RESUMO
BACKGROUND: Serum urea correlates very well to salivary urea, which can be used as a low-cost, easily accessible and noninvasive diagnostic method for screening patients in early stages of kidney disease, especially in developing countries where resources are limited, giving the possibility of establishing secondary prevention programs later. METHODS: One hundred and one people were enrolled: 11 with creatinine clearance (CrCl) less than 10 ml/min per 1.73 m(2); 10 with CrCl of 11-20 ml/min per 1.73 m(2); 12 with CrCl of 21-50 ml/min per 1.73 m(2); 26 with CrCl of 51-80 ml/min per 1.73 m(2) and 42 with CrCl of 81-170 ml/min per 1.73 m(2), to analyze salivary urea test accuracy through construction of a receiver operating characteristic curve. RESULTS: Salivary urea cutoff point of 20 mg/dL and a CrCl of 80 ml/min per 1.73 m(2) showed sensitivity (S) of 0.98, specificity (SP) of 0.29, pretest probability (PPT) of 0.58, positive predictive value (PPV) of 0.66, negative predictive value (NPV) of 0.92, posttest positive probability (PTPP) of 0.66 and posttest negative probability (PTNP) of 0.09. A cutoff point of 40 mg/dL and a CrCl of 80 ml/min per 1.73 m(2) showed S=0.80, SP=0.71, PPT=0.58, PPV=0.80, NPV=0.71, PTPP=0.79 and PTNP=0.28. A cutoff point of 100 mg/dL and a CrCl of 80 ml/min per 1.73 m(2) showed S=0.22, SP=1, PPT=0.58, PPV=1, NPV=0.48, PTPP=1 and PTNP=0.52. Receiver operating characteristic curve analysis showed that the best cutoff point for salivary urea was 40 mg/dL. CONCLUSION: The salivary urea test has a great capacity to discriminate patients with chronic kidney disease from healthy people, and it was shown that the best cutoff point is 40 mg/dL.