Galectin-3 is a protein from the group of β-galactoside binding lectins with pro-inflammatory properties, presented as a novel biomarker of pathological conditions, such as various types of cancer, especially thyroid cancer , acute and chronic heart failure [26, 27], chronic pancreatitis , prediabetes state and apparent diabetes , depression  and many others. It has multiple functions including cells adhesion, inflammation, and extracellular matrix formation, differentiation, proliferation, embryogenesis, and host–pathogen interactions [29–31]. As a lectin, it binds carbohydrates but it is characterized by a wide range of ligands and, in addition to simple sugars and disaccharides such as galactose, lactose or N-acetyllactosamine, it also binds large carbohydrate molecules, such as cell surface N-glycans and advanced glycation products that are intensely generated during hyperglycemia [17, 18]. Moreover, the protein is found to cause cellular and systemic insulin resistance . Therefore, we expected that galectin-3 would also reflect the various pathological states typical for the geriatric population.
The correlation of galectin 3 with BMI values has been demonstrated repeatedly in different study groups [30–32]. Our study confirmed this result. Similar correlations with glucose and triglycerides have been previously described, and these relationships are especially often observed in women with polycystic ovary syndrome and pregnant women with preeclampsia [32, 33]. These correlations were also described based on studies on cardiac patients  and with hypothyroidism . We describe these relationships probably for the first time for geriatric persons.
According to our observations, galectin-3 inversely correlates with GFR, indicating a relationship with kidney damage. Many studies have already concluded that galectin-3 strongly affects this organ's fibrosis and it has even been suggested that its measurement will allow to predict renal failure and patients’ survival [35, 36].
Moreover, galectin 3 affects fibrosis of many other organs, including the heart after myocardial infarction (MI), hence the need to lower the level of this glycoprotein in order to avoid heart failure after MI . Our studies show that metformin is a drug that is effective in reducing plasma GAL3 levels. Our results are consistent with the results of Asensio-Lopez et al . The practical conclusion is that the administration of metformin in geriatric patients is beneficial not only in regulating carbohydrate metabolism but also in reducing the likelihood of heart failure threatening elderly persons.
In our study, galectin 3 is significantly higher in elderly patients with diabetes type 2. It would seem that galectin 3 is a very interesting diagnostic tool, and many authors present it as a novel biomarker, np. Alves, Hrynchyshyn, Kanukurti, King [26, 27, 29, 32]. However, doubts as to its diagnostic value are raised by the multitude of pathological conditions associated with an elevated level of this protein. When determining its elevated level, the geriatricians cannot state whether the cause is diabetes, heart failure, pancreatitis, cancer or depression. It seems that this parameter is not specific, thus is of little diagnostic value.
The second of the analyzed receptors for AGE, i.e., receptor for advanced glycation end products (RAGE), is the protein responsible for inflammation, apoptosis, reactive oxygen species (ROS) signaling, proliferation and autophagy . sRAGE is a soluble form of protein that circulates in blood plasma competing for ligand, thus having RAGE-reversing, anti-inflammatory properties [18, 38]. We expected a correlation between sRAGE and a fluorescent AGE, pentosidine, glucose or HbA1c. These expectations have not been met but it turned out that sRAGE correlates with triglycerides and SR-BI in our study population. Both of these parameters are related to lipid metabolism, TAG obviously, SR-BI due to the fact that, in addition to AGE, it also binds HDL, mediating selective uptake of cholesterol ether and HDL-dependent cholesterol efflux . So far, no one has associated SR-BI and sRAGE. Only Marshe et al, who found that sRAGE is related to another scavenger receptor CD36 and blocks CD36 -mediated uptake of LDL modified by hypochlorous acid and in this way reduces foam cell formation . Thus, it appears that both sRAGE and SR-BI have an inhibitory effect on the development of atherosclerosis and it is beneficial for the patient if the levels of these proteins, due to their mutual correlation, are high.
In our study group, the negative correlation between SR-BI and HDL was also observed. This is an expected result, as it has been proven that elevating SR-BI expression causes increase in HDL clearance, reduction in plasma HDL levels, and increase in biliary cholesterol levels .
The analysis revealed also that SR-BI correlates with one more parameter - Lectin-type oxidized LDL receptor 1 (LOX-1). In the latest publications (2020 and 2021), LOX-1 is presented as a promising target for early diagnosis and cardiovascular risk prediction [39–41]. Such enthusiastic announcements made us expect to find higher levels of LOX-1 in people with stroke, after a myocardial infarction or with obesity, but no such association was found in our geriatric population. Perhaps due to a small enough study group or too high degree of atherosclerosis in the entire study group. The correlation found between LOX-1 and SR-BI is, yet, a novelty in the subject literature and it is worth continuing research on the cause-and-effect relationship between these two receptors, since they correlate with each other.
Through modeling, we found that for SR-BI, HDL is a good predictor. West reported this relationship for subjects with hyperalphalipoproteinemia . This is the first time that this dependence is described for the geriatric population. However, this finding is not clinically valuabl, because it is still more convenient to measure HDL levels in patients’ blood serum .
Numerous data from in vitro experiments indicate that metformin inhibits expression of LOX-1. This is evidenced, for example, by the data of Hung et al. who studied this dependence in human umbilical vein endothelial cells (HUVECs)  or Shiu et al. who examined the relationship on human aortic endothelial cells incubated with AGE-BSA , and Ouslimani et al. who investigated bovine aortic endothelial cells . The above-mentioned studies showed that metformin protects against oxLDL-induced endothelial apoptosis, oxLDL-induced intracellular calcium rise and mitochondrial dysfunction  as well as reduces, in dose-dependent manner, expression of LOX-1 both in stimulated (by either glucose or AGE) and in unstimulated cells . No study of this relationship has been found in patients’ trials, thus we are probably the first to report it. Our research shows, however, that treatment with metformin alone does not cause a large difference in the concentration of LOX-1 in the serum, similarly with insulin, only the combination of metformin with insulin seems to cause that patients demonstrate the significantly reduced LOX-1 serum level. This is a very important finding that can help geriatricians make therapeutic decisions.
Free amine groups are potentially available to react with a reducing sugar, therefore, this is an important parameter for us regarding the potential for glycation. In addition, Valencia et al. found that free amine content measured using OPA correlates strongly with RAGE binding affinity to AGE . Fracesso expected differences between the amount of free amine content in the serum and tissues of rats after myocardial infarction compared to controls but did not find them . It seems that this is the first time when this is observed in geriatric patients. Higher levels of serum free amino groups were found in the study patients with ischemic heart disease and after stroke. These are diseases associated with atherosclerosis. We expect that if there were more people in the test group with history of myocardial infarction, we would also see a difference with the control group. Therefore, it can be concluded that patients with cardiovascular diseases have an increased level of free amine content in the serum, and thus a greater potential for glycation and probably greater binding affinity RAGE to AGE. To increase the credibility of these statements, the study should be repeated on a larger geriatric population.
In addition to free amine content, we also observe higher AGE and pentosidine values in patientswith ischemic heart disease. This was an expected result, as it was previously reported that circulating fluorescent AGE and pentosidine levels have been associated with cardiovascular disease [47–51]. The same applies to the relationship of pentosidine with arterial hypertension. These dependencies are so strong that IHD and GFR can be considered as a predictor of fluorescent AGE content, and IHD and hypertension as a predictor of high pentosidine levels. These correlations have been described earlier [50, 52, 53]. This is a logical consequence of the fact that glycation causes the formation of cross-links between collagen fibers and other proteins responsible for biochemical properties in the arteries, making them more rigid, unable to dilate properly, and manifesting in hypertension .
It is stated that the kidney normally clears circulating AGE but these products accumulate both in diabetic and not diabetic nephropathy . Both fluorescent AGEs and pentosidines correlate positively with creatinine and negatively with GFR in our study. A correlation similar to the one we found was reported by Dozio et al based on patients with chronic kidney disease , Yoshida based on patients who underwent renal transplantation  and Schalkwijk comparing hemodialysed patients with normal subjects . We would like to point out that there were generally no subjects with renal failure in our study group, so one may wonder whether fluorescent AGE/ pentosidine is not an indicator showing subclinical renal changes.
We do not try to present pentosidine or fluorescent AGEs as a diagnostic tool but we want to pay attention to the low cost, speed and simplicity of AGE determination by measuring fluorescence compared to other methods, such as enzyme immunoassay or liquid chromatography with tandem mass spectrometry (LC-MS/MS). For research purposes, this method seems to be sufficient.
The glycation pathway is associated with reactive oxidative species (ROS) and with peroxidation. ROS peroxidise membrane unsaturated fatty acids, leading to generate reactive aldehydes as advanced lipid peroxidation end products (ALE). Resulting reactive aldehydes react with proteins to cause alteration of protein structure to exacerbate complication of diseases like diabetes and atherosclerosis . One of the classic products of lipid peroxidation is malondialdehyde which is easily determined in serum and tissues . For many years, laboratory animals and human studies have analyzed the effect of hypoglycemic therapies on oxidative stress and the intensity of lipid peroxidation. Depending on the research model, the results vary, for example, Sotoudeh showed in the study with diabetic rats that metformin reduces MDA levels , Kocer claimed the same while looking at the results of women with polycystic ovaries  yet Alrefai reported that metformin does not lower MDA in obese patients with T2DM . Similarly, in the case of insulin, there have been reports of both an anti-inflammatory role  as well as that hyperinsulinemia seems to cause exaggerating oxidative stress . Bunck et al. confirmed that insulin therapy alone for one year does not cause a decrease in MDA levels . Many researchers recommend a combination therapy with insulin as the best therapeutic approach for reducing oxidative stress in patients who need glycemic control [60, 62, 64]. The results of our study on the group of geriatric patients confirm that treatment with both metformin and insulin causes a decrease in the concentration of the lipid peroxidation marker.