Introduction

Patients with chronic kidney disease (CKD) are prone to malnutrition or protein-energy wasting (PEW), which is a strong risk factor for poor prognoses in patients undergoing maintenance hemodialysis (MHD) [1]. The worldwide prevalence of PEW ranges from 11 to 54% in patients with CKD stages 3–5 [2], and between 28 and 54% in patients undergoing dialysis [3]. The prevalence of PEW varies in different studies, depending on the assessment method. The Seven-Point Subjective Global Assessment (SGA) [4] and Geriatric Nutritional Risk Index (GNRI) [5] are widely used tools for PEW assessment. Abnormalities in growth hormone (GH)/insulin-like growth factor-1 (IGF-1) are considered significant factors in the development of skeletal muscle wasting in patients with CKD via inflammation [6]. Studies have shown that lower serum IGF-1 levels are significantly associated with biochemical and anthropometric markers of malnutrition [7, 8]. Researchers have also observed an inverse association between serum IGF-1 levels and nutritional status assessed by traditional SGA in patients undergoing MHD [9]. Since serum IGF-1 levels are strongly related to age or sex, serum IGF-1 SDSs were calculated using the method described by Isojima et al. to minimize the bias of these confounding factors [10].

To the best of our knowledge, no study has analyzed the relationship between standardized IGF-1 levels and indicators of comprehensive nutritional assessment such as the 7-point SGA and GNRI, in patients with MHD. Hence, we designed the present cross-sectional study to explore the relationship between IGF-1 SDS levels and the overall nutritional status indicators in these patients.

Methods and materials

Population and study design

This was a single-center, observational, cross-sectional study of patients undergoing MHD at Guangzhou Red Cross Hospital in March 2021. Patients included (1) had end-stage renal disease (ESRD); (2) performed maintenance hemodialysis regularly for more than 3 months, 3 times a week, and for four hours each time; (3) aged 18–80 years; and (4) provided informed consent. The exclusion criteria were: (1) neurological or mental illness or inability to complete the questionnaire; (2) history of acute heart failure, severe infection, or malignant tumor within 3 months; (3) history of surgery within 3 months; (4) pacemaker use; and (5) refusal to participate in the study.

Research methodology

Demographic, clinical, and laboratory parameters

Data on the following demographic and clinical parameters were collected: history of diabetes, sex, age, and dialysis vintage. Venous blood samples were collected by nurses shortly before the hemodialysis sessions. They were sent to the clinical laboratory of Guangzhou Red Cross Hospital within 2 h. Routine blood tests, serum biochemical tests, and enzyme immunity measurements were performed. Further, the following parameters were measured: Serum creatinine, Kt/V, serum albumin, serum prealbumin, serum IGF-1, serum interleukin-6 (IL-6), serum ceruloplasmin, and serum hypersensitive C-reactive protein (hs-CRP).

Measurement and calculation of IGF-1

Serum IGF-1 was measured by high-performance liquid chromatography-mass spectrometry using a Thermo Q Exactive Focus instrument (Thermo Fisher Scientific, Waltham, MA, USA). The assay was calibrated using standards prior to testing. All reagents used in this study (including inoculated strains, reagent wedges, calibrators, and dilutions) were obtained from the same batch. The intra-group coefficients of variation ranged from 2.4% to 6.3%, and the inter-group coefficients of variation ranged from 3.0% to 7.6%. The sensitivity was 20 μg/L, and the upper limit of detection was 1600 μg/L. The SDS of serum IGF-1 was calculated using the following formula [10]:

$${\text{SDS}} = {\text{(measurement\, - \,mean)/standard deviation (SD)}}{.}$$

Nutritional indicators

Body mass

A body composition analyzer (Bodystat5000, UK) was used to measure the body composition of patients. According to the manufacturer's guidelines, measurements were performed in the supine position, with electrodes attached to the hands and feet on the side without a fistula for hemodialysis. Resistance (R in ohms) and reactance (Xc in ohms) values were recorded at 50 kHz. Patients with implantable electronic devices (e.g., electronic heart pacemakers) were excluded. Lean body mass, phase angle, extracellular water (ECW), ECW%, intracellular water (ICW), ICW%, and the ECW/ICW ratio were measured during the process.

Calculation of nutritional indices

According to the 7-point SGA, a total score of 6 to 7 was classified as normal nutritional status, 3 to 5 as mild to moderate malnutrition, and 1 to 2 as severe malnutrition. A 7-point SGA score ≤ 5 points was used as the diagnostic criterion for malnutrition or PEW [11].

The GNRI was calculated according to the following formula:

$${\text{GNRI}} = 1.489 \times {\text{serum albumin (}}g/L) + 41.7 \times ({\text{actual body weight/ideal body weight}})$$

where the ideal body weight was calculated as \(22\;({\text{kg/m}}^{2} ) \times {\text{height}}\) [12]. If the actual body weight was above the ideal body weight, the value of “(actual body weight/ideal body weight)" was set to 1. At present, some studies suggest that a GNRI in MHD patients of < 91.2 can be defined as a risk of malnutrition [13].

The mCI was calculated using parameters including sex, age, spKt/V (for urea clearance), and pre-hemodialysis creatinine level, using the following formula [14]:

$$\begin{aligned} {\text{MCI}}\;{\text{(mg/kg/day)}} & = 16.21 + 1.12 \times ({\text{0 for women; 1 for men}}) - 0.06 \times {\text{age }}({\text{years}}) - 0.08 \times SPKt/V\;{\text{for urea}} \\ & + 0.009 \times {\text{pre-hemodialysis creatinine}}\;{\text{(umol/L}}). \\ \end{aligned}$$
Grip strength and pinch strength

Grip strength and pinch strength were measured using the BASELINE digital Grip Force Tester (12-0091, Fabrication Enterprises Inc., USA) and the BASELINE digital Pinch Force tester (12-0081, Fabrication Enterprises Inc., USA). The participants were instructed to apply as much grip or pinch force as possible on the instruments with the dominant hand or the hand without fistulae. The measurement was repeated thrice, and the maximum value was recorded.

Anthropometry

Upper arm circumference was measured according to the criteria developed by Frisancho [15]. The upper arm muscle circumference was calculated according to the following formula:

$${\text{upper arm muscle circumference (cm)}} = {\text{upper arm circumference (cm)}} - 3.14 \times {\text{triceps skinfold thickness (cm)}}.$$

Parameters of hemodialysis

Patients in this study were treated on a Braun Dialog + (B. Braun Co., Ltd., Melsungen, Germany) dialysis machine using a REXEED-15L high-throughput polysulfone membrane dialyzer (Asahi Kasei Corp., Tokyo, Japan) with a membrane area of 1.5 m2, dialysis blood flow rate of 200–300 ml/min, dialysis fluid flow rate of 500 ml/min, and dialysis duration of 4 h.

Statistical analysis

Continuous variables with a normal distribution were described as \(\overline{x }\) ± s, and variables with a non-normal distribution were described as the median (25–75% interquartile range). Categorical variables were described as percentages. Patients were divided into three groups based on their serum IGF-1 SDS levels. Differences in clinical and demographic data among the three groups were assessed by one-way analysis of variance, Kruskal–Wallis test, or χ2 test, as appropriate. Spearman analyses were used to evaluate the correlation between the IGF-1 SDS and nutritional indicators. Univariate analyses were conducted to select potential explanatory variables with dependent variables of malnutrition defined as SGA ≤ 5 and GNRI ≤ 91.2. In multivariate binary logistic regression analyses, potentially relevant variables or known to be important in the physiology of malnutrition: sex, age, dialysis vintage, Kt/V (≥ 1.2), DM, IL-6, phosphorus, and 25-OH-D values, were included to determine whether IGF-1 SDS levels was independently associated with malnutrition. SPSS (version 22.0; IBM Corp, Armonk, NY, USA) was used for statistical analyses. Significance was set at p < 0.05.

Results

Characteristics of the study population

A total of 155 patients with MHD (Fig. 1) were included (male/female = 90:65), with a median dialysis vintage of 28.0 (11.0, 55.0) months and an average age of 66 (65.5 ± 13.0) years. The sample selection flowchart is shown in Fig. 1. The baseline characteristics of the total study population and those after stratification by serum IGF-1 tertile are presented in Table 1. The median serum IGF-1 level was 186.0 (140.0, 248.7) ng/mL. The median of IGF-1 SDS levels were levels − 0.1 (− 0.6 to 0.6). Malnourished patients, as defined by the 7-point SGA and the GNRI, accounted for 28.4% (n = 44) and 25.3% (n = 39) of the study population, respectively. Patients with lower serum IGF-1 levels were older and had lower levels of serum creatine, serum ceruloplasmin, serum albumin, serum prealbumin, handgrip strength, pinch strength, upper arm circumference, upper arm muscle circumference, phase angle, modified Creatinine Index (mCI), ICW, higher levels of ECW%, serum IL-6, and higher malnutrition risk, as defined by the SGA and GNRI values (all p < 0.05).

Fig. 1
figure 1

Sample selection flowchart of the study

Full size image
Table 1 Baseline Characteristics of the study population and after stratification by tertiles of serum IGF-1 SDS
Full size table

Correlations between IGF-1 SDS and nutritional indicators

In Spearman’s analyses, we observed that the levels of serum IGF-1 SDS were positively correlated with the values of grip strength (r = 0.32, p < 0.001), pinch strength (r = 0.28, p < 0.001), phase angle (r = 0.33, p < 0.001), lean body mass (r = 0.30, p < 0.001), serum albumin (r = 0.42, p < 0.001), serum prealbumin (r = 0.52, p < 0.001), mCI (r = 0.52, p < 0.001), ICW (r = 0.21, p = 0.012), and were negatively correlated with the levels of ECW% (r = − 0.22, p = 0.010) (Fig. 2). There were no associations between ECW (r = 0.13, p = 0.141), ICW% (r = 0.02, p = 0.800), and ECW/ICW (r = − 0.11, p = 0.180).

Fig. 2
figure 2

Spearman’s correlation analyses between levels of grip strength (a), pinch strength (b), phase angle (c), lean body mass (d), serum albumin (e), serum prealbumin (f), mCI (g), ICW (h), ECW% (i), and serum levels of IGF-1 SDS

Full size image

Associations between the serum IGF-1 and nutrition risk defined by different nutritional indicators

The results of univariate binary logistic regression analysis are shown in Table 2. We found that the lowest tertile of IGF-1 SDS levels (SGA ≦ 5: OR = 13.71, 95% CI = 4.31–43.61, p < 0.001; GNRI ≦ 91.2: OR = 7.21, 95% CI = 2.62–19.87, p < 0.001), mCI, and ICW values were related to malnutrition defined by the 7-point SGA and GNRI. Higher levels of serum hs-CRP and IL-6 were related to the malnutrition as defined by the GNRI.

Table 2 Correlations between nutritional indicators and clinical parameters
Full size table

The multivariate binary logistic regression results are presented in Table 3. After partial or full adjustment for sex, age, dialysis vintage, Kt/V (≥ 1.2), diabetes mellitus(DM), serum IL-6, serum 25 hydroxyvitamin D3, and serum phosphorus levels, we found that the lowest tertile of IGF-1 SDS levels ( SGA ≦ 5: OR = 9.69, 95% CI = 2.49–37.62, p = 0.001; GNRI ≦ 91.2: OR = 5.71, 95% CI = 1.64–19.89, p = 0.006) were independently positively associated with malnutrition defined as SGA ≤ 5 and GNRI ≤ 91.2.

Table 3 Associations between the serum IGF-1 and malnutrition defined by different nutritional indicators
Full size table

Discussion

Our study showed that the patients with lower serum IGF-1 SDS levels had lower serum levels of creatine, ceruloplasmin, albumin, and prealbumin; higher serum levels of IL-6; lower values of grip strength, pinch strength, upper arm circumference, upper arm muscle circumference, phase angle, and modified Creatinine Index (mCI); and higher malnutrition defined by the two nutrition indicators, SGA and GNRI. After adjusting for confounding factors in multivariate binary logistic regression models, we found that the lowest quartile of IGF-1 SDS levels were independently associated with malnutrition defined by SGA and GNRI.

Disorder of the GH/IGF-1 axis may occur at any stage of CKD. Notably, IGF-1 is mainly produced in the liver. Circulating insulin-like growth factor-binding proteins (IGFBPs), which are transport proteins, modulate IGF-1 bioavailability, prolong its half-life, and regulate its activity in target tissues [16]. Normal total IGF levels in uremia are thought to result from decreased IGF degradation caused by enhanced IGFBP binding. Binding to IGFBP protects IGF1 from metabolic degradation, but could also inhibit IGF1 interaction with its receptor, resulting in impaired IGF1 bioactivity [17]. It has been shown that serum IGF-I levels vary with age, gender, puberty, physiological status, and ethnicity [18]. In one study, serum IGF-1 was evaluated as the SDS of IGF-1 levels based on a Japanese population reference range established according to age and sex [10].

GH and IGF-1 resistance may adversely affect metabolism and nutritional status in adults with ESRD [19]. In another study, researchers observed that protein catabolism was also associated with impaired IGF-I signaling pathways in ESRD-related sarcopenia [20]. Serum IGF-1 is a positive marker of skeletal muscle strength and mass in patients undergoing hemodialysis [21].

In addition, phase angle (PhA) is a parameter obtained from direct measurements of bioelectrical impedance analysis (BIA). It is widely used as a marker of cellular health and has been recognized as a valuable measure for nutritional assessment, reflecting both muscle mass and muscle function [22, 23]. Extracellular body water (ECW) and intracellular body water (ICW) measured using BIA have also been introduced as markers, and studies have shown that the ECW/ICW ratio is associated with malnutrition [24]. The IGF-deficient group has been found to have a lower PhA, ICW%, and higher ECW values [25]. Other researchers have also found a correlation between serum IGF-1 and biochemical and anthropological parameters of malnutrition in patients with MHD [26, 27]. Our results showed that serum albumin, serum prealbumin, grip strength, pinch strength, body phase angle, lean body mass, mCI, and ICW were all significantly and positively correlated with serum IGF-1 SDS, and ECW% was negatively correlated with serum IGF-1 SDS, which is consistent with the results of a previous study [24,25,26,27]. However, our study did not find a correlation between serum IGF-1 SDS and the ECW/ICW ratio.

However, the underlying factors associated with these findings are complex. First, IGF-1 induces protein synthesis and myogenesis by activating the Akt/mTOR pathway, resulting in the growth and repair of skeletal muscles [20]. Together with our results, these findings indicate that serum IGF-1 levels may reflect protein anabolism and skeletal muscle function in MHD patients. Second, lower serum IGF-1 levels may reflect inflammatory status. Chronic inflammation has been reported to disrupt the GH/IGF-1 axis through relative GH and/or IGF-1 insufficiency, peripheral resistance to GH/IGF-1 receptors, inhibition of GH/IGF-1 signaling, dysregulation of IGF-binding proteins, reduced IGF-1 bioavailability, and altered gene regulation through the microRNA system [28,29,30,31,32].

Clinical data showed significantly higher levels of circulating IL-6 and IL-6 receptors in older individuals [33], with IL-6 levels being a significant predictor of sarcopenia [34]. Lastly, there is a negative correlation between plasma levels of CRP and IL-6 and rates of mixed muscle and myosin heavy chain protein synthesis in a population-based study of the general community, further illustrating the likelihood of inflammation contributing to reduced protein synthesis in sarcopenia [35]. Researchers have observed that inflammation and malnutrition in patients with ESRD often coexist and are mutually causal, a condition termed malnutrition–inflammation complex syndrome (MICS) [36]. Possible causes of MICS include comorbidities, oxidative and carbonyl stress, nutrient loss through dialysate, anorexia, loss of appetite, uremic toxins, decreased clearance of inflammatory cytokines, and volume overload [37]. In conclusion, a possible mechanism for attenuated skeletal muscle protein synthesis in CKD may be suppression of insulin/IGF-1 signaling due to metabolic acidosis, upregulated pro-inflammatory cytokine expression, and malnutrition due to anorexia [38].

Finally, in persistent systemic and tissue inflammation, the impairment of IGF-I-mediated signaling pathways preferentially increases muscle protein catabolism and inhibits muscle protein anabolism, resulting in excessive muscle protein loss, high hospitalization rates, and high rates of cardiovascular events and mortality among patients with CKD and ESRD [37, 39]. In our study, we also found that the group with a lower IGF-1 SDS had higher IL-6 levels; therefore, the results support the above notion.

This study has some limitations. First, this was a cross-sectional observational study, and causal conclusions could not be drawn. Second, given that our study had a small sample size, bias cannot be ruled out. Third, our results cannot exclude the possibility of residual confounding. Lastly, we did not measure serum IGFBPs and growth hormone levels; therefore, the relationship between the GH/IGF-1 axis and nutritional status in patients with MHD remains unclear.

Although a crossover clinical trial and randomized controlled trial have shown that the application of recombinant human growth hormone and IGF-1 could improve protein synthesis and metabolism in patients with ESRD [40, 41], these studies included only a small number of patients with short follow-up periods. However, there are currently no prospective studies with large enough sample sizes or sufficient evidence to determine the potential long-term side effects of recombinant IGF-1 or GH supplementation in patients with MHD.

Our study confirmed that higher serum IGF-1 SDS levels are significantly associated with better nutritional status as assessed by biochemical markers, anthropometric measures, body composition parameters, and comprehensive nutrition assessment. Future observational and interventional studies with larger sample sizes and multicenter are still needed to provide a theoretical basis for routine clinical application of IGF-1 preparations to improve nutritional status in patients with ESRD and dialysis.