Comment on ‘A sham-controlled, randomized trial of spinal cord stimulation for the treatment of pain in chronic pancreatitis’ by Gulisano et al.

We read with interest the sham-controlled randomized trial of spinal cord stimulation (SCS) to assess pain response in patients with chronic pancreatitis (Gulisano et al., 2024). We are grateful to and applaud the authors' efforts who have conducted a challenging study in a population not routinely considered for SCS and for which there is scarce evidence on the use of this intervention, limited to case reports and small case series (Bieze et al., 2024). Despite the authors' best efforts and unquestionable value of this addition to the SCS literature, we have some concerns with the study that should be highlighted.

We are unaware of any studies that evaluated high-frequency SCS (1000 Hz or other) in patients with chronic pancreatitis and as such its efficacy for this population is unknown. Besides the ability to produce paraesthesia-free stimulation and therefore enable patient blinding, we are unsure as to why this frequency was selected by the authors if no evidence of potential effect was previously available. We would also query whether stimulation at 75% subthreshold of sensation would be as effective as at the level of or above sensation threshold. A main limitation of the current evidence base of SCS and sham-controlled trials is that spinal cord fibre activation has not been evaluated; how is it possible to determine if what the patients received was actually an ‘active intervention’ if it was not confirmed whether spinal cord activation occurred (Mekhail et al., 2024)?

We note there was no eligibility criterion for baseline pain intensity. An entry criterion of ≥4 for chronic pain clinical trials has been recommended (Langford et al., 2023). The baseline pain score in Gulisano et al. was 5.2 ± 1.9 and while we acknowledge that patients with pain intensity levels ≥5 are considered for SCS in routine clinical practice, these baseline scores are considerably lower than previous reports in the same population (Bieze et al., 2024) or SCS studies in a neuropathic pain population evaluated in sham-controlled trials (Duarte et al., 2020). Despite already low pain intensity levels at baseline, patients in the study reported approximately 40% reduction in pain intensity during the open-label extension to 12-month follow-up. This reduction represents a clinically meaningful change and as the authors mention, larger than a 20% response observed during sham-controlled phases of pain therapies. Due to the limited evidence of SCS in this population, it would be of interest to understand if clinically meaningful improvements were also observed in other patient-reported outcomes measures collected (Levy et al., 2023).

We note that after anatomical positioning and on-table testing of paraesthesia pain mapping topography, the stimulation was left ‘activated’ until the next day. Only following confirmation of adequate mapping and x-ray position was the sub-perception or sham mode selected which can introduce expectation bias. The crossover phase occurred during the subsequent screening trial period; therefore, it was not possible to ascertain in advance if any of these patients would respond to any form of SCS therapy. The reasons to proceed to permanent SCS implant when no differences were observed between sham stimulation and 1000 Hz stimulation should be further explored. If there were limited benefits, why did 12 out of 16 patients elect to have another procedure of potential limited value and with associated risks? We suggest that future sham-controlled randomized trials start by evaluating treatment response to the type of SCS being investigated during an on-table trial or screening trial phase, followed by a wash-out period of sufficient duration to allow return to baseline scores. Randomization to active or sham arms would only then ensue. Confirmation that SCS therapy achieved the desired effect in the active intervention (i.e. spinal cord activation) or was definitely not achieving this effect in a sham intervention that still uses stimulation is essential to evaluate the true effect of SCS.

Finally, we agree with the authors that the evidence for the use of different SCS stimulation paradigms in patients with chronic pancreatitis is underwhelming and further studies are warranted. We wish to emphasize that the findings from the current study apply to SCS at 1000 Hz only and are not generalizable to other forms of SCS therapy.

There was no funding received in relation to this letter.

JWK is an advisory board member for Boston Scientific, Medtronic, Abbott and Saluda Medical. RVD has previously received consultancy fees from Mainstay Medical, Medtronic and Saluda Medical outside the submitted work. He is an employee of Saluda Medical. SE reports consultancy fees from Medtronic, and Mainstay Medical outside the submitted work. He has received department research funding from the National Institute of Health Research, Saluda Medical and Medtronic. ST has received consultancy fees from Boston Scientific, Mainstay Medical, and Saluda Medica outside the submitted work. He has received department research funding from the National Institute for Health Research, Boston Scientific, Saluda Medical and Mainstay Medical.