Childhood obesity is a growing problem in the current world. Worldwide, its prevalence was about 340 million in 2016, including Africa. In the last 4 decades, rates have tripled all over the world, from 4% to 18%, with comparable gender distribution. Currently, obesity is considered a major contributor to disease-related morbidity and mortality more frequently than stunted growth. It has been incriminated in early onset respiratory problems, psychiatric troubles, cardiovascular disorders, hyperlipidemia, bony fractures, insulin resistance, type 2 diabetes mellitus (DM), nonalcoholic fatty liver disease (NAFLD) and premature death (WHO 2018).
In the pediatric age group, the WHO and the American Academy of Pediatrics postulated the current definitions of overweight and obesity based upon the body mass index (BMI)-for-age. BMI is an easy noninvasive tool for large scale clinical screening. Still, BMI cannot distinguish between lean and fat mass, it relies on body weight regardless of body composition (Reilly et 2010). Generally accepted, the anthropometric measures were recommended as indirect indicators of body composition, and to evaluate adiposity in children as well as in adults (Sakuno et al 2014).
The underlying problem of obesity is an excessive fat collection, with the most predilection sites being under the skin in the subcutaneous tissues, and surrounding the internal organs, described as the visceral fat. Adipose tissue distribution together with increased visceral adiposity seems to be linked to complications of obesity (Sakuno et al 2014, Dong et al 2014). Yet, it is very crucial to determine the body content of fats in overweight and obese children. Suggested methods included waist circumference (WC) or imaging techniques such as ultrasonography, dual X-ray absorptiometry (DEXA), magnetic resonance imaging (MRI), and computed tomography (CT) (Schwenzer et 2009, Li et 2012).
The primary aim of the current work was to assess a group of overweight and obese children clinically, anthropometrically, and sonographically, to detect the most sensitive predictors for body fat content and obesity complications. Secondarily, to find the best cutoff values for measured sonographic parameters of subcutaneous fat (SCF) and visceral fat in overweight and obese children.
This was a cross-sectional study conducted on a group of overweight and obese children and adolescents with simple obesity, and a control group of average weight peers. Participants were recruited from Diabetes, Endocrine& Metabolism Pediatric Unit (DEMPU) Cairo University during the period from March till October 2016. Patients included from both sexes, aged between 3-15 years, with a body mass index (BMI) above 85% for age and sex (Standard Egyptian Growth 2008). Patients with endocrinal causes of obesity (as hypothyroidism, Cushing syndrome), syndromic obesity, receiving long term steroid therapy or drugs modifying liver functions or having a chronic liver disease were excluded from the study.
The study protocol was approved by the University-research ethics committee. Written informed consent was obtained from the parent/guardian after an explanation of the study.
The entire study group was subjected to the following
The following parameters were measured and recorded (Sakuno et al 2014):
Subcutaneous fat (SCF) is the distance from the skin to the linea alba, measured on the hemisternal line, 1 cm above the umbilicus, utilizing the linear transducer in a longitudinal section (Figure 1-A). The smallest measurement was referred to as minimum S.
Visceral fat including:
Preperitoneal fat (PPF) is the distance from the linea alba to the anterior parietal peritoneum, measured on the hemi-sternal line, 1 cm above the umbilicus, utilizing the linear transducer in a longitudinal section (Figure 1-B). The largest measurement was referred to as maximum P.
Intraperitoneal fat (IPF) was measured with the convex transducer in three ways, as follows:
Abdominal wall fat index (AFI): was calculated by dividing maximum P by minimum S.
Liver echo pattern to detect non-alcoholic fatty liver disease (NAFLD), it was graded as follows:
Besides, the following blood tests were performed for cases only:
Cases were considered susceptible to obesity’s comorbidities (Gahagan, 2020) when any of the following criteria was observed:
Data were collected and tabulated. Statistical Package for Social Science (SPSS, Chicago, IL 60606-6412, USA) program version 17.0 was used for data analysis. Mean and standard deviation (SD) or median and interquartile range (IQR) were used for quantitative data, while frequency and percentage were done for qualitative data. Differences in anthropometric, dietetic, clinical and laboratory characteristics between the cases and controls and between cases with predictors of comorbidities and those without were tested using Student’s t-test or Mann Whitney U test for numerical data and by Chi-square test for categorical data. Spearman correlations were used to test the linear association of measured sonographic parameters with the anthropometric and laboratory results of the cases. Receiver operator characteristic (ROC) curve was drawn to calculate the best cutoff points of measured sonographic parameters, using those of the healthy group as a control. The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and accuracy for each test was calculated. A two-sided P value
The study included 52 cases (10 overweight & 42 obese) children/adolescents, their mean age was 9.17 ± 3.07 years (range 3-15). The healthy control group was 41 children, with a mean age of 7.71±2.7 years (range 3-14). Cases and control were comparable regarding the sex, but the control group was younger than the cases. Demographic data, dietetic history and blood pressure measurements of cases and controls are shown in table 1.
History of depression episodes and asthmatic attacks, family history of obesity, hyperlipidemia, DM, hypertension, cardiac, and hepatic affection were reported in cases more frequently than in controls. No observed difference regarding breastfeeding duration, food allergies, intake of supplements, animal protein, and vegetable. Eating at home was significantly higher in controls while eating outside was significantly higher in cases (29% vs 7%, P= 0.009). No difference observed in either early or late sleeping, and duration of sleep.
On examination, the mean systolic and diastolic blood pressure were significantly higher in cases than in controls (115.90±10.21 vs 104.90±6.79, P Table 2 showed the anthropometric measures, ultrasound parameters and hepatic echogenicity in cases and controls. NAFLD was detected in 11 obese patients 6 had mild, 5 had moderate and one had a severe degree. ALT was high in 4 cases with increased hepatic echogenicity. In overweight and obese children, the anthropometric measures and the sonographic parameters were comparable as regards gender distribution, data not shown.
As regards to the laboratory parameters of cases: abnormal TC, TG, LDL & HDL were found in 13 (25%), 14 (26.9%), 15 (28.8%), 21 (40.4%) patients, respectively. Mean FBS was 94.9+12.1 (71-125), and 13 cases have a level above normal values. In addition, elevated ALT & AST were found in 8 (15.4%) and 10 (19.2%) patients, respectively, data not shown.
Pearson correlations were done between sonographic parameters versus the anthropometric and laboratory parameters of cases, which is shown in table 3. SCF was positively correlated with PPF, IPF1, 2, 3 and negatively correlated with AFI. IPF were positively correlated with each other and negatively with AFI, data not shown.
Cases having one or more predictors of obesity’s comorbidities were 46 (88.5%) cases 10 (19.2%) have single predictor and 36 (69.2%) have multiple predictors, and the remaining 6 cases do not have any. Hypertension was presumed from a single blood pressure measurement. The anthropometric and sonographic parameters were compared between cases without and those with individual obesity’s comorbidity, as shown in table 4. Anthropometric measures were significantly higher in cases with Acanthoses nigricans, NAFLD, and asthma, while sonographic parameters were significantly higher in cases with NAFLD. All parameters showed no statistical significance when compared between cases who had one or more predictor of obesity’s comorbidities and those without.
ROC curves were done to identify the best cutoff values for measured sonographic parameters at 95% confidence level, table 5.
Most of studied cases were pre-pubertal.
Among studied cases, observed risk factors for obesity included the short duration of exclusive breastfeeding and unhealthy food habits. Claimed triggers for obesity included breastfeeding duration, diet intake, physical activity, nighttime sleep duration, and time spent watching television (Dev et al 2013). Coincidence with civilization excess fatty food intake together with a lack of physical activity, cause an imbalance between consumed calories and expended calories (WHO 2018).
The studied cases were mainly pre-pubertal, and noticeably, their anthropometric measures and sonographic parameters were comparable as regards gender distribution. It was established that the developmental pattern of adipose tissue accumulation and distribution is gender-specific and changes rapidly during puberty (Taylor et 2010).
Among study participants, the anthropometric measures, and the ultrasound parameters were significantly higher in overweight and obese children, with remarkable correlations between them. Only WC correlated positively with all measured fat indices, it was the best anthropometric indicator of subcutaneous and visceral fat. BMI, TSF, and HC were positively correlated with most of the fat indices, while MAC, and WHR showed less frequent correlations. Our results were agreed with Moreno and colleagues (2002), who reported that WC alone is a good indicator of central adiposity in children. The utility of BMI was debatable with many limitations reported. The accuracy of BMI to predict adiposity in children was tested by a meta-analysis, showing good specificity with poor sensitivity, with failure to identify a considerable number of children with excess body fat percentage (Javed et 2015). On the contrary, other reports observed a significant correlation of BMI and the sonographic fat indices in children (Semiz et al 2007 Sakuno et al 2014), and in adults as well (Nadeem et 2018). Some reports advised the use of MAC as a simple clinical tool for excess fat in children, with comparable sensitivity, and specificity to BMI and WC (Reilly et 2010, Craig et 2014). In adult studies, skinfold thickness and arm circumference were accurate methods for body fats assessment when compared to ultrasound (Nadeem et 2018). Prior observation using DEXA showed that WHR did not show correlation with adiposity in children, being affected with age (Mandato et 2005).
In our overweight and obese group, there was a positive correlation between measured subcutaneous fat and visceral fat. This was agreed by Pirimoglu and colleagues (2019), who reported a strong correlation between visceral fat and subcutaneous fat in children with BMI >85th.
Among our cases, the prevalence of predictors of obesity’s comorbidities was 88.5%, with about 70% having multiple comorbidities. Detected metabolic comorbidities in order of frequency were dyslipidemia, metabolic syndrome, acanthoses nigricans, hypertension, and NAFLD. Noticeably, cases showed significantly prevalent positive family history of obesity and its complications than in controls. The link of metabolic aberration, with underlying insulin resistance, and obesity is well established (Juárez-López et 2010).
In our study, SCF was the only fat indices that positively correlated with cholesterol levels. Also, no considerable difference observed regarding all measured parameters and the presence of dyslipidemia. Jung and coworkers (2016) found a positive correlation between abdominal fat and TG. In the other side, Semiz et al. (2008) reported no association between lipid profile and abdominal fat thickness, among their studied prepubertal and pubertal children.
Among the study group, both metabolic syndrome and acanthoses nigricans were observed in 32.7% of cases. Anthropometric measures were significantly higher in cases with acanthoses nigricans, but not in cases with metabolic syndrome. The measured sonographic parameters showed no significant difference between cases with both conditions. In previous studies in children, WC was considered a good predictor for insulin resistance, and the metabolic syndrome (Moreno et 2002). The relation between insulin resistance and body fat have been studied previously. In pre-pubertal obese children, some found no correlation between abdominal fat thickness measured by ultrasound and HOMA IR, indicating insulin resistance (Semiz et al. 2008, Peçanha et 2018), while another report identified a positive correlation (Reinehr and Wunsch 2010).
Our obese patients had significantly higher systolic and diastolic blood pressure as compared to the healthy group. BMI was the only anthropometric measure that was significantly high among hypertensive cases. This is in line with previous studies that found a positive correlation between obesity and high blood pressure. They postulated that excess weight is correlated to hypertension and pre-hypertension as well (Genovesi et al 2010, Ferreira and Ayodas 2010).
In our study, hepatic echogenicity denoting NAFLD was observed in eleven cases (21%) and not in the average weight children, with ALT values above reference level in four out of them. Among studied NAFLD cases most of the anthropometric measures and ultrasound parameters were significantly higher than in cases with normal hepatic echogenicity. The prevalent echogenic liver in ultrasound examination of obese children has been reported previously, in up to 80% of cases, which coincidence with augmented obesity rates (El-Koofy et al 2011 Sakuno et al 2014, Felix et al 2016). WC measurement, reflecting abdominal obesity was recommended for predicting NAFLD. A meta-analysis has reported the significant role of central obesity on NAFLD, independent of Obesity (Pang et al. 2015). Moreno and colleagues (2002) mentioned the superiority of WC over BMI in predicting liver steatosis.
In our group, depression episodes, and asthmatic attacks were frequently observed. Most anthropometric measures were significantly higher among cases with asthma, while SCF was significantly lower in cases with depression. In earlier reports, the link between obesity and psychological problems as depression and anxiety have been elaborated (Gupta 2017), as well as with asthmatic episodes (WHO 2018).
In our work, assessment of the body fat relied upon ultrasonography measurement. Its advantages included high accuracy, low cost, non-invasive nature, and short testing time (Wagner 2013). It has a very good correlation to MRI and CT in the detection of subcutaneous and abdominal fat, as well as diagnosis of hepatic affection (Widhalm and Ghods 2010, Von Schnurbein et 2011). It is said in a previous study that ultrasound is a reliable and easier method of measuring abdominal fat and the actual measurement of abdominal fat by ultrasound are more informative than other anthropometric measurements (Premanath et 2017). On the opposite side, some previous studies in obese children showed no benefit of imaging modalities over anthropometry to assess body fat (Koot et 2013).
The strength of our study arises from the current lack of cutoff values for visceral and SCF thicknesses in overweight and obese children. Besides, the need for a simple tool to predict body fat, as well as obesity complications.
This study has some limitations. The control group was younger than the cases. It is a single-center, with a few participants included. Blood pressure was measured only once, owing to the cross-sectional nature of our study.
In conclusion, WC was the best anthropometric indicator for both subcutaneous fat and visceral fat, followed by BMI, TSF, HC, MAC, and WHR. Ultrasound is a reliable method to assess body fat distribution in overweight and obese children, it is a simple, cheap, noninvasive, and valuable technique. Fat parameters were significantly associated with NAFLD cases.