Serum visfatin characteristic peak screening coronary heart disease study based on Raman spectroscopy

Abstract

This study used Raman spectroscopy to explore differences between serum visfatin and its subcomponents in healthy people and coronary heart disease (CHD) patients. To explore whether serum visfatin can be used as screening biomarker for CDH. In total, 60 CHD patients and 57 healthy people were enrolled from Xinjiang Production and Construction Corps Seventh Division Hospital. Serum samples were collected and detected by Raman spectroscopy, and the characteristic peaks of serum visfatin were analyzed. The spectral peaks analysis showed, the peaks at 850.95, 1265.92, 1336.99, 1446.48, 1603.83, and 1657.69 cm⁻¹ were stronger in the CHD patient sample, and the spectral peaks at 1000.52, 1154.51, and 1515.53 cm⁻¹ were stronger in the healthy sample. Clinical sample verification revealed that the 1265.92 and 1446.48 cm⁻¹ peaks had certain distinguishing effects on CHD patients and healthy people.

1. Introduction

Raman spectroscopy is a vibration spectroscopy technique with the advantages of no damage to the sample and simple operation. Raman spectroscopy is relatively insensitive to water and can be used to quickly analyze samples containing water.^[1] Raman spectroscopy has been widely used in the biomedical field.^[1–4] Serum Raman spectroscopy combined with multiple classification algorithms was used to implement an auxiliary diagnosis method for breast cancer, which will aid in the early diagnosis of breast cancer patients.^[1] Raman scattering spectra of the serum of tumor-bearing mice at different stages were obtained, a recognition model of surface-enhanced Raman scattering spectra was constructed using the principal component analysis-representation coefficient-based k-nearest centroid neighbor algorithm, and based on the principal component analysis features was used to successfully differentiate the surface-enhanced Raman scattering spectra of the serum of tumor-bearing mice at different stages with a classification accuracy of 100%.^[3] The characteristic Raman peak of potassium-sodium cobaltinitrite was used to distinguish it from other kinds of matter, and its intensity was used to monitor the amount of potassium ions in human serum.^[4] The Raman spectra of several cells from gastrointestinal cancer patients were measured, and the Raman spectra of stomach cancer cells were similar to those of normal cells, but the Raman intensity of cancer cells was much lower than that of normal cells, and even some lines disappeared.^[2] The correlation analysis revealed that for all the serum components, the correlation coefficients between the Raman spectra and all the enzymatic test results were significantly >94%. Discrimination results of healthy subjects and gastric cancer patients show that 87.5 ± 2.5% of healthy and gastric cancer patients are diagnosed properly.^[5]

Visfatin is a newly identified proinflammatory adipokine expressed predominantly in visceral fat, and previous studies have suggested that visfatin plays a role in low-grade inflammation and the regulation of lipid metabolism.^[6] Low vaspin concentrations seemed to correlate with coronary artery disease (CAD) severity.^[7] The results suggested that the visfatin-1535C > T polymorphism might be associated with a reduced risk of CAD in a Chinese population.^[6] The serum visfatin level was positively correlated with CAD severity in patients with high synergy between percutaneous coronary intervention with taxus and cardiac surgery scores.^[8] These findings suggest that increased peripheral blood visfatin concentrations may be a risk marker for CAD.^[9] Visfatin levels were significantly greater in patients with SCF than in controls.^[10] A study of 208 elderly patients with coronary heart disease (CHD) revealed that the levels of serum Hcy, vaspin, and visfatin vary according to the type of CHD and are correlated with the degree of coronary artery stenosis. As such, these serum levels can be used as sensitive indicators for early detection and disease evaluation.^[11]

In the CHD population, pregnant women are also a special group due to their unique characteristics. The occurrence of major coronary events is increased among women with preeclampsia.^[12] A study of 517,504 individuals showed that a history of miscarriage or recurrent miscarriage is associated with a greater risk of subsequent CHD.^[13] Between January 1998 and October 2018, at 16 UK specialized cardiac obstetric clinics, CAD delivery showed that women with established CAD had relatively low rates of adverse cardiac events during pregnancy. The rates of adverse obstetric and neonatal events are greater, highlighting the importance of multidisciplinary care.^[14] To improve our understanding of the risk factors related to CHD, it is very important to establish a perfect diagnostic method. In this study, the serum samples of healthy individuals and patients with CHD were analyzed by Raman spectroscopy to provide evidence for clinical diagnosis.

2. Methods

2.1. Samples

Patients with CHD and healthy people were enrolled from the Xinjiang Production and Construction Corps Seventh Division Hospital from March 2020 to July 2023. This study was approved by the Xinjiang Production and Construction Corps Seventh Division Hospital Ethics Committee (No: IRB2019-BG-001). And all patients and healthy people signed informed consent forms.

There were 28 male patients and 32 female patients, for a total of 60 CHD patients with a mean age of 67.7 years. There were 33 healthy males and 24 healthy females, for a total of 57 healthy people with a mean age of 36.7 years. Peripheral blood specimens (4 mL each) were extracted from subjects and injected into procoagulant collection tubes to extract serum. The serum was stored immediately at −80 °C until analysis. All the serum specimens were stored in a −80 °C freezer and transported to the Laboratory of the Third Military Medical University of Chongqing for experiments, after which the test data were obtained.

Coronary heart disease patients inclusion criteria: CHD patients that were diagnosed as CHD or acute coronary syndrome in the department of cardiology were enrolled. Willing to participate in this project and signed the informed consent form. Healthy people inclusion criteria: Healthy physical examination population and had no cardiovascular disease. Willing to participate in this project and signed the informed consent form. CHD patients exclusion criteria: Patients who were not CHD patients or did not want to participate in this study.

Healthy people exclusion criteria: Healthy people who did not want to participate in this study.

2.2. Raman spectroscopy

Sample detection was carried out on a HORIBA XploRA PLUS, and the data were analyzed with the LabSpec 6 spectrum package. The test conditions were as follows: laser wavelength, 532 nm; gate, 1800 g/mm; objective, 100×; acquisition time, 30 seconds; acquisition time, once; laser power, 100%; spectral range, 400 to 1800 cm⁻¹. Each sample was tested 3 times, and all spectra were normalized for analysis.

2.3. Statistical analysis

GraphPad Prism 10 statistical software (GraphPad Software, Inc., San Diego) was used to analyze the age data and the receiver operating characteristic curve.

Raman spectra were analyzed by LabSpec 6 and Simca-14.1.

3. Results

3.1. Raman spectroscopy of serum visfatin normalization processing and characteristic peaks

In this study, a total of 60 CHD patients and 57 healthy people were enrolled, and serum specimens from CHD patients and healthy people were collected and subjected to HORIBA XploRA PLUS testing. The test conditions were as follows: laser wavelength, 532 nm; gate, 1800 g/mm; objective, 100×; acquisition time, 30 seconds; acquisition time, once; laser power, 100%; spectral range, 400 to 1800 cm⁻¹. Figure 1A shows a typical original Raman spectrum of visfatin, which was obtained from healthy human serum. From the spectrum, there was a wide fluctuation range, and it was not easy to compare the differences. To solve this problem, we normalized the results by the external standard method, and the specific normalized spectrum is shown in Figure 1B. After normalization, the peak graph became steeper.

Figure 1.

Raman spectroscopy for serum visfatin detection. (A) The original Raman spectra of serum visfatin and healthy samples. (B) Raman spectrum of the same healthy serum sample after normalization. (C) Comparison of the Raman spectra of serum visfatin from CHD patient samples and healthy control samples (red indicates CHD samples, blue indicates healthy samples). (Raman displacement unit: cm⁻¹). CHD = coronary heart disease.

Then, the serum samples of CHD patients and healthy people were analyzed by Raman spectroscopy, and the Raman spectra of visfatin were subjected to normalization. By comparing the spectra of CHD patients and healthy people, we found that the characteristic peaks of serum visfatin were most similar. However, there were several differences: the spectral peaks at 850.95, 1265.92, 1336.99, 1446.48, 1603.83, and 1657.69 cm⁻¹ were more intense in the CHD patient sample, and the spectral peaks at 1000.52, 1154.51, and 1515.53 cm⁻¹ were more intense in the healthy sample (Fig. 1C).

3.2. Principal component analysis of serum visfatin

To clarify the clinical significance of these 9 differential peaks, we compared the detection data of 60 CHD patients and 57 healthy people to obtain more detailed analysis results. The results showed that only the area under the curve (AUC) values at 1265.92 cm⁻¹ and 1446.48 cm⁻¹ were 0.6097 and 0.6142, respectively (Fig. 2B and D). The total Raman intensity data are shown in Figure 3A and B, and these 2 peaks had certain distinguishing effects on CHD patients and healthy people (Fig. 3C). Other peaks, such as those at 850.95, 1336.99, 1603.83, 1657.69, 1000.52, 1154.51, and 1515.53 cm⁻¹, had the AUC values below 0.6 and had poor distinguishing effects (Figs. 2 and 3C).

Figure 2.

ROC analysis of CHD patients and healthy people. (A) ROC analysis of the 850.95 peaks in CHD patients and healthy people. (B) ROC analysis of 1265.92 peaks in CHD patients and healthy people. (C) ROC analysis of 1336.99 peaks in CHD patients and healthy people. (D) ROC analysis of 1446.48 peaks in CHD patients and healthy people. (E) ROC analysis of 1603.83 peaks in CHD patients and healthy people. (F) ROC analysis of 1657.69 peaks in CHD patients and healthy people. (G) ROC analysis of 1000.52 peaks in CHD patients and healthy people. (H) ROC analysis of 1154.51 peaks in CHD patients and healthy controls. (I) ROC analysis of 1515.33 peaks in CHD patients and healthy people. CHD samples (n = 60), healthy samples (n = 57). CHD = coronary heart disease, ROC = receiver operating characteristic.

Figure 3.

Raman intensity analysis. (A) Raman intensity analysis of 1265.92 cm⁻¹ for CHD patients and healthy people. (B) Raman intensity analysis of 1446.48 cm⁻¹ for CHD patients and healthy people. (C) Raman intensity analysis of 60 CHD patients and 57 healthy controls. * P < .05 indicates statistical significance. CHD = coronary heart disease.

4. Discussion

Raman spectroscopy is a vibration spectroscopy technique with the advantages of avoiding damage to the sample and simple operation.^[1] Serum Raman spectroscopy combined with multiple classification algorithms was used to implement an auxiliary diagnosis method for breast cancer,^[1] and stomach cancer cells were measured.^[2] Therefore, Raman spectroscopy is widely used in the biomedical field.^[1–4] In this study, we detected serum visfatin, which is a newly identified proinflammatory adipokine.^[6] The serum visfatin level was positively correlated with CAD severity in patients with high SYNTAX scores.^[8] Visfatin levels were significantly greater in patients with SCF than in controls.^[10]

The detection of CHD by Raman technology has been a newly developed technology in recent years, and more exploration is needed to identify more suitable detection sample types and biomarkers. Bingyan Li et al proposed diagnosing CHD with human urine based on surface-enhanced Raman spectroscopy (SERS).^[15] Huinan Yang et al also proposed a CHD diagnosis method using human urine samples based on SERS.^[16]

Our study also provides a reference for exploration in this field. Serum visfatin was detected by Raman spectroscopy in the CHD group and healthy group. After examining CHD patient and healthy human samples, we found that the peaks at 50.95 1265.92, 1336.99, 1446.48, 1603.83, and 1657.69 cm⁻¹ were stronger in the CHD patient sample, and the spectral peaks at 1000.52, 1154.51, and 1515.53 cm⁻¹ were stronger in the healthy sample (Fig. 1C). To evaluate this phenomenon more accurately, more samples were used for verification. The results showed that only the 1265.92 and 1446.48 cm⁻¹ peaks had a certain screening effect, and the AUC values were 0.6097 and 0.6142, respectively (Fig. 2B and D). Raman intensity analysis also revealed that these 2 peaks were significantly different (P < .05) (Fig. 3C).

5. Conclusions

Serum visfatin levels were associated with CHD incidence and had a certain effect on CHD screening.

Author contributions

Conceptualization: Lujia Yang, Chaohui Lv, Xiayan Liu.

Data curation: Lujia Yang, Chaohui Lv, Zhongwei Yang, Feng Shi.

Formal analysis: Lujia Yang, Chaohui Lv, Zhongwei Yang, Xiayan Liu.

Funding acquisition: Zhongwei Yang, Feng Shi.

Methodology: Lujia Yang, Feng Shi.

Resources: Feng Shi, Xiayan Liu.

Software: Zhongwei Yang.

Writing – original draft: Lujia Yang, Chaohui Lv, Zhongwei Yang, Xiayan Liu.

Writing – review & editing: Lujia Yang, Chaohui Lv, Xiayan Liu.