Developing new classification criteria for diffuse idiopathic skeletal hyperostosis: back to square one

Objective. To revise the definition of DISH and suggest a classification that may better represent our current knowledge of this entity allowing earlier diagnosis. Methods. Seven rheumatologists and an orthopaedic surgeon suggested a list of 63 parameters that might be included in a future classification of DISH. Participants rated their level of agreement with each item, expressed in percentages. In a second session, participants discussed each item again and re-rated all parameters. Thirty items that were granted >= 50% support on average were considered valid for a third round. A questionnaire listing these 30 items was mailed to 39 rheumatologists and orthopaedic surgeons worldwide with a request to answer categorically if they agreed on an item to be included as a criterion for a future classification of DISH. Items were regarded as perfect consensus when at least 95% of the respondents agreed and were regarded as consensus when at least 80% agreed. Results. There was perfect consensus for 2 (6.7%) of the 30 parameters and consensus for another 2 parameters. These items were ossification and bridging osteophytes in each of the three segments of the spine and exuberant bone formation of bone margins. Conclusion. At present there is no agreement about the inclusion of extraspinal, constitutional and metabolic manifestations in a new classification of DISH. Investigators with an interest in this condition should be encouraged to restructure the term DISH in an attempt to establish a more sophisticated definition.