1995 Volvo Award in clinical sciences. The diagnostic accuracy of magnetic resonance imaging, work perception, and psychosocial factors in identifying symptomatic disc herniations.

STUDY
DESIGN: This was a prospective study of patients (study group) with symptomatic disc herniations and asymptomatic volunteers (control group) matched for age, sex, and work-related risk factors.
OBJECTIVE: To determine the prevalence of disc herniation in a matched group of asymptomatic volunteers and to access the diagnostic accuracy of magnetic resonance imaging, work perception, and psychosocial factors in identifying symptomatic disc herniations. SUMMARY OF BACKGROUND DATA: Disc herniations have been reported to occur in 20-36% of asymptomatic volunteers. A valid comparison of asymptomatic individuals and patients with disc herniations has not been performed.
METHODS: Forty-six patients with low back pain and sciatica severe enough to require a discectomy were compared with 46 age-, sex-, and risk factor-matched (heavy lifting, twisting and bending, vibration, and sedentary activity) asymptomatic volunteers. Both groups had a complete clinical and magnetic resonance imaging examination and completed a questionnaire to assess differences in the psychosocial and work perception profiles. The prevalence and the severity of morphologic alterations (disc herniation, disc degeneration, and neural compromise) was analyzed by tow independent radiologists in a blinded fashion. Differences between both groups regarding MRI findings, work perception (occupational mental stress, intensity of concentration, job satisfaction, and job-related resignation) and psychosocial factors (anxiety, depression, self-control, social support, and marital status) were compared using multivariate techniques. Stepwise discriminating analysis was used to identify the best discriminating variables within the magnetic resonance image, work perception, and psychosocial categories in terms of the diagnostic accuracy to predict group membership (study [pain] or control [no pain] group).
RESULTS: Matched controls had significantly more risk factors than a group of normal individuals. The present study has presented evidence that an age-, gender-, and occupational risk factors-matched group of asymptomatic patients shows a high incidence rate of disc herniations (76%). Although significantly less than the symptomatic group incidence of 96%, this represents a much higher prevalence rate than generally expected and reported in other studies of unmatched asymptomatic volunteers. Patients had more severe disc herniations (disc extrusions) than asymptomatic volunteers (35% vs. 13%). There was no significant differences regarding disc degeneration between both groups (96% vs. 85%). The only substantial morphologic difference between both groups was the presence of a neural compromise (83% vs. 22%), which was highly significant (P < 0.0001). There were significant differences between both groups regarding work perception (occupational mental stress, intensity of concentration, job satisfaction, and resignation; P < 0.027) and psychosocial factors (anxiety, depression, self-control, marital status; P < 0.0001). The best single predictor of a group membership was the extent of neural compromise. A combination of this factor with occupational mental stress, depression, and marital status was the best predictive model. With this model, the false-negative rate (potential overtreatment of disc morphology) was reduced by more than half compared with morphologic factors (nerve root compression) alone (22% vs. 11%).
CONCLUSIONS: In an age-, sex-, and risk factor-matched group of asymptomatic individuals, disc herniation had a substantially higher prevalence (76%) than previously reported in an unmatched group. Individuals with minor disc herniations (i.e., protrusion, contained discs) are at a very high risk that their magnetic resonance images are not a causal explanation of pain because a high rate of asymptomatic subjects (63%) had comparable morphologic findings. The only highly significant difference between the study group and control group regarding morphologic fi