ASSESSMENT OF LEPTIN-TO-ADIPONECTIN RATIO IN PREDICTION OF INSULIN RESISTANCE AND NUTRITION STATUS IN A GERIATRIC FEMALE POPULATION

The phenomenon of demographic ageing has been experienced for a long time, affecting many countries around the world. This direction of change is particularly noticeable in the countries of the European Union (EU), in North America and Japan. According to data from the World Health Organization (WHO), between 2015 and 2050 the proportion of the world’s population over 60 years will nearly double from 12% to 22% (1). Data provided by Eurostat show that, in 2017, 19% of the EU population was over 65 years old (2). It is also worth paying attention to the phenomenon of feminisation, which is clearly visible in older age groups. This is due to the longer, average life expectancy of women and the excess mortality of men (3, 4). In developed countries, women outlive men by a margin of four to ten years; in the developing world 58% of older people are women (5).

The ageing of societies has serious economic, social and medical consequences. Patients from the oldest age groups are characterised by polypathology and polypharmacy and a lack of clinical symptom specificity (6). The main health problems in the elderly population include hypertension, ischemic heart disease, diabetes, a variety of cancer types, asthma, dementia syndromes, depression and osteoarticular disorders. However, more and more often attention is also paid to disturbances in the nutritional status of elderly individuals, such as obesity or malnutrition. It is also worth adding that hospital malnutrition is the most frequent iatrogenic complication of therapy carried out in these facilities. Malnutrition is defined as inadequate nutritional status and undernourishment, particularly protein energy undernutrition (7). The prevalence of malnutrition varies. Kaiser et al., in their retrospective analysis from 12 countries regarding malnutrition, indicates that this problem affects approximately 23% of all older people (8).

Adipose tissue is a connective tissue consisting predominantly of adipocytes. It is diffused in different parts of the body - beneath the skin (subcutaneous adipose tissue) and around internal organs (visceral). Distinctive locations present various structures, biological functions, gene expression, metabolic and intrasecretory activity. There is considerable diversity in the amount and localisation of body fat in men and women (9, 10). Premenopausal Caucasian females mainly accumulate subcutaneous fat, but males have a higher amount of visceral adipose tissue - this characteristic adipose tissue distribution pattern is caused by oestrogens (11). This gender variation in the amount of visceral fat distribution reduces at older ages as postmenopausal women’s visceral adipose tissue increases by 50%, which is strongly associated with a greater risk of cardiovascular diseases and diabetes mellitus type 2 (9, 11). Oestrogens influence adipose tissue in many ways including impacting the production of adipokines: resistin, adiponectin, leptin and angiotensinogen, as well as suppressing inflammatory signalling and improving insulin function (11).

Visceral adipose tissue (VAT) plays a crucial role in the initiation of insulin resistance, type 2 diabetes mellitus, arterial hypertension and coronary heart disease due to the oversecretion of leptin and diminished production of adiponectin (12). Leptin plays a substantial role in glucose and triglyceride metabolism. It reduces appetite, lowers food intake and stimulates thermogenesis. Leptin counteracts insulin by suppressing its secretion, enhancing liver glycogenolysis and glucose synthesis, increasing free fatty acid oxidation and lipolysis as well as by inhibiting lipogenesis. On the other hand, many indications of adiponectin present the opposite functions to leptin (13). Adiponectin expresses anti-atherogenic, anti-inflammatory and anti-diabetic properties. Adiponectin increases lipoprotein lipase activity and enhances of the free fatty acids uptake and their oxidation in skeletal muscles. It inhibits liver gluconeogenesis, improving tissue insulin sensitivity. Adiponectin also prevents the adhesion of monocytes to blood vessels by inhibiting TNF-α (14). There are many studies which indicate that the leptin-to-adiponectin ratio (LAR) may serve as a very sensitive biomarker of the atherosclerotic process or tissue insulin sensitivity in obese individuals (11, 15). Chou et al. observed a positive correlation between the LAR, the C-reactive protein concentration and the HOMA-IR index irrespective of obesity or other metabolic risk factors in the progression of cardiovascular disease (11). The juxtaposition of both hormones mentioned in the ratio may more accurately predict atherosclerotic plaque formation than standard threat factors such as hypertension, hyperglycemia and dyslipidemia (16).

Interestingly, leptin concentration decreases markedly after the menopause. However, the obese population expresses a high leptin level in adipose tissue and in peripheral blood circulation; this indicates leptin resistance since high leptin levels did not reduce excess adiposity (13). The aim of the study was to evaluate selected adipokines and biochemical parameters as well as the leptin-to-adiponectin ratio (LAR) in respect to insulin resistance and the degree of malnutrition in the geriatric population. Since adipokines are essential for many physiological and pathological processes in the human body, we also assessed the diagnostic power of selected adipokines in the prediction of insulin resistance and malnutrition, according to HOMA-IR formula and Mini Nutritional Assessment (MNA) scale, respectively, in these individuals.

MATERIAL AND METHODS

Compliance with ethical standards

The study protocol was approved by the Bioethics Committee Collegium Medicum in Bydgoszcz; the Nicolaus Copernicus University in Torun, Poland (permission no. KB/470/2016; WN707). The design and procedures of the study were carried out in accordance with the principles of the Declaration of Helsinki. All subjects provided written informed consent to participate in this study after a full explanation of the study had been provided.

Study population

This cross-sectional study included 88 elderly women with a median age of 75 years (interquartile range 69 – 81 years). All patients were hospitalised in scheduled mode at the Geriatrics Clinic of University Hospital No. 1 in Bydgoszcz, Poland between March 2017 and November 2018. The purpose of hospitalisation was to perform a comprehensive geriatric assessment. Patients suffering from dehydration, edema, liver disease, chronic kidney disease at more than stage 2, deformation of upper limbs, cancer, bone marrow proliferative disorders, cachexia, or severe dementia were excluded from the study. Additional exclusion criteria were male gender, chronic immobilisation, prior stroke, eGFR < 60 ml/min./1.73 m2 and insulin therapy.

The patients’ dominant disease was arterial hypertension (in 70 subjects). Additionally, 28 patients had overt diabetes. According to the Geriatric Depression Scale (GDS) severe depression was identified in six cases. Eleven patients had anemia (four developed microcytic and hypochromic anemia but seven had normocytic and normochromic anemia). The average number of drugs taken permanently was five (range 0 – 15 drugs). The patients were given a comprehensive geriatric assessment (CGA) using standardised scales and tools used in geriatrics as well as a thorough assessment of their nutritional status. Anthropometric measurements along with hand dynamometry and blood biochemistry tests were carried out. Detailed data are presented in Tables 1 and 2.

Nutritional status and anthropometric phenotype of the cohort

The assessment of nutritional status was based on the Mini Nutritional Assessment (MNA), Nutrition Risk Screening 2002 (NRS 2002), the Geriatric Nutritional Risk Index (GNRI), the Malnutrition Risk Scale (SCALES), the total lymphocyte count (TLC), and the serum albumin level g/dl (17-21).

The Mini Nutritional Assessment (MNA) scale is a comprehensive, non-invasive, well-validated screening tool for the assessment of malnutrition in elderly individuals. The MNA comprises of 18 items, including the anthropometrical measurements BMI, MAC, and CC, weight loss, a global assessment (six questions related to lifestyle, medication, and mobility), a dietary questionnaire (eight questions related to the number of meals, types of food, and fluid intake), and a subjective assessment (self-perception of health and nutrition). The MNA assigns points on nutritional adequacy with a maximum score of 30 points. The MNA score distinguishes between elderly patients with adequate nutrition (scores of 24 and more), risk of malnutrition (between 17 and 23.5) and protein-calorie undernutrition (lower than 17) (22).

Anthropometric parameters were also measured: body weight (kg), height (cm) using a digital medical scale with height measurement Seca M787 (Germany) with an accuracy of 0.1 kg. Waist circumference (cm), hip circumference (cm), mid-arm circumference (MAC) (cm) and calf circumference (CC) (cm) were measured to the nearest 0.1 cm using a measuring tape Seca 203 (Germany). The body mass index (BMI, kg/m2) and waist-hip ratio (WHR) were calculated according to patterns currently in force. Using the WHO recommendations in relation to BMI, normal weight was considered as the BMI of 18.5 – 24.9 kg/m2, overweight as 25.0 – 29.9 kg/m2 and obesity was qualified in cases of the BMI ≥ 30 kg/m2 (23). In addition, a hand grip strength measurement (kg) was carried out using a Kern MAP 80K1S digital dynamometer (Germany) with an accuracy of 0.1 kg, and skinfolds (mm) were measured using a Harpenden skinfold caliper with software, Baty (United Kingdom) with an accuracy of 0.2 mm. The thickness of the four skinfolds, biceps, triceps - triceps skinfold thickness (TSF) - subscapular and suprailiac was measured in the standard manner. Using the software, the percentage of body fat BF (%), fat body mass FBM (kg) and lean body mass LBM (kg) were calculated. Each of the measurements - hand grip strength and thickness of skin folds - was performed three times with the mean value calculated. All measurements were performed by one nurse at the geriatric ward in standard conditions. Mid arm muscle circumference (MAMC, cm) was calculated from the MAC and TSF using a standard formula: MAMC = MAC – (3.1415 × TSF).

Sample collection and laboratory analysis

Venous blood collection was performed in the morning (7.00 – 8.30) according to the standards for material collection for laboratory tests. Standard blood sampling protocols were followed; patients were fasting, after 30 minutes of rest and after a 12-hour night-time fasting.

Basic laboratory blood tests, such as: albumin level, high-sensitivity C-reactive protein (hs-CRP), aspartate aminotransferase (AST), alanine aminotransferase (ALT), fasting glucose level, fasting insulin concentration, lipid profile (total cholesterol, LDL-cholesterol, HDL-cholesterol, triglycerides), folic acid and vitamin B12 (cobalamin), complete blood count (CBC) with automatic white blood cell count, were carried out during the routine testing done when admitting a patient to a ward. Serum albumin and other parameters were determined using automated analysers applying biochemical methods.

In order to determine the concentrations of adiponectin and leptin, the blood was collected in a 4.0 ml tube (BD Vacutainer® Plus Plastic Serum Tubes) without anticoagulant. It was centrifuged at 3000 × g for 15 minutes at +4ºC and subjected to further analytical procedures. Serum aliquots were centrifuged and then stored at –80ºC until analysis.

Enzyme-linked immunosorbent assay (ELISA) adipocytokines measurement

The serum concentrations of leptin and adiponectin were measured by enzyme linked immunosorbent assay (ELISA) using a commercial kit according to the manufacturer’s instructions: leptin (Human LEP protein); total adiponectin (Human ADP protein); BioVendor Research and Diagnostic products, Czech Republic. The levels of adipocytokines including leptin and adiponectin were determined at the Department of Pathophysiology at Collegium Medicum in Bydgoszcz. The assays were run by an individual blinded to the clinical data of the patients.

The non-HDL value was calculated by the use of formula:

non-HDL-C = total cholesterol – HDL-cholesterol

and

The insulin resistance index HOMA-IR was calculated using an online calculator.

The LAR (leptin-to-adiponectin ratio) was calculated as the concentration of leptin divided by the adiponectin concentration.

The estimated glomerular filtration rate (eGFR, (mL/min/per 1.73 m2) was estimated according to the 4-Variable Modification of Diet in Renal Disease (MDRD) study formula as follows:

eGFR = 186 × (serum creatinine (mg/dL)–1.154) × (age–0.203) × (0.742 gender index for female).

Statistical analysis

All statistical analyses were conducted using Statistica v. 13.1 (StatStoft®, Cracow, Poland). The assumption of normality of the data was tested by the Shapiro-Wilk test. Data are presented as percentages or medians and interquartile ranges (IQR) as well as selected results are given as mean ± standard deviation. Statistical differences between subgroups were determined with the U Mann-Whitney test or ANOVA Kruskal-Wallis test. Receiver operated characteristic (ROC) curves were used to assess the diagnostic predictive capacity of the selected biomarkers. The area under the curve (AUC) was calculated in order to estimate the diagnostic accuracy. Optimal cut-off points were determined according to Youden criteria. ROC curves for a single laboratory parameter were established and the areas under the curve with a 95% confidence interval were determined (AUC, 95% CI thresholds with sensitivity and specificity). We designed the ROC curves in order to estimate the diagnostic accuracies of the investigated variables for the prediction of malnutrition expressed by the MNA scale and insulin resistance presented in HOMA-IR form in the geriatric population. All probability (P) values < 0.05 were deemed significant.

RESULTS

Baseline characteristics

Tables 1 and 2 give an overview of the baseline characteristics of the study population. Between March 2017 and November 2018, 88 elderly women (100%) were recruited to the study with the median age of 75 years (interquartile range 69 – 81). Twenty-seven (30.7%) patients were between 60 – 69, 33 (37.5%) cases were between 70 – 79 and 28 (31.8%) subjects were between 80 – 89. Based on BMI, we observed two underweight cases, 20 patients (22.7%) had normal weight but 75% of patients (66 cases) were overweight or obese. According to the MNA scale, 56 patients were well-nourished, 29 were at risk of malnutrition and only three cases were malnourished.

Table 1. Indicators related to the state of nutrition.

Values are presented as: n (%), M, mean; SD, standard deviation; MNA, Mini Nutritional Assessment (points); NRS 2002, Nutrition Risk Screening 2002 (points); GNRI, Geriatric Nutritional Risk Index; SCALES, Malnutrition Risk Scale; TLC, total lymphocyte count (cells/mm3).

Table 2. Anthropometric data of the study population.

Values are presented as: M, mean; SD, standard deviation, BMI, body mass index; MAC, mid-arm circumference; CC, calf circumference; MAMC, mid-arm muscle circumference; HGS, handgrip strength; WHR, waist hip ratio; TSF, triceps skinfold thickness; BF, body fat; LBM, lean body mass; FBM, fat body mass.

Age-groups

Among the 88 women who participated in the present study, 27 (were 60 – 69 years old, 33 were 70 – 79 years old, and 28 were 80 – 89 years old. Table 3 presents the median and interquartile range (IQR) results for the principal variables categorised by age groups. We observed significant decreasing of albumin, alanine aminotransferase (ALT), insulin, and triglycerides concentrations with age (P = 0.0002, P = 0.0011, P = 0.0173, P = 0.0380, respectively). Interestingly, the risk of insulin resistance based on HOMA-IR index lowered with age (P = 0.0057). Additionally, the adiponectin level increased in those aged 70 – 79 years but by age 80 – 89 years it dropped slightly, which is probably the reason why dependence of adiponectin concentration did not reach the level of significance (P = 0.0851). The leptin-to-adiponectin ratio demonstrates a tendency to significance at age 70 – 79 years when the level of LAR decreases but at age 80 – 89 years it slightly rises (P = 0.0697). However, a significantly higher concentration of adiponectin and a lower LAR level were noted in those aged 70 – 79 years than in at age 60 – 69 years. Additionally, a significantly lower concentration of fasting glucose was observed in those aged 80 – 89 years than in at age 70 – 79 years.

Table 3. Adiponectin, leptin concentrations, leptin-to-adiponectin ratio, biochemical data as well as carbohydrate and lipid status according to age in the geriatric population.

LAR, leptin-to-adiponectin ratio; hsCRP, high sensitivity C-reactive protein; AST, aspartate aminotransferase; ALT, alanine-aminotransferase; HOMA-IR, homeostasis model assessment of insulin resistance; LDL-cholesterol, low-density lipoprotein; HDL-cholesterol, high-density lipoprotein.

Nutritional states (Table 4)

According to the MNA scale, more than half of the participants (63.6%) were identified as well-nourished but 33% were at risk of malnutrition. Three subjects (3.4%) were malnourished, and thereby were excluded from the statistical analysis. The concentrations of albumin, aspartate aminotransferase, alanine aminotransferase and total cholesterol were significantly lower in those patients at risk of malnutrition than in the well-nourished subjects (P = 0.0001, P = 0.0089, P = 0.0001, P = 0.0483, respectively). There was a tendency towards a lower LAR status and folic acid concentration (P = 0.0614, P = 0.0788) in the cases at risk of malnutrition.

Table 4. Measured parameters of two groups of patients according to Mini Nutritional Assessment (MNA) classification.

LAR, leptin-to-adiponectin ratio; hsCRP, high sensitivity C-reactive protein; AST, aspartate aminotransferase; ALT, alanine-aminotransferase; HOMA-IR, homeostasis model assessment of insulin resistance; LDL-cholesterol, low-density lipoprotein; HDL-cholesterol, high-density lipoprotein.

Categories of insulin resistance (Table 5)

Furthermore, we hypothesised that the concentrations of adiponectin, leptin, the LAR, as well as selected carbohydrates and lipid parameters, can vary according to the HOMA-IR score. We split a study group into two subgroups by HOMA-IR status. Thirty-two patients had normal insulin sensitivity, but 56 cases presented with insulin resistance. Significantly higher concentrations of leptin, LAR, hsCRP, fasting glucose, insulin (P = 0.0037, P = 0.0090, P = 0.0087, P = 0.0018, P = 0.0001, respectively) in the insulin resistant subgroup in respect to normal sensitivity insulin cases were noted. An essential tendency towards a lower concentration of total cholesterol (P = 0.0828) and a significantly lower concentration of folic acid were noted in insulin resistant patients as compared to those subjects normally responding to insulin (P = 0.0289).

Table 5. Serum parameters of two groups of patients according to homeostasis model assessment of insulin resistance (HOMA-IR) score.

LAR, leptin-to-adiponectin ratio; hsCRP, high sensitivity C-reactive protein; AST, aspartate aminotransferase; ALT, alanine-aminotransferase; LDL-cholesterol, low-density lipoprotein; HDL-cholesterol, high-density lipoprotein.

Categories of leptin-to-adiponectin ratio (Table 6)

Based on the division of all participants according to the leptin-to-adiponectin ratio, we obtained three subgroups: with low (< 3.0), moderate (≥ 3.0 – 4.99) and high ≥ 5.0) LAR scores. With an increase in the leptin-to-adiponectin ratio, the concentration of adiponectin decreases (P = 0.0152), but leptin and hsCRP significantly elevate (P = 0.0001, P = 0.0007, respectively). Interestingly, an increased concentration of insulin and HOMA-IR index have been noted with simultaneous elevation of the leptin-to-adiponectin ratio (P = 0.0003, P = 0.0010, respectively). Surprisingly, the leptin-to-adiponectin ratio was associated with vitamin B12 and folic acid concentrations. A significant reduction in the folic acid concentration was accompanied by an increase in the leptin-to-adiponectin ratio (P = 0.0176). The highest concentration of vitamin B12 was in cases with a low LAR score, in the moderate group its concentration was lower, however, in patients with an LAR above 5 the concentration of vitamin B12 was higher compared to the moderate LAR subgroup (P = 0.0008). Essential tendencies towards higher concentrations of alanine aminotransferase, fasting glucose, and triglycerides (P = 0.0860, P = 0.0918, P = 0.0872, respectively) with elevation of the leptin-to-adiponectin ratio were reported.

Table 6. Adiponectin and leptin values as well as carbohydrates and lipids according to the leptin/adiponectin ratio (LAR).

hsCRP, high sensitivity C-reactive protein; AST, aspartate aminotransferase; ALT, alanine-aminotransferase; HOMA-IR, homeostasis model assessment of insulin resistance; LDL-cholesterol, low-density lipoprotein; HDL-cholesterol, high-density lipoprotein.

Tests of sensitivity and specificity

The borderline of a diagnostic application of a test based on the area under the ROC curve (AUCROC) ≥ 0.5; P < 0.05 were reached for all analysed parameters except adiponectin in respect to insulin resistance prediction (Fig. 1). Based on the AUCROC, Youden Index cut-off points were specified to maximise the sum of sensitivity and specificity. LAR was the most accurate biomarker with an area under the curve, AUCROC = 0.705; (95% CI: 0.573 – 0.814; P = 0.0004). Using the Youden Index cut-off value, we identified LAR at 3.85 with sensitivity of 75% and specificity of 75% as the best cut-off value in order to discriminate between patients with adequate insulin sensitivity and those with insulin resistance. The LAR presents the highest AUCROC, the next significant values demonstrate those for leptin (0.687; 95% confidence interval 0.567 – 0.807; P = 0.0022); glucose (0.668; 95% confidence interval 0.593 – 0.817; P = 0.0086); hsCRP (0.668; 95% confidence interval 0.549 – 0.787; P = 0.0055) and folic acid (0.651; 95% confidence interval 0.519 – 0.781; P = 0.0246). The Youden Index cut-off value for leptin was 9.12 ng/mL with 68% specificity and 78% sensitivity; for glucose it was 89 mg/dL with 60% specificity and 77% sensitivity; for hsCRP it was 1.22 mg/L with 70% specificity and 65% sensitivity and for folic acid it was 8.3 ng/mL with 82% specificity and 50% sensitivity.

Fig. 1. A graph showing the six ROC curves for leptin, adiponectin, leptin-to-adiponectin ratio, fasting glucose, folic acid, hsCRP with different values of the area under the ROC curve (AUCROC). The most accurate indicator for insulin resistance according to HOMA-IR (< 1.5 normal) and with insulin resistance (HOMA-IR ≥ 1.5) presents an area of 0.705 under the curve, which was reached for the leptin/adiponectin ratio (P = 0.0004).

The second scheme (Fig. 2) was performed in order to establish which analysed parameters may precisely predict malnutrition risk, identified according to the MNA scale in the geriatric group. The areas under the ROC curve for leptin, adiponectin, LAR, glucose, total cholesterol, folic acid, albumin, and hsCRP were estimated. The greatest diagnostic power present albumin concentration with AUCROC =0.761; 95% confidence interval 0.658 – 0.862; P < 0.0001) and then LAR 0.718; 95% confidence interval 0.643 – 0.828; P = 0.0006); total cholesterol 0.631; 95% confidence interval 0.505 – 0.758; P = 0.0422) in discriminating between well-nourished patients and those at risk of malnutrition. Youden Index cut-off points were identified for albumin 4 g/dL with 92% specificity and 53% sensitivity, LAR 4.67 with 75% specificity and 58% sensitivity and total cholesterol 165 mg/dL with 65% specificity and 65% sensitivity.

Fig. 2. A graph showing the eight ROC curves for leptin, adiponectin, leptin-to-adiponectin ratio, fasting glucose, total cholesterol, folic acid, albumin concentration, hsCRP with different values of the area under the ROC curve (AUCROC). The most accurate indicator for malnutrition risk revealed an area of 0.761 under the curve, which was reached for albumin (P < 0.0001).

DISCUSSION

It is worth emphasising that body composition alters with age and it is connected with impairment of energy homeostasis and abnormalities in carbohydrate and lipid metabolism. Obesity, particularly visceral adiposity, is a powerful determining factor for cardiovascular disease (CVD), insulin resistance, chronic low-grade inflammation and, finally, type 2 diabetes mellitus development (24-26). Adipose tissue is considered as a bioactive endocrine organ in the human body producing several adipokines, which are involved in many physiological processes including the regulation of homeostasis, appetite, and lipid and carbohydrate metabolism (25). There are two major adipokines which metabolically work in an opposite manner: adiponectin and leptin. Adiponectin supports adequate insulin sensitivity; it protects against insulin resistance. It improves fatty acid and glucose metabolism (25, 27, 28). However, leptin activates liver glycogenolysis and glucose synthesis, enhancing free fatty acid oxidation and lipolysis as well as by inhibiting lipogenesis. Additionally, leptin activates neutrophils, monocytes and natural killer cells and stimulates the production of pro-inflammatory cytokines and then exerts a proatherogenic effect (27). Ferenc at al. claimed that a chronic local low-grade inflammation in the IUGR liver stimulates the development of insulin resistance (29).

The assumption of this study was to analyse several adipokines, lipid and carbohydrate metabolism parameters as well as albumin, folic acid and vitamin B12 concentrations in a geriatric cohort in respect to the risk of malnutrition expressed by the MNA scale and insulin resistance presented by the HOMA-IR index. Due to the fact that adiponectin and leptin levels are higher in women, we eliminated gender-related variabilities and considered only the female population (27, 28, 30, 31). Also, patients with edema and water-electrolyte imbalance were excluded due to the fact that both abnormalities are frequent in the elderly, and both influence weight and other anthropometric values.

Ageing-associated changes in body composition, adipokine levels and insulin sensitivity

The study group was divided into three subgroups according to age: 27 were 60 – 69 years old, the 70 – 79 year-old group was made up of 33 women, and 28 cases were 80 – 89 years old. In our study, we observed that the adiponectin, leptin concentrations and leptin-to-adiponectin ratio did not significantly differ between the three age subgroups. However, a significantly higher concentration of adiponectin and a lower LAR level were noted in those aged 70 – 79 years than in at age 60 – 69 years. According to our data we speculate that, in the case of females ageing, it is associated with a lower risk of endothelial dysfunction, insulin resistance, hyperglycemia, hypertriglyceridemia, but increases the prevalence of hypoalbuminemia and malnutrition. Our results are extremely interesting and crucial due to the fact that the number of other studies, which analyse the population aged 80 – 89 years, is limited. Our observation is consistent with Paolisso et al. who observed that glucose tolerance and insulin sensitivity were better maintained in healthy centenarians than in elderly individuals aged over 75 years (32). Takayama et al. analysed a representative centenarian group and observed that the morbidity of diabetes mellitus type 2 was merely 6.0%, which was lower in respect to people in their 60s (15.3%) and 70s (14.7%) in Japan (33). Taken together, we speculate that during ageing, compensatory mechanisms are mobilised and the risk of metabolic disorders lowers. The profile of adipokine secretion changes with age, thus, the secretion of leptin drops but adiponectin production elevates. This status is associated with a better metabolic index followed by higher high-density lipoprotein-cholesterol (HDL-C) levels and lower hemoglobin A1c, and is negatively correlated with C-reactive protein in 66 female centenarians (34), which is in line with our results for the women aged 80 – 89 years as well as this phenotype suggests improvement in endothelial function (35). Peng and co-workers observed that higher level of HOMA-IR was significantly associated with frailty in the elderly group, but this association was not seen in the middle-aged population (36).

Malnutrition of ageing and its indicators in elderly women

The prevalence of malnutrition in the senior population is high. Malnutrition is associated with an increased tendency to infection, sarcopenia, impairment of wound healing, reduced quality-of-life as well as shorter survival rates (24, 25, 37). The mortality in malnourished patients was 12.4% versus 4.7% in respect to well-nourished patients (38). Easily applicable and sensitive indicators of malnutrition still need to be found. In the geriatric population, the mini-nutritional assessment (MNA) score is a common clinical scale used for the assessment of the nutritional phenotype. In the present study, the concentration of albumin, aspartate aminotransferase, alanine aminotransferase and total cholesterol were significantly lower in patients at risk of malnutrition (MNA scale: 17 – 23.5 points) than well-nourished subjects (MNA scale: > 24 points). There was a tendency towards a lower LAR status and folic acid concentration in cases at risk of malnutrition. According to our study, we suggest that the albumin level is a sensitive biomarker of a malnourished state. Also, this suggests that individuals who are MNA-scored as at risk of malnutrition are conclusively protein malnourished. Albumin is a well-known negative acute-phase protein and its level is connected with selected pro-inflammatory agents and drugs, which are metabolised by the liver. Many diseases including hepatic failure, burns sepsis, trauma, post-surgery states and cancer lead to decreased albumin levels, thus it becomes an unsuitable serum biomarker for malnutrition. Primarily in the elderly group, this is due to normally existing co-morbidities (37). Kuzuya et al. demonstrated that applying a serum albumin concentration < 3.5 g/dL as a determinant for malnutrition presents a low specificity for recognising the nutritional phenotype in the elderly group. Authors claim that up to 80% of the patients were improperly defined as malnourished (39). In order to support our study, we carried out additional statistical analysis, designing ROC curves to estimate the prediction values for analysed parameters in order to distinguish between well-nourished and at risk of malnutrition cases. We found that the optimal cut-off for albumin was 4 g/dL with 53% sensitivity and 92% specificity, for LAR it was 4.67 with 58% sensitivity and 75% specificity and for total cholesterol 165 mg/dL with 65% sensitivity and 65% specificity. However, identification of the cut-off values for these parameters in the geriatric population need confirmation, since the adequate thresholds require longer-term studies on large sized populations to validate them. Additionally, the LAR and total cholesterol could serve as better nutritional markers than fasting glucose and insulin concentrations in these contexts. The LAR decreases as undernourishment becomes more pronounced. We speculate that the LAR is a more stable indicator of nutritional status then albumin concentration in the senior population.

Leptin-to-adiponectin ratio of ageing and its indicators in elderly women

To our knowledge, our study is the first one which analyses different leptin-to-adiponectin ratios (low 3, moderate ≥ 3.0 – 4.99 and high ≥ 5) in the context of analysed parameters in the geriatric population and demonstrated the dependencies between them. It is well known that when the LAR increases, the adiponectin concentration is reduced but the leptin level increases. Interestingly, with the LAR elevates the concentrations of hsCRP and insulin and the HOMA-IR index increase, thus, the risk of hyperglycemia, and hypertriglyceridemia increases. The stated observations are consistent with those of Kotani and co-workers. They also observed that the LAR is positively associated with adiposity, and most studies supported the concept (16). However, in our study we observed that the concentrations of vitamin B12 and folic acid decrease with a simultaneous increase of the LAR. Interestingly, we observed that a higher LAR indicates a proatherogenic and prodiabetic phenotype in elderly women as well as being associated with less folic acid and vitamin B12 absorption or the impaired metabolism of both components. The novel aspect of this study is an identification of single diagnostic parameters, which can prognosticate an insulin resistance and malnutrition in the elderly population. Our study indicates that the leptin-to-adiponectin ratio may be used as a suitable, non-invasive prognostic biomarker of insulin resistance. LAR reached the most accurate AUCROC =0.705; (95% CI: 0.573 – 0.814; P = 0.0004) for insulin resistance prediction, with a cut-off value at 3.85 with 75% sensitivity and 75% specificity. Our results are in line with those of Kang and co-workers. However, their study was conducted in individuals with metabolic syndrome (31). Leptin and adiponectin exert a contrary effect on glucose, lipid metabolism and inflammation regulation. Both adipokines cooperate in dyslipidemia and insulin resistance risk as well as in type 2 diabetes mellitus development (27, 35). An adequate level of adiponectin maintains the endothelium in a quiescent state, on the other hand, leptin supports endothelial dysfunction and enhances the pro-atherogenic index. Furthermore, our analysis using ROC confirmed that the LAR had a better ability to classify insulin resistance in the elderly than adiponectin or leptin considered independently.

The authors acknowledge the following limitations of the study. The disadvantage was the relatively small sample size in the age brackets used in our study. The study group consisted of elderly females who had comorbid diseases including overt diabetes, but insulin therapy was not used in those patients. Due to the cross-sectional design of this study, it is not possible to establish a causal relation between the LAR and insulin resistance and malnutrition.

Conclusions

According to our results, we suggest that the LAR may be considered as an efficient parameter in monitoring insulin resistance status. Thus, the juxtaposition of leptin and adiponectin in the ratio more precisely identifies the atherogenic, diabetic, and inflammatory profile regardless of the analysed population. Our findings suggest that leptin and adiponectin imbalance, expressed by an increase in LAR levels, may play a crucial role in the development of insulin resistance in the geriatric population. In conclusion, we suggest that the leptin-to-adiponectin ratio is more suitable as a marker of insulin resistance and malnutrition status in the elderly than both adipokines separately.

Acknowledgements: We would like to thank all nurses at the Geriatrics Clinic of University Hospital No. 1 in Bydgoszcz, and participants of this study for their patience, time and interest. Special thanks also to Barbara Goralczyk and Aleksandra Rutkowska for support in collecting data.

Conflict of interests: None declared.

REFERENCES

1. World Health Organization. Ageing and health. Available from: http://www.who.int/news-room/fact-sheets/detail/ ageing-and-health. Accessed August 14, 2019.
2. Eurostat. Population structure and ageing. Available from: http://ec.europa.eu/eurostat/statistics-explained/index .php/Population_structure_and_ageing. Accessed August 14, 2019.
3. Wieczorowska-Tobis K. Why males live shorter? [in Polish]. Now Lekarskie 2012; 81: 386-389.
4. White A, McKee M, de Sousa B, et al. An examination of the association between premature mortality and life expectancy among men in Europe. Eur J Public Health 2014; 24: 673-679.
5. Women’s UN Report Network. Challenges facing older women - the feminization of aging. Available from: https://wunrn.com/2017/02/challenges-facing-older-women-the-feminization-of-aging/. Accessed August 14, 2019.
6. Masnoon N, Shakib S, Kalisch-Ellett L, Caughey GE. What is polypharmacy? A systematic review of definitions. BMC Geriatr 2017; 17: 230. doi:10.1186/s12877-017-0621-2
7. Chen CC, Schilling LS, Lyder CH. A concept analysis of malnutrition in the elderly. J Adv Nurs 2001; 36: 131-142.
8. Kaiser MJ, Bauer JM, Ramsch C, et al. Frequency of malnutrition in older adults: a multinational perspective using the mini nutritional assessment. J Am Geriatr Soc 2010; 58: 1734-1738.
9. Gomez-Hernandez A, Beneit N, Diaz-Castroverde S, Escribano O. Differential role of adipose tissues in obesity and related metabolic and vascular complications. Int J Endocrinol 2016; 2016: 1216783. doi:10.1155/2016/1216783
10. Fuente-Martin E, Argente-Arizon P, Purificacion R, Argente J, Chowen JA. Sex differences in adipose tissue: it is not only a question of quantity and distribution. Adipocyte 2013; 2: 128-134.
11. Chou HH, Hsu LA, Wu S, Teng MS, Sun YC, Ko YL. Leptin-to-adiponectin ratio is related to low grade inflammation and insulin resistance independent of obesity in non-diabetic Taiwanese: a cross-sectional cohort study. Acta Cardiol Sin 2014; 30: 204-214.
12. Doumatey AP, Bentley AR, Zhou J, Huang H, Adeyemo A, Rotimi CN. Paradoxical hyperadiponectinemia is associated with the metabolically healthy obese (MHO) phenotype in African Americans. J Endocrinol Metab 2012; 2: 51-65.
13. Park HK, Ahima RS. Physiology of leptin: energy homeostasis, neuroendocrine function and metabolism. Metabolism 2015; 64: 24-34.
14. Esmaili S, Hemmati M, Karamian M. Physiological role of adiponectin in different tissues: a review. Arch Physiol Biochem 2020; 126: 67-73.
15. Iwan-Zietek I, Ruszkowska-Ciastek B, Michalska M, et al. Association of adiponectin and leptin-to-adiponectin ratio with the function of platelets in morbidly obese patients. J Physiol Pharmacol 2016; 67: 555-561.
16. Kotani K, Sakane N. Leptin: adiponectin ratio and metabolic syndrome in the general Japanese population. Korean J Lab Med 2011; 31: 162-166.
17. Guigoz Y, Vellas B, Garry PJ. Mini nutritional assessment: a practical assessment tool for grading the nutritional status of elderly patients. Facts Res Gerontol 1994; 4: 15-59.
18. Kondrup J, Rasmussen HH, Hamberg O, Stanga Z. Nutritional risk screening (NRS 2002): a new method based on an analysis of controlled clinical trials. Clin Nutr 2003; 22: 321-336.
19. Bouillanne O, Morineau G, Dupont C, et al. Geriatric Nutritional Risk Index: a new index for evaluating at-risk elderly medical patients. Am J Clin Nutr 2005; 82: 777-783.
20. Morley JE. Death by starvation. a modern American problem? J Am Geriatr Soc 1989; 37: 184-185.
21. DeChico RS, Matarase LE, Seidner D, Steiger E. Biochemical assessment and moitoring of nutritional status. In: The Biology and Practice of Current Nutritional Support. Latifi R, Dudrick JS, (eds.) Georgetown, Landes Bioscience, 2003, 126-144.
22. Vellas B, Guigoz Y, Garry PJ, et al. The Mini Nutritional Assessment (MNA) and its use in grading the nutritional state of elderly patients. Nutrition 1999; 15: 116-122.
23. World Health Organization. Obesity: Preventing and Managing the Global Epidemic: Report of a WHO Consultation on Obesity, Geneva, 3 – 5 June 1997. Available from: https://apps.who.int/iris/handle/10665/63854. Accessed August 14, 2019.
24. Liguori I, Russo G, Aran L, et al. Sarcopenia: assessment of disease burden and strategies to improve outcomes. Clin Interv Aging 2018; 13: 913-927.
25. Arai Y, Kamide K, Hirose N. Adipokines and aging: findings from centenarians and the very old. Front Endocrinol (Lausanne) 2019; 10: 142. doi: 10.3389/fendo.2019.00142
26. Zawiejska A, Wender-Ozegowska E, Grewling-Szmit K, Brazert M, Brazert J. Short-term antidiabetic treatment with insulin or metformin has a similar impact on the components of metabolic syndrome in women with gestational diabetes mellitus requiring antidiabetic agents: results of a prospective, randomised study. J Physiol Pharmacol 2016; 67: 227-233.
27. Coimbra S, Brandzo Proenca J, Santos-Silva A, Neuparth MJ. Adiponectin, leptin, and chemerin in elderly patients with type 2 diabetes mellitus: a close linkage with obesity and length of the disease. Biomed Res Int 2014; 2014: 701915. doi: 10.1155/2014/701915
28. Guo Y, Liu L, Wang J. Adiponectin and the risk of new-onset atrial fibrillation: a meta-analysis of prospective cohort studies. Biosci Rep 2019; 39: BSR20182284. doi: 10.1042/BSR20182284
29. Ferenc K, Pietrzak P, Wierzbicka M, Matyba P, Grzesiuk E, Gajewski Z, Zabielski R. Alterations in the liver of intrauterine growth retarded piglets may predispose to development of insulin resistance and obesity in later life. J Physiol Pharmacol 2018; 69: 211-218.
30. Song HJ, Oh S, Quan S, et al. Gender differences in adiponectin levels and body composition in older adults: Hallym aging study. BMC Geriatr 2014; 14: 8. doi: 10.1186/1471-2318-14-8
31. Kang DR, Yadav D, Koh SB, Kim JY, Ahn SV. Impact of serum leptin to adiponectin ratio on regression of metabolic syndrome in high-risk individuals: The ARIRANG study. Yonsei Med J 2017; 58: 339-346.
32. Paolisso G, Gambardella A, Ammendola S, et al. Glucose tolerance and insulin action in healthy centenarians. Am J Physiol Endocrinol Metab 1996; 270: E890-E894.
33. Takayama M, Hirose N, Arai Y, et al. Morbidity of Tokyo-area centenarians and its relationship to functional status. J Gerontol A Biol Sci Med Sci 2007; 62: 774-782.
34. Arai Y, Nakazawa S, Kojima T, et al. High adiponectin concentration and its role for longevity in female centenarians. Geriatr Gerontol Int 2006; 6: 32-39.
35. Gluszewska A, Gryglewska B, Rewiuk K, et al. Arterial structure and function and its short- and long-term changes after bariatric surgery. J Physiol Pharmacol 2019; 70: 909-919.
36. Peng PS, Kao TW, Chang PK, Chen WL, Peng PJ, Wu LW. Association between HOMA-IR and frailty among U.S. middle-aged and elderly population. Sci Rep 2019; 9: 4238. doi: 10.1038/s41598-019-40902-1
37. Bharadwaj S, Ginoya S, Tandon P, et al. Malnutrition: laboratory markers vs nutritional assessment. Gastroenterol Rep (Oxf) 2016; 4: 272-280.
38. Correia MI, Waitzberg DL. The impact of malnutrition on morbidity, mortality, length of hospital stay and costs evaluated through a multivariate model analysis. Clin Nutr 2003; 22: 235-239.
39. Kuzuya M, Izawa S, Enoki H, Okada K, Iguchi A. Is serum albumin a good marker for malnutrition in the physically impaired elderly? Clin Nutr 2007; 26: 84-90.