Development and validation of a mobile application for point of care evaluation of growth failure.

OBJECTIVES: Lack of systematic evaluation of short stature results in unnecessary work-up on one hand while missing pathology on the other. We have developed a mobile application that guides work-up based on age, auxology (height, BMI, and corrected standard deviation score), and skeletal maturation with an aim of reducing the diagnostic errors. Aim of this study is to develop and validate a mobile application for point of care evaluation of short stature. METHODS: The application was developed (n=400) and validated (n=412) on children and adolescents (2-18 years of age) presenting to our Pediatric Endocrinology Clinic with short stature. Height standard deviation score thresholds determining the need for workup were derived from Receiver Operating Characteristics (ROC) curve. Student's t-test and ROC curves were used to identify the most appropriate parameter differentiating constitutional delay of growth and puberty (CDGP) from pathological and nutritional from endocrine causes. The validation of the application involved comparing the application predicted and clinical diagnosis at each step of the algorithm. RESULTS: The mobile application diagnosis was concordant with clinical diagnosis in 408 (99.0%) with discordance in four (two with CDGP labeled as growth hormone deficiency [GHD] and two with GHD labeled as CDGP). CONCLUSIONS: Mobile application guided short stature assessment has a high concordance with the clinical diagnosis and is expected to help point of care short stature evaluation.

Predictive Value of IAP 2015, IAP 2007 and WHO Growth Charts in Identifying Pathological Short Stature.

OBJECTIVE: To compare the diagnostic accuracy of IAP 2015, WHO and IAP 2007 growth charts in identifying pathological short stature in Indian children. METHODOLOGY: The predictive value of the growth charts for pathological short stature was assessed in 500 (266 boys) short subjects (age 5-17.9 years) presenting to our pediatric endocrine clinic. RESULTS: WHO, IAP 2015, IAP 2007 criteria classified 500, 410 (82%) and 331 (66.2%) subjects short respectively. A total of 218 (43.6%) subjects had a pathological cause. Two out of 90 subjects short by WHO criteria but normal as per IAP 2015 had a pathological cause (2.2%) whereas 38 out of 79 subjects short as per WHO and IAP 2015 criteria but normal by IAP 2007 had pathological short stature. The diagnostic measures of IAP 2015 and IAP 2007 charts in identifying pathological short stature showed a sensitivity 99.1% and 81.7%, negative predictive value 97.8% as against 76.3%, positive predictive value 52.7% and 53.8%, and specificity of 31.2% and 45.7%, respectively. CONCLUSION: IAP 2015 growth charts are superior in identifying pathological growth failure compared to WHO and IAP 2007.

Metabolic Impact of Subclinical Hypothyroidism in Obese Children and Adolescents.

OBJECTIVE: Subclinical hypothyroidism is common in children and adolescents with obesity and has been considered to be its effect with no need for treatment. Its metabolic impact has not been evaluated. Therefore the present study was conducted to determine the metabolic impact of obesity related subclinical hypothyroidism. METHODS: Retrospective record review of obese children and adolescents between 5 and 18 y of age presenting to pediatric endocrine clinic was done. Four hundred four obese children and adolescents [251 boys, 11.8 (3.2); 5.1-18 y, BMI SDS 2.4 (0.7); 1.4-6.6] were assessed regarding thyroid functions, adiposity (clinical and DXA derived) and metabolic complications. RESULTS: Subclinical hypothyroidism was observed in 122 (30.2%) and was associated with higher fat percentage [49.2 (5.8) vs. 47.2 (6.4) p = 0.009], android to gynoid ratio [1.1 (0.1) vs. 1.0 (0.1), p = 0.007] and alanine aminotransferase (ALT) levels [49.3 (31.5) vs. 40.8 (38.1), p = 0.04]. Subjects with subclinical hypothyroidism had 1.9 times greater odds of having non-alcoholic steatohepatitis (47.3% vs. 31.8%, p = 0.005) with no difference in the prevalence of dyslipidemia, dysglycemia or hypertension. Subclinical hypothyroidism was the only determinant of non-alcoholic steatohepatitis on binomial logistic regression (WALD = 11.04, p = 0.001) with no impact of BMI SDS, waist circumference SDS, fat percentage or android to gynoid ratio. Thyroid stimulating hormone (TSH) was the most important determinant of ALT on linear regression (B = 3.027, p < 0.005). CONCLUSIONS: Obesity related subclinical hypothyroidism predisposes to increased ALT and non-alcoholic steatohepatitis independent of severity adiposity. The impact of thyroid supplementation in this setting needs to be explored.