Diabetic sensorimotor peripheral neuropathy is the most common complication of diabetes and results in potentially debilitating symptoms, including numbness, tingling, and, frequently, neuropathic pain. Approximately 20% of persons with diabetes will develop painful diabetic neuropathy (PDN) with paresthesia, burning, and shooting pain (1).

Currently, there are no disease-modifying treatments for PDN. Therapeutic goals include symptom management along with behavioral modifications to mitigate further damage (2). Neuropathic pain medications are recommended, including gabapentinoids, serotonin-norepinephrine reuptake inhibitors, tricyclic antidepressants, and opioids. Adherence to commonly prescribed PDN medications is poor due to inadequate pain relief or intolerable side effects.

Spinal cord stimulation (SCS) involves a surgically implanted device delivering mild electrical pulses to modulate chronic pain pathways. High-frequency (10-kHz) SCS provides superior pain relief for chronic back and leg pain, and recent data demonstrate that it also results in substantial pain relief for PDN patients (3,4). This randomized controlled trial evaluated the long-term impact of 10-kHz SCS for PDN patients with refractory symptoms.

Methods have been described previously (4). Participants had symptoms for at least 12 months that were refractory to medications, lower limb pain ≥5 on the 10-cm visual analog scale (VAS), HbA1c ≤10%, and BMI ≤45 kg/m2. Participants were eligible for crossover at 6 months if they had <50% pain relief, they were dissatisfied with treatment, and the investigator deemed it medically appropriate. Temporary trial SCS evaluated eligibility for permanent device implant (Nevro Corp., Redwood City, CA), with success defined as ≥50% pain relief. Neurologists trained investigators to perform comprehensive neurological examinations assessing lower limb motor strength, reflexes, and sensation, including pinprick and 10-g monofilament tests. Paired t tests assessed mean percent change from baseline within treatment groups. Categorical variables were compared between treatment groups using Fisher exact test.

In total, 216 patients were randomized 1:1 to continued conventional medical management (CMM) (n = 103) or the addition of 10-kHz SCS to CMM (n = 113). Treatment groups were well matched for baseline characteristics (4). Among participants assigned 10-kHz SCS + CMM, 104 proceeded to temporary trial SCS and 90 received permanent device implants. In the CMM group, 95 completed 6-month follow-up and 81% (77 of 95) crossed to 10-kHz SCS compared with none from the 10-kHz SCS + CMM arm (P < 0.001). Sixty-four participants received permanent device implants after crossover.

Mean lower limb pain VAS was 7.6 cm (95% CI 7.2–7.9) for 10-kHz SCS + CMM patients at baseline, 1.7 cm (95% CI 1.3–2.1) at 6 months, and maintained at 1.7 cm (95% CI 1.3–2.1) to 12 months, representing 77.1% mean pain relief (95% CI 71.8–82.3; P < 0.001) (Fig. 1A). At both 6 and 12 months, 86% (72 of 84) were treatment responders, defined as those with at least 50% pain relief from baseline (Fig. 1B). For the crossover group, mean baseline lower limb pain VAS was 7.2 cm (95% CI 6.8–7.6) with no change at 6 months but improvement after crossover, similar to the originally assigned 10-kHz SCS group: mean 70.3% pain relief (95% CI 63.4–77.1, P < 0.001), lower limb pain VAS score of 2.0 cm (95% CI 1.6–2.4) (Fig. 1A), and 84% responders (49 of 58) (Fig. 1B).

Figure 1

Pain and neurological results over 12 months. A: Average lower limb pain scores over time for 10-kHz SCS + CMM participants (n = 84, left) and CMM participants with crossover after 6 months (n = 58, right). Participants rated pain on a 10-cm VAS, with 0 representing “no pain” and 10 being the “worst pain imaginable.” Left and right lower limbs were each rated separately, and the scores were averaged together for each participant. Error bars: 95% CI. B: Proportion of pain responders, defined as those with at least 50% pain relief from baseline, at 3, 6, 9, and 12 months for 10-kHz SCS + CMM participants (n = 84, left) and CMM participants with crossover after 6 months (n = 58, right). C: Proportion of participants over time who investigators reported to have improvement on neurological examination for 10-kHz SCS + CMM participants (n = 76, left) and CMM participants with crossover after 6 months (n = 52, right). Assessment included motor strength and reflex testing as well as sensory testing for light touch, pinprick, and 10-g monofilament. All follow-up assessments were compared with baseline, and the investigator categorized motor, reflex, and sensory separately as “improvement,” “no change,” or “deficit.” Overall neurological improvement was defined as an improvement in motor, reflex, or sensory function without a deficit in any category.

Figure 1

Pain and neurological results over 12 months. A: Average lower limb pain scores over time for 10-kHz SCS + CMM participants (n = 84, left) and CMM participants with crossover after 6 months (n = 58, right). Participants rated pain on a 10-cm VAS, with 0 representing “no pain” and 10 being the “worst pain imaginable.” Left and right lower limbs were each rated separately, and the scores were averaged together for each participant. Error bars: 95% CI. B: Proportion of pain responders, defined as those with at least 50% pain relief from baseline, at 3, 6, 9, and 12 months for 10-kHz SCS + CMM participants (n = 84, left) and CMM participants with crossover after 6 months (n = 58, right). C: Proportion of participants over time who investigators reported to have improvement on neurological examination for 10-kHz SCS + CMM participants (n = 76, left) and CMM participants with crossover after 6 months (n = 52, right). Assessment included motor strength and reflex testing as well as sensory testing for light touch, pinprick, and 10-g monofilament. All follow-up assessments were compared with baseline, and the investigator categorized motor, reflex, and sensory separately as “improvement,” “no change,” or “deficit.” Overall neurological improvement was defined as an improvement in motor, reflex, or sensory function without a deficit in any category.

Investigators reported neurological improvements, particularly improved sensory function, maintained over 12 months for the majority of patients with 10-kHz SCS: 68% (52 of 76) of participants originally assigned to SCS and 62% (32 of 52) of participants after crossover (Fig. 1C). Insensate feet limit activities of daily living and may result in debilitating sequelae, including injury from falling, foot ulceration, and lower limb amputation.

There were eight procedure-related infections (5.2%): three resolved with conservative treatments and patients continued in the study, while five (3.2%) required surgical explant of the device. There were no explants for loss of efficacy. Two participants (1.3%) had the location of the implantable pulse generator revised, and one participant (0.6%) experienced lead migration that required a revision procedure; all three continued in the study.

Findings for the crossover group replicated the findings from the original implant group, providing a cumulative sample of 154 implanted patients with long-term data. In addition to a higher proportion of pain responders compared with pharmacotherapy or low-frequency SCS (5), 10-kHz SCS does not induce paresthesia, an advantage for PDN patients with uncomfortable paresthesia at baseline. Additionally, sleep disturbance due to pain, a common ailment for PDN patients, markedly improved by mean 61.7% (95% CI 55.9–67.5) with 10-kHz SCS. This study, the largest randomized controlled trial conducted for SCS treatment of PDN, demonstrates substantial, durable pain relief and potentially disease-modifying neurological improvements over 12 months, providing high-quality evidence in support of 10-kHz SCS for PDN patients with refractory symptoms.

Clinical trial reg. no. NCT03228420, clinicaltrials.gov

Acknowledgments. The authors are grateful to Joe Massaro, professor of biostatistics, mathematics, and statistics at Boston University and director of biostatistics and data management at Harvard Clinical Research Institute, for his independent statistical analyses for this publication. The authors are also indebted to Brian Levy for his thoughtful review and expert feedback on the manuscript.

Funding and Duality of Interest. This study was funded by Nevro Corp. E.A.P. has received consulting fees from Abbott, Medtronic, Neuros Medical, Nevro, Saluda Medical, and Vertos and research support from Medtronic, Neuros Medical, Nevro, ReNeuron, and Saluda Medical. J.A.S. has received research support from Boston Scientific, Nevro, Saluda Medical, and Vertiflex. J.L.W. has received consulting fees from Lilly and Calibr. S.M.S. has received research support from Nevro. K.A. has received consulting fees from Nevro, Nalu, Saluda Medical, Biotronik, and Medtronic and research support from Nevro, Biotronik, Vivex, Saluda Medical, and SPR Therapeutics. M.N.G. has received consulting fees from Nevro, Saluda Medical, and Avanos and research support from Abbott, Saluda Medical, Neuros, Nevro, Nalu, and Avanos. J.X. has received research support from Nevro and National Institutes of Health K08 grant CA228039. C.Y. has received research support from Nevro. A.N. has received consulting fees from Flowonix and Nevro and research support from Nevro. D.G.P. has received consulting fees from Abbott, AIS, Allergan, Amgen, and CornerLoc and research support from Abbott, Biotronik, Flowonix, Nevro, Nuvectra, and Vertiflex. N.D.M. has received consulting fees from Nevro, Salix Pharma, Biodelivery Sciences, and Averitas and research support from Boston Scientific, Medtronic, and Nevro. D.S. has received consulting fees from Abbott, Boston Scientific, Nevro, Vertos, and Vertiflex and research support from Abbott, Biotronic, Nevro, Vertos, and Vertiflex. S.P.L. has received consulting fees from Nevro. J.H.G. has received consulting fees from Stratus Medical and Abbott and research support from SPR Therapeutics and Mainstay Medical. C.E.A. has received consulting fees from Nevro, Vertex, Lilly, Pfizer, and Teva and research support from Allergan, Amgen, DSI, Lilly, Novartis, and Teva. C.E.N. has received consulting fees from Siemens Healthineers, Horizon Therapeutics, Nevro, Neurogastrx, and Exelixis. R.S.T. has received consulting fees from Medtronic, Nevro, and Saluda Medical. N.A.M. has received consulting fees from Boston Scientific, Sollis Therapeutics, Saluda Medical Medical, Nevro, Abbott (formerly Spinal Modulation), Vertos Medical, Nuvectra, and Relievant Medsystems; is a Medical Monitor for Mainstay’s RESTORE clinical trial; and has received research support from Avanos “Halyard,” Mallinckrodt, Mesoblast, and Neuros Medical. E.S.B. and D.L.C. are employees of Nevro Corp. No other potential conflicts of interest relevant to this article were reported.

Author Contributions. E.S.B., K.A., M.N.G., V.G., R.H.B., C.E.A., C.E.N., R.S.T., D.L.C., and N.A.M. contributed to the conception and design of the study. E.A.P., T.G.S., J.A.S., J.L.W., S.M.S., K.A., M.N.G., J.X., C.Y., A.N., D.G.P., K.C.T., M.J.C., V.G., R.H.B., N.D.M., D.S., S.P.L., D.J.D., K.A.S., J.H.G., M.T.B., N.J.H., A.F.I., P.C., and P.W.W. recruited subjects and acquired data. E.A.P., E.S.B., C.E.A., C.E.N., R.S.T., D.L.C., and N.A.M. were involved in the analysis and interpretation of the data along with an independent biostatistician. All authors reviewed the manuscript critically and approved of its content. E.A.P. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Prior Presentation. Parts of this study were presented in abstract form at the American Diabetes Association 81st Scientific Sessions, held virtually 25–29 June 2021; the North American Neuromodulation Society Mid-Year Meeting, Orlando, FL, 15–17 July 2021; the American Society of Pain & Neuroscience Annual Conference, Miami Beach, FL, 22–25 July 2021; the Neuromodulation Society of Australia and New Zealand 14th Annual Scientific Meeting, held virtually 14–15 August 2021; the European Association for the Study of Diabetes Annual Meeting, held virtually 28 September to 1 October 2021; the Congress of Neurological Surgeons Annual Meeting, Austin, TX, 16-20 October 2021; and the Diabetic Foot Conference, San Francisco, CA, 21-23 October 2021.

1
Abbott
CA
,
Malik
RA
,
van Ross
ER
,
Kulkarni
J
,
Boulton
AJ
.
Prevalence and characteristics of painful diabetic neuropathy in a large community-based diabetic population in the U.K
.
Diabetes Care
2011
;
34
:
2220
2224
2
Pop-Busui
R
,
Boulton
AJ
,
Feldman
EL
, et al
.
Diabetic neuropathy: a position statement by the American Diabetes Association
.
Diabetes Care
2017
;
40
:
136
154
3
Kapural
L
,
Yu
C
,
Doust
MW
, et al
.
Comparison of 10-kHz high-frequency and traditional low-frequency spinal cord stimulation for the treatment of chronic back and leg pain: 24-month results from a multicenter, randomized, controlled pivotal trial
.
Neurosurgery
2016
;
79
:
667
677
4
Petersen
EA
,
Stauss
TG
,
Scowcroft
JA
, et al
.
Effect of high-frequency (10-kHz) spinal cord stimulation in patients with painful diabetic neuropathy: a randomized clinical trial
.
JAMA Neurol
2021
;
78
:
687
698
5
Gupta
M
,
Knezevic
NN
,
Abd-Elsayed
A
,
Ray
M
,
Patel
K
,
Chowdhury
B
.
Treatment of painful diabetic neuropathy—a narrative review of pharmacological and interventional approaches
.
Biomedicines
2021
;
9
:
9
Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at https://www.diabetesjournals.org/journals/pages/license.