Association of FGF-2 Concentrations with Atheroma Progression in Chronic Kidney Disease Patients

Abstract

Background and objectives

Atherosclerosis is highly prevalent in CKD. The rate of progression of atherosclerosis is associated with cardiovascular events. Fibroblast growth factor 2 (FGF-2) is a member of the FGF family with potentially both protective and deleterious effects in the development of atherosclerosis. The role of circulating FGF-2 levels in the progression of atherosclerosis in CKD is unknown.

Design, setting, participants, & measurements

We used a multicenter, prospective, observational cohorts study of 481 patients with CKD. We determined the presence of atheroma plaque in ten arterial territories by carotid and femoral ultrasounds. Progression of atheromatosis was defined as an increase in the number of territories with plaque after 24 months. Plasma levels of FGF-2 were measured by multiplex analysis. A multivariable logistic regression analysis was performed to determine whether plasma FGF-2 levels were associated with atheromatosis progression.

Results

Average age of the population was 61 years. The percentage of patients in each CKD stage was 51% in stage 3, 41% in stages 4–5, and 8% in dialysis. A total of 335 patients (70%) showed plaque at baseline. Atheromatosis progressed in 289 patients (67%). FGF-2 levels were similar between patients with or without plaque at baseline (79 versus 88 pg/ml), but lower in patients with atheromatosis progression after 2 years (78 versus 98 pg/ml; P<0.01). In adjusted analyses, higher plasma FGF-2 was associated with lower risk of atheromatosis progression (odds ratio [OR], 0.86; 95% confidence interval [95% CI], 0.76 to 0.96; per 50 pg/ml increment). Analysis of FGF-2 in tertiles showed that atheroma progression was observed for 102 participants in the lowest tertile of FGF-2 (reference group), 86 participants in the middle tertile of FGF-2 (adjusted OR, 0.70; 95% CI, 0.40 to 1.20), and 74 participants in the lowest tertile of FGF-2 (adjusted OR, 0.48; 95% CI, 0.28 to 0.82).

Conclusions

Low FGF-2 levels are independently associated with atheromatosis progression in CKD.

Introduction

CKD is a growing public health problem worldwide and, over the past decade, it has developed into an area of intensive clinical and epidemiologic research. Patients with CKD, even at early stages, demonstrate higher risk for the development of cardiovascular disease (1,2), enhancing morbidity and mortality in this population (3). A variety of traditional cardiovascular risk factors such as hypertension, diabetes, dyslipidemia (2), and nontraditional variables such as endothelial cell dysfunction (4), inflammation (2), low vitamin D levels (5,6), or hyperphosphatemia (7–9) begin to act very early in the course of CKD, accelerating the pathogenesis of cardiovascular disease. Taking into account that traditional and CKD-specific risk prediction equations explain only a portion of the cardiovascular disease associated with CKD, the search for new risk factors for cardiovascular disease in patients with CKD continues.

Atherosclerotic cardiovascular disease is highly prevalent in patients with CKD (7) and its progression is closely associated with the progression of CKD (10). The role of atherosclerosis in the increased cardiovascular mortality in patients with CKD seems clear, although its contribution in late stages is controversial. Indeed, there are differences in the type of cardiovascular events and the associated factors along the progression of CKD. Thus, in early stages, there is a high mortality rate due to ischemic events, related to atherosclerosis (11). However, in patients on dialysis, most of the deaths seem to be caused by heart failure and sudden cardiac death, although subclinical atherosclerosis appears to be influencing the higher susceptibility of the myocardium of those patients to electrolyte imbalances (12). Indeed, atheroma burden also independently predicts cardiovascular events in dialysis, and the rate of atherosclerotic plaque formation predicts cardiovascular events in those patients (13). Therefore, the determination of factors that affect atherosclerosis progression in CKD could lead to therapeutic targets specific for patients with CKD.

Fibroblast growth factor 2 (FGF-2 or bFGF) is a member of a large family of heparin-binding proteins (14), and plays an important role in the pathogenesis of atherosclerosis, but it also has cardioprotective actions. FGF-2 affects essential biologic activities of vascular cells such as differentiation, proliferation, and migration, showing a dual role in the cardiovascular system. The expression of FGF-2 and its receptors in a normal vessel wall is highly beneficial for the maintenance of the vascular homeostasis and protection of endothelial cells (15). However, FGF-2 and its receptors play a role in the inflammatory process, intimal thickening, and intraplaque angiogenesis (16), stimulating proliferation and migration of vascular smooth muscle cells (VSMC) (14,17) and development of vasa vasorum in atherosclerotic lesions (18), therefore accelerating atherosclerotic plaque growth.

There are scarce clinical data investigating the role of circulating levels of FGF-2 in atherosclerosis. Only a few studies have demonstrated that levels of FGF-2 did not correlate with carotid obstruction grade (19) or coronary artery calcification in the general population (4). However, to date, no investigation has been conducted analyzing the role of circulating FGF-2 levels in atheromatosis progression. In this work, we analyzed the association between FGF-2 and progression of atheromatosis burden in a subpopulation of the National Observatory of Atherosclerosis in Nephrology (NEFRONA) study.

Materials and Methods

Design and Study Population

The protocol of the study was approved by the ethics committee of each hospital and all patients were included after signing informed consent. This research followed the principles of the Declaration of Helsinki. The design and objectives of the NEFRONA study have been already published in detail (20,21). Briefly, 2445 patients with CKD (937 in CKD stage 3, 820 in stages 4–5, and 688 in dialysis) without a history of previous cardiovascular disease who were 18–75 years of age were enrolled from 81 Spanish hospitals between October of 2009 and June of 2011, with a scheduled follow-up visit after 24 months. Patients from the NEFRONA study who had hemodynamically significant stenotic carotid plaque, ankle-brachial index (ABI) <0.7 at baseline, a cardiovascular event or received a renal allograft along the 2 year follow-up, or died after the first ultrasound exploration were excluded from the follow-up. Thus, 1555 were followed for 24 months. Out of those, 481 had available plasma samples to measure FGF-2 levels. Samples from the other patients had already been spent to measure other biomarkers in sub-studies within the NEFRONA study (22,23). In any case, both populations did not differ in most of the parameters of the study, although there was a lower proportion of patients in dialysis than in the total sample. Consequently, high sensitivity C-reactive protein levels were lower and 25-hydroxyvitamin D, total, and LDL cholesterol levels were higher in the analyzed sample (Supplemental Table 1).

Determination of FGF-2 in Plasma Samples

FGF-2 levels were measured in duplicate in frozen plasma samples after the end of the recruitment period. The detection was performed with multiplex kits (Milliplex MAP; Merck Millipore), an assay specially designed to identify ten different biomarkers in a small sample by an ELISA-like method. The assay uses internally color-coded microspheres coated with specific primary antibodies. After a molecule from a test sample is captured by the bead, an additional biotinylated antibody, built to identify another epitope of the molecule, is introduced. Thus, the emission of the bead identifies the compound and the intensity if the second antibody quantifies the amount of the compound. The multiplex panel included FGF-2, Eotaxin, GM-CSF, Fractalkine, IFN-γ, MDC, IP-10, MCP-1, MIP-1β, and VEGF. The purpose of the study was to identify whether those analytes, which have been previously associated with cardiovascular disease, were associated with atherosclerosis progression in CKD. Out of those, only FGF-2 showed an association with the outcome, after Bonferroni correction for multiple testing.

In order to determine whether the method of recollection of the samples was interfering with the results, we used the multiplex kit to measure 40 samples from in-house patients, frozen immediately after extraction, and compared the results against the same samples stored at 4°C for 24 hours (in order to mimic the conditions of the samples during the shipment from the extraction place to the biobank in which they were stored). The Pearson correlation coefficient was 0.99 (P<0.001), showing that the handling of the samples before storage did not affect the levels of FGF-2. The interassay coefficient of variability of this analyte is 4.8%.

Clinical Data and Laboratory Examinations

Current health status, medical history, baseline cardiovascular risk factors, and drug use information was obtained at baseline. A physical examination was performed, consisting of anthropometric measures, standard vital tests, and ABI measurements as previously described (24). A pathologic ABI was described as ≤0.9 or ≥1.4. Biochemical data were obtained from a routine fasting blood test within 3 months from the vascular study. Previous and current smokers were classified as ever smokers. Diagnosis of dyslipidemia was obtained from the clinical history. GFR was estimated using the Modification of Diet in Renal Disease Study formula (MDRD-4).

Ultrasound Imaging

B-mode ultrasound of the carotid and femoral arteries was performed using the Vivid apparatus (General Electric) equipped with a 6–13 MHz broadband linear array probe, as previously described (25). Briefly, ultrasound imaging was performed with the participant in a supine position and the head turned 45° contralateral to the side of the probe to evaluate carotid plaques. The presence of atheromatous plaques was defined as an intima-media thickness (IMT)>1.5 mm protruding into the lumen, according to the American Society of Echocardiography Consensus Statement (26) and the Mannheim IMT Consensus (27).

The presence of atheromatous plaques was explored in ten territories (both internal, bulb, and common carotids, and both common and superficial femoral arteries) by a single reader in blinded mode, using semiautomatic software (EchoPAC Dimension; General Electric Healthcare). Therefore, a score from 0 to 10 was assigned to every patient both at basal and at the follow-up exploration 2 years later; 0 meaning no plaques in any territory and 10 meaning the presence of atheroma plaque in all ten territories explored. To assess intraobserver reliability, a sample of 20 individuals was measured between three and five times on different days by a reader unaware of patients’ clinical history. An intraclass correlation coefficient of 0.8 was obtained for plaque assessment, indicating very good intraobserver reliability. IMT was only measured in arterial regions without plaque and calculated as the average between left and right sides. When a territory presented with plaque, the IMT value was censored to be 1.5 mm in that territory. The average of the IMT value of the ten territories was calculated and presented as the patient’s IMT value.

Evaluation of Progression

Atheromatosis progression was defined as an increase in the number of territories showing plaque with respect to the baseline visit, as previously used in the Multi-Ethnic Study of Atherosclerosis study (28). The analysis was performed in ten arterial territories, yielding a score of 1–10 both at the baseline (SB) and at the follow-up (SFU) visit. The outcome variable (atheromatosis progression) was defined as an increase in the number of territories with plaque with respect to the basal visit. Thus, if SFU−SB>0, the patient was assigned to the group of progressors. If SFU−SB=0, the patient was assigned to the group of nonprogressors. We did not observe a negative value (SFU−SB) in any patient.

Statistical Analyses

Univariable relationship between the levels of the ten compounds measured in the multiplex analysis with plaque presence and progression at 24 months was analyzed by t test for normally distributed variables, and Mann–Whitney U test in nonparametric analysis. In order to account for the multiple testing problem, Bonferroni correction was applied. Thus, only P values ≤0.01 were considered statistically significant in the univariate analysis. Significant variables in univariable analyses and potential confounders were used to develop appropriate multivariable logistic regression models. A forward step procedure was used to build the multivariable model, including the variable showing maximum contribution identifying those patients with 24-month atheromatosis progression, according to the likelihood ratio test. Those variables without a statistically significant contribution, but modifying in >10% the value of the coefficients of any of the significant variables when removed from the model, were considered confounders and included in the final model. Possible first-degree interactions between FGF-2 and different parameters were tested. A statistical significance level of 0.05 was used. A sensitivity analysis using FGF-2 as a categorical variable was also built. All analyses were made using a standard statistical package (SPSS 24.0).

Results

The characteristics of the population, grouped by tertiles of FGF-2, are shown in Table 1. The median age was 61 years, and the median (25th to 75th percentiles) plasma FGF-2 levels were 85 (55–138) pg/ml. Table 2 shows the univariable analysis of factors associated with atheromatosis progression. Of the 289 patients that progressed, 157 (54%) did so in one territory, 75 (26%) in two, 36 (13%) in three, 14 (5%) in four, four (1%) in five, two (0.7%) in six, and one (0.3%) in seven new territories were affected after 24 months. The univariable analysis (Table 2) showed that patients in whom atherosclerosis progressed were significantly older, more frequently smokers, and had higher blood levels of triglycerides, glucose, and high-sensitivity C-reactive protein, with a higher percentage of patients with diabetes and hypertension. Furthermore, the IMT values and the percentage of patients with plaque at baseline was also higher in progressors.

Table 1.

Baseline characteristics of 481 participants in the NEFRONA cohort study according to plasma fibroblast growth factor 2 (FGF-2) concentration

Characteristic	All (n=481)	Lower Tertile, 3–64 pg/ml (n=160)	Middle Tertile, 65–119 pg/ml (n=160)	Higher Tertile, 120–865 pg/ml (n=161)
Age, yr	61 (52–68)	61 (53–68)	61 (50–68)	61 (50–68)
Men, N (%)	295 (61)	107 (67)	94 (59)	94 (58)
History of smoking, N (%)	283 (59)	91 (61)	88 (55)	98 (61)
Dyslipidemia, N (%)	336 (70)	107 (67)	114 (71)	115 (71)
Diabetes mellitus, N (%)	137 (29)	43 (27)	51 (32)	43 (27)
Hypertension, N (%)	442 (92)	150 (94)	147 (92)	145 (90)
Systolic BP, mm Hg	142 (20)	142 (20)	142 (20)	141 (20)
Diastolic BP, mm Hg	81 (10)	81 (11)	81 (10)	81 (11)
Total cholesterol, mg/dl	180 (156–207)	179 (148–206)	184 (165–209)	180 (157–206)
LDL cholesterol, mg/dl	104 (85–126)	102 (75–124)	106 (93–128)	104 (87–128)
HDL cholesterol, mg/dl	47 (38–59)	47 (39–58)	47 (37–61)	47 (39–61)
Triglycerides, mg/dl	127 (90–171)	133 (89–171)	133 (91–180)	117 (88–164)
Glucose, mg/dl	98 (88–114)	99 (87–111)	99 (90–129)	95 (89–112)
25-hydroxyvitamin D, ng/ml	16.7 (11.9–21.4)	16.1 (12.0–21.0)	17.5 (12.2–21.9)	15.7 (11.1–21.2)
Intima-Media Thickness, mm	0.86 (0.67–1.1)	0.86 (0.71–1.14)	0.86 (0.65–1.1)	0.84 (0.64–1.1)
HsCRP, mg/L	2.12 (1.13–4.52)	2.12 (1.04–4.52)	1.97 (1.15–4.36)	2.16 (1.21–4.73)
CKD stage, N (%)
3	243 (51)	58 (36)	89 (56)	96 (60)
4–5	199 (41)	71 (44)	67 (42)	61 (38)
Dialysis	39 (8)	31 (20)	4 (2)	4 (2)
Plaque present at baseline, N (%)	335 (70)	115 (72)	106 (66)	114 (71)
FGF-2, pg/ml	85 (55–138)	43 (26–55)	85 (73–100)	164 (138–204)

Dyslipidemia: diagnosis of dyslipidemia in the clinical history. Data are median (25th to 75th percentile) for quantitative variables non-normally distributed, mean (SD) for normal variables, and number (percentage) for categorical variables. HsCRP, high-sensitivity C-reactive protein.

Table 2.

Baseline characteristics of 481 participants in the NEFRONA cohort study according to atherosclerosis progression over the subsequent 2 years

Characteristic	No (n=192)	Yes (n=289)	P (No versus Yes)
Age, yr	56 (42–65)	63 (56–69)	<0.001
Men, N (%)	109 (57)	186 (64)	0.10
Ever smoker, N (%)	100 (52)	183 (63)	0.01
Dyslipidemia, N (%)	126 (66)	210 (73)	0.11
Diabetes mellitus, N (%)	40 (21)	97 (34)	0.002
Hypertension, N (%)	171 (89)	271 (94)	0.04
Systolic BP, mm Hg	137 (18)	145 (21)	<0.001
Diastolic BP, mm Hg	80 (10)	81 (11)	0.29
Total cholesterol, mg/dl	179 (156–202)	183 (157–213)	0.31
LDL cholesterol, mg/dl	102 (85–123)	105 (86–129)	0.39
HDL cholesterol, mg/dl	48 (40–61)	47 (37–58)	0.15
Triglycerides, mg/dl	109 (82–163)	137 (103–176)	0.001
Glucose, mg/dl	95 (86–105)	102 (91–123)	<0.001
25-hydroxyvitamin D, ng/ml	17.2 (11.8–22.3)	16.2 (11.9–20.7)	0.54
IMT, mm	0.72 (0.59–1.01)	0.92 (0.75–1.14)	<0.001
HsCRP, mg/L	1.87 (1.02–3.77)	2.49 (1.23–5.17)	0.01
CKD stage, N (%)			0.36
3	104 (54)	139 (48)
4–5	72 (38)	127(44)
Dialysis	16 (8)	23 (8)
Plaque presence at baseline, N (%)	109 (57)	226 (78)	<0.001

Dyslipidemia: diagnosis of dyslipidemia in the clinical history. Data are median (25th to 75th percentile) for quantitative variables non-normally distributed, mean (SD) for normal variables, and number (percentage) for categorical variables. IMT, average intima-media thickness; HsCRP, high-sensitivity C-reactive protein.

As seen in Figure 1, A and B, FGF-2 levels did not differ among patients with or without plaque at baseline nor between patients with different atheromatosis burden. However, Figure 1C shows that FGF-2 levels were significantly lower in patients in which atherosclerosis progressed after 2 years of follow-up compared with patients where the plaque burden remained stable. In addition, FGF-2 levels were significantly lower in patients with more advanced CKD stages (Figure 2).

Figure 1.

Plasma levels of fibroblast growth factor 2 (FGF-2) are higher in patients with atheromatosis progression after 2 years. (A) Plasma levels in patients with atheromatosis versus no atheromatosis at baseline. (B) Plasma levels in patients with no atheromatosis, mild atheromatosis (one to three territories with plaque), and severe atheromatosis (more than three territories with plaque. (C) Plasma levels in patients in which atheromatosis progressed after 2 years versus those in which it remained stable.

Figure 2.

Plasma levels of fibroblast growth factor 2 (FGF-2) are lower in patients with more advanced stages of CKD.

The unadjusted analysis modeling atherosclerosis progression (Table 3, model 1) shows that higher levels of FGF-2, assessed as a continuous variable, are associated with lower progression of atherosclerosis. Adjustment for age and sex (model 2) and by several known risk factors for atheromatosis progression (model 3) shows that the effect of FGF-2 remains unmodified. A sensitivity analysis using FGF-2 as a categorical variable (tertiles) yielded similar results. The possible interactions between FGF-2 levels and several parameters (including CKD stage) were tested and disregarded as nonsignificant. The Hosmer and Lemeshow P values show an excellent goodness of fit of the logistic regression model, so the model is considered well calibrated.

Table 3.

Associations of plasma fibroblast growth factor 2 concentration with atherosclerosis progression

	Model 1		Model 2		Model 3
Variables	Odds Ratio	P Value	Odds Ratio	P Value	Odds Ratio	P Value
FGF-2, per 50 pg/ml	0.86 (0.76–0.96)	0.008	0.85 (0.75–0.95)	<0.01	0.86 (0.76–0.96)	0.02
Hosmer–Lemeshow P value	0.94		0.76		0.22
FGF-2 lowest tertile, 3–64 pg/ml	Reference		Reference		Reference
FGF-2 middle tertile, 65–119 pg/ml	0.61 (0.37–0.99)	0.05	0.65 (0.39–1.01)	0.11	0.70 (0.40–1.20)	0.20
FGF-2 highest tertile, 120–865 pg/ml	0.42 (0.26–0.68)	<0.001	0.43 (0.26–0.72)	0.001	0.48 (0.28–0.82)	<0.01
Hosmer–Lemeshow P value	1.00		0.19		0.69

First and second rows: model using levels of FGF-2 as a continuous variable. Third to sixth rows: model using levels of FGF-2 as tertiles. Model 1: unadjusted analysis; model 2: adjusted by age and sex; model 3: model 2 plus smoking status, diabetes, dyslipidemia, plaque presence at baseline, stage of CKD, intima-media thickness, body mass index, serum levels of glucose, total cholesterol, high-sensitivity C-reactive protein, 25-hydroxyvitamin D, and phosphate. All models included 431 patients (262 progressors and 169 nonprogressors). The lowest tertile of FGF-2 included 143 patients (102 progressors and 41 nonprogressors). The middle tertile of FGF-2 included 143 patients (86 progressors and 47 nonprogressors). The highest tertile of FGF-2 included 145 patients (74 progressors and 71 nonprogressors).

Discussion

This work shows that FGF-2 levels are associated with atherosclerosis progression in patients with CKD. This effect is independent of other parameters known to affect the progression of atherosclerosis in CKD, such as age, the degree of renal function, the levels of cholesterol, and smoking status. Therefore, FGF-2 could be considered as a marker and potential target in atherosclerosis progression in patients with CKD.

The role of FGF-2 in atherosclerosis is unclear, and preclinical data show that beneficial and detrimental effects could coexist. On the one hand, FGF-2 may be deleterious because of its potential to increase plaque size. Thus, FGF-2 and FGF receptors are expressed in VSMC (29) and FGF-2 induces VSMC proliferation in vitro (17). Furthermore, FGF-2 promotes VSMC migration by inducing matrix metalloproteases 2 and 9 expression in baboon aortic explants (30) and by regulating TRAIL expression in mice undergoing the cuff injury model (31). In addition, FGF-2 could regulate VSMC phenotypic transformation from the contractile to the synthetic phenotype, a necessary step in the early stages of atherosclerosis (32). Thus FGF-2 has been shown to act synergistically with PDGF, inducing the phenotypic change in vitro (33). In any case, and although FGF-1 levels increase in atherosclerotic plaques versus normal artery tissue, FGF-2 levels have been reported to remain normal in human atherosclerotic artery samples (29).

On the other hand, the effects of FGF-2 in endothelial cells point to a beneficial effect on atherosclerosis. Atherosclerotic disease is a progressive disorder that develops over a long period of time, triggered locally by endothelial dysfunction. FGF-2 increases proliferation and migration of endothelial cells (34) and furthermore, maintains their integrity (15,35), antagonizing endothelial dysfunction. In addition, administration of FGF-2 improves endothelial function, decreasing vascular endothelial adhesion molecule expression and macrophage infiltration early on in the course of experimental atherosclerosis (36).

In vivo studies in experimental models of atherosclerosis have yielded controversial results. Thus, Che et al. (37) showed that endothelial overexpression of FGF-2 accelerates atherosclerosis in a transgenic mouse model. Raj et al. (38) also showed that pharmacologic inhibition of FGF receptor signaling with a tyrosine kinase activity inhibitor attenuated atherosclerosis in ApoE null mice. However, the effects observed in the former studies could be attributed to signaling of any of the 23 FGF species described so far. In experiments using FGF-2, direct administration of the compound did not increase neointimal formation after balloon injury in pigs (39), dogs (40), and rabbits (41), and it even decreased neointimal formation in a rat model of balloon injury (42) and in a rabbit model of poor distal runoff circulation (43). In any case, the potential different effects of FGF-2 on experimental atherosclerosis could be explained by the different effect of the compound depending on the stage of evolution of plaque, from type 1 plaques with relatively normal histology and endothelial dysfunction, to complicated type 6 plaques with thrombus. Thus, in early stages, increased FGF-2 levels can prevent the formation of plaque by reducing endothelial dysfunction (36). In contrast, in advanced lesions where endothelial dysfunction is no longer an issue, FGF-2 promitogenic actions can induce plaque growth and even rupture by increasing the density of the vasa vasorum and matrix metalloprotease synthesis (44,45). In our study, lower FGF-2 levels are associated with the appearance of new plaques, but were not associated with the baseline presence of atheroma plaques, agreeing with the notion of a protective role for FGF-2 in the early stages of atheroma plaque formation. Indeed, clinical trials using FGF-2 as a treatment for revascularization have not shown any effect aggravating atherosclerotic disease (46,47).

Our results disclosing a higher odds ratio (OR) for atheromatosis progression in patients with CKD stages 4–5 and on dialysis than in patients with CKD stage 3 may appear discordant with those of Rigatto et al. (48) observing lower progression of atheromatosis in later CKD stages. In our study, the OR for atheromatosis progression was higher in patients with CKD stages 4–5 and on dialysis compared with patients with CKD stage 3. However, the design of both studies differ, and those differences may account for the discrepancies. First, our cohort specifically excluded patients with CKD with previous cardiovascular disease, whereas one third of patients in Rigatto’s cohort had a previous cardiovascular event. Furthermore, our study explored the femoral territories, which were not tested in Rigatto’s study. Finally, a possible interaction between FGF-2 levels and CKD stage could also explain the discrepancy, although that interaction was tested and disregarded as nonsignificant.

Another result from our study shows that FGF-2 levels are lower in more severe CKD stages, reaching very low levels in patients on dialysis. There is little information about the role of FGF-2 in CKD. Experimental data show that FGF-2 could be a therapeutic tool to reduce CKD. FGF-2 administration reduces functional and structural damage in experimental CKD (49). Furthermore, FGF-2 mediates the improvement of renal function induced by the administration of mesenchymal stem cells in a rat model of CKD (50). Thus, FGF-2 could have a potential role in the progression of CKD, although further research is warranted.

The main strength of this study is the relatively large number of patients with longitudinal observations, which allows us to make associations controlling for multiple confounders. A second strength is that the vascular exploration was performed by the same team and evaluated by a single reader. Nevertheless, our study also has some limitations. The main one is that out of the total sample of the patients followed for atherosclerosis progression, only a portion had an available plasma sample to measure FGF-2 levels. Therefore, the subpopulation might not be representative of the full cohort, as some parameters differed between patients with an available sample and the rest of the cohort. However, the number of patients is still high enough to adjust for many potential confounding factors. Furthermore, patients that died or suffered a cardiovascular event were also unavailable for the follow-up visit, alongside patients that underwent a kidney transplant. Therefore, we cannot exclude survival bias because individuals with poor cardiovascular health were excluded. Another limitation of the study is the fact that the detection method is a multiplex assay, and results should be validated with a method specific for FGF-2. Furthermore, the observational nature of the study and the unclear biologic function of the measured circulating FGF-2 are additional limitations.

In summary, lower FGF-2 levels are associated with atheromatosis progression in patients with CKD. Further research is needed to determine whether FGF-2 could be a therapeutic target to reduce atheromatosis progression in kidney patients.

Disclosures

None.