Predicting Responses to Interventional Pain Management Techniques for Chronic Low Back Pain: A Single-Center Observational Study (PReTi-Back Study).

Background: Exploring factors linked to the outcomes of certain interventional pain management techniques may optimize the selection of candidates for those procedures. Our hypothesis is that factors that influence responses to interventional therapies for chronic low back pain (CLBP) can be identified by analyzing a prospective cohort.

Objectives: Our main aim is to identify the factors that may be associated with adult patients' responses to interventional therapies for the treatment of CLBP after 4 weeks of follow-up. Secondary objectives include the development of a predictive model and the establishment of a predictive score.

Study design: The PReTi-Back (Predicting REsponse to interventional Therapies In chronic BACK pain) study is an observational prospective single-center study, employing a nonprobability-sampling method.

Setting: Our population consists of adult outpatients with CLBP in a chronic pain unit of a tertiary hospital. The procedures we evaluated included epidural steroid injections, medial branch blocks and denervations, dorsal root ganglion blocks, and pulsed radiofrequency.

Methods: Ratings on the Numeric Pain Rating Scale (NPRS) and Oswestry Disability Index (ODI) were measured at the baseline and after 4 weeks of follow-up. The primary outcome of the study was composite and was evaluated at 4 weeks. A positive response to an intervention was defined as the simultaneous occurrence of a decrease of at least 2 points in the NPRS score and a decrease of at least 20% in the ODI score. A predictive model was constructed using logistic regression analysis, which incorporated 14 variables selected in advance. A predictive score was developed based on the odds ratios of the model variables.

Results: Four hundred patients were recruited. Of these patients, 368 completed follow-up, 49 were excluded, and 319 were included in the analysis. The interventional therapies provided a positive response to 85 patients (26.6%) at 4 weeks. Listhesis, radicular compression, and satisfaction with previous interventional therapies were positively associated with the positive response, and their ORs were close to 2. Meanwhile, obesity and persistent spinal pain syndrome type 2 (PSPS-2) had negative associations with the outcome, presenting ORs close to 0.5. The models were statistically significant and exhibited satisfactory goodness of fit. The area under the curve was 0.67 (95% CI, 0.60-0.74). Both models exhibited low sensitivity but high specificity. The synthesis of the prediction score had little impact on its discriminatory capacity.

Limitations: The subgroup analysis revealed that both listhesis and radicular compression were associated with the response to epidural therapies but not with the response to medial branch therapies. The score was efficient in ruling out those who would not benefit from intervention (scores of 0 or one), but its main limitation was that it was less effective in identifying those who might respond favorably (scores ≥ 2).

Conclusions: Patients satisfied with previously performed interventional therapies or who exhibit findings of radicular compression or listhesis on imaging show approximately twice the likelihood of experiencing a positive response to short-term IMPT than do patients without those characteristics. Patients who are obese or have PSPS-2 exhibit approximately a 50% lower likelihood of short-term response than do patients without these conditions.