Background: Testing for microalbuminuria (MA) is an important tool for detection of the earliest clinical manifestation of diabetic nephropathy Dipstick test for MA is commonly used for screening MA but this dipstick test has not been validated in Thai patients with DM.
Objective: To evaluate the diagnostic accuracy of the dipstick test for MA in random spot urine samples of type 2 diabetic mellitus patients, using urinary albumin to creatinine ratio (UACR) as the reference standard method.
Material and method: Type 2 diabetic mellitus patients from the out-patient department of seven public hospitals were recruited. Random spot urine samples from all patients were screened for microalbuminuria by Combi-Screen dipstick test and UACR.
Results: A total of 6,223 urine samples from 899 diabetic patients were screened for MA. From UACR criteria, these urine samples were classified as normoalbuminuria (4,016 samples, 64.5%), MA (1,795 samples, 28.8%) and macroalbuminuria (412 samples, 6.6%). The dipstick test for MA had an overall sensitivity of 83.7%, specificity of 92.6% with a positive predictive value of 83.4% and a negative predictive value of 92.7%. The area under the receiver operating characteristics curve of the dipstick test for MA is 0.9427.
Conclusion: The dipstick test for MA may be a useful method to initially screen for MA in Thai patients with type 2 DM.