A multidimensional nomogram combining clinical factors and imaging features to predict 1-year recurrence of low back pain with or without radicular pain after spinal manipulation/mobilization

BackgroundIn this retrospective study, we aimed to develop a nomogram to predict recurrence during a 1-year period of spinal manipulation/mobilization (SM/M) in patients with low back pain (LBP) with greater pain intensity, more severe comorbid conditions, or a neuropathic component.MethodsA total of 786 consecutive patients with LBP treated with SM/M as primary therapy were divided into training (n = 545) and validation (n = 241) sets. Cox regression analyses were used to assess the relative value of clinical factors and lumbar magnetic resonance imaging features associated with recurrence during the 1-year period. Predictors of recurrence with significant differences were used to construct a nomogram in the training set. We evaluated the performance of the model on the training and validation sets to determine its discriminative ability, calibration, and clinical utility. The prognostic value of the nomogram for predicting recurrence was assessed using Kaplan-Meier analysis and time-dependent receiver operating characteristic analyses.ResultsA nomogram comprising hospitalization time, previous history of LBP, disease duration, lumbar range of motion, lower extremity tendon reflex, muscle strength, ratio of herniation to uncompressed dural sac area, and Pfirrmann classification was established for recurrence during a 1-year period after SM/M in patients with LBP. Favorable calibration and discrimination were observed in the nomogram training and validation sets (C-index 0.753 and 0.779, respectively). Decision curve analysis confirmed the clinical utility of the nomogram. Over a 1-year period, the nomogram showed satisfactory performance in predicting recurrence in LBP after SM/M.ConclusionWe established and validated a novel nomogram that can accurately predict a patient's risk of LBP recurrence following SM/M. This realistic prognostic model may aid doctors and therapists in their decision-making process and strategy optimization for non-surgical treatment of LBP using SM/M.