Estimates of sensitivity and specificity of electric pulp testing depend on pulp disease spectrum: a modelling study.

AIM: To demonstrate how the spectrum of diseased pulps may influence sensitivity and specificity in diagnostic studies on pulp status. METHODOLOGY: An original sample from a previous study consisting of 59 teeth scheduled for root canal treatment was used where the relationship between the response to electric pulp testing and the visual status of the pulp was evaluated. To alter the spectrum of diseased pulps, a hypothetical sample of asymptomatic teeth with deep caries lesions was added to the original sample. Sensitivity and specificity were then compared for the two samples. RESULTS: In the original sample of 59 teeth, sensitivity was 72% and specificity 90%. When the spectrum of diseased pulps was altered, sensitivity decreased to 67% and specificity increased to 97%. The change in disease spectrum also decreased the prevalence of necrotic pulps. CONCLUSIONS: The spectrum of diseased pulps included in a diagnostic study on the accuracy of electric pulp testing, and indirectly also disease prevalence (here pulp necrosis), influences estimates of sensitivity and specificity. This implies that estimates of diagnostic accuracy from one study with a particular tooth population spectrum may not apply to another tooth population with a different disease spectrum.

Diagnosis of the condition of the dental pulp: a systematic review.

The aim of this systematic review was to appraise the diagnostic accuracy of signs/symptoms and tests used to determine the condition of the pulp in teeth affected by deep caries, trauma or other types of injury. Radiographic methods were not included. The electronic literature search included the databases PubMed, EMBASE, The Cochrane Central Register of Controlled Trials and Cochrane Reviews from January 1950 to June 2011. The complete search strategy is given in an Appendix S1 (available online as Supporting Information). In addition, hand searches were made. Two reviewers independently assessed abstracts and full-text articles. An article was read in full text if at least one of the two reviewers considered an abstract to be potentially relevant. Altogether, 155 articles were read in full text. Of these, 18 studies fulfilled pre-specified inclusion criteria. The quality of included articles was assessed using the QUADAS tool. Based on studies of high or moderate quality, the quality of evidence of each diagnostic method/test was rated in four levels according to GRADE. No study reached high quality; two were of moderate quality. The overall evidence was insufficient to assess the value of toothache or abnormal reaction to heat/cold stimulation for determining the pulp condition. The same applies to methods for establishing pulp status, including electric or thermal pulp testing, or methods for measuring pulpal blood circulation. In general, there are major shortcomings in the design, conduct and reporting of studies in this domain of dental research.