J Pediatr Endocr Met 2015; 28(5-6): 589–596
Irine-Ikbale Sakou*, Theodora Psaltopoulou, Theodoros N. Sergentanis, Kyriaki Karavanaki, Feneli Karachaliou, Ioannis Ntanasis-Stathopoulos, Stamatios Tzanninis, Triantafyllia Sdogou, Donald Greydanus and Artemis Tsitsika
Insulin resistance and cardiometabolic risk factors in obese children and adolescents: a hierarchical approach Abstract Background: This study aims to globally assess the network of insulin resistance (IR)-related factors in a sample of overweight and obese Greek youths. Methods: A total of 185 subjects were examined, and IR was quantified by homeostasis model assessment (HOMAIR). Multivariate hierarchical approach was performed, and five distinct levels were recognized, namely, immutable demographic features and early life parameters, current anthropometric measures, IR, unfavorable clinical conditions, and social parameters. Two analyses were performed based on HOMA-IR cut-off values (3.16 and, as an alternative, 3.99). Results: Obesity was associated with IR (adjusted OR = 3.19, 95% CI: 1.12–9.09). IR steadily predicted low HDL (adjusted OR = 5.75, 95% CI: 1.58–20.87), hypertriglyceridemia (adjusted OR = 10.28, 95% CI: 1.18–89.55), and systolic hypertension. At the alternative analysis, IR was also associated with older age, older age at menarche, hyperuricemia, and low school grades.
*Corresponding author: Irine-Ikbale Sakou, Adolescent Health Unit (AHU), University of Athens, Second Department of Pediatrics, “P&A Kyriakou”, Children’s Hospital, Athens, Greece, Phone: (0030) 2107710824, Fax: (0030) 210-7710824, E-mail: [email protected] Theodora Psaltopoulou, Theodoros N. Sergentanis, Ioannis Ntanasis-Stathopoulos and Stamatios Tzanninis: Department of Hygiene, Epidemiology and Medical Statistics, Medical School, National University of Athens, Greece Kyriaki Karavanaki and Triantafyllia Sdogou: Diabetic Clinic, Second Department of Pediatrics, “P&A Kyriakou”, Children’s Hospital, National and Kapodistrian University of Athens School of Medicine, Athens, Greece Feneli Karachaliou: Department of Endocrinology, “P&A Kyriakou”, Children’s Hospital, Athens, Greece Donald Greydanus: Michigan State University, Department of Pediatrics and Human Development, College of Human Medicine, MI, USA Artemis Tsitsika: Adolescent Health Unit (AHU), University of Athens, Second Department of Pediatrics, “P&A Kyriakou”, Children’s Hospital, Athens, Greece
Conclusion: Emerging on the grounds of obesity, IR confers risks for dyslipidemia and hypertension at a relatively early age. Along with weight loss, interventions targeted at IR are required to prevent cardiometabolic risk in adolescence. Keywords: adolescents; cardiometabolic risk factors; insulin resistance; obesity. DOI 10.1515/jpem-2014-0431 Received October 13, 2014; accepted January 21, 2015; previously published online March 7, 2015
Introduction Obesity among children and adolescents has become a major health problem today and represents the most important chronic disease in these age groups (1). In Greece, the prevalence of overweight and obese children aged 6 to 12 years is high (31%), particularly in boys (2). Obesity is associated with the development of metabolic complications, including insulin resistance (IR), hypertension and dyslipidemia, which increase the risk for type 2 diabetes and early cardiovascular disease (3). IR is one of the important consequences of obesity, and is defined as the inability of insulin to increase the glucose uptake by skeletal muscle and adipose tissue and the inability to suppress hepatic glucose production and output (4). This condition has a multifactorial etiology, including genetic predisposition, hormonal factors, and environmental influences. Several factors, such as small for gestational age birth weight, low and high birth weight, low rates of breast-feeding, excessive weight gain during childhood, puberty, premature pubarche, ethnicity and gender, have all been shown to be associated with IR risk in childhood (4–6). IR in obese youth is related to cardiometabolic risk factors in a variety of ethnic groups (7, 8). Other disorders, such as fatty liver disease, which
Brought to you by | New York University Bobst Library Technical Services Authenticated Download Date | 5/20/15 1:56 PM
590 Sakou et al.: Insulin resistance in obese adolescents are usually accompanied by increased levels of hepatic enzymes and polycystic ovary syndrome in adolescent girls, are also associated with IR (4, 6). From a methodological point of view, IR represents a rather challenging notion. On the one hand, obesity represents the primary risk factor for IR emergence (9) and may also be predicted by immutable demographic factors and early life parameters (10). On the other hand, IR per se may interact synergistically with the obesity-related risk factors, thus contributing to various unfavorable clinical conditions. Moreover, interference with social parameters, such as educational attainment, cannot be excluded (11, 12). The dynamic network of these risk factors indicates that a multivariate stepwise model may not adequately describe the phenomenon and that a hierarchical approach may be preferable (13, 14). The hierarchical approach implies the classification of variables into distinct levels, with each ‘lower’ level exerting its effects upon the ‘higher’ levels. In our analysis, parameters were clustered into five levels as follows: immutable demographic features and early life parameters, current anthropometric measures, IR, unfavorable clinical conditions, and social parameters (Figure 1). Standards for IR in children, with definitions for normal and abnormal levels, have not been formally established (4, 15, 16). Although a number of methods for IR evaluation have been used, the homeostasis model assessment of IR (HOMA-IR) is a widely used surrogate measure of IR in children (17). In Greece, studies have focused on the epidemiology, prevalence and risk factors of childhood obesity, as well as
on the frequency of metabolic syndrome (2, 18–20). To our knowledge, no studies in Mediterranean countries have addressed the interplay of IR with a wider range of risk factors – both cardiometabolic and socio-demographic. The aim of this study was to globally assess the network of IR-related factors in a sample of overweight and obese Greek children and adolescents.
Patients and methods A total of 185 five children and adolescents aged 8 to 19 years with a mean age of 13.6±2.2 years were assessed at the Adolescent Health Unit (AHU), between January 2008 and December 2012, for evaluation of increased weight-obesity. AHU is a model public health unit located in Athens, which provides primary, secondary, and preventive health care services in adolescents all over the country. We performed a search in the electronic databases of the AHU with the diagnosis ‘overweight’ or ‘obesity’. The identified medical records of patients were reviewed by the research team, after which data on age, gender, weight, height, birth weight, gestational age of delivery, breastfeeding, number of siblings, blood pressure, biochemical parameters [e.g., fasting blood glucose and insulin levels, fasting lipid profile, serum levels of serum glutamic-pyruvic transaminase (SGPT), serum glutamic oxaloacetic transaminase (SGOT), and uric acid], menarche age, hirsutism, acne, and educational attainment (school grades) were collected. Given the retrospective review of the patient charts, data from some examined parameters were occasionally missing. The protocol of this study was approved by the Ethics Committee of “P&A Kyriakou” Children’s Hospital. Due to the retrospective retrieval of data, consent was not required based on the Helsinki Declaration of 1961.
Social parameters (school grades) Unfavorable clinical conditions (hypertension, dyslipidemia, hyperuricemia, hypertransaminasemia, acne, hirsutism)
Insulin resistance (HOMA-IR)
Current anthropometric measures (BMI, weight, height)
Immutable demographic features and early life parameters (sex, age, siblings, birth weight, gestational age, breastfeeding, menarche age)
Figure 1: The conceptual hierarchical frame work of the study.
Brought to you by | New York University Bobst Library Technical Services Authenticated Download Date | 5/20/15 1:56 PM
Sakou et al.: Insulin resistance in obese adolescents 591
Evaluation of risk factors – Definitions The body mass index (BMI) was calculated [weight (kg)/ height (m2)], and obesity was defined as a BMI above the 95th percentile for age and gender (21). IR was assessed using the HOMA-IR, which was calculated using the following formula: HOMA-IR = [fasting glucose (mmol/L) × fasting insulin (μU/mL)÷22.5] (22). As commonly used, we adopted the 3.16 cut-off for our analysis (15). Additionally, we analyzed our data with the 3.99 cut-off (16) to ensure an objective and more inclusive approach. Arterial hypertension was defined as systolic blood pressure (SBP) and/or diastolic blood pressure (DBP) ≥ 95th percentile for age, gender and height (23). Abnormal fasting levels of lipids and other biochemical parameters were defined as follows: Triglycerides (TG) ≥ 150mg/dL (24); High-density lipoprotein cholesterol (HDL-C)
Irine-Ikbale Sakou*, Theodora Psaltopoulou, Theodoros N. Sergentanis, Kyriaki Karavanaki, Feneli Karachaliou, Ioannis Ntanasis-Stathopoulos, Stamatios Tzanninis, Triantafyllia Sdogou, Donald Greydanus and Artemis Tsitsika
Insulin resistance and cardiometabolic risk factors in obese children and adolescents: a hierarchical approach Abstract Background: This study aims to globally assess the network of insulin resistance (IR)-related factors in a sample of overweight and obese Greek youths. Methods: A total of 185 subjects were examined, and IR was quantified by homeostasis model assessment (HOMAIR). Multivariate hierarchical approach was performed, and five distinct levels were recognized, namely, immutable demographic features and early life parameters, current anthropometric measures, IR, unfavorable clinical conditions, and social parameters. Two analyses were performed based on HOMA-IR cut-off values (3.16 and, as an alternative, 3.99). Results: Obesity was associated with IR (adjusted OR = 3.19, 95% CI: 1.12–9.09). IR steadily predicted low HDL (adjusted OR = 5.75, 95% CI: 1.58–20.87), hypertriglyceridemia (adjusted OR = 10.28, 95% CI: 1.18–89.55), and systolic hypertension. At the alternative analysis, IR was also associated with older age, older age at menarche, hyperuricemia, and low school grades.
*Corresponding author: Irine-Ikbale Sakou, Adolescent Health Unit (AHU), University of Athens, Second Department of Pediatrics, “P&A Kyriakou”, Children’s Hospital, Athens, Greece, Phone: (0030) 2107710824, Fax: (0030) 210-7710824, E-mail: [email protected] Theodora Psaltopoulou, Theodoros N. Sergentanis, Ioannis Ntanasis-Stathopoulos and Stamatios Tzanninis: Department of Hygiene, Epidemiology and Medical Statistics, Medical School, National University of Athens, Greece Kyriaki Karavanaki and Triantafyllia Sdogou: Diabetic Clinic, Second Department of Pediatrics, “P&A Kyriakou”, Children’s Hospital, National and Kapodistrian University of Athens School of Medicine, Athens, Greece Feneli Karachaliou: Department of Endocrinology, “P&A Kyriakou”, Children’s Hospital, Athens, Greece Donald Greydanus: Michigan State University, Department of Pediatrics and Human Development, College of Human Medicine, MI, USA Artemis Tsitsika: Adolescent Health Unit (AHU), University of Athens, Second Department of Pediatrics, “P&A Kyriakou”, Children’s Hospital, Athens, Greece
Conclusion: Emerging on the grounds of obesity, IR confers risks for dyslipidemia and hypertension at a relatively early age. Along with weight loss, interventions targeted at IR are required to prevent cardiometabolic risk in adolescence. Keywords: adolescents; cardiometabolic risk factors; insulin resistance; obesity. DOI 10.1515/jpem-2014-0431 Received October 13, 2014; accepted January 21, 2015; previously published online March 7, 2015
Introduction Obesity among children and adolescents has become a major health problem today and represents the most important chronic disease in these age groups (1). In Greece, the prevalence of overweight and obese children aged 6 to 12 years is high (31%), particularly in boys (2). Obesity is associated with the development of metabolic complications, including insulin resistance (IR), hypertension and dyslipidemia, which increase the risk for type 2 diabetes and early cardiovascular disease (3). IR is one of the important consequences of obesity, and is defined as the inability of insulin to increase the glucose uptake by skeletal muscle and adipose tissue and the inability to suppress hepatic glucose production and output (4). This condition has a multifactorial etiology, including genetic predisposition, hormonal factors, and environmental influences. Several factors, such as small for gestational age birth weight, low and high birth weight, low rates of breast-feeding, excessive weight gain during childhood, puberty, premature pubarche, ethnicity and gender, have all been shown to be associated with IR risk in childhood (4–6). IR in obese youth is related to cardiometabolic risk factors in a variety of ethnic groups (7, 8). Other disorders, such as fatty liver disease, which
Brought to you by | New York University Bobst Library Technical Services Authenticated Download Date | 5/20/15 1:56 PM
590 Sakou et al.: Insulin resistance in obese adolescents are usually accompanied by increased levels of hepatic enzymes and polycystic ovary syndrome in adolescent girls, are also associated with IR (4, 6). From a methodological point of view, IR represents a rather challenging notion. On the one hand, obesity represents the primary risk factor for IR emergence (9) and may also be predicted by immutable demographic factors and early life parameters (10). On the other hand, IR per se may interact synergistically with the obesity-related risk factors, thus contributing to various unfavorable clinical conditions. Moreover, interference with social parameters, such as educational attainment, cannot be excluded (11, 12). The dynamic network of these risk factors indicates that a multivariate stepwise model may not adequately describe the phenomenon and that a hierarchical approach may be preferable (13, 14). The hierarchical approach implies the classification of variables into distinct levels, with each ‘lower’ level exerting its effects upon the ‘higher’ levels. In our analysis, parameters were clustered into five levels as follows: immutable demographic features and early life parameters, current anthropometric measures, IR, unfavorable clinical conditions, and social parameters (Figure 1). Standards for IR in children, with definitions for normal and abnormal levels, have not been formally established (4, 15, 16). Although a number of methods for IR evaluation have been used, the homeostasis model assessment of IR (HOMA-IR) is a widely used surrogate measure of IR in children (17). In Greece, studies have focused on the epidemiology, prevalence and risk factors of childhood obesity, as well as
on the frequency of metabolic syndrome (2, 18–20). To our knowledge, no studies in Mediterranean countries have addressed the interplay of IR with a wider range of risk factors – both cardiometabolic and socio-demographic. The aim of this study was to globally assess the network of IR-related factors in a sample of overweight and obese Greek children and adolescents.
Patients and methods A total of 185 five children and adolescents aged 8 to 19 years with a mean age of 13.6±2.2 years were assessed at the Adolescent Health Unit (AHU), between January 2008 and December 2012, for evaluation of increased weight-obesity. AHU is a model public health unit located in Athens, which provides primary, secondary, and preventive health care services in adolescents all over the country. We performed a search in the electronic databases of the AHU with the diagnosis ‘overweight’ or ‘obesity’. The identified medical records of patients were reviewed by the research team, after which data on age, gender, weight, height, birth weight, gestational age of delivery, breastfeeding, number of siblings, blood pressure, biochemical parameters [e.g., fasting blood glucose and insulin levels, fasting lipid profile, serum levels of serum glutamic-pyruvic transaminase (SGPT), serum glutamic oxaloacetic transaminase (SGOT), and uric acid], menarche age, hirsutism, acne, and educational attainment (school grades) were collected. Given the retrospective review of the patient charts, data from some examined parameters were occasionally missing. The protocol of this study was approved by the Ethics Committee of “P&A Kyriakou” Children’s Hospital. Due to the retrospective retrieval of data, consent was not required based on the Helsinki Declaration of 1961.
Social parameters (school grades) Unfavorable clinical conditions (hypertension, dyslipidemia, hyperuricemia, hypertransaminasemia, acne, hirsutism)
Insulin resistance (HOMA-IR)
Current anthropometric measures (BMI, weight, height)
Immutable demographic features and early life parameters (sex, age, siblings, birth weight, gestational age, breastfeeding, menarche age)
Figure 1: The conceptual hierarchical frame work of the study.
Brought to you by | New York University Bobst Library Technical Services Authenticated Download Date | 5/20/15 1:56 PM
Sakou et al.: Insulin resistance in obese adolescents 591
Evaluation of risk factors – Definitions The body mass index (BMI) was calculated [weight (kg)/ height (m2)], and obesity was defined as a BMI above the 95th percentile for age and gender (21). IR was assessed using the HOMA-IR, which was calculated using the following formula: HOMA-IR = [fasting glucose (mmol/L) × fasting insulin (μU/mL)÷22.5] (22). As commonly used, we adopted the 3.16 cut-off for our analysis (15). Additionally, we analyzed our data with the 3.99 cut-off (16) to ensure an objective and more inclusive approach. Arterial hypertension was defined as systolic blood pressure (SBP) and/or diastolic blood pressure (DBP) ≥ 95th percentile for age, gender and height (23). Abnormal fasting levels of lipids and other biochemical parameters were defined as follows: Triglycerides (TG) ≥ 150mg/dL (24); High-density lipoprotein cholesterol (HDL-C)
