Demographics and PainDETECT as Predictors of 24-Month Outcomes for 10 kHz SCS in Nonsurgical Refractory Back Pain.

Background: Nonsurgical refractory back pain (NSRBP) is broadly defined as chronic refractory back pain in patients who have not had previous spine surgery and, because they are deemed inappropriate candidates for surgery, are reliant on conventional medical management (CMM), which often provides poor long-term outcomes. High-frequency spinal cord stimulation (10kHz SCS) has demonstrated high rates of pain relief and improvements in functioning in patients with NSRBP. However, despite the use of temporary trial stimulation to select patients who will respond to therapy, some patients fail to achieve long-term therapy response with permanent implants. Prediction analysis founded on patients' baseline characteristics may enrich the appropriate selection of patients for permanent implantation.

Objectives: To examine baseline patient characteristics to predict long-term pain and functional responses to treatment with 10 kHz SCS for NSRBP.

Study Design: A retrospective analysis of baseline patient characteristics as predictors of 24-month pain and functional outcomes from a previous multicenter randomized controlled trial of 10 kHz SCS in patients with NSRBP.

Patients: Patients diagnosed with chronic, neuropathic, axial, low back pain refractory to CMM who had had no previous spine surgery, were deemed unsuitable candidates for it according to a spine surgeon, were implanted with 10kHz SCS and continued with CMM for up to 24 months.

Methods: The baseline characteristics of and 24-month outcomes in the 125 implanted patients who participated in the NSRBP randomized controlled trial (RCT) were included in this analysis. The baseline characteristics included demographics, baseline pain on the visual analog scale (VAS), baseline function based on the Oswestry Disability Index (ODI), mental health according to the patient health questionnaire-9 (PHQ-9), neuropathic pain as measured by PainDETECT, and each patient's temporary trial response. Patient response at 24 months was defined as absolute change from the baseline on the VAS and ODI, and each patient was also classified as a pain responder (achieving at least a 50% decrease in VAS pain score from the baseline) and a function responder (at least a 10-point decrease in ODI or a 24-month score of no more than 20 points). Multivariate prediction models based on regression and classification and regression tree (CART) techniques were developed using the response variables discussed above as the dependent variables and the baseline characteristics as the independent variables.

Results: Different factors contributed to pain and functional outcomes. Patients presenting with neuropathic pain (PainDETECT >= 19) and female gender had higher odds of being pain responders to 10 kHz SCS therapy than did males and those without neuropathic pain. Both higher age and depression score (PHQ-9) independently reduced the odds that a patient would be an ODI responder. Years since diagnosis, the reason the patient was deemed unsuitable for spine surgery, and pain etiology were not predictive of pain or functional outcomes.

Limitations: A retrospective sub-analysis of a single pragmatic randomized controlled trial.

Conclusions: There may be an opportunity to increase pain relief and functional improvement if additional patient screening accompanies the temporary lead trial. The presence of neuropathic pain, female gender, age, and depression had some predictive value, but this analysis demonstrates the treatment efficacy of 10 kHz SCS across a wide range of patients with NSRBP.