional groups (1796 –1675 cm⁻¹) showed a significant relationship with serum triglyceride levels before and after RYGB. Additionally, various biochemical parameters exhibited strong correlations with other spectral regions, implying that the serum spectral profile can indicate biochemical variations at different post-surgery stages.

Trial registration

Brazilian Registry of Clinical Trials (Rebec), September 5, 2022, protocol RBR-26chs2g.

Peer Review reports

Background

Bariatric-metabolic surgery, specifically Roux-en-Y gastric bypass (RYGB), continues to be indicated for the treatment of severe obesity [1]. RYGB involves modifications to the digestive system and consistently results in significant weight loss and improved quality of life [2]. However, weight loss after surgery can vary significantly between individuals, making it challenging to predict precisely how much weight someone will lose, especially in the long term [3]. Moreover, dietary restrictions and nutrient malabsorption resulting from RYGB can lead to substantial changes in anthropometric, biochemical, and metabolic parameters, potentially compromising nutritional status [4, 5].

In this context, methodological approaches for nutritional and prognostic assessment of patients undergoing RYGB, particularly in the first months after surgery, are imperative [6]. Early monitoring is crucial to promptly identify and address potential issues, improving long-term outcomes. Fourier transform infrared spectroscopy (FTIR) has been increasingly applied in biological analysis and is a possible tool for analyzing clinical parameters and detecting molecular alterations in health biomarkers [7]. This technique absorbs infrared radiation through covalent bonds in organic and inorganic molecules, generating characteristic molecular vibrations of specific functional groups. The resulting spectrum acts as a biochemical fingerprint of the sample, providing detailed molecular information [8]. The infrared (IR) spectrum is divided into three regions based on wavenumbers: far-infrared (far-IR, < 400 cm⁻¹), mid-infrared (mid-IR, 400–4000 cm⁻¹), and near-infrared (near-IR, 4000–13000 cm⁻¹). The mid-IR region is the most commonly used for sample analysis, which was used in this research. The distinctive pattern produced by mid-IR radiation occurs because the sample absorbs specific frequencies corresponding to its chemical structure. The peaks of the mid-IR spectrum are unique for each sample, making this type of spectroscopy valuable for a wide range of applications [9].

Recent studies demonstrate the effectiveness of FTIR in identifying molecular alterations in blood samples associated with diseases such as metabolic syndrome and cancer [10, 11]. Multivariate data analysis methods, such as Principal Component Analysis (PCA) and Partial Least Squares Discriminant Analysis (PLS-DA), are crucial for extracting meaningful information from FTIR spectra. These techniques help identify relevant patterns and highlight molecular changes [12, 13].

A study using FTIR to analyze serum samples from individuals with and without obesity found differences in absorbance between the spectra of the two groups, particularly in the functional groups of proteins and lipids [14]. Positive correlations have also been observed between FTIR spectra and biochemical parameters, such as aspartate aminotransferase, alanine transaminase, triglycerides, glucose, total cholesterol, high-density lipoprotein, low-density lipoprotein, and insulin.

Continuous patient engagement and effective post-operative monitoring are crucial for the success of these procedures. Monitoring with FTIR spectroscopy facilitates the timely identification of lipid profile changes, providing essential insights into patients’ metabolic states post-surgery. This enhances adherence to post-surgical guidelines and enables prompt interventions when necessary. While most studies present results six months to a year after bariatric surgery [1, 15], communicating initial findings is critical for understanding trends and making timely adjustments to treatment approaches, which can significantly impact long-term outcomes. Hence, this study aims to evaluate the use of FTIR in early monitoring to identify alterations in the biochemical parameters of individuals undergoing RYGB.

Methods

Study design and participants

This prospective study was conducted with participants from the Bariatric and Metabolic Surgery Program of a university hospital in Brazil. From February to June 2020, individuals were invited to participate in the study voluntarily. The convenience sample included patients aged 18 to 60 with a BMI > 40 kg/m² or > 35 kg/m² with comorbidities. Pregnant women and patients with pacemakers or other metallic structures were excluded, following the recommendations of the European Society for Clinical Nutrition and Metabolism (ESPEN) [16]. Evaluations were performed at an average of 24.0 ± 20.5 days before (T0) and 72.0 ± 19.5 days after (T1) RYGB. The study was approved by the Research Ethics Committee of the university hospital, number CAAE 59075722.7.0000.5071, following the Code of Ethics of the World Medical Association (Declaration of Helsinki), to ensure the ethical conduct and protection of participants’ rights. Participants were fully informed of the study objectives and methodology, and informed consent was obtained from all individuals included.

Anthropometry and body composition

Body weight was measured using an anthropometric scale with an accuracy of 0.05 kg, and height was measured using a wall-mounted stadiometer with an approximation of 0.1 cm. BMI was calculated by dividing weight (kg) by height (m2).

Fat-free mass (FFM) and fat mass (FM) (kg) were obtained by bioelectrical impedance analysis using the Biodynamics 450® analyzer (Biodynamics Co., Shoreline, WA, USA) at a single frequency of 50 kHz. FFM was calculated using the formula for people with obesity proposed by Segal & colleagues [17] and expressed in kg. FM was calculated using the following formula: FM = total body weight – FFM, also expressed in kg.

Biological material and biochemical parameters

Blood samples were collected after an 8–12 h fast, and the serum was collected immediately after centrifugation and stored in a freezer at -80 °C for later analysis.

The following parameters were analyzed using commercial kits (Wiener Lab, Santa Fé, Argentina): albumin; transthyretin (TTR); alpha-1-acid glycoprotein (AGP); C-reactive protein (CRP); aspartate aminotransferase (AST); alanine aminotransferase (ALT); alkaline phosphatase (ALP); glucose; triglycerides; total cholesterol (TC); high-density lipoprotein (HDL) and low-density lipoprotein (LDL).

Sample preparation and FTIR spectrum acquisition

Samples were thawed at room temperature (22 °C ± 0.2 and 51% ± 2.1 humidity) for 30 min and homogenized using a vortex mixer. Then, 10 µL were pipetted onto aluminum plates and dried at room temperature for at least two hours in triplicate. Spectra were acquired in triplicate using the ALPHA II spectrometer (Bruker, Germany) in the spectral range of 4000 to 400 cm− 1, with an attenuated total reflectance (ATR) accessory with a diamond crystal, resolution of 4 cm− 1 and 32 scans for each background and sample. For each analysis, the diamond sampling window was cleaned with Milli-Q® water and 70% (v/v) ethanol and dried with absorbent paper tissue.

Spectral data processing

The spectra were evaluated using Matlab software (R2023b). The mean of the spectral triplicates was calculated, followed by baseline correction (adaptive iteratively reweighted Penalized Least Squares algorithm - airPLS) [18] and smoothing with a Savitzky-Golay filter (5-point window). In addition, other preprocessing methods were tested to generate models with better group discrimination ability, such as vector normalization, multiplicative scatter correction (MSC), and second derivative (15-point window and second-degree polynomial).

Multivariate analysis

Multivariate analysis was conducted to discriminate the samples. For this purpose, the spectra were truncated and analyzed in the 3800–900 cm− 1 wavenumber range. PCA was used to explore interactions between samples and variables, identifying patterns and information that explain the most significant variation in the data [12].

A supervised PLS-DA approach was used to classify samples from groups T0 and T1. For the development of classification models, the data were randomly divided into training sets (70%) and test sets (30%), keeping T0 and T1 from the same patient in either the training or test set.

The training samples were used in training and cross-validation (Monte Carlo method) to determine the model’s optimal number of latent variables (LV). External test samples, however, were not included in training and were reserved for evaluating the model’s predictive ability. To evaluate the performance of the PLS-DA models, the sensitivity (SENS), specificity (SPEC), error rate (ER) and accuracy (ACC) parameters were used, measuring the ability to identify positive cases correctly, the ability to correctly identify negative cases, the false positive rate, the overall error rate and the overall accuracy of the classifications, respectively.

Furthermore, the PLS-DA model enables the interpretation of relevant variables through an analysis known as the VIP score. A VIP value above 1 is commonly used as a threshold to identify variables with significant contribution [19, 20].

Statistical analysis

Analyses were performed using GraphPad Prism 8 software. Data normality was tested using the Kolmogorov-Smirnov test and expressed as mean ± standard deviation or median (25–75%). For comparative tests, data were analyzed by paired t-test or Wilcoxon test, with a significance level of 5%.

Spearman’s correlation coefficient was calculated between the set of biochemical parameters and the correspondingly selected spectral regions of interest, which are indicated in Table 1.

Table 1 Spectral regions and molecular assignments
Full size table

For each selected region, the area under the curve (AUC) of the mean spectra, pre-processed by second derivative (Savitzky-Golay method, 21-point window), was calculated [25]. The correlations found were classified as weak (0.30–0.50), moderate (0.50–0.70) or strong (0.70-1.00) [26].

Results

The final sample comprised 29 participants (86.2% female, n = 25) who attended both scheduled evaluations before and after the surgery. Participants had a mean age of 41.2 ± 7.8 years and a mean height of 162.0 ± 8.7 cm. The most prevalent comorbidities found in the participants at baseline (T0) were hypertension (70.3%), type 2 diabetes mellitus (42.3%), and dyslipidemia (60.4%).

Table 2 indicates that all anthropometric and body composition parameters decreased significantly between the two time points (p < 0.001). Regarding biochemical parameters, serum albumin, TTR, AGP, CRP, ALP, glucose, triglycerides, TC, and LDL decreased significantly (p < 0.05). At the same time, AST increased significantly (p < 0.05). Notably, only serum TTR was below the reference value after RYGB.

Table 2 Biochemical, anthropometric, and body composition parameters before and after RYGB
Full size table

Principal Component Analysis (PCA) with the first two main components accounted for 43.84% of the variance in the original data, indicating a tendency for separation between the groups. Most T0 samples were located on the negative side of PC1, while T1 samples were predominantly on the positive side of PC1 (Fig. 1a). The loadings of the PCA model for the biochemical parameters (Fig. 1b) indicated that the main variables influencing the T0 samples towards the negative side of PC1 were TC, albumin, CRP, high-density lipoprotein (HDL), glucose, and triglycerides, which decreased after RYGB.

Fig. 1
figure 1

Score plots and loading biochemical parameters before and after RYGB. RYGB: Roux-en-y gastric bypass. T0: 29.3 ± 23.9 days before RYGB; T1: 81.3 ± 32.9 days after RYGB. Score plots PCA model (a). Loadings of PCA model with scaled biochemical parameter data (b). 1: Transthyretin (TTR); 2: Albumin; 3: C-reactive protein (CRP); 4: alpha-1-acid glycoprotein (AGP); 5: aspartate aminotransferase (AST); 6: alanine aminotransferase (ALT); 7: alkaline phosphatase (ALP); 8: total cholesterol (TC); 9: high-density lipoprotein (HDL); 10: low-density lipoprotein (LDL); 11: glucose; 12: triglycerides

Full size image

The spectral profile with the average spectra of the T0 and T1 groups is shown in the Fig. 2. The spectra display characteristic bands of biological matrices attributed to proteins (mainly amide I and II, 1687 –1478 cm− 1), carbohydrates and amino acids (1195 –905 cm− 1), and lipids (3358 –2816 cm− 1, 1796 –1685 cm− 1, 1484 –1420 cm− 1, and 1195 –1133 cm− 1). A PCA model of the spectra, using the scores plot with the first two principal components, explained 72.48% of the total variance of the original data. However, it did not show a separation of the samples from T0 and T1 in the principal component space, which is why this analysis was not presented.

Fig. 2
figure 2

Average spectra from FTIR of serum samples before and after RYGB. RYGB: Roux-en-y gastric bypass. T0: 29.3 ± 23.9 days before RYGB; T1: 81.3 ± 32.9 days after RYGB

Full size image

However, employing Partial Least Squares Discriminant Analysis (PLS-DA) yielded a model with an accuracy of 77.8%. The model demonstrated an error rate of 5.0% for the training set and 22.2% for the test set (Table 3).

Table 3 Performance measures and characteristics of the PLS-DA model for before and after RYGB
Full size table

From the PLS-DA model, it was possible to evaluate the most relevant spectral variables for distinguishing between groups through the VIP score. Figure 3 shows regions with a VIP score > 1, indicating that regions with significant VIP scores are primarily attributed to lipids, proteins, and carbohydrates, demonstrating the contribution of the entire spectral information.

Fig. 3
figure 3

VIP score analysis of spectral variables using the PLS-DA model

Full size image

Table 4 shows the correlation coefficients of the selected regions with the biochemical parameters. At T0, there was a positive correlation of regions 2, 3, and 9 with serum AGP, regions 4, 7, and 10 with serum AST, and regions 2, 3, 6, and 9 with triglycerides concentrations. At T1, there was a significant correlation between region 2 and serum total and LDL cholesterol, and region 3 with serum TTR, ALT, AST, and triglycerides concentrations. Only triglycerides concentrations significantly correlated with region 3 at both time points.

Table 4 Pearson correlation values between biochemical parameters and spectral regions before and after RYGB
Full size table

Discussion

The spectral region characteristic of lipid functional groups, located at 1796 –1675 cm− 1 (region 3), showed a significant correlation with serum triglycerides concentration both before and after RYGB. This region also includes aldehyde groups related to oxidative stress [24]. Although the correlation between spectral regions and oxidative stress parameters was not evaluated in this study, these findings open new perspectives for investigations into oxidative stress in obese patients undergoing surgical treatment, using mid-infrared spectroscopy to explore specific absorption bands.

Other biochemical parameters also showed significant correlations with specific spectral areas, although not in the same region at both time points. The correlations identified in region 2 (3011 –2816 cm⁻¹) with triglycerides, TC, and LDL highlight the relevance of this region for investigating lipid dynamics and are consistent with literature data reporting changes in lipid CH2 antisymmetric bands (2975 –2952 cm⁻¹) in obese individuals enrolled in a bariatric surgery program compared to eutrophic individuals (p = 0.037), associating these bands with structural and metabolic changes in lipids [27]. Previous studies observed a significant correlation between serum triglycerides and the absorbance peak at 1457 cm⁻¹, corresponding to region 6 (1484 –1420 cm⁻¹), as well as changes in the wavelength range of 1181 –1131 cm⁻¹, corresponding to region 9 (1195 –1133 cm⁻¹), between obese individuals and controls (BMI < 25) [14, 27].

FTIR demonstrated its ability to detect early changes in the lipid profile of individuals undergoing RYGB. This can be crucial for re-evaluating and potentially adjusting treatment approaches, as well as guiding early therapeutic decisions in clinical settings. In a study that investigated FTIR spectra of blood serum from individuals with and without obesity, significant differences in lipid composition were observed between the groups, identified by analysis of absorbance and peak positions associated with lipid functional groups, highlighting specific variations in wavenumbers 1750 cm− 1 (lipid C = O bonds) and 2970 –2950 cm− 1 (lipid CH3 bonds) [14]. The authors also observed a significant correlation between serum triglycerides and specific wavenumbers 1750 cm− 1 and 2700–3000 cm− 1, respectively, to lipid C = O and CH3 bonds. These findings are similar to the present study and contribute to a more comprehensive understanding of the use of FTIR for detecting alterations in lipid profiles in obesity.

The significant correlations of spectral information with serum triglycerides concentrations observed before and after RYGB may be partly explained by the ability of FTIR to detect vibrations of functional groups such as C = O and CH and, consequently, changes in the spectral regions associated with these compounds [21, 22, 28]. Lipids, especially triglycerides, represent a significant portion of blood serum content, and alterations in their molecular composition resulting from RYGB include modifications in the quantity and composition of circulating lipids [29, 30]. Alterations in nutrient absorption and lipid metabolism induced by RYGB might be reflected in FTIR spectra.

Concerning the PCA analysis, a clear separation of serum samples from patients before and after RYGB was not observed. Factors such as intra- and inter-individual variability may hinder the achievement of a clear spectral distinction at different time points [31, 32]. In the context of RYGB, this suggests that variations in spectral characteristics exist even within a single individual before and after RYGB. Before surgery, factors such as inadequate diet, sedentary lifestyle, and health status may be reflected in FTIR spectra. After RYGB, despite improving health status, metabolic and hormonal changes and alterations in post-surgical diet may contribute to the complexity of spectra and specific sample characteristics [2, 4, 33, 34]. However, the results from the PLS-DA analysis suggest that considering the different spectral regions and their characteristics, FTIR was effective in differentiating individuals before and after RYGB, identifying true positives (sensitivity), true negatives (specificity), and the overall rate of correct classification (accuracy). In a study evaluating the use of FTIR in the diagnosis of COVID-19, a sensitivity rate of 85% and a specificity rate of 83% were observed, suggesting that FTIR is an essential tool for rapid, low-cost, and non-invasive diagnosis [35].

It is important to note that other significant correlations were also observed, although with variations at different time points. These variations may be attributed to the complex metabolic and physiological changes induced by RYGB, which may affect the serum concentrations of biochemical parameters and, in turn, the spectral characteristics of serum samples, disrupting the relationships between spectral characteristics and evaluated parameters [2, 4, 31, 32].

The significant changes in anthropometric and body composition parameters, as well as the changes in biochemical parameters, are consistent with the expected effects of RYGB [36, 37]. We observed that serum concentrations of CRP, glucose, TC, and LDL, altered before RYGB, were within the normal range after surgery. These findings highlight the efficacy of RYGB in improving metabolic and lipid profiles, contributing to an overall improvement in health status. However, it is essential to note that the serum concentration of TTR, a protein potentially used to identify nutritional risk [38], was below the reference value at T1. In this context, although not observed in this study, new perspectives could be opened for FTIR in evaluating molecular changes characteristic of protein nutritional status.

We acknowledge the importance of biochemical analysis as the gold standard for assessing the health conditions of bariatric patients. However, our data demonstrate that, even in the absence of established reference values, FTIR spectroscopy emerges as a promising approach. It offers valuable insights that can complement traditional methods, serving as an auxiliary tool for analyzing molecular changes in response to RYGB.

This study has certain limitations, as the small sample size may have reduced the statistical power of some analyses, limiting the generalizability of the results and constraining the interpretation of the evaluated sample. Additionally, the chemometric model may have been potentially influenced by patient comorbidities. Furthermore, data collection was affected by the suspension of elective surgeries funded by the Brazilian Unified Health System during the COVID-19 pandemic and the lack of follow-up during the first two years, resulting in deviations from the planned time points. Participant dropouts for personal reasons further contributed to the reduced sample size.

The study followed hospital protocols requiring an 8-to-12-hour fasting before lipid profile measurement. Although variations within this fasting range may introduce minor fluctuations in lipid levels, evidence suggests that these differences are minimal and unlikely to significantly impact the overall lipid profile interpretation [39, 40]. At the same time, evidence highlights the importance of standardizing the fasting timing before sample collection to manage variability [40].

While FTIR provides a rapid and efficient method for lipid analysis, it has intrinsic limitations should be considered, such as potential interference from overlapping absorbance bands, variations in sample types [41], and the need for calibration to ensure accurate results. Nevertheless, the laboratory where the analysis was conducted adhered strictly to the equipment manufacturer’s guidelines for sample preparation, handling, storage conditions, and regular equipment calibration and validation to ensure data accuracy and reliability.

Conclusion

The results indicate that FTIR can effectively monitor biochemical changes in patients undergoing RYGB surgery. The spectral range associated with lipid functional groups (1796 –1675 cm⁻¹) showed a significant relationship with serum triglyceride levels before and after RYGB. Additionally, various biochemical parameters exhibited strong correlations with other spectral regions, implying that the serum spectral profile can indicate biochemical variations at different stages post-surgery.

Data availability

The data supporting the findings of this research are securely maintained under the supervision and responsibility of the Head of the Department of Laboratory of Obesity and Nutrition Studies at the Federal University of Espírito Santo, Brazil. For further details regarding the data and materials used, including the raw data, please contact Dr. Fabiano Kenji Haraguchi (fabiano.haraguchi@ufes.br).

Abbreviations

AGP:

Alpha-1-acid glycoprotein

ALP:

Alkaline phosphatase

ALT:

Alanine aminotransferase

AST:

Aspartate aminotransferase

AUC:

Area under the curve

BMI:

Body mass index

CRP:

C-reactive protein

ESPEN:

European society for clinical nutrition and metabolism

FFM:

Fat-free mass

FM:

Fat mass

FTIR:

Fourier transform infrared

HDL:

High-density lipoprotein

LDL:

Low-density lipoprotein

PCA:

Principal component analysis

PLSA-DA:

Partial least squares discriminant analysis

RYGB:

Roux-en-Y gastric bypass

SENS:

Sensitivity SPEC: specificity ER: error rate ACC: accuracy

T0:

Preoperative period, approximately one month before surgery

T1:

Postoperative period, approximately two months after surgery

TC:

Total cholesterol

TTR:

Transthyretin

References

  1. Biter LU, ‘t Hart JW, Noordman BJ, Smulders JF, Nienhuijs S, Dunkelgrün M, et al. Long-term effect of sleeve gastrectomy vs Roux-en-Y gastric bypass in people living with severe obesity: a phase III multicentre randomised controlled trial (SleeveBypass). Lancet Reg Health Eur. 2024;22:38:100836.

    Article  Google Scholar 

  2. Steenackers N, Vanuytsel T, Augustijns P, Tack J, Mertens A, Lannoo M, et al. Adaptations in gastrointestinal physiology after sleeve gastrectomy and Roux-en-Y gastric bypass. Lancet Gastroenterol Hepatol. 2021;6(3):225–37.

    Article  PubMed  Google Scholar 

  3. Pereira SS, Guimarães M, Monteiro MP. Towards precision medicine in bariatric surgery prescription. Rev Endocr Metab Disord. 2023;24:961–77.

    Article  PubMed  PubMed Central  Google Scholar 

  4. Schmatz R, Bitencourt MR, Patias LD, Beck M, Alvarez GD, Zanini D, et al. Evaluation of the biochemical, inflammatory and oxidative profile of obese patients given clinical treatment and bariatric surgery. Clin Chim Acta. 2017;465:72–9.

    Article  CAS  PubMed  Google Scholar 

  5. Pucci A, Batterham RL. Mechanisms underlying the weight loss effects of RYGB and SG: similar, yet different. J Endocrinol Invest. 2019;42(2):117–28.

    Article  CAS  PubMed  Google Scholar 

  6. Meneses E, Zagales I, Fanfan D, Zagales R, McKenney M, Elkbuli A. Surgical, metabolic, and prognostic outcomes for Roux-en-Y gastric bypass versus sleeve gastrectomy: a systematic review. Surg Obes Relat Dis. 2021;17(12):2097–106.

    Article  PubMed  Google Scholar 

  7. De Bruyne S, Speeckaert MM, Delanghe JR. Applications of mid-infrared spectroscopy in the clinical laboratory setting. Crit Rev Clin Lab Sci. 2018;55(1):1–20.

    Article  PubMed  Google Scholar 

  8. Duygu D, Baykal T, Açikgöz İ, Yildiz K. Fourier transform infrared (FT-IR) spectroscopy for biological studies. Gazi Univ J Sci. 2009;1(3):117–21.

    Google Scholar 

  9. Nandiyanto AB, Oktiani R, Ragadhita R. How to read and interpret FTIR spectroscope of organic material. IJoST. 2019;4(1):97–118.

    Article  Google Scholar 

  10. Lilo T, Morais CLM, Shenton C, Ray A, Gurusinghe N. Revising Fourier-transform infrared (FT-IR) and Raman spectroscopy towards brain cancer detection. Photodiagnosis Photodyn Ther. 2022;38:102785.

    Article  CAS  PubMed  Google Scholar 

  11. Souza NMP, Machado BH, Koche A, Furtado LBFS, Becker D, Corbellini VA, et al. Detection of metabolic syndrome with ATR-FTIR spectroscopy and chemometrics in blood plasma. Spectrochim Acta Mol Biomol Spectrosc. 2023;5:288:122135.

    Article  Google Scholar 

  12. Wold S, Esbensen K, Geladi P. Principal component analysis. Chemom Intell Lab Syst. 19871;2(1–3):37–52.

  13. Kennard RW, Stone LA. Computer aided design of experiments. Technometrics. 1969;1(11):137–48.

    Article  Google Scholar 

  14. Guleken Z, Çeçen S, Ceylan Z, Jakubczyk P, Depciuch J. Application of Fourier transform infrared spectroscopy to detect biochemical changes in blood serum of obese patients. J Biophotonics. 2023;16(6):e202200388.

    Article  CAS  PubMed  Google Scholar 

  15. Coral RV, Bigolin AV, Machry MC, Menguer RK, Pereira-Lima JC, Contin I, et al. Improvement in muscle strength and metabolic parameters despite muscle Mass loss in the initial six months after bariatric surgery. Obes Surg. 2021;31(10):4485–91.

    Article  PubMed  Google Scholar 

  16. Kyle UG, Bosaeus I, De Lorenzo AD, Deurenberg P, Elia M, Gómez JM, et al. Composition of the ESPEN Working Group. Bioelectrical impedance analysis–part I: review of principles and methods. Clin Nutr. 2004;23(5):1226–43.

    Article  PubMed  Google Scholar 

  17. Segal KR, Gutin B, Presta E, Wang J, Van Itallie TB. Estimation of human body composition by electrical impedance methods: a comparative study. J Appl Physiol. 1985;58(5):1565–71.

    Article  CAS  PubMed  Google Scholar 

  18. Zhang ZM, Chen S, Liang YZ. Baseline correction using adaptive iteratively reweighted penalized least squares. Analyst. 2010;135(5):1138–46.

    Article  CAS  PubMed  Google Scholar 

  19. Chong IG, Jun CH. Performance of some variable selection methods when multicollinearity is present. Chemometr Intell Lab Syst. 2005;78(1–2):103–12.

    Article  CAS  Google Scholar 

  20. Ballabio D, Consonni V. Classification tools in chemistry. Part 1: linear models. PLS-DA Anal Methods. 2013;5(16):3790–8.

    Article  CAS  Google Scholar 

  21. Naseer K, Ali S, Qazi JA. ATR-FTIR spectroscopy as the future of diagnostics: a systematic review of the approach using bio-fluids. Appl Spectrosc Rev. 2021;56(2):85–97.

    Article  CAS  Google Scholar 

  22. Li Z, Lv H, Li T, Si G, Wang Q, Lv J, et al. Reagent-free simultaneous determination of glucose and cholesterol in whole blood by FTIR-ATR. Spectrochim Acta Mol Biomol Spectrosc. 2017;5:178:192–7.

    Article  Google Scholar 

  23. Barth A. Infrared spectroscopy of proteins. Biochim Biophys Acta. 2007;1767(9):1073–101.

    Article  CAS  PubMed  Google Scholar 

  24. Guleken Z, Kuruca SE, Ünübol B, Toraman S, Bilici R, Sarıbal D, et al. Biochemical assay and spectroscopic analysis of oxidative/antioxidative parameters in the blood and serum of substance use disorders patients. A methodological comparison study. Spectrochim Acta Mol Biomol Spectrosc. 2020;240:118625.

    Article  CAS  Google Scholar 

  25. Kołodziej M, Chrabąszcz K, Pięta E, Piergies N, Rudnicka-Czerwiec J, Bartosik-Psujek H, et al. Spectral signature of multiple sclerosis. Preliminary studies of blood fraction by ATR FTIR technique. Biochem Biophys Res Commun. 2022;593:40–5.

    Article  PubMed  Google Scholar 

  26. Mukaka MM. Statistics corner: a guide to appropriate use of correlation coefficient in medical research. Malawi Med J. 2012;24(3):69–71.

    CAS  PubMed  PubMed Central  Google Scholar 

  27. Yılmaz A, Bahtiyar N, Doğan Mollaoğlu A, Zengin K, Taskin HE, Karimova A, et al. Mitochondrial common deletion level in adipose tissue is not Associated with obesity but is Associated with a Structural Change in triglycerides as revealed by FTIR Spectroscopy. Med Princ Pract. 2024;33(1):74–82.

    Article  PubMed  Google Scholar 

  28. Jessen TE, Höskuldsson AT, Bjerrum PJ, Verder H, Sørensen L, Bratholm PS, et al. Simultaneous determination of glucose, triglycerides, urea, cholesterol, albumin and total protein in human plasma by Fourier transform infrared spectroscopy: direct clinical biochemistry without reagents. Clin Biochem. 2014;47(13–14):1306–12.

    Article  CAS  PubMed  Google Scholar 

  29. Psychogios N, Hau DD, Peng J, Guo AC, Mandal R, Bouatra S, et al. The human serum metabolome. PLoS ONE. 2011;16(2):e16957.

    Article  Google Scholar 

  30. Heffron SP, Parikh A, Volodarskiy A, Ren-Fielding C, Schwartzbard A, Nicholson J, et al. Changes in lipid Profile of obese patients following contemporary bariatric surgery: a Meta-analysis. Am J Med. 2016;129(9):952–9.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  31. Cheung KT, Trevisan J, Kelly JG, Ashton KM, Stringfellow HF, Taylor SE, et al. Fourier-transform infrared spectroscopy discriminates a spectral signature of endometriosis independent of inter-individual variation. Analyst. 2011;21(10):2047–55.

    Article  Google Scholar 

  32. Margaritelis NV, Nastos GG, Vasileiadou O, Chatzinikolaou PN, Theodorou AA, Paschalis V, et al. Inter-individual variability in redox and performance responses after antioxidant supplementation: a randomized double blind crossover study. Acta Physiol (Oxf). 2023;238(4):e14017.

    Article  CAS  PubMed  Google Scholar 

  33. Reid RE, Oparina E, Plourde H, Andersen RE. Energy Intake and Food habits between Weight Maintainers and regainers, five years after Roux-en-Y gastric bypass. Can J Diet Pract Res. 2016;77(4):195–8.

    Article  PubMed  Google Scholar 

  34. Hosseini-Esfahani F, Kazemi-Aliakbar M, Koochakpoor G, Barzin M, Khalaj A, Valizadeh M, et al. Diet quality and anthropometric indices of patients undergone bariatric surgery: the prospective Tehran obesity treatment study. BMC Surg. 2023;23(1):125.

    Article  PubMed  PubMed Central  Google Scholar 

  35. Nascimento MHC, Marcarini WD, Folli GS, da Silva Filho WG, Barbosa LL, Paulo EH, et al. Noninvasive Diagnostic for COVID-19 from Saliva Biofluid via FTIR Spectroscopy and Multivariate Analysis. Anal Chem. 2022;8(5):2425–33.

    Article  Google Scholar 

  36. Braga GB, Bortoli AM, Brito BB, Salaroli LB, Lopes AB, Haraguchi FK et al. Roux-en-Y gastric bypass reduces body parameters but does not alter diet quality during six months follow-up. J Hum Growth Dev. 2023; 33(2):164–172.

  37. Kermansaravi M, Valizadeh R, Shahsavan M, Maleknia SA, Eghbali F, Pazouki A, et al. Metabolic and bariatric surgery in patients with class I obesity; a two-year follow-up. BMC Surg. 2024;24(1):6.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  38. Bortoli AM, Braga GB, Brito BB, Moraes RAG, Miguel GPS, Pedrosa RG, et al. Decrease phase angle one year after Roux-en-Y bypass and sleeve gastrectomy is related to risks to protein nutritional status. Clin Nutr ESPEN. 2022;52:138–43.

    Article  CAS  PubMed  Google Scholar 

  39. Warnick GR, Nakajima K. Fasting versus nonfasting triglycerides: implications for laboratory measurements. Clin Chem. 2008;54(1):14–6.

    Article  CAS  PubMed  Google Scholar 

  40. Nordestgaard BG, Langsted A, Mora S, Kolovou G, Baum H, Bruckert E et al. Fasting is not routinely required for determination of a lipid profile: clinical and laboratory implications including flagging at desirable concentration cut-points—a joint consensus statement from the European Atherosclerosis Society and European Federation of Clinical Chemistry and Laboratory Medicine. Eur Heart J. 2016; 1;37(25):1944-58.

  41. Petibois C, Gouspillou G, Wehbe K, Delage JP, Déléris G. Analysis of type I and IV collagens by FT-IR spectroscopy and imaging for a molecular investigation of skeletal muscle connective tissue. Anal Bioanal Chem. 2006;386(7–8):1961–6.

    Article  CAS  PubMed  Google Scholar 

Download references

Acknowledgements

We thank the National Council for Scientific and Technological Development (CNPQ), Coordination for the Improvement of Higher Education Personnel (CAPES), Espírito Santo Research and Innovation Support Foundation (FAPES), Management of Teaching and Research of the Cassiano Antônio Moraes University Hospital at the Federal University of Espírito Santo.

Funding

This research was supported by Espírito Santo Research and Innovation Support Foundation, (FAPES), Grant #2022–41989, and Coordination for the Improvement of Higher Education Personnel (CAPES), Grant # 88887.691108/2022-00.

Author information

Authors and Affiliations

  1. Postgraduate Program in Nutrition and Health, Health Sciences Center, Federal University of Espírito Santo, Av. Mal. Campos, 1468 - Maruípe, Vitória, 29047-100, ES, Brazil

    Amanda Motta de Bortoli, Blanca Elena Guerrero Daboin, Beatriz Bobbio de Brito, Andressa Bolsoni Lopes & Fabiano Kenji Haraguchi

  2. Postgraduate Program in Chemistry, Center of Exact Sciences, Federal University of Espírito Santo, Av. Fernando Ferrari, 514 - Goiabeiras, Vitória, 29075-910, ES, Brazil

    Márcia Helena Cassago Nascimento & Paulo Roberto Filgueiras

  3. Department of Integrated Health Education, Health Sciences Center, Federal University of Espírito Santo, Av. Maruípe, 1386 - Maruípe, Vitória, 29047-185, ES, Brazil

    Luiza Recla Pessotti & Fabiano Kenji Haraguchi

  4. Postgraduate Program in Physiological Sciences, Health Sciences Center, Federal University of Espírito Santo, Av. Mal. Campos, 1468 - Maruípe, Vitória, 29047-100, ES, Brazil

    Valerio Garrone Barauna

Authors
  1. Amanda Motta de Bortoli
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Márcia Helena Cassago Nascimento
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Blanca Elena Guerrero Daboin
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Beatriz Bobbio de Brito
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Luiza Recla Pessotti
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Paulo Roberto Filgueiras
    View author publications

    You can also search for this author inPubMed Google Scholar

  7. Andressa Bolsoni Lopes
    View author publications

    You can also search for this author inPubMed Google Scholar

  8. Valerio Garrone Barauna
    View author publications

    You can also search for this author inPubMed Google Scholar

  9. Fabiano Kenji Haraguchi
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

AM de B wrote the main manuscript text. AM de B, MHCN, BEGD, BB de B, and LRP data curation, formal analysis, and investigation. PRF, ABL, VGB, and FKH conceptualization, data curation, methodology, and supervision. VGB and FKH designed the study, funding acquisition, project administration, and resources. All authors reviewed the manuscript.

Corresponding author

Correspondence to Fabiano Kenji Haraguchi.

Ethics declarations

Ethics approval and consent to participate

The study was approved by the Research Ethics Committee of the university hospital, number CAAE 59075722.7.0000.5071, following the Code of Ethics of the World Medical Association (Declaration of Helsinki). Trial Registration: Brazilian Registry of Clinical Trials (Rebec) on September 5, 2022, under the number RBR-26chs2g. Participants were fully informed of the study objectives and methodology, and informed consent was obtained from all individuals included.

Consent for publication

Participants were fully informed of the study objectives and methodology, and informed consent was obtained from all individuals included.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Motta de Bortoli, A., Cassago Nascimento, M., Guerrero Daboin, B. et al. Application of fourier transform infrared vibrational spectroscopy in identifying early biochemical changes in lipid profiles of individuals undergoing Roux-en-y gastric bypass. BMC Surg 25, 33 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12893-024-02707-6

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12893-024-02707-6

Keywords

BMC Surgery

ISSN: 1471-2482

Contact us