Electrical stimulation as an adjunctive therapy for diabetic A systematic review and meta‐analysis

Authors: Xiaodong Lan, Zhenjia Huang, Yan Zheng, Zhiyong Huang, Yong Tang, Tao Zhou, Chao Wang, Yan Ma, Dan Li

Abstract

Diabetic ulcers are chronic wounds that are notoriously difficult to treat, leading to significant physical and psychological distress and increased healthcare costs. Their multifactorial aetiology necessitates long‐term interdisciplinary collaboration and various complementary treatment measures. While numerous studies suggest that electrical stimulation (ES) positively impacts diabetic ulcer healing, the robustness and consistency of these findings require further evaluation to optimize clinical application. We searched databases including PubMed, the Cochrane Library, Embase, Web of Science and the China National Knowledge Infrastructure (CNKI). Only randomized clinical trials (RCTs) comparing ES treatment to placebo or conventional treatment were included. Extracted information included objective healing measures and data for assessing effect sizes. Ten RCTs involving 451 patients met inclusion criteria. ES improved ulcer healing rate compared to control or placebo (MD 20.37, 95% CI: 16.89–23.85, p <0.001) and increased the number of healed ulcers (RR 1.45, 95% CI: 1.18–1.78, p <0.001), with both results being statistically significant. The observed benefits are likely due to the positive effects of ES on the vascular and neurological functions of the lower limbs in patients with diabetic ulcers. Both low‐frequency, moderate‐intensity alternating current and low‐intensity or high‐voltage direct current have demonstrated efficacy in promoting ulcer healing. The results suggest ES may be a promising approach of managing diabetic ulcers. However, the optimal method of ES application remains undetermined; therefore, high‐quality and large‐scale studies are essential.

INTRODUCTION

Diabetic ulcers (DUs) are a serious complication in patients with diabetes mellitus, typically associated with peripheral neuropathy, vascular disease and infections. ^1^ , ^2^ These challenging‐to‐heal ulcers often lead to amputation, significantly impacting patients' quality of life and incurring substantial healthcare costs. ^3^ , ^4^ , ^5^ Conventional treatments, including offloading interventions, antibiotics, medications, advanced dressings and surgical debridement, frequently fail to effectively address chronic ulcers characterized by poor circulation and sensory loss. ^1^ , ^6^

In recent years, electrical stimulation (ES) therapy has garnered significant attention and is considered a promising adjunctive treatment for DUs. ^7^ The medical use of ES dates back to ancient civilizations, where electrically charged organisms like electric eels were employed to alleviate pain and enhance blood circulation. ^8^ , ^9^ In contemporary research, ES is defined as the application of an electrical current through an electrode placed on or near the wound. ^9^ ES has been suggested to facilitate wound healing through various mechanisms. ^10^ It is thought to mimic or enhance the natural bioelectric currents within the wound site, thereby promoting the proliferation, migration and functional activation of fibroblasts, epithelial cells and keratinocytes. ^11^ , ^12^ , ^13^ It may also modulate inflammation by recruiting immune cells and altering cytokine expression, including interleukin‐1 (IL‐1), tumour necrosis factor‐alpha (TNF‐α) and interleukin‐6 (IL‐6). ^14^ Furthermore, ES could improve blood flow, enhance nutrient delivery and facilitate waste removal, all of which support tissue repair. ^15^ It is believed to reduce bacterial colonization by disrupting bacterial cell walls and altering local pH, with increased peroxyl radicals exerting antimicrobial properties. ^16^ ES might provide analgesia by releasing endogenous opioid peptides. ^17^

The types of electrical current used in ES therapy include direct current (DC), alternating current (AC) and pulsed current (PC), with several subtypes for each. Nearly all types of ES have shown positive effects on skin wound healing. ^18^ Notably, different types of electrical currents produce varying effects at different stages of wound healing, thereby offering diverse clinical treatment options. ^19^ Additionally, ES is a non‐invasive therapy that can be administered at home and combined with other treatments to enhance its efficacy. ^20^ Technological advancements have introduced innovative ES therapies providing new strategies and options for the treatment of DUs. ^18^ Several clinical randomized controlled trials have investigated the effects of different types and intensities of ES on diabetic ulcer healing. The majority of studies have confirmed the positive impact of ES‐assisted therapy. ^21^ , ^22^ , ^23^ However, the variability in ES parameters (current type, intensity, frequency and duration) and the severity of diabetic peripheral neuropathy and vascular disease result in differing wound healing responses. ^24^ Consequently, many questions remain regarding the optimal application of ES therapy in clinical settings for diabetes. Therefore, this systematic review aims to critically evaluate published randomized trials on the efficacy and safety of ES therapy for DUs and to identify the most effective ES approaches for treating these wounds.

METHODS

This study was conducted in strict adherence to the guidelines outlined in the Cochrane Handbook for Systematic Reviews (http://www.cochranelibrary.com/). All procedures for systematic reviews and meta‐analyses complied with the standards set by the Preferred Reporting Items for Systematic Reviews and Meta‐Analyses (PRISMA) statement.

Literature search

The search terms utilized included ‘ulcer’, ‘foot ulcer’, ‘diabetes mellitus’, ‘diabetes complications’, ‘diabetic foot’, ‘electric stimulation therapy’, ‘electric field’, ‘direct current’ and ‘alternating current’. These subject terms or keywords were employed to identify relevant published studies. The databases searched encompassed PubMed, Cochrane Library, Embase, Web of Science and the China National Knowledge Infrastructure (CNKI) Database, covering publications from their inception until April 2024. The search strategy adhered to the protocol defined in the Cochrane System Evaluator's Handbook (see Data S1 for detailed search strategies). Additionally, a manual search of the bibliographies of pertinent articles and existing reviews was conducted to identify studies not captured in the electronic database searches.

Inclusion and exclusion criteria

Inclusion criteria

**Type of ** Randomized controlled trial (RCT). Subjects: Diabetic patients of any age and gender; the experimental group received ES‐assisted routine wound care for ulcers, while the control group received either ES‐placebo‐assisted or routine wound care only. There were no restrictions on the ulcer conditions, blood glucose levels or medications used by the two groups. **Effectiveness ** The primary outcomes include the ulcer healing rate (calculated as [(Initial area–Final area)/Initial area] × 100%), healing time, number of healed ulcers. The secondary outcomes consist of skin temperature, local skin perfusion pressure, blood flow, tissue oxygen saturation (StO₂), plantar vibration perception threshold (VPT), ankle‐brachial index (ABI) and pain assessment. **Data ** Sufficient data must be available for meta‐analysis. Language: No restrictions on language.

Exclusion criteria

Studies comparing ES with other adjuvant therapies.Retrospective studies, reviews, case reports and conference literature.Statistical data from original studies that could not be translated or applied.Duplicate publications or literature not accessible in its original form through various channels.

Literature screening and data extraction

The literature search and screening were performed independently by two researchers. Articles were initially screened by title, followed by abstract, and finally, the full text if necessary. Publications that were clearly irrelevant to the study's purpose were excluded, and duplicates were identified and removed. Full‐text articles meeting the inclusion criteria were obtained for detailed review. At each stage, disagreements between the reviewers were resolved through discussion, involving senior reviewers and corresponding author when necessary.

Detailed information from each included study was independently extracted and summarized by the two reviewers using a predesigned data collection form. Data from repeated publications were included only once. Efforts were made to obtain any missing data by contacting the study authors. The data collection form the information about the studies (first author, sample size, year of publication and country); characteristics of the study population (age, gender and ulcer status); details of the interventions (ES device and current type, intensity and duration); primary outcomes (ulcer healing rate, ulcer healing time and number of healed ulcers); secondary outcomes (skin temperature, local skin perfusion pressure, blood flow, StO2, VPT, ABI and pain); and quality indicators (randomization, allocation concealment, blinding and data completeness).

Quality assessment

The included clinical controlled studies were independently assessed by two trained researchers using the Cochrane Risk of Bias Tool, which covers seven (1) generation of randomized sequences, (2) allocation concealment, (3) blinding of investigators and subjects, (4) blinding of study outcomes, (5) integrity of outcome data, (6) selective reporting of outcomes and (7) other biases. Disagreements were resolved through negotiation and discussion until a consensus was reached, with unresolved issues escalated to senior reviewers and corresponding authors.

Statistical analysis

Meta‐analysis was performed using R 4.3 software. Measurement information was expressed as mean x¯ and standard deviation (s), with the mean difference (MD) used as the effect indicator for continuous data, and count data were expressed as numbers (n), using relative risk (RR) as the effect indicator. Furthermore, 95% confidence intervals were calculated for both MD and RR. Heterogeneity was assessed using the Q‐test (significance level α = 0.10) and quantified with the I ^2^ test. If p ≤0.1 and I ^2^ ≤50%, it indicated homogeneity among the study results, and a fixed‐effects model was used for the meta‐analysis. If there was significant heterogeneity (p >0.1 or I ^2^ >50%), a random‐effects model was applied. Publication bias was assessed quantitatively using Egger's test, with a p‐value <0.05 indicating the presence of publication bias, and a p‐value ≥0.05 indicating its absence. Additionally, a sensitivity analysis was conducted to evaluate the robustness of the findings.

RESULTS

Results of literature search and screening

The initial search across various databases identified 1170 articles. After removing duplicates and irrelevant studies, 87 articles remained. Manual searches of relevant references did not yield any additional articles for inclusion in the meta‐analysis. Based on titles and abstracts, 43 articles met the eligibility criteria. After full‐text reviews, 10 articles ^25^ , ^26^ , ^27^ , ^28^ , ^29^ , ^30^ , ^31^ , ^32^ , ^33^ , ^34^ were selected for inclusion, none of which were multicentre clinical RCTs (Figure 1).

Characteristics of the included studies

Table 1 summarizes the information of the 10 RCTs included in this study. All 10 studies were randomized, controlled, single‐centre trials. In six studies, the control group received sham ES, while in the other four studies, the control group underwent routine wound care. Geographically, the trials were conducted in the United States (four), Iran (three), China (two) and Sweden (one). A total of 451 diabetic patients (229 in the ES group and 222 in the control group) participated in these studies, with ages ranging from 30 to 82 years, and the majority being male (280, 63.08%). Most patients had foot ulcers (including ankle ulcers), with some having ulcers on the lower leg. Ulcer sizes varied significantly, ranging from 1.6 to 28.2 cm^2^.

Table 2 summarizes the specific details of the ES interventions utilized in the included studies. With the exception of one study that did not provide details on the ES device used, ^32^ all other studies employed percutaneous ES instruments. The specific devices varied across studies, with electrode placement differing across experiments. In four studies, cathode electrode pads were placed around the wounds while anode electrode pads were placed proximally on the leg. ^26^ , ^30^ , ^31^ , ^32^ In one study, electrode pads were positioned on either side of the nerve trunk innervating the ulcerated area. ^29^ Another study placed pads on the Taixi acupoint of the ankle, ^33^ one on the fibula head of the affected limb, ^34^ and another used special socks with integrated electrode pads. ^27^ In two studies, electrode pads were placed around the ulcer wound. ^25^ , ^28^ The types of current included biphasic square‐wave ^25^ , ^26^ , ^29^ or sine‐wave pulses, ^28^ , ^34^ twin peak monophasic pulse, ^27^ cathodic DC ^30^ , ^31^ , ^32^ and high‐voltage DC. ^33^ Parameter settings such as voltage and current intensity, pulse width, frequency and treatment regimen varied greatly. Generally, ES therapy employs AC with a low‐frequency (8–100 Hz), moderate‐intensity (15–30 mA) and various pulse widths (100–1000 μs) ^25^ , ^26^ , ^28^ , ^29^ , ^34^ and DC with low intensity (1–4 mA) ^30^ , ^31^ , ^32^ or high voltage (150–350 V). ^33^ However, the duration of most included studies was 4 weeks, ^26^ , ^28^ , ^29^ , ^30^ , ^31^ , ^32^ , ^33^ with two studies extending to 12 weeks ^25^ , ^27^ and one study to 3 weeks. ^34^ Most studies administered ES therapy in hospitals with trained personnel. Only three studies involved patients using home‐based ES devices after receiving instruction. ^25^ , ^27^ , ^33^ Most studies reported primary outcomes including ulcer healing rate (proportion of ulcers reduced in area), ulcer healing time or the number of healed ulcers. Four studies also examined local skin temperature, perfusion pressure, blood flow, StO2 and plantar VPT. ^28^ , ^30^ , ^31^ , ^33^ Additionally, two studies measured nitric oxide (NO), vascular endothelial growth factor (VEGF), hypoxia‐inducible factor 1 (HIF‐1) and soluble VEGF receptor 2 (sVEGFR‐2) in blood or wound exudate, while one study evaluated the ABI and pain. ^34^

Bias risk assessment results

According to the quality assessment results (Table 3), all 10 studies provided detailed descriptions of the randomization methods used for study participants. Group concealment was explicitly implemented in four studies, ^25^ , ^30^ , ^32^ , ^33^ blinding of subjects and relevant staff was applied in four studies, ^25^ , ^27^ , ^30^ , ^33^ and blinding of outcome assessors was conducted in six studies. ^25^ , ^26^ , ^27^ , ^28^ , ^30^ , ^33^ None of the studies had missing data. However, four studies lacked a detailed description of specified outcomes, rendering the risk of selective reporting unclear. ^29^ , ^30^ , ^31^ , ^32^ All 10 studies explicitly described their funding sources and conflict of interests, indicating a low risk of other biases. Overall, the risk of bias was low for all included studies (Figure 2).

Systematic evaluation and meta‐analysis results

Ulcer healing rate

All studies provided data on changes in the cross‐sectional area of ulcers, with the ulcer healing rate as the primary endpoint. There was no significant heterogeneity between them (p = 0.5, I ^2^ = 0.0%). Consequently, a meta‐analysis using a fixed‐effects model showed that the ulcer healing rate in patients with DUs treated with ES increased significantly by 20.37% compared to the control group (95% confidence interval [CI]: 16.89–23.85%, p <0.001), as shown in Figure 3.

Number of healed ulcers

Four studies reported the number of patients with DUs who were completely healed following treatment. ^25^ , ^26^ , ^27^ , ^29^ Minimal heterogeneity was observed between these studies (p = 0.2, I ^2^ = 39.8%). A meta‐analysis using a fixed‐effects model indicated a significant increase in complete ulcer healing in patients treated with ES compared to controls, with a relative risk (RR) of 1.45 (95% CI: 1.18–1.78, p <0.001), as illustrated in Figure 4.

Other outcomes

Mohajeri‐Tehrani et al. ^30^ found that ES increased foot skin temperature by 11.5% ± 1.6% and leg skin temperature by 6.22% ± 3.45% during the 12th session. In the study by Asadi et al., ^31^ foot skin temperature changes in the ES group were significantly greater than in the control group at session 12 (0.66 ± 0.23°C vs. 0.25 ± 0.27°C, p = 0.007).

On the first day of the Petrofsky et al.'s ^28^ study, skin blood flow in and around the wound increased to 102.3% ± 25.3% with heat alone and to 152.3% ± 23.4% with combined ES and heat. Zulbaran‐Rojas et al. observed that after 4 weeks of treatment, blood perfusion pressure decreased in both the ES and placebo groups (ES: 71.0 ± 12.2 to 65.1 ± 19.1 mmHg; 78.1 ± 18.6 to 72.2 ± 18.1 mmHg), while StO2 increased (ES: 69.3 ± 21.3 to 72.3% ± 21.5%; 73.1 ± 13.5 to 73.9% ± 12.1%). These changes did not reach statistical significance within or between groups. However, in the moderate‐to‐severe peripheral artery disease (PAD) subgroup, StO2 significantly improved (baseline: 47.3% ± 20.1% to week 63.3% ± 22.3%, p = 0.015) in the ES group. Additionally, ES resulted in a significant improvement in plantar VPT after 2 weeks (20.2 ± 14.0 vs. 32.1 ± 9.2 volts, p = 0.049). ^33^ One study demonstrated that the ABI significantly improved in the ES group compared to the control group (0.86 ± 0.25 vs. 0.75 ± 0.03, p = 0.047). What's more, pain relief was substantially greater in the ES group, as indicated by lower Visual Analogue Scale (VAS) scores (2.69 ± 0.18 vs. 4.98 ± 0.36, p <0.000). ^34^

In addition, two studies examined the expression of key bioactive factors in blood or wound secretions. Mohajeri‐Tehrani et al. ^30^ reported a significant increase in blood levels of VEGF (109.28 ± 67.30 vs. 34.79 ± 13.20 pg/mL, p = 0.005) and NO (44.21 ± 14.00 vs. 35.25 ± 11.00 pg/mL, p = 0.040) due to ES. Asadi et al. ^32^ observed that ES positively influenced the release of localized wound HIF‐1 (+61.98 vs. −3.85 pg/mL, p = 0.009) and VEGF (+36.77 vs. +4.15 pg/mL, p = 0.007) at the first section, but had no effect on the release of soluble sVEGFR‐2 (−0.06 vs. −0.03 pg/mL, p = 0.46) and NO (−0.06 vs. −0.03 μM, p = 0.14).

Adherence and adverse reactions

Zulbaran‐Rojas et al. ^33^ reported a high patient adherence rate to daily home ES therapy (93.9%). Peters et al. ^27^ demonstrated that in the ES treatment group, patients with high adherence achieved a 71% healing rate, whereas those with low adherence had a healing rate of only 50%. None of the studies mentioned adverse events associated with ES treatment, such as skin redness, pain, rash or allergies. However, two studies reported participant withdrawals due to severe infections. ^27^ , ^33^

Sensitivity analysis and publication bias

The sensitivity analysis results indicated robustness in the meta‐analysis for both the ulcer healing rate and the number of healed ulcers (see Data S2 for sensitivity analysis). Egger's test results revealed no publication bias for the ulcer healing rate (p = 0.93), but there was evidence of publication bias for the number of healed ulcers (p <0.05).

DISCUSSION

Previous reviews have suggested that ES has beneficial effects on various stages of skin wound healing in both chronic and acute conditions. ^7^ , ^18^ , ^19^ , ^35^ Our study further confirms the beneficial effects of ES therapy on DUs, highlighting that ES significantly improves the healing rate of DUs and the number of healed ulcers. These therapeutic effects can be attributed to the positive influence of ES on blood vessels and nerve function. This is demonstrated by favourable changes in secondary outcomes such as blood flow, plantar VPT and ABI. Furthermore, these outcomes may be related to the increased release of VEGF, NO and HIF‐1 in the blood or ulcer exudate. ^30^ , ^32^

Multiple reviews conclude that the selection of ES devices and parameters is critical to the treatment outcomes of DUs. ^21^ , ^22^ , ^23^ , ^24^ , ^35^ Almost all devices generating electrical currents used in our study essentially involved percutaneous ES, which produces various physiological effects by stimulating tissues such as nerves, muscles and blood vessels. ^10^ Typically, electrodes are positioned around the wound; however, their placement may vary according to the specific requirements of each device. Some devices are designed specifically for nerve stimulation, ^29^ , ^34^ while others incorporate acupuncture stimulation ^33^ or integrate electrode pads into wearable devices attached to specific acupuncture points. ^27^ In recent years, innovative ES therapy devices, including frequency rhythm ES systems, wireless microcurrent stimulation devices, biofeedback ES devices and microbattery‐embedded bioelectric dressings, have demonstrated promising outcomes in the management of chronic wounds. ^36^ , ^37^ , ^38^ , ^39^ The intelligence and miniaturization of these devices have improved patient compliance with ES therapy and have contributed to its wider application.

Currently, most studies utilize AC or DC to assist in treating DUs. ^25^ , ^26^ , ^27^ , ^28^ , ^29^ , ^33^ However, due to the wide variety of current application parameters and modes, establishing a standardized treatment approach remains challenging, consistent with previous reviews and meta‐analyses. ^21^ , ^22^ , ^23^ , ^35^ For AC, symmetric biphasic pulse currents with low‐frequency, moderate‐intensity and various pulse widths significantly enhance the healing of DUs. However, Baker et al. ^26^ found that symmetric biphasic square‐wave currents were ineffective for wound healing, while asymmetric currents were beneficial. ^25^ , ^28^ , ^29^ They suggest that asymmetric currents create a specific polarity in deep tissues, but we find this explanation unconvincing. Regarding DC, evidence suggests a generally effective ES treatment involves using low‐intensity cathodal DC by placing the negative electrode at the wound edge as the active electrode and adjusting the current intensity to the sensory threshold for 1 h per day, 3 days per week. ^30^ , ^31^ , ^32^ Conversely, an animal study indicated that anodal DC might be more effective than cathodal DC in healing skin wounds. ^40^ This discrepancy could be due to different wound types and varying magnitudes of DC. Additionally, patients with type 2 diabetes ulcers exposed to pulsating electrostatic fields exhibited significantly faster wound healing rates compared to localized ES treatments. ^41^ Studies and reviews have shown that combining ES with other wound treatments, such as negative pressure, heat therapy or decompression therapy, can create a synergistic effect and improve outcomes in chronic, non‐healing wounds. ^26^ , ^28^ , ^42^ , ^43^

Peripheral neuropathy and vascular disease in diabetic patients may reduce the effectiveness of ES therapy. ^44^ Peripheral neuropathy alters the bioelectrical properties of the skin, diminishing ES conduction effectiveness and potentially causing patients to inaccurately perceive the current intensity, necessitating parameter adjustments. ^45^ This was corroborated by Baker et al., ^26^ who documented lower healing rates. Interestingly, Zulbaran‐Rojas et al. ^33^ found that in patients with DUs and moderate‐to‐severe PAD, ES's effect on improving blood flow was more pronounced. The authors suggest that since the healing process of these wounds is more hindered, any intervention improving blood flow and tissue oxygenation likely results in a greater relative improvement. Over extended periods, ES therapy can provide relief from local ischemia caused by peripheral arterial disease. ^46^ This has been indirectly confirmed by the use of ES in patients with severe peripheral arterial occlusive disease (PAOD) and ischemic wounds below the knee. ^47^ , ^48^ As well, Tan et al. ^34^ reported that ES treatment could improve the ABI, indicating that ES might restore some functionality in lower extremity arterial diseases.

In our study, adherence was identified as a potentially important factor affecting wound healing in ES therapy, although this conclusion is drawn from data reported in only two trials. Home‐use devices improved patient compliance, resulting in a higher wound healing rate of 71%. ^27^ Similar findings have been observed in other wound types, suggesting that patients with good adherence tend to receive longer and more frequent ES treatments, indicating a dose effect for ES. ^49^ , ^50^ ES therapy is considered a non‐invasive treatment with no reported side effects. ^24^ , ^35^ However, several studies included in our analysis reported infection‐related withdrawal symptoms, ^27^ , ^33^ suggesting a potential infection risk associated with ES therapy. This might be related to the frequent use of electrode pads, wires and other components during treatment. Therefore, special attention must be paid to the sterilization of relevant electrical auxiliary components during ES therapy. In the future, it would be best to integrate ES therapy with wound dressings or wearable devices, which might help reduce the risk of infection. Energy self‐sufficient devices such as nanogenerators, photovoltaics and thermoelectrics provide inspiration and solutions for ES therapy. ^51^ , ^52^ , ^53^

Despite the encouraging results of this study, several pressing issues remain in the practical application of ES‐assisted treatment for DUs. First, the most effective selection of ES treatment parameters needs to be studied and standardized over larger sample sizes and longer time periods. Second, individual differences necessitate tailored ES strategies for wound treatment. Therefore, when applying ES therapy, the characteristics of different patients and their wounds must be considered. Combining multiple therapy modalities with ES therapy can create a synergistic effect, leading to better outcomes. The effects and synergistic mechanisms of these combined treatments need further investigation.

Study limitations

This study had several deficiencies and Variations in Physical Therapy Parameters: Different studies used varying physical therapy parameters (e.g., current intensity, frequency, treatment duration), leading to inconsistent standards and potentially contributing to unexplained heterogeneity.Small Sample Sizes and Experimental Designs: The small sample sizes and different experimental designs of the included studies may limit the generalizability and reliability of the results.Lack of Blinding: Some studies lacked blinding in their design, which could introduce potential bias.Insufficient Patient Information: Limited information from study authors on patient characteristics hindered effective subgroup analysis.Geographical Concentration: The geographical concentration of study sites in the United States and Iran may have introduced some degree of publication bias.

Despite these limitations, our study provides valuable insights that warrant further validation through large‐scale, multicentre clinical trials. Addressing these challenges and refining ES therapy protocols will be critical for advancing their clinical application in the treatment of DUs.

CONCLUSION

Our systematic evaluation and meta‐analysis demonstrated that ES treatment significantly improved the healing rate of DUs and the number of healed ulcers without notable adverse effects. These therapeutic effects can be attributed to the beneficial impact of ES on the vascular and neurological function of the lower limbs in patients with DUs. Specifically, both low‐frequency, moderate‐intensity AC and low‐intensity or high‐voltage DC have been shown to be effective in promoting ulcer healing. Nevertheless, large‐scale and long‐term studies are still required to refute full concerns.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest.

Supporting information