Gabapentinoids reduce neuropathic pain by depressing dorsal horn sensitivity through multiple mechanisms. They inhibit neurotransmitter release by binding to the α2δ subunit of calcium voltagegated channels. They also inhibit movement of the α2δ subunit from dorsal root ganglion neurons to central terminals in the dorsal horn, which disrupts the reuse of the α2δ subunit from endosomal compartments. Gabapentinoids inhibit thrombospodin-mediated processes and stimulate the uptake of glutamate by excitatory amino acid transporters. In addition, the mode of action of gabapentinoids has been suggested to involve inhibition of descending serotonergic facilitation, stimulation of descending inhibition, anti-inflammatory actions, and influence on the affective component of pain [23].

Gabapentin was first approved by the FDA as an anticonvulsant in 1993. It has been reported to be effective in alleviating pain and sleep interference associated with DPN [24] and has been approved for the treatment of peripheral neuropathic pain such as postherpetic neuralgia and DPN in the United Kingdom and Australia; however, the USA has approved gabapentin only for the treatment of postherpetic neuralgia [24-26]. In a phase II study of 165 patients who experienced painful DPN for 1 to 5 years and had a pain score of ≥40 mm on the Short-Form McGill Pain Questionnaire visual analogue scale (VAS), the daily pain score at the study end point was significantly lower in patients treated with gabapentin than in those treated with placebo for 8 weeks; the scores decreased from 6.4 to 3.9 in the gabapentin-treated group and from 6.5 to 5.1 in the placebo group (P<0.001) [24]. In another study of 147 patients with painful DPN and an average daily pain (ADP) score ≥4, the ADP score decreased more in patients treated with gabapentin extended-release formula than in those treated with placebo (gabapentin, −2.76; placebo, −1.38; P=0.001) [26]. In a study that compared gabapentin with amitriptyline, gabapentin was more effective in relieving pain caused by DPN [27].

Pregabalin is a potent and selective ligand for the α2δ subunit of the voltage-gated calcium channel and has been approved for the treatment of painful DPN, postherpetic neuralgia, and spinal cord injury in the USA, Europe, and Canada [1,28]. In a multicenter study of 146 patients with painful DPN for 1 to 5 years and a pain score of ≥40 mm on the Short-Form McGill Pain Questionnaire or an ADP score of ≥4 on an 11-point pain numerical rating scale (NRS), pregabalin significantly reduced the mean pain score compared with placebo (pregabalin, from 6.5 to 4.0; placebo, from 6.1 to 5.3; P<0.0001) [29]. In a study of 338 patients with a 1 to 5-year history of DPN and an average weekly pain score of ≥4 on an 11-point NRS, pregabalin at 300 or 600 mg/day significantly reduced the mean pain score compared with placebo (difference from placebo: 300 mg/day, −1.26; 600 mg/day, −1.45; P=0.0001) [30]. In addition, pregabalin not only relieved pain associated with DPN and but also improved sleep interference and the profile of mood states scores in patients with DPN [29-31]. By contrast, some studies have reported that pregabalin is not more effective than placebo for improving DPN [32,33]. The possible explanations for the lack of effect of pregabalin on painful DPN include the use of low-dose pregabalin, a placebo response in people with painful DPN, or a carryover effect. However, in meta-analysis of studies of patients with painful DPN, pregabalin was more effective than placebo in improving neuropathic pain associated with DPN (change in least squares [LSs] mean change in mean pain score: –0.56 to –1.26) [34-36].

Mirogabalin is a novel selective oral ligand for the α2δ subunit and is known to have a slower dissociation rate from the α2δ subunit than pregabalin. Mirogabalin has been shown to improve painful DPN in phase II and III trials [37-39]. In a phase II study, mirogabalin and pregabalin were administered to 452 adults with painful DPN for more than 6 months, and the LS mean difference in the change in ADP score from the baseline was evaluated. The effects of pregabalin and placebo were similar after the 3-week treatment, but mirogabalin at a dose of 15, 20, or 30 mg/day was significantly better than placebo in inducing a change in ADP score (–0.94, –0.88, and –1.01, respectively) [37]. In a phase III study of 834 Asian patients, a mirogabalin dose of 15, 20, or 30 mg/day reduced the ADP score by –1.34, –1.47, and –1.81, respectively; this pattern suggests that mirogabalin relieved DPN pain in a dose-dependent manner [39].

The side effects reported for gabapentinoids include dizziness, somnolence, gastrointestinal complaints, sedation, ataxia, peripheral edema, headache, and postural hypotension, etc. [34,37,40]. Because of these side effects, caution is needed when prescribing this drug, especially in elderly patients. Furthermore, gabapentin and pregabalin are excreted primarily unchanged in the urine, renal dose adjustment is required [41]. The recommended maximal daily dose of gabapentin is 1,500 mg in people with grade 3 chronic kidney disease (CKD), 700 mg in those with grade 4 CKD, 300 mg in those with grade 5 CKD, and 100 to 300 mg after dialysis in dialysis patients. For pregabalin, the recommended maximal daily doses are 300, 150, and 75 mg for people with grade 3, grade 4, and grade 5 CKD, respectively, and 75 to 150 mg after dialysis for dialysis patients [42].

Neurotransmitters such as serotonin and norepinephrine are involved in the descending inhibitory nociceptive pathway. Serotonin–norepinephrine reuptake inhibitors (SNRIs) inhibit reuptake of serotonin and norepinephrine, which increases descending inhibition and reduces pain associated with DPN [43]. Among the available SNRIs, duloxetine is recommended for the initial symptomatic treatment of painful DPN, and venlafaxine is also used for the treatment of painful DPN [1]. In Korea, duloxetine is currently available, but venlafaxine is not.

Duloxetine, a potent SNRI, reduced neuropathic pain in an animal model and significantly improved painful physical symptoms associated with depression in a study of patients with major depressive disorder [44,45]. Duloxetine is therefore expected to be effective for treating painful DPN. In a study that compared the effects of duloxetine and placebo in 457 patients with a score of ≥3 on the MNSI, duloxetine at a dose of 60 or 120 mg/day improved pain from the first week of administration and was continuously effective throughout the study compared with placebo. The differences in ADP scores compared with placebo were −1.17 (95% confidence interval [CI], −1.84 to −0.50) for 60 mg/day and −1.45 (95% CI, −2.13 to −0.78) for 120 mg/day [46]. Meta-analysis of studies that compared duloxetine and placebo concluded that duloxetine is effective in improving painful DPN [47,48].

Another SNRI, venlafaxine, has also been shown to be effective in improving painful DPN by reducing VAS pain intensity by 50% after 6 weeks of administration in a study of 244 patients with painful DPN (P<0.001 vs. placebo) [49]. In studies that have compared the effects of venlafaxine and other drugs, venlafaxine was as effective as imipramine and carbamazepine for improving pain associated with DPN [50,51].

Nausea, somnolence, dizziness, constipation, and loss of appetite have been reported as common side effects of duloxetine and venlafaxine, but they are considered relatively safe drugs to use [46,49].

TCAs are thought to relieve pain by inhibiting the reuptake of noradrenaline and serotonin in presynaptic neurons and by antagonizing N-methyl-D-aspartate receptors, which mediate hyperalgesia and allodynia [52-54]. TCAs include tertiary amines such as imipramine and amitriptyline, and secondary amines such as desipramine and nortriptyline; secondary amines are thought to be better tolerated than tertiary amines [55].

A small (n=29) randomized, crossover study using “active” placebo, which mimics the side effects of amitriptyline, showed the superiority of amitriptyline in relieving pain associated with DPN compared with placebo [56]. In later, larger studies, amitriptyline was used as active comparator to gabapentin, pregabalin, and duloxetine, and was shown to have similar effects in relieving DPN pain as these other drugs [57-59].

Despite their promising effects and low cost, TCAs should be prescribed cautiously because of the side effects such as dry mouth, constipation, urinary retention, and orthostatic hypotension. TCAs are contraindicated in patients with glaucoma and cardiac conduction disturbances [55,60].

Increased oxidative stress caused by free radical formation and defects in antioxidant defense related to hyperglycemia is thought to cause endoneurial hypoxia and nerve dysfunction, which contribute to DPN [61]. Preclinical studies have shown that administration of the antioxidant alpha-lipoic acid reduces neurovascular abnormalities associated with DPN [62-65].

In a clinical study, intravenous alpha-lipoic acid and placebo were administered for 3 weeks to 328 patients with symptomatic DPN and intravenous alpha-lipoic acid induced significant improvement in pain associated with DPN compared with placebo. The total symptom score in the feet decreased by –4.5±3.7 (–58.6%) points in the group treated with alpha-lipoic acid at a dose of 1,200 mg (P=0.003) and by –5.0±4.1 (–63.5%) points at a dose of 600 mg (P<0.001) [66]. By contrast, a comparison of the effects of oral alpha-lipoic acid and placebo given for 7 months after intravenous alpha-lipoic acid administration for 3 weeks in 509 patients found no significant difference in total symptom score between alpha-lipoic acid and placebo [67]. High doses may be necessary to see the effects of alpha-lipoic acid on painful DPN, but it has not received FDA approval for DPN treatment. Nevertheless, alpha-lipoic acid is the most often prescribed agent as a monotherapy or as a combination therapy in Korea [22].

Carbamazepine, which is used in the treatment of simple and complex partial seizures of epilepsy and generalized convulsions, inhibits the secretion of neurotransmitters such as glutamine by blocking presynaptic voltage-sensitive sodium channels in the central nervous system (CNS) [68]. In a study that compared the effects of carbamazepine, venlafaxine, and pregabalin in 257 patients with clinically definite DPN, carbamazepine significantly decreased the VAS score (from 74.5 to 39.6, P=0.0001) and showed similar effects as venlafaxine and pregabalin in improving scores for sleep, mood, and work interference [51].

Oxcarbazepine, a derivative of carbamazepine that has better safety and tolerability, has also been reported to be effective in improving pain associated with DPN [69,70]. In a meta-analysis, the percentage of patients with DPN whose pain was “improved” or “very much improved” was 45.9% in those treated with oxcarbazepine and 30.1% in those treated with placebo. However, the quality of evidence to support the effectiveness of oxcarbazepine for treating DPN was very low and the serious adverse effects occurred more frequently in oxcarbazepine than placebo (8.3% vs. 2.5%) [71]. In this regard, oxcarbazepine should be prescribed carefully in patients with DPN.

Capsaicin is an alkaloid extracted from red chili peppers and reduces painful stimuli to the CNS by removing substance P from vanilloid nerve receptors [72]. Therefore, capsaicin was expected to be effective for treating painful DPN. Topical capsaicin formulations in the form of 0.025% capsaicin gel or 0.075% capsaicin lotion were not found to be superior to placebo in relieving painful DPN [73,74]. However, in a study in which an 8% capsaicin patch or placebo patch was applied to painful areas of the feet for 30 minutes in 369 patients with painful DPN, the percentage reduction in ADP score from the baseline to weeks 2 to 8 was significantly higher for the 8% capsaicin patch than placebo (–27.4% vs. –20.9%, P=0.025) [75]. Accordingly, in 2020, the FDA approved the 8% capsaicin patch for the treatment of neuropathic pain associated with DPN of the feet in adults [76]. Application site reactions such as a burning sensation and application site pain have been reported as side effects of capsaicin patch [73-75]. In Korea, only the lower concentrated topical capsaicin is available.

Antioxidant supplementations such as benfotiamine, a fat-soluble derivative of thiamine, and gamma linolenic acid, an omega-6 fatty acid, were used in some countries for DPN treatment. Even though these drugs have potential mechanism to prevent or treat DPN [77,78] and very low chance to induce side effects, but large-scale randomized controlled trials to show its efficacy on DPN are scarce. Aldose reductase inhibitor has been developed as a promising drug because it targeted a rate-limiting enzyme of polyol pathway. However, it could not address symptomatic improvement of DPN in clinical trials, but it only improved nerve conduction velocity [79,80]. Another pharmacological agent to be used for painful DPN is opioid drug such as tapentadol and tramadol. However, considering lack of longterm treatment data and side effects [81] this class of drug is hardly recommended for the treatment of DPN [82].

Transcutaneous electrical nerve stimulation (TENS) is a noninvasive method of neuromodulation using cutaneous adhesive electrodes to apply pulsed electrical stimulation that stimulates A-beta fibers with the goal of indirectly inhibiting nociceptive transmission in the spinal cord dorsal horn [87-89]. TENS also contributes to pain relief by inducing the release of endogenous opioids [87,90]. Electrical stimulation applied in TENS can be modified in terms of frequency (stimulation rate), intensity and duration: Low-frequency TENS is defined as ≤10 Hz and often used at higher intensities inducing motor contraction, while high-frequency TENS is defined as ranging up to 50 or 100 Hz and above and used at lower intensities [91-93]. When low-frequency TENS was applied to 19 patients suffering from mild-to-moderate symptomatic DPN, the new total symptom score (NTSS-6) VAS decreased significantly compared to placebo [94]. When traditional TENS (80 Hz) or acupuncture-like TENS (2 Hz) was applied to five patients suffering from DPN, significant pain relief occurred in all five patients, but there was no placebo in this study [95]. In addition, in a study comparing micro-TENS (2 Hz) and placebo in 41 diabetic patients, micro-TENS did not show significant improvement in pain relief compared to placebo [90]. No specific adverse reactions were reported in patients treated with TENS in the studies [90,95].

Spinal cord stimulation (SCS) is a treatment for chronic pain that uses electrical stimulation of the dorsal columns of the spinal cord. SCS includes low-frequency SCS (frequency 10–100 Hz, pulse width 100–1,000 ms, and amplitude 1–10 mA) that applies paresthesia-based stimulation and high-frequency SCS (frequency 1–10 kHz, pulse width 30–150 ms, and amplitude 1–5 mA) that applies stimulation below the paresthesia threshold [96]. Although the mechanisms of action of SCS have not been clearly identified, it is thought to involve both spinal and supraspinal effects and has been shown to be effective for the treatment of various neuropathic pain such as failed back surgery syndrome and complex regional pain syndrome [97-100].

In a study that compared the effects of SCS and placebo for 6 months in 60 patients with DPN refractory to medical therapy, the VAS score decreased significantly only in the SCS group (Table 2) [101]. Similarly, in a study that analyzed the effect of SCS in 36 patients whose DPN did not improve with conventional therapy, the percentages of patients who achieved ≥50% pain relief were 59% in the SCS group and 7% in the best medical treatment group; these findings indicate that SCS was more effective for DPN pain relief [102]. In a study that analyzed the effect of high-frequency SCS on 216 patients with DPN refractory to medical therapy and a VAS score ≥5, the percentages of patients who achieved ≥50% pain relief were 79% in the SCS group and 5% in the conventional medical management group (P<0.001) [103]. The FDA approved SCS devices for the treatment of painful DPN in 2022.

The side effects reported for SCS include infection and wound dehiscence; one patient who complained of postdural puncture headache experienced a lethal subdural hematoma 3 days after the procedure [101-104].