The real performance of bitewing radiography and fiber-optic transillumination in approximal caries diagnosis

For both general practitioners and researchers in caries diagnostics, the true validity of available diagnostic techniques is of considerable interest. Yet, for both bitewing radiography and fiber-optic transillumination, this is still not accurately known, nor is it clear which of the two techniques performs best clinically. This study's purpose was to estimate the clinical performance of the two techniques in diagnosing approximal caries lesions in low-caries-prevalence populations. Clinical studies that compare the two techniques were selected from literature. We determined 2 x 2 contingency tables from these studies and calculated one overall contingency table. The cut-off for decay was at dentinal caries. Assuming that erroneous outcomes from both techniques are mutually independent, we expressed diagnostic sensitivity and specificity of the two techniques as functions of the contingency table cell contents, with caries prevalence as the parameter. Because the caries prevalence was unknown, every sensitivity and specificity value was possible. From the conditions that sensitivity, specificity, and caries prevalence are always between one and zero, a limited range of sensitivity and specificity values was obtained. Three situations were examined: Bitewing radiography specificity is 1, fiber-optic transillumination specificity is 1, and both specificities are equal. Under these conditions, the bitewing radiography sensitivity was between 1.00 and 0.71 +/- 0.01, and the fiber-optic transillumination sensitivity was between 0.70 +/- 0.01 and 0.50 +/- 0.02. Both specificities were between 1.00 and 0.99. We concluded that the two techniques have comparable specificities, but that the fiber-optic transillumination sensitivity is significantly lower than that for bitewing radiography.