Comparison of the determination of superoxide dismutase and antioxidant capacity in neurological patients using two different procedures

As oxidative stress in relation with neurological diseases has become an im portant point in recent research, simple methods to be used in epidemiologi cal studies and clinical practice are required. The hypothesis that the ana lytical methods used in research laboratories (RLM) can be used interchange ably with commercial kits (CKM) for SOD and TRAP is tested. Both methods we re compared using linear transformations of the RLM measurements into the C KM scales. Data were obtained from Alzheimer's, Parkinson's, and vascular d ementia patients and controls. The lack of fit and the run's test of residu als were not significant, but the same sign method detected significant non linearities (P < 0.000001 for SOD, P < 0.01 for TRAP). The intragroup CVs o f both methods were comparable for TRAP, while in the RLM determinations of SOD resulted in <50% of those obtained with the CKM. The ANCOVA comparison of the regression parameters across the clinical groups resulted significa nt for SOD (P < 0.0001) and not significant for TRAP. Both methods agree in describing the features of the clinical groups, but the degree of agreemen t at the individual concentration was poor and they could not be readily in tercalibrated. Normal and pathological values should be obtained independen tly for the CKM to insure their applicability to large populations. (C) 200 0 Elsevier Science B.V. All rights reserved.