First episode of acute low back pain – an exploratory cluster analysis approach for early detection of unfavorable recovery

Abstract

Purpose: To identify recovery patterns in patients with a first episode of acute low back pain (LBP) and to define risk factors for unfavorable outcome.Methods: One hundred and eight patients (55 male, 53 female; mean age = 40.8, SD 14.2 years) rated pain (NRS) and disability [Oswestry Disability Index (ODI)] before the first treatment and 1 week, 1, 3, 6, and 12 months later. Hierarchical cluster analysis identified recovery patterns based on NRS data. Clusters were compared for age, NRS and ODI at baseline, pain reduction in the first week, gender, radicular signs and traumatic onset using one-way ANOVA (post hoc Bonferroni) and χ2 tests.Results: The cluster analysis revealed four clusters: moderate baseline pain/fast recovery; high baseline pain/fast recovery; high baseline pain/persistent mild pain; high baseline pain/persistent high pain. These clusters differed in baseline NRS [F(3,104) = 39.61, p < 0.001], baseline ODI [F(3,104) = 12.17, p < 0.001], pain reduction in the first week [F(3,104) = 11.51, p < 0.001] and in radicular signs [χ2(3) = 9.20, p = 0.027].Conclusions: These results suggest that an initial and regularly repeated assessment of pain intensity and functional disability is important. Initial pain intensity does not seem to be a prognostic factor per se, as it did not negatively affect recovery provided that it decreased early in treatment.Implications for RehabilitationPrediction of outcome is particularly important in patients with a first episode of acute LBP as one third did not completely recover.Pain intensity and functional disability should be initially assessed and regularly repeated in the first phase of treatment.High initial pain intensity and disability combined with small pain reduction during the first week might predict unfavorable outcome and require adequate treatment.