Background: Most patients undergo physical examination. However, low back pain can only rarely be clearly diagnosed as having an irrefutable pathoanatomic cause on the basis of the classical battery of noninvasive physical examination procedures. However, physical examination findings are also used to predict the prognosis of treatment of chronic low back pain. A systematic review of the clinical literature is needed for an understanding of findings in this area.
Objectives: To establish whether various physical examination procedures can be used to predict treatment outcome in the conservative (noninvasive) treatment of chronic low back pain.
Data sources: From 910 titles published between 1986 and 1998, 10 original research reports were selected. These were obtained by means of a computerized MEDLINE search through use of various combinations of the key words low back pain, treatment, physical findings, predictors, prognosis, prognostic factors, prognostic indicators, and predictors and outcome(s); manual search strategies were also used.
Data synthesis: Articles that fitted the objectives of this review were retained and systematically reviewed for the prognostic value of the relevant tests/observations. In addition, results were studied in relation to type of outcome variable and type of treatment.
Results: The most commonly investigated tests were lumbar range-of-motion tests, which were found to have a clear-cut significant predictive value in 3 of 9 studies. Further analyses in relation to definition of outcome (back-to-work or reduction of symptoms) were not possible because study results were usually not clearly reported for each definition and because therapies were multimodal rather than distinctly well defined.
Conclusions: This area has not been fully investigated. Therefore, there is no satisfactory answer to the question of whether some physical examination tests have a prognostic value in the conservative treatment of chronic low back pain.