Objective: In quantitative analysis of serum proteins turbidimetric and nephelometric methods are the most common. Aim of this study was to compare the nephelometric method with the turbidimetric method for serum and urine samples in our clinical laboratory and to analyse the performance of the turbidimetric method.

Material and Methods: 24 hours collected urine albumin ( n=146), spot urine albumin (n=78), serum ASO (n=30), serum RF (n=48), serum IgA (n=24), IgM (n=24), IgG (n=24) and IgE (n=74) were analyzed with Behring BN II (Siemens, Germany) and Architect C 16000 (Abbott, USA). The relationship between the turbidimetric and nephelometric methods was demonstrated by applying the non – parametric method of Passing&Bablok. Each parameter's regression coeffiency and regression equation with intercept and slope value, Standart error, Standart error of mean (Syx) were checked. Correlation coefficients of serum plasma proteins and microalbumin levels were determined by linear regression analysis.

Results: There wasn't any significant difference between nephelometric and turbidimetric methods (correlation coefficiency: 0.996, 0.999, 0.991, 0.967, 0.930, 0.847, for 24 hour and spot microalbumin, Ig A, M, G and E respectively). % 95 CI was (0.995-0.997) for 24 hour microalbumin and for (0.999-1) for spot microalbumin results. For Ig A, M, G and IgE test results % 95 CI lower and upper limits were (0.976-0.996), (0.913-0.987), (0.828-0.972) and (0.757-0.904), respectively. Intercept and slope values were (-1.58, 1.07), (0.40, 0.92), (-1.63, 1.2),(-0.95, 1.22),(112.83, 1.06) and (-12.24, 1.19), respectively.

Conclusion: Our results indicated that the turbidimetric and nephelometric methods were compatible.