Low pain scores after spinal cord stimulation (SCS) trials are predictive of successful SCS implants with high sensitivity, according to a recent study. The authors, Charles Odonkor (senior author), Vwaire Orhurhu and colleagues from Harvard Medical School, Boston, USA, report that post-SCS trial pain scores less than or equal to 4.9 had greater than 50% probability of a successful permanent SCS implant (97.14% sensitivity, 44.5% specificity; ROC=0.71).
Further to this, the authors found that post-SCS trial pain scores between 4 and 7 were associated with a significantly higher probability of a successful SCS implant among patients without spine surgery, compared to those who had previously undergone spine surgery. Separate gender analyses also evidenced that females with pain scores between 5 and 7 had a higher probability of a successful SCS implant.
According to Odonkor et al, permanent SCS implants remains a standard intervention used to provide pain relief for chronic pain patients with failed back surgery syndrome (FBSS), complex regional pain syndrome (CRPS) and peripheral vascular disease. The authors acknowledge that several studies have attempted to determine predictors of pain reduction after SCS implantation. For instance, in terms of patient demographic factors, lower age, lack of a psychological comorbidity, and a lower BMI have been associated with better SCS outcomes in previous studies. Regarding implantation-related factors, a shorter interval between onset of symptoms to implantation, decreased complication rates, an increased area covered by the SCS leads and an increased number of leads have previously been found to increase efficacy.
However, prior to the current study, Odonkor and colleagues emphasise that “it remained unclear to what extent pain reduction during the trial period is associated with long-term success of the permanent implant”. With this in mind, the authors investigated if pain scores obtained during the SCS trial serve as reliable predictors of ultimate success of a permanent SCS implant.
Odonkor and team retrospectively identified 88 patients (57% female, median age: 52.5 years) who underwent both an SCS trial and a subsequent percutaneous permanent implantation from 2015–2018. The authors report a mix of paraesthesia-based versus paraesthesia-free programming implants. Pain scores were obtained at baseline, post-SCS trial period, and post-permanent SCS implantation. Successful permanent SCS implant was defined as patients who had more than 50% pain reduction in pain scores (documented as numeric pain rating [NRS] scores).
Of the entire cohort, 70 (79%) had successful permanent SCS implantation. The primary author, Orhurhu also reports that there were no statistical significant differences in demographic variables between patients who receive successful versus unsuccessful SCS implantations, with the exception of overrepresentation of patients with a history of spinal surgery among the subgroup who received unsuccessful permanent SCS implants (p=0.02).
In terms of the association between post-SCS trial pain scores and successful permanent SCS implant, pain scores were significantly lower among patients with successful implants compared to those without (pain scores: 2.14; 95% confidence interval [CI]: 177–2.52 vs. 3.88; 95% CI: 3.11–4.67; p<0.0001).
However, while maintaining that—in the current cohort of patients—males and patients with surgical history with higher pain scores had a lower probability of successful SCS implant comparted to females and nonsurgical patients, respectively, the authors allude to several limitations present in the current study. They write, “As with all other retrospective studies, our work cannot determine a causative relationship between any independent variable and the SCS outcome.”
Odonkor and colleagues put forward that confounding biases, the small sample size and the nature of a single-centre study may compromise the study’s generalisability, and findings may not be transferable to other settings. Furthermore, they highlight an important shortcoming, of which they note is “particularly relevant in studies of patient pain experiences”. Referring to pain as a multifactorial, heterogeneously experienced phenomenon, the authors postulate that the same intervention may yield drastically different results in separate patients. Moreover, they allude to the lack of validity of numeric rating scores as a means of quantifying pain. “Clinicians and researchers should recognise that numeric pain scores are not entirely indicative of clinical pain reduction,” they write.
Thus, Odonkor and team surmise that not only should the measurement of such scores be standardised more systematically, but future studies should aim to consider patients’ perceived responses to pain intervention as an additional measure of success.