Resistin in Early Diabetic Chronic Kidney Disease: Exploring the Link with Nutritional Status and Cardiovascular Outcome

Abstract

High resistin levels have been associated with malnutrition, inflammation, and cardiovascular risk in patients with chronic kidney disease (CKD). This study aimed to elucidate the relationship between serum resistin levels and the Patient-Generated Subjective Global Assessment (PG-SGA), a validated nutritional assessment tool in this population. It also investigates the role of resistin as a potential predictor of cardiovascular mortality in early-stage diabetic CKD. Prospective observational study that included 217 type 2 diabetic patients with mild to moderate CKD. Patients were divided into three groups according to PG-SGA: well-nourished (category A), moderately malnourished or suspected of being malnourished (category B), or severely malnourished (category C). The severely malnourished group had higher resistin levels, and resistin was positively correlated with IL-6, PG-SGA score, left ventricular mass index, and pulse pressure, while negatively correlating with vitamin D and estimated glomerular filtration rate (eGFR). We found that high resistin levels (HR = 1.350; 95% CI 1.187–1.535), PG-SGA greater than 10 (HR = 4.858; 95% CI 1.664–14.185), and higher HOMA-IR (HR = 1.099; 95% CI 1.007–4.001) were significant independent predictors of cardiovascular mortality. The study suggests that high resistin levels are associated with malnutrition in mild to moderate CKD and independently predict cardiovascular mortality in this population.

1. Introduction

Cardiovascular disease is the leading cause of death in chronic kidney disease (CKD) [1]. Alongside the classic cardiovascular risk factors, such as hypertension, dyslipidemia, diabetes, and insulin resistance, inflammation and malnutrition, are recognized factors associated with increased morbidity and mortality in these patients [2, 3].

Resistin, an adipokine discovered two decades ago, is characterized as a 12.5 kilodalton cysteine-rich protein, which has been notably implicated in insulin resistance, particularly in murine models [4]. Nevertheless, human data presents a divergence of findings concerning its association with insulin resistance and obesity [5]. Paradoxically, resistin emerges as a pivotal participant in the immune system, being acknowledged as a pro-inflammatory adipokine [6]. While its principal source of secretion is adipocytes, it is noteworthy that monocytes, macrophages, bone marrow cells, and cardiomyocytes can also contribute to its secretion [7, 8, 9].

Among individuals with CKD, resistin demonstrates a correlation with inflammation as numerous studies elucidate an association between its serum levels and serum tumor necrosis factor-alpha and interleukin 6 (IL-6) levels [10, 11]. Elevated levels of resistin have also been linked to malnutrition in dialysis patients [12, 13]. Among elderly nondiabetic CKD patients, elevated circulating levels of resistin appear to be an independent predictor of cardiovascular and all-cause mortality [14]. Furthermore, resistin exhibits a robust association with mortality and graft loss in kidney transplant recipients [15].

The kidney disease outcomes and quality initiative (KDOQI) recommends the subjective global assessment (SGA) instrument as the gold standard for nutritional assessment in CKD, a tool that is now available as the web-based program Scored Patient-Generated Subjective Global Assessment (PG-SGA), which has already been validated in this population [16, 17, 18].

This study aims to elucidate the relationship between serum resistin levels and malnutrition, explore its association with inflammation, and evaluate the potential of resistin as a prognostic factor for cardiovascular mortality in early diabetic CKD.

2. Materials and methods

2.1 Study population

In this observational prospective study, we included patients with type 2 diabetes who were recruited between January 2015 and June 2022 with a diagnosis of mild to moderate CKD (stages 2–3) in a stable clinical condition followed in our outpatient clinic.

The exclusion criteria were as follows: age 75 years, estimated glomerular filtration rate (eGFR) ≤29 or > 90 mL/min, type 1 diabetes, nondiabetic renal disease, and known neoplastic, infectious, or chronic inflammatory diseases.

The study procedure encompassed data collection and patient interviews to evaluate nutritional status through the utilization of the PG-SGA. This tool includes objective and subjective information. The first section focuses on patient grading and comprises data concerning the patient’s medical history, covering factors such as weight variations, changes in dietary habits, gastrointestinal symptoms, physical activities, and functional capacity. The second section, based on physician grading, encompasses information related to the patient’s disease status such as comorbidities and advanced age, metabolic demands, as well as findings from a physical examination, including observations of subcutaneous fat loss and indications of muscle wasting. Patients were then categorized into well-nourished (category A), moderately malnourished or suspected of being malnourished (category B), or severely malnourished (category C). The PG-SGA score ranges from 0 to 36, with a higher score indicating more severe malnutrition.

Serum samples were collected from fasting patients. Plasma resistin levels were determined by enzyme-linked immunosorbent assay. GFR was estimated using the National Kidney Foundation recommended Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation [19].

Approval was obtained from the local ethics committee. All principles of the declaration of Helsinki of 1975 were followed and study procedures were only conducted after obtaining patients’ written informed consent.

2.2 Statistical analysis

Categorical variables are expressed as frequencies and percentages, while continuous variables are presented as mean and standard deviation (SD) or median and interquartile range (IQR).

We used one-way analysis (ANOVA) and post hoc analysis with the Scheffe test to compare the groups. Pearson’s correlation test was performed to measure the association between the identified variables and resistin levels.

Univariate and multivariate Cox regression analyses were performed to determine the association between serum resistin levels and cardiovascular mortality.

Statistical analysis was performed using ‘Statistical Package for the Social Science’ version 29.0 for Windows (SPSS, Chicago, IL, USA). A P-value less than 0.05 was considered statistically significant.

3. Results

Two hundred and seventeen (217) patients with type 2 diabetes and CKD stage 2–3 were included in the study after confirming they did not meet any of the exclusion criteria.

The mean age was 56.01 ± 24.05 years, of whom 60.4% (131) were male. 36.9% (n = 80) of patients had hypertension and the mean follow-up was 54.30 ± 24.05 months.

Table 1 shows patients’ demographic and clinical parameters, and a comparison between groups. 36.9% (n = 80) of patients had hypertension.

	Category A; (well-nourished); N = 71 (32.7%)	Category B; (moderately nourished); N = 58 (26.7%)	Category C; (malnourished); N = 88 (40.6%)	ANOVA; p-value	Category; A vs. -C; p-value
Age, years	49.08 ± 26.34	54.30 ± 18.83	60.92 ± 20.61	< 0.001	< 0.001
eGFR, mL/min/1.73m^2	48.48 ± 23.93	40.06 ± 22.60	29.16 ± 15.34	< 0.001	< 0.001
PG-SGA score, median	3	9	11	< 0.001	< 0.001
Resistin, ng/mL	4.49 ± 3.18	5.63 ± 3.38	6.25 ± 3.47	0.004	0.004
IL-6, pg./mL	4.56 ± 3.33	6.65 ± 3.21	7.23 ± 3.32	< 0.001	< 0.001
HOMA-IR	1.52 ± 1.53	2.19 ± 1.70	2.46 ± 1.90	0.008	0.009
Hemoglobin, g/dL	12.97 ± 1.61	12.30 ± 1.51	11.94 ± 1.60	< 0.001	< 0.001
Ferritin, ng/mL	194.80 ± 134.37	169.65 ± 167.76	189.36 ± 162.42	0.544	0.992
LVMI, g/m^2	95.59 ± 24.54	108.82 ± 27.66	115.19 ± 23.27	< 0.001	< 0.001
Pulse pressure, mmHg	56.58 ± 21.37	60.99 ± 18.63	71.60 ± 17.51	< 0.001	< 0.001
Total cholesterol, mg/dL	163.04 ± 70.78	148.07 ± 74.21	115.29 ± 80.61	< 0.001	< 0.001
Calcium, mg/dL	9.05 ± 1.11	8.92 ± 1.06	8.58 ± 1.14	0.020	0.026
Phosphorus, mg/dL	3.91 ± 0.74	4.17 ± 0.82	4.47 ± 1.18	0.001	0.001
25(OH)D3, ng/mL	21.52 ± 6.38	20.85 ± 7.13	18.46 ± 8.40	0.024	0.035
PTH, ng/mL	115.23 ± 88.21	135.36 ± 70.73	176.22 ± 89.54	< 0.001	< 0.001

Table 1.

Patients’ demographic and clinical characteristics of each cohort.

Values are means ± SD unless specified otherwise. eGFR: estimated glomerular filtration rate HOMA-IR: homeostatic model assessment for insulin resistance; IL-6: interleukin-6; LVMI: left ventricular mass index; PG-SGA: Patient-Generated Subjective Global Assessment; PTH: parathyroid hormone.

A total of 43 (19.8%) patients died during the course of the study.

The severely malnourished group had lower eGFR (29.16 vs. 48.48 mL/min/1.73m², p < 0.001); higher levels of resistin (6.25 vs. 4.49 ng/mL, p = 0.004), parathormone (176.22 vs. 115.23 pg./mL, p < 0.001), phosphorus (4.47 vs. 3.91 mg/dL, p = 0.001), HOMA-IR (2.46 vs. 1.52, p = 0.009) and IL-6 (7.23 vs. 4.56 pg./ml, p < 0.001); lower levels of hemoglobin (11.94 vs. 12.97 g/dL, p < 0.001), calcium (8.58 vs. 9.05 mg/dl, p = 0.026) and 25(OH)D3 (18.46 vs. 21.52 ng/mL, p = 0.035); higher left ventricular hypertrophy (LVMI 115.19 vs. 95.59 g/m², p < 0.001) and higher pulse pressure (71.60 vs. 56.58, p = 0.005) when compared to the well-nourished group. No statistical differences were observed between sexes.

Pearson’s correlation analysis revealed significant associations between resistin and various biochemical and clinical parameters. A moderate positive correlation was observed between resistin and IL-6 (r = 0.591, p < 0.001), as well as between resistin and LVMI (r = 0.489, p < 0.001). Additionally, weaker positive correlations were identified between resistin and both pulse pressure (r = 0.310, p < 0.001) and PG-SGA score (r = 0.217, p < 0.001). Conversely, a strong negative correlation was observed between resistin and vitamin D (r = −0.854, p < 0.001), and a weaker negative correlation was found between resistin and eGFR (r = −0.246, p < 0.001). No association was found between resistin and HOMA-IR and total cholesterol.

On multivariable modeling, we found that higher resistin levels (HR = 1.350; 95% CI 1.187–1.535), PG-SGA score greater than 10 (HR = 4.858; 95% CI 1.664–14.185), and higher HOMA-IR (HR = 1.099; 95% CI 1.007–4.001) were significant independent predictors of cardiovascular mortality when adjusted for age, IL-6 and eGFR (Table 2).

	HR (CI 95%)	P-value
Resistin levels	1.350 (1.187–1.535)	< 0.001
PG-SGA score
0–10 points	Ref.
> 10 points	4.858 (1.664–14.185)	0.004
HOMA-IR	1.099 (1.007–4.001)	0.002

Table 2.

Independent predictors of cardiovascular mortality by multivariate cox regression in early CKD.

Adjusted to age, eGFR and IL-6. HOMA-IR: homeostatic model assessment for insulin resistance; HR: hazard ratio; IL-6: interleukin-6; PG-SGA: Patient-Generated Subjective Global Assessment.

4. Discussion

Resistin has gained attention as a potential link between malnutrition, inflammation, and cardiovascular risk in patients with CKD [12, 20, 21].

Increased resistin levels may be involved in the development of malnutrition-inflammation states and, consequently, protein-energy wasting, a condition of paramount significance in the context of managing this population [21, 22]. Our results corroborate this, showing higher plasma resistin levels in the malnourished group and a positive association with both IL-6 and PG-SGA scores. Increased resistin levels were also correlated with left ventricular hypertrophy and pulse pressure, both of which are independent markers of cardiovascular risk in CKD [23, 24].

A strong inverse correlation was identified between resistin and vitamin D levels. Multiple investigations have reported similar findings, thereby bolstering the hypothesis that vitamin D may play a role in modulating resistin [25, 26]. Nevertheless, a study conducted by Vaidya A. et al. documented a positive correlation between these factors, complicating the formulation of their interrelationship [27].

A negative association between resistin and eGFR was also established, consistent with several data suggesting that this relationship represents a physiological mechanism due to the elimination of resistin by the kidneys [28, 29]. Stepien et al. also observed a negative correlation between resistin and eGFR, although this difference did not reach statistical significance [30]. They attributed this nonsignificant result to factors such as the sample size and the early stage of CKD. In fact, the comparison of a group with early-stage CKD to a group without CKD may account for this finding.

Given resistin’s known association with inflammation and the well-established link between inflammation and the pathogenesis of atherosclerosis, it has been advocated that resistin could be a useful predictive biomarker of adverse cardiovascular outcome in this population. Our study explored this idea, showing that higher serum resistin levels are independently associated with cardiovascular mortality in early diabetic CKD patients. PG-SGA score greater than 10 was also associated with cardiovascular mortality. Similar result was obtained by Rodrigues et al., who enrolled 146 women with gynecologic cancer and found a significant association between PG-SGA score (> 10 points versus 0–10 points) and all-cause mortality within 1 year [31].

Several limitations of this study need to be acknowledged. Firstly, it was a small-sample observational, prospective study, potentially compromising the strength of some statistical analysis. Secondly, resistin levels were assessed on a single occasion, and confounding factors at the time of collection not taken into account in this study may have influenced our results. Thirdly, it is important to note that this study is prognostic rather than an etiologic study.

5. Conclusion

The study suggests that high resistin levels are associated with malnutrition in mild to moderate CKD and independently predict cardiovascular mortality in this population.

Conflict of interest

The authors declare no conflict of interest.