Association of limited joint mobility in childhood and adolescent diabetes mellitus patients with HBA1C, microalbuminuria, and retinopathy at Tygerberg Hospital, Western Cape, South Africa
Date
2020-03
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Stellenbosch : Stellenbosch University
Abstract
ENGLISH SUMMARY: BACKGROUND: Limited joint mobility (LJM) is the earliest clinically apparent long-term complication of diabetes mellitus (DM) in children and adolescents. The importance of LJM in young diabetic patients is that it is related to serious microvascular complications, especially retinopathy and nephropathy.
OBJECTIVES: The primary objective of this study was to determine the relationship of LJM with blood HbA1c levels, microalbuminuria, and diabetic retinopathy. The secondary objective was to determine the prevalence of LJM in all children with DM.
METHODS: This was a retrospective folder review done at the Tygerberg Hospital paediatric diabetes clinic over a10 year period. Pre-pubertal patients were included if they had DM for ≥5 years, and pubertal patients were included if they had DM for ≥2 years. The following data was recorded: Age, sex, type of DM, duration of DM, absence/presence of LJM, HbA1c, insulin regimen, urine albumin/Cr, and absence/presence of retinopathy.
RESULTS: Of the 129 patients enrolled in the study, only 74 patients fulfilled the entry criteria. They had Type 1 diabetes (T1DM), with a median age of 8 years (IQR 4- 10.75) at diagnosis. LJM occurred in 60.8% (45/74). Of these, 37 had HbA1c of 8.5% and above, 2 had retinopathy, and a total of 12 patients had microalbuminuria (> 2mg/mmol). There was no statistically significant association between LJM and HbA1c [OR=1.038 (95% CI 0.303 to 3.550)], retinopathy (P value 1.000), and microalbuminuria, (p value 0.594). Sex, the type of insulin regimen, and age at diagnosis also did not influence the outcome [OR=0.988,95% CI (0.365,2.673), p value 0.202,OR=1.063,95% CI (0.877,1,290), respectively)]. A borderline association was observed between LJM and duration on DM (p-value=0.053). This association is not confirmed in the multiple regression model that accounts for the effect of potential confounders [OR=1.215 (95% CI 0.941, 1.569)].
The diagnostic performance of LJM as a screening tool for poor glycaemic control is as follows: Sensitivity = 82.2%, (95% CI 71.1-93.3) %, specificity = 17.24 % (95% CI 3.49-30.99%), Positive Predictive Value=60.66%, 95% (CI 48.4%, 72.91%), and Negative Predictive Value=38.46%, 95% CI (12.02%, 64.91%). ROC model is not better than chance (95% CI 0.413,0.592), and the p value of 0.9534 also confirms this. LJM has a good sensitivity and poor specificity.
CONCLUSION: Although most studies describe LJM as a common and early long-term complication in diabetes patients, it is not a useful screening test for poor glycaemic control or microvascular complications. Diabetes duration seems to be an important determinant of LJM. A larger multicentre prospective study to better investigate this relationship should be performed. Knowing that there is an association between LJM and microvascular complications would provide an inexpensive clinical alternative to investigations not readily available in a low resource setting, for the early identification of poor control and increased risk of other microvascular complications.
"Geen opsomming biskikbaar"
"Geen opsomming biskikbaar"
Description
Thesis (MMed)--Stellenbosch University, 2020
Keywords
Diabetes, Diabetes mellitus, Limited joint mobility, Albuminuria, Microcirculation disorders, Diabetes in children, UCTD