Review on the Screening of Urine Glucose in School Children and Adolescents With Obesity for Early Diagnosis of Type 2 Diabetes Mellitus in Hong Kong

Introduction: Obesity and type 2 diabetes mellitus (T2DM) are growing health concerns. A ten-fold increase of T2DM was noted in the Hong Kong paediatric population from 1997 to 2007. T2DM is often asymptomatic at presentation, but complications can emerge rapidly, especially in youngsters. Experience in Japan, Korea and Taiwan suggests that urine glucose screening is a practical and non-invasive screening tool for identification of T2DM. The Hong Kong Student Health Service (SHS) offers yearly health checks for students and is a good platform for screening of T2DM since attendance rate is over 90% for primary school students and over 70% for secondary school students. Method: In 2005, SHS and the Hong Kong Paediatric Society formulated a protocol on urine glucose screening for early diagnosis of T2DM in students with obesity. Students between the ages of 10–18 years old with age- and sex- specific body mass index (BMI) >97th percentile were recruited. Those screened positive for glycosuria were referred to paediatric departments for workup under a standardized protocol, whilst those who screened positive for both glucose and ketones were referred to the emergency departments. Students enrolled from school year 2005–2006 to 2017–2018 were included. Demographic data, clinical presentation, investigatory results and co-morbidities were captured using a structured reply letter. Results: A total of 219,276 eligible students attended SHS in the years specified and 216,528 students (99%) completed urine glucose screening. 381 (0.18%) students were tested positive for urine glucose; 18 (4.7%) had concomitant urine ketones. In total 120 students had T2DM, 41 had pre-diabetes [impaired fasting glucose and /or impaired glucose tolerance] and 126 turned out normal. 43 students defaulted the referrals and 51 students had known diabetes. 21 students (17.5%) were started on insulin therapy upon diagnosis. A significant proportion of students with T2DM had co-morbidities including raised alanine amino-transferase (57%), hypercholesterolaemia (59%), and hypertension (13%). Five students (4.2%) had microalbuminuria at presentation. Of those with ketonuria, two students had serum glucose of over 20mmol/L and required fluid resuscitation ± insulin infusion in high dependency unit. Conclusion: Our pick up rate for T2DM from students with obesity aged 10–18 years using urine glucose is 0.05% (120/216,528). According to the Hong Kong Childhood Diabetes Registry, the crude incidence of T2DM for this age group was 6.16 /100,000/year over the study period, which equates to 506 new cases of T2DM. Thus 24% of the new T2DM cases were diagnosed by this program and many had associated co-morbidities at diagnosis. Our study shows that urine glucose testing is an inexpensive and simple test that allows for early diagnosis and treatment of T2DM in the primary care setting in this at risk population.