The reliability of clinical methods for recording temporomandibular joint sounds.

Measurement errors in recording temporomandibular joint sounds may originate from variation between observers and from variation in the phenomenon. Laboratory settings enable various procedures to be used to minimize both sources of variation. These procedures yield some excellent intra- and inter-examiner reliabilities, but this does not imply that dentists in a clinical setting are likely to evaluate temporomandibular joint sounds in a comparable way. This study was designed to evaluate clinical joint sound assessment methods (palpation and stethoscopy) without using special precautions to minimize variance. An attempt was made to quantify the signal variance. Within- and between-examiners agreement is estimated for both methods in a sample of 44 non-patients. The results show that two clinically experienced craniomandibular disorders specialists were able to reach fair to good agreement on the identification of (the number of) temporomandibular joint sounds. There was some disagreement with respect to the number of reciprocal clicks. Compared with the palpation technique, stethoscopy is more sensitive, especially with regard to crepitation. Based on the electronically recorded sounds, both examiners appeared to be overconsistent. It is concluded that the use of both palpation and stethoscopy in clinical settings can be justified but that both methods have limitations. When, in a given clinical setting, these limitations are acceptable, there appears to be no need for extra-sensitive but expensive electronic recording devices.