Significant efficacy of short-course, low-concentration betamethasone mouthwash therapy for severe erosive oral lichen planus: a randomized controlled trial

Abstract

Objectives

To evaluate the short-term efficacy of low-concentration betamethasone mouthwash for severe erosive oral lichen planus (EOLP).

Materials and method

In this randomized, investigator-blind, positive-controlled trial, OLP patients with erosive lesions received betamethasone mouthwash (0.137 mg/mL) or dexamethasone mouthwash (0.181 mg/mL) three times daily for 2 or 4 weeks and were followed up for 3 months to observe recurrence. The primary outcome was the week-2 reduction in erosive area.

Results

Fifty-seven participants were randomized to betamethasone (n = 29) and dexamethasone (n = 28). At week 2, participants using betamethasone (n = 28) experienced a greater reduction in erosive area than gargling with dexamethasone (n = 26). Similarly, secondary outcomes, including the healing proportion of erosions, reduced pain level, reduction in atrophic area, Thongprasom score, and recurrence interval, showed the superiority of betamethasone. At week 4, betamethasone (n = 7) was not superior to dexamethasone (n = 15) in further reducing lesional area and pain level. No serious adverse events were documented.

Conclusions

The 0.137 mg/mL compound betamethasone mouthwash exhibited significant efficacy in rapidly enhancing erosion healing within 2 weeks and extending the recurrence interval with a good safety profile.

Clinical relevance

This study proved the significant efficacy of short-course 0.137 mg/mL betamethasone mouthwash therapy for treating erosion and pain, providing a novel topical agent for patients with severe EOLP.

Trial registration

This study was prospectively registered at the International Clinical Trials Registry Platform (ChiCTR1800016507) on 5 June 2018.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00784-023-05051-w.

Introduction

Lichen planus is a common chronic inflammatory mucocutaneous disease with unclear etiology. The lesions often involve only the oral mucosa (oral lichen planus, OLP), characterized by multisite or bilateral symmetrical white reticular lesions sometimes accompanied by atrophic, bullous, or erosive lesions [1, 2]. The global incidence of OLP is approximately 1.01%, and erosive oral lichen planus (EOLP) accounts for 32.46% of cases [3]. Unlike self-limited cutaneous lesions, oral erosions are difficult to self-heal and susceptible to recurrence, frequently causing painful symptoms in patients.[1, 2] The World Health Organization (WHO) has considered OLP to be an oral potentially malignant disorder since 2007 [4]. As a potential risk factor in oral carcinogenesis, 1.05–2.85% of erosions may transform into malignant tumors, imposing a heavy psychological burden on patients with EOLP [5].

According to recent European S1 guidelines, OLP treatment options comprise first-line, second-line, and third-line treatments [2]. Topical steroids are the preferred drugs against localized erosions, such as clobetasol, fluocinonide, triamcinolone, dexamethasone, and betamethasone [6], while for widespread or multisite erosions of severe EOLP, the most effective medicines are oral corticosteroids [1, 2, 7]. For severe EOLP patients with contraindications or recalcitrant erosions to oral prednisone or methylprednisolone, the remaining first-line treatments, cyclosporine [8, 9] and retinoids [10, 11], and some second-line systematic treatments, such as azathioprine [12], methotrexate [13] and mycophenolate mofetil [14, 15], can be considered. However, the efficacy of most systemic therapies has not been validated by randomized controlled trials (RCTs), and long-term use could cause a spectrum of side effects, such as hepatotoxicity, nephrotoxicity, myelotoxicity and increased risk of infection or malignant transformation [1, 7, 9, 12, 13, 15]. Among the first-line treatments, topical steroids have been investigated in several prospective studies for treating severe EOLP patients, 42.9 to 93.3% of whom had a good response in decreasing clinical scores and 64.3 to 93.7% of whom had pain relief [16–21]. It is worth noting that the concentration of topical steroids, commonly used in dermatology, might be so high as to cause systemic absorption since the oral mucosa is 4 to 4000 times as permeable as the skin [22]. In addition, long-term application, usually for 3 to 6 months, leaves patients susceptible to fungal infections, mucosal atrophy, and glucocorticoid resistance [23–26]. To prevent oral candidiasis secondary to immunosuppression, antifungal agents are often associated with topical corticosteroids, such as miconazole and chlorhexidine [27].

Hence, there is a clinical need for an effective topical agent with a short course of treatment and low concentration to reduce systemic and local side effects. We previously reported the safety and noticeable efficacy of the intralesional compound betamethasone for the treatment of EOLP, and we assumed that soluble betamethasone sodium phosphate (BSP) and slightly soluble betamethasone dipropionate (BD) cooperatively exerted rapid and sustained anti-inflammatory effects [28, 29]. From this, we proposed a novel hypothesis that short-course, low-concentration betamethasone mouthwash had a dual effect on facilitating fast erosion healing and prolonging the recurrence-free period. We therefore designed this RCT to compare the short-term efficacy of 0.137 mg/mL betamethasone mouthwash for the treatment of severe EOLP with dexamethasone, the most frequently used glucocorticoid mouthwash [26].

Materials and methods

Study design and participants

This investigator-initiated, single-center, randomized, investigator-blind, positive-controlled trial was approved by the Ethical Committee of the West China Hospital of Stomatology, Sichuan University (granted number: WCHSIRB-D-2018-046-R) and was prospectively registered at the International Clinical Trials Registry Platform (ChiCTR1800016507). This study was conducted in accordance with the Declaration of Helsinki. Signed informed consent was obtained from all participants. There was no critical change to methods or outcomes after trial commencement.

Participants diagnosed with OLP by clinical and histopathological analysis at West China Hospital of Stomatology according to modified WHO criteria were sequentially enrolled in our department [30]. Inclusion criteria were as follows: (a) aged 18 to 60 years; (b) single erosion area ≥ 1 cm², or total area of multiple erosions ≥ 1 cm², or erosion area < 1cm² with the total area of surrounding atrophic lesions ≥ 1 cm² [31, 32]; (c) disease duration ≥ 2 months; and (d) volunteered to follow up.

Exclusion criteria were (a) allergic to glucocorticoids; (b) received systemic therapy for 3 months or topical treatment for 2 weeks before participating; (c) had comorbidities, such as uncontrolled hypertension, cardiovascular diseases, blood diseases, neuropsychiatric disorders, or other systemic diseases requiring taking medications; (d) presence of lichenoid lesions caused by amalgam fillings or certain drugs; (e) extraoral lesions of lichen planus; (f) other oral mucosal diseases; and (g) pregnant or breast-feeding women.

Withdrawal criteria were as follows: (a) lost to follow-up; (b) occurrence of adverse events (AEs) determined by the investigator to be unfit for participation; (c) unwilling to keep on participating in the study; (d) developing oral or systemic diseases in the exclusion criteria after the intervention; and (e) poor compliance.

Intervention

All participants received 5 mL of compound betamethasone mouthwash (0.137 mg/mL, experimental group) or dexamethasone mouthwash (0.181 mg/mL, control group), applied thrice daily (3 min each) for 2 or 4 weeks.

The betamethasone mouthwash was prepared by diluting 1 mL of compound betamethasone injection (DIPROSPAN®, MSD Merck Sharp &Dohme AG, Luzern, Switzerland) containing 6.43 mg BD (equivalent to 5 mg betamethasone) and 2.63 mg BSP (equivalent to 2 mg betamethasone) with 50 mL of sterile water (dilution ratio 1:50). The dexamethasone mouthwash was prepared by diluting 1 mL of dexamethasone sodium phosphate (DSP) injection (Kingyork Pharmaceutical Co., Ltd, Tianjin, China) containing 5 mg DSP (equivalent to 3.8 mg dexamethasone) with 20 mL of sterile water (dilution ratio 1:20).

All patients were instructed on how to measure volume both in written and oral forms. Compliance was evaluated by the gargling days until dosage for 2 weeks was used up, and 14 ± 2 days was regarded as good compliance. Drinking and eating were forbidden for 30 min after gargling. Spicy foods and other treatments, except for rescue medicine, were not allowed during treatment. In addition, 0.12% chlorhexidine mouthwash was applied three times daily before intervention against oral candidiasis in both groups.

The trial included two stages of a 2-week treatment period and one stage of a 3-month observation period for recurrence. When patients were free of erosion and pain symptoms after the first 2 weeks of treatment, they discontinued intervention and entered the observation period. If not, the second 2-week treatment was given. The participants without erosion and pain after 4 weeks of treatment also entered the observation period. If the erosions were still visible at week 4, alternative therapy was given, such as oral glucocorticoids and intralesional injection [2, 33].

Outcomes

The primary outcome was the week-2 reduction in erosive area, defined as the difference in erosive area between baseline and week 2. The lesional area was calculated by multiplying the greatest length diameter (mm) and vertical greatest width diameter (mm) using a calibrated dental probe [29, 34].

Secondary outcomes were as follows. (a) Week-4 reduction in erosive area, the erosive area difference between week 2 and week 4. (b) Reduced pain level, using a numeric rating scale (NRS) [35] to estimate the difference in spontaneous pain before and after treatment. (c) Healing proportion of erosions, presented as participants with erosion healing after treatment/allocated participants × 100%. (d) Reduction in atrophic area and reticular area. (e) Thongprasom sign score, ranging from 0 to 5 [31]. (f) Recurrence, including the recurrence interval, recurrence erosive area, and recurrence rate, which was defined as the proportion of participants who experienced recurrent erosions within 3 months.

Safety outcomes included the incidence and severity of AEs. At each subsequent clinic visit, participants were asked to report any local or systemic discomfort and abnormality. We carefully checked for oral fungal infection signs and took oral swabs for direct microscopic examination of fungi before and after therapy.

Randomization and blinding

Participants were randomized 1:1 to the experimental or control group according to the allocation sequence, which was randomly generated by SPSS 20.0.0 (IBM, Armonk, USA) and hidden in enclosed opaque envelopes. Only one investigator not engaged in clinical visits could access the sequence. As a participant was recruited, an envelope was opened sequentially by investigator A, who was not involved in the outcome assessment. Another investigator B blinded to the intervention information was responsible for the outcome measurement [29, 34].

Statistical analysis

By reference to our pre-experimental results, the mean difference (MD) of the primary outcome in both groups was assumed to be 18 cm², with a standard deviation (SD) of 30. For 90% power at a one-sided 2.5% significance level, allowing for 20% withdrawal, a sample size of 50 (25 per group) was needed.

Analyses were conducted by two independent investigators who were blinded to the allocation and outcome evaluation using SPSS 22.0.0 and STATA 15.1 (Stata Corporation, College Station, TX, USA). Quantitative data are presented as the median (Q1 [first quartile]–Q3 [third quartile]) and were analyzed using the t test (paired t test or independent t test) or rank-sum test (Wilcoxon test or Mann-Whitney U test). Categorical data are expressed as numbers (%) and were analyzed by the Pearson chi-square test or Fisher’s exact test. The main analysis was carried out on the per-protocol (PP) principle to evaluate the actual efficacy of cases without protocol violations or unassessable outcomes.

For the primary outcome, MD with 95% confidence interval (95% CI) and p value before and after each treatment arm (within-group comparison) or between two groups before and after treatment (between-group comparison) was calculated. If the reduction in lesional area was negative in some aggravating cases, the data of reduction area was adjusted as zero, and the adjusted MD was presented. Secondary outcomes of quantitative data were analyzed by similar methods. For the healing proportion of erosions and recurrence rate, the risk ratio (RR) with 95% CI and p value was calculated.

Intention-to-treat (ITT) analyses will be performed as sensitivity analyses for the results of PP analyses. The missing data were regarded as invalid responses to treatment, and the method of last observation carried forward (LOCF) was applied to supplement the missing data [36]. Logistic regression was applied to analyze local and systemic risk factors for erosion healing, including erosive area (mm²), number of oral erosions, number of affected sides, lesion position (buccal mucosa and/or tongue), Thongprasom score, perilesional supragingival dental calculus (presence or absence), age, and sex, and the forest plot was generated by R language 4.1.2 (R Foundation for Statistical Computing, Vienna, Austria). A two-sided p value < 0.05 was considered statistically significant.

Results

Recruitment and participant characteristics

From July 2018 to June 2021, 68 consented to participate in this trial, 57 of whom were eligible; 29 patients were randomized to betamethasone, and 28 were randomized to the control (Fig. 1). Baseline characteristics were well balanced between the treatment groups (Table 1).

Fig. 1.

CONSORT flow chart (a) and design schematic (b) of this randomized controlled trial. ITT, intention to treat; PP, per-protocol

Table 1.

Baseline characteristics of the 57 patients randomized to receive dexamethasone or betamethasone

Baseline	BET (n = 29)	DEX (n = 28)	p value
Male sex (n, %)	9 (31.03%)	12 (42.86%)	0.36a
Age (years)	54.0 [47.0; 58.5]	49.0 [41.5; 59.5]	0.19b
Duration of OLP (months)	8.0 [3.0; 13.5]	6.0 [3.0; 12.0]	0.48b
Number of oral erosions	3.0 [2.0; 4.5]	3.0 [2.0; 4.0]	0.88b
Number of affected sides	2.0 [1.0; 2.0]	2.0 [1.0; 2.0]	0.87c
Lesion position (n, %)			0.70a
Buccal mucosa	22 (75.9%)	19 (67.9%)
Tongue	3 (10.3%)	5 (17.9%)
Buccal mucosa and tongue	4 (13.8%)	4 (14.3%)
Thongprasom score	4.0 [4.0; 4.0]	4.0 [4.0; 4.0]	0.42c
Oral erosive surface (mm2)	50.0 [17.5; 85.5]	36.0 [21.3; 65.8]	0.56b
Oral atrophic surface (mm2)	172.0 [108.5; 267.3]	184.5 [106.5; 291.9]	0.80b
Oral reticular surface (mm2)	771.0 [470.0; 1749.5]	1072.0 [900.0; 1415.0]	0.18b
Oral pain (NRS score)	3.0 [2.0; 3.0]	3.0 [2.0; 3.0]	0.75b

Qualitative data are presented as numbers (%), and quantitative data are expressed as the median and interquartile range [first quartile; third quartile]. ^aPearson chi-square test; ^bMann-Whitney test; ^cIndependent samples t test. BET betamethasone, DEX dexamethasone, NRS numeric rating scale

At week 2, one betamethasone and two control participants withdrew from treatment due to time conflicts, and 28 betamethasone (96.6%) and 26 control (92.9%) patients attended the visit. The median gargling day was 14.0 [13.0;14.0] for betamethasone and 14.0 [14.0;15.0] for the control, indicating good compliance of all participants. Nineteen betamethasone and seven control patients achieved the disappearance of erosion and pain. However, three control participants complained of no treatment response (reduction of erosive surface ≤ 0 mm²) and asked for alternative therapy. Thus, nine betamethasone (31.0%) and sixteen control (57.1%) patients who had no healing erosions continued the second 2-week treatment, but two betamethasone and one control patient were lost to follow-up at week 4. In total, nine participants (15.8%) withdrew from this trial.

Week-2 reduction in erosive area

At week 2, significant effectiveness in reducing erosive area was found in both the betamethasone (n = 28; MD 44.77, 95% CI 26.95–62.58; p < 0.001) and control groups (n = 26; adjusted MD 27.60, 95% CI 2.47–52.72; p = 0.001) (Table 2). However, the MD of the between-group comparison revealed the superiority of betamethasone over the control in rapidly reducing erosive area (21.65, 95% CI 2.36–40.94; p = 0.022) (Fig. 2a and Table 3).

Table 2.

The PP analysis of lesional area and pain level of each group before and after the treatment

Intervention		Erosive area (mm2)	Atrophic area (mm2)	Reticular area (mm2)	Pain level (NRS score)
After the first 2-week treatment (week 2)
BET (n = 28)	Week 0	44.0 [17.3; 82.3]	170.5 [106.8;280.9]	828.0 [465.0; 1872.3]	3.0 [2.0; 3.0]
	Week 2	0.0 [0.0; 9.5]	38.0 [10.1; 88.3]	485.0 [217.5; 973.0]	0.0 [0.0; 0.0]
	Unadjusted MD (95% CI)	44.77 (26.95 to 62.58)	182.23 (103.53 to 260.93)	419.91 (45.94 to 793.88)	2.54 (2.16 to 2.92)
	p valuea	<0.001b	<0.001b	<0.001b	<0.001b
DEX (n = 26)	Week 0	36.0 [19.8; 57.3]	177.0 [105.0; 280.6]	1101.0 [900.0; 1446.5]	2.5 [2.0; 3.0]
	Week 2	19.0 [0.0; 52.6]	85.0 [48.8; 151.5]	960.5 [628.8; 1131.8]	2.0 [0.0; 2.0]
	Unadjusted MD (95% CI)	23.12 (−2.40 to 48.63)	89.75 (28.60 to 150.90)	233.04 (2.53 to 463.54)	1.27 (0.65 to 1.89)
	Adjusted MD (95% CI)	27.60 (2.47 to 52.72)	102.67 (51.40 to 153.95)	236.46 (7.11 to 465.82)	1.38 (0.82 to 1.94)
	p valuea	0.001b	<0.001b	<0.001b	<0.001b
After the second 2-week treatment (week 4)
BET (n = 7)	Week 2	18.0 [8.0;71.0]	110.0 [32.0; 120.0]	560.0 [120.5; 1125.0]	1.0 [0.0; 2.0]
	Week 4	9.0 [0.0;24.0]	56.0 [0.0; 63.0]	560.0[63.0; 820.0]	0.0 [0.0; 2.0]
	Unadjusted MD (95% CI)	17.46 (−9.40 to 44.33)	57.14 (7.68 to 106.60)	240.36 (−325.63 to 806.34)	0.29 (−0.74 to 1.31)
	p valuea	0.089c	0.008c	0.189c	0.317b
DEX (n = 15)	Week 2	32.5 [5.0;72.0]	103.5 [62.0; 156.0]	957.0 [580.0; 1065.0]	2.0 [1.0; 2.0]
	Week 4	8.0 [0.0; 57.0]	55.0 [24.0; 180.0]	765.0[500.0; 936.0]	1.0 [0.0; 1.0]
	Unadjusted MD (95% CI)	8.20 (−22.74 to 39.14)	−50.30 (−203.31 to 102.71)	82.93 (−291.21 to 457.08)	0.87 (0.32 to 1.42)
	Adjusted MD (95% CI)	22.10 (−4.99 to 49.19)	48.90 (−4.99 to 102.80)	198.80 (−125.67 to 523.27)	0.93 (0.41 to 1.46)
	p valuea	0.532c	0.152b	0.074b	0.003c

Data are expressed as the median and interquartile range [first quartile; third quartile]. ^aThe within-group differences were compared between the start and the end of treatment of each group. ^bWilcoxon signed ranks test. ^cPaired sample t test. BET betamethasone, DEX dexamethasone, MD mean difference, NRS numeric rating scale, PP per-protocol

Fig. 2.

Median erosive area and the number of erosion-free cases during treatment with betamethasone (BET) or dexamethasone (DEX) and a 3-month off-treatment period. a Changes in patients completing the first 2-week treatment and the recurrence of 2-week healing patients. b Changes in patients receiving the second 2-week therapy and the recurrence of 4-week healing patients

Table 3.

The comparison of lesional area and pain level between the two groups before and after the treatment

Intervention	Reduction in erosive area (mm2)	Reduction in atrophic area (mm2)	Reduction in reticular area (mm2)	Reduced pain level (NRS score)
PP analysis at week 2
BET (n = 28)	31.6 [16.8; 75.5]	125.5 [81.4; 187.9]	194.0 [136.0; 710.3]	2.0 [2.0; 3.0]
DEX (n = 26)	18.0 [7.5; 38.0]	69.0 [36.3; 161.3]	196.5 [61.5; 298.8]	1.0 [1.0; 2.0]
MD (95% CI)	21.65 (2.36 to 40.94)	92.48 (7.74 to 177.22)	186.87 (−8.43 to 382.17)	1.27 (0.69 to 1.84)
p valuea	0.022b	0.013b	0.166b	<0.001b
ITT analysis at week 2
BET (n = 29)	31.0 [16.1; 72.0]	123.0 [74.0; 184.8]	193.0 [126.0; 698.5]	2.0 [2.0; 3.0]
DEX (n = 28)	15.5 [3.0; 36.0]	68.0 [30.3; 147.3]	168.0 [23.3; 287.3]	1.0 [0.25; 2.0]
MD (95% CI)	21.76 (3.17 to 40.34)	92.61 (10.47 to 174.74)	189.04 (−1.29 to 379.36)	1.27 (0.69 to 1.85)
p valuea	0.017b	0.010b	0.121b	<0.001b
PP analysis at week 4
BET (n = 7)	8.0 [0.0; 32.0]	60.0 [25.0; 91.0]	57.5 [12.0; 305.0]	0.0 [0.0; 0.0]
DEX (n = 15)	14.0 [0.0; 36.0]	36.0 [0.0; 90.0]	70.0 [0.0; 390.0]	1.0 [0.0; 2.0]
MD (95% CI)	9.26 (−20.99 to 39.52)	107.44 (−46.40 to 261.28)	157.42 (−253.84 to 568.69)	−0.58 (−1.31 to 0.15)
p valuea	>0.999b	0.407b	>0.999b	0.142b
ITT analysis at week 4
BET (n = 10)	1.13 [0.0; 21.5]	36.0 [0.0; 69.3]	15.5 [0.0; 154.3]	0.0 [0.0; 0.0]
DEX (n = 21)	2.0 [0.0; 28.3]	11.0 [0.0; 57.0]	21.0 [0.0; 250.5]	0.0 [0.0; 1.0]
MD (95% CI)	6.37 (−15.0 to 28.24)	75.93 (−34.52 to 186.37)	109.01 (−185.34 to 403.36)	−0.46 (−1.02 to 0.09)
p valuea	0.917b	0.492b	0.819b	0.124b

Data are expressed as the median and interquartile range [first quartile; third quartile]. ^aThe between-group differences were compared between the betamethasone and dexamethasone groups. ^bMann-Whitney test. BET betamethasone, DEX dexamethasone,ITT intention-to-treat analysis, MD mean difference, NRS numeric rating scale, PP per-protocol

Week-4 reduction in erosive area

Neither the MD of betamethasone (17.46, 95% CI −9.40 to 44.33; p = 0.089) nor the control (adjusted MD 22.10, 95% CI −4.99 to 49.19; p = 0.532) showed a further significant reduction in erosive area before and after the second 2-week treatment (Table 2). Betamethasone (n = 7) did not demonstrate the superiority of narrowing erosive area over control (n = 15) in weeks 3 to 4 (MD 9.26, 95% CI −20.99 to 39.52; p > 0.999) (Fig. 2b and Table 3).

Healing proportion of erosions

At week 2, the healing proportion of erosions in betamethasone patients (n = 19, 67.9%) was significantly higher than that in control patients (n = 7, 26.9%) (Fig. 2), and the RR between the treatment arms was 2.52 (95% CI 1.27-4.99; p = 0.003) (Fig. 3). At week 4, a significant between-group difference in the healing proportion of erosions was not observed (RR 1.61, 95% CI 0.48–5.33; p = 0.630) (Table 4).

Fig. 3.

Significant efficacy of 2 weeks of betamethasone mouthwash for severe erosive oral lichen planus. a, b Two participants suffered extensive erosions surrounded by atrophic and reticular lesions on the buccal mucosa before treatment. c, d Their erosive lesions completely healed, atrophic lesions nearly disappeared and reticular lesions significantly regressed at week 2

Table 4.

Analysis of the healing proportion of erosions and recurrence

Intervention	Healing proportion of erosions		Recurrence
	PP analysis (%)	ITT analysis (%)	Participantsa (%)	Interval daya,b (Mean ± SD)
Participants receiving 2-week treatment
BET	19/28 (67.9%)	19/29 (65.5%)	6/19 (31.6%)	40.3 ± 14.7
DEX	7/26 (26.9%)	7/28 (25.0%)	4/7 (57.1%)	20.8 ± 9.4
Effect size (95% CI)	RR 2.52 (1.27–4.99)	RR 2.62 (1.31–5.24)	RR 0.55 (0.22–1.39)	MD 19.50 (4.56–34.44)
p value	0.003c	0.002c	0.369d	0.048e
Participants receiving 4-week treatment
BET	3/7 (42.9%)	3/10 (30.0%)	0/3 (0.0%)	None
DEX	4/15 (26.7%)	4/21 (19.0%)	2/4 (50.0%)	24.5 ± 6.4
Effect size (95% CI)	RR 1.61 (0.48–5.33)	RR 1.57 (0.43–5.74)	N/A	N/A
p value	0.630d	0.652d	N/A	N/A

^aITT analysis was not applicable. ^bThe recurrence intervals were calculated by the period between the healing and recurrence of erosions at the same mucosal site. ^cPearson chi-square test. ^dFisher's exact test. ^eIndependent samples t test. BET betamethasone, DEX dexamethasone, ITT intention-to-treat analysis, MD mean difference; N/A not applicable, PP per-protocol, RR risk ratio, SD standard deviation

Reduced pain level

At week 2, the reduced pain level was significantly more pronounced in the betamethasone group than in the control group (MD 1.27, 95% CI 0.69–1.84; p < 0.001). However, further pain reduction did not differ between the two groups during the second 2-week treatment (MD -0.58, 95% CI −1.31 to 0.15; p = 0.142) (Table 3).

Reduction in atrophic area and reticular area

Patients from the betamethasone group experienced a greater reduction in atrophic area than controls (MD 92.48, 95% CI 7.74–177.22; p = 0.013) at week 2 but no significant superiority over controls after the second 2-week treatment (MD 107.44, 95% CI −46.40 to 261.28; p = 0.407). While both groups could reduce the white reticular area, there was no evidence for the superiority of betamethasone over the control in reducing the reticular area at week 2 (MD 186.87, 95% CI −8.43 to 382.17; p = 0.166) or at week 4 (MD 157.42, 95% CI −253.84 to 568.69; p > 0.999) (Table 3).

Thongprasom score

At week 2, the Thongprasom score in the betamethasone group (2.0 [2.0; 4.0]) showed a more significant reduction than that in the control group (4.0 [2.0; 4.0]) (MD −1.04, 95% CI −1.66 to −0.41; p = 0.002) while at week 4, there was no significant difference between the betamethasone (4.0 [1.0; 4.0]) and control groups (4.0 [2.0; 4.0]) (MD −0.74, 95% CI −1.97 to 0.48; p = 0.203).

Recurrence

After the first 2-week treatment withdrawal, six betamethasone (31.6%) and four control (57.1%) patients had recurrence of erosions, with no significant between-group difference (RR 0.55, 95% CI 0.22–1.39). Patients from the betamethasone group experienced more erosion-free days than control patients (MD 19.50, 95% CI 4.56–34.44; p = 0.048) (Table 4). No recurrence was detected in patients receiving 4 weeks of betamethasone, and two control patients experienced recurrence 20 and 29 days after dexamethasone withdrawal (Fig. 2). At 3 months, the relapsed erosive areas in the betamethasone (n = 6) and dexamethasone (n = 6) groups were 8.5 [3.8; 26.1] and 13.3 [5.0; 23.5] mm², respectively, and no significant difference was detected (MD 2.00, 95% CI −22.77 to 26.77; p = 0.872). Compared to prior to treatments, 70.9% [12.4%; 84.7%] of the reduction of relapsed erosion area after 3 months in betamethasone and 43.6% [12.5%; 79.2%] of reduction in control revealed no significant difference (MD 0.12, 95% CI −0.27 to 0.51; p = 0.571). In addition, six patients experiencing failure of dexamethasone mouthwash were cured with oral prednisone as an alternative, and three did not suffer from recurrence during the three-month follow-up period.

Sensitivity analysis

The sensitivity analyses demonstrated that the p values of ITT analyses yielded similar results in primary and secondary outcomes (Table 3, Figs S1-S8 and Table S1 in Supplementary Information).

Risk factor analysis

Figure 4 indicates that no significant local and systemic risk factors for 2-week erosion healing were identified (p > 0.05) among the two treatment groups and all patients.

Fig. 4.

Forest plot of risk factors for erosion healing at week 2

Safety

During the whole study period, no serious AEs were reported. Two betamethasone participants (6.9%) experienced slight dryness of the mouth, and one control participant (3.6%) had a buccal burning sensation. These symptoms occurred over the first 2-week period and resolved spontaneously within 1 week. No case of oral candidiasis was identified before and after treatment based upon both clinical signs and direct microscopic examination. Telephone follow-up was performed for six patients lost to follow-up, and no AE was reported.

Discussion

The compound betamethasone mouthwash in this RCT was a novel pragmatic agent for people suffering from severe EOLP. We revealed for the first time that short-term use of low-concentration betamethasone mouthwash could effectively and safely promote erosion healing and symptom resolution. Compared with 0.181 mg/mL dexamethasone, 0.137 mg/mL betamethasone significantly reduced the erosive area, atrophic area, and pain level, increased the healing proportion of erosions, and prolonged the recurrence interval after 2 weeks of treatment. Sensitivity analyses produced consistent findings with the above results.

There is currently no published RCT comparing therapeutic effects on OLP lesions between betamethasone and dexamethasone, both long-acting corticosteroids. Dexamethasone 0.1–1.0 mg/mL mouthwash is widely used by OLP patients and has been designed as the positive control in many RCTs [24, 26, 37]. According to glucocorticoid potency conversion, the dose of 0.137 mg betamethasone is approximately equivalent to the dose of 0.137–0.171 mg dexamethasone [38, 39]. Considering the highest equivalent concentration to betamethasone (> 0.171 mg/mL), efficacy and side effects of different concentrations reported in previous studies [26, 40, 41], and convenient dilution (ratio 1:20), we used 0.181 mg/mL dexamethasone as the control.

The risk factor analyses demonstrated that the efficacy of promoting erosion healing was not significantly influenced by potential local and systemic factors. The superiority of betamethasone over dexamethasone might be explained in the following three aspects. First, 0.137 mg/mL betamethasone exhibited much higher anti-inflammatory potency than 0.181 mg/mL dexamethasone. Although these two concentrations have not been included in the common potency ratings of topical corticosteroids [42], it is generally known that 0.1% DSP cream, 0.1% betamethasone valerate (BMV) cream, and 0.05% BD cream are graded as low potency (Class VII), moderate potency (Class V), and high potency (Class III), respectively, indicating the higher potency of betamethasone than dexamethasone [43]. Second, the compound betamethasone contains not only rapidly hydrolyzed BSP but also slowly metabolized BD that could be retained in erosions with less influence from saliva dilution and flush. Kopylova et al. found that betamethasone was detectable in all human urine samples (n =20) 28 days after Diprospan® injection, and the complete elimination extended up to 48 days [44]. Interestingly, the mean interval of erosion recurrence in our betamethasone patients was 40.3 days, suggesting that there was a longer hydrolysis period of BD in the mouth. Third, betamethasone might promote erosion recovery and reduce the patient’s symptoms by blocking the association of local lesions with systemic factors. Although not clear, the pathogenesis of OLP is generally considered T-cell-mediated epithelial barrier damage, which is responsible for the occurrence of local erosions and symptoms [2]. Eating might be too painful for patients to receive a normal diet and adequate nutrition, causing weight loss, impairment in immune cell activity, and mental disorders, such as stress, anxiety, and depression [2]. However, these psychological disorders as potential predisposing factors, contributing to activation of the host immunological responses toward local oral mucosa, might in turn cause exacerbation or recalcitrance of erosions, forming a vicious cycle [45, 46]. The rapid anti-inflammatory and pain-relieving actions of betamethasone help disrupt this vicious cycle as early as possible (and earlier than dexamethasone) (Fig. 5) and recover regular diet and nutrient uptake to maintain proper immune cell functionality [47–49].

Fig. 5.

Schematic diagram of the vicious circle of oral lichen planus (OLP) between inflammation and psychological disorders. Patients with erosive OLP tend to suffer pain and psychological disorders. Stress results in the release of cortisol to promote Th2 cell differentiation and mast cell degranulation, secreting cytokines for basement membrane disruption. Depression and anxiety can elevate the levels of proinflammatory cytokines to enhance immune cell migration. Under the effects of psychological disorders, oral keratinocytes are more prone to be destroyed by cytotoxic T cells, aggravating inflammation, pain, and psychological disorders. The fast anti-inflammatory and analgesic effects of betamethasone could block this vicious cycle, facilitating a rapid healing process

Previous RCTs of applying topical betamethasone for EOLP have demonstrated its efficacy [17, 21, 29, 50, 51]. Hegarty et al. found that 22 patients had a mean 42.91% reduction in lesional area and a 29.73% reduction in visual analogue scale (VAS) after using a 6-week 0.005% BSP mouthrinse (0.5 mg tablet dissolved in 10 mL water) without AE. Ezzatt and Helmy showed that 15 patients receiving 0.1% BMV cream had significant reductions in clinical scoring (mean ± SD 63.41% ± 18.10%) and VAS (95.42% ± 4.12%) at week 4, and 3 patients (20.0%) had a burning sensation or taste alteration. Samimi et al. detected that 0.05% BD ointment decreased the median erosive surface by 98.8% in 36 patients at month 3, 30 of whom (83.3%) reported AEs. The above findings suggested that the higher the concentration or potency of betamethasone, the better the anti-inflammatory effect and the higher the risk of AEs. In this study, 28 participants using 2 weeks of betamethasone had significant reductions in erosive area (88.16% ± 22.52%) and pain level (92.26% ± 19.50%), with a low rate of adverse reactions (7.1%), displaying its comparable efficacy of potent cream/ointment and good safety with short course and low concentration. Apart from the dual-acting betamethasone ingredients, the diluted liquid dosage not only had better access to all oral lesional sites to improve inflammation conditions adjacent to erosions but could rapidly contact erosions by directly releasing active molecules from water. In addition, mouthwash was much easier to use for any oral location and to reduce drug residence time in the mouth [16, 28].

Both types of mouthwash had an acceptable safety profile. The reported dry mouth and burning sensation were common side effects of topical corticosteroids, which might be due to elevated nervous excitability [50, 52]. Given the characteristics of high efficacy and low toxicity, betamethasone mouthwash could be a good candidate for the treatment of EOLP. The significant effects on enhancing erosion healing and extending the recurrence interval could reduce the revisit frequency of patients, especially during the current COVID-19 pandemic. We did not monitor betamethasone plasma levels with the long excretion period of BD and no impact on plasma cortisol levels in previous studies [44, 51]. Of note, because most of the data followed non-normal distributions in this trial, the median might look dissimilar in the two arms, such as the erosive surface at baseline. Therefore, we also added the mean and SD in the supplementary figures (Figs. S1-S8) to display data variability.

Although over 20 scoring systems have been used to grade the severity of OLP, such as the Thongprasom score and Reticular–Erythematous–Ulceration (REU) system, there is no gold standard scoring system or uniform definition of severe EOLP [32, 53, 54]. Referencing our previous method of defining refractory OLP [55], we proposed the inclusion criteria for severe EOLP based on the classic Thongprasom score (≥ 4) [31] to facilitate quick identification by oral clinicians without complicated algorithmic summation.

The primary study limitation was the lack of patient-blind implementation with the different mouthwash physical appearances. Second, there was a larger difference in the number of patients receiving the second 2-week treatment because 67.9% of betamethasone patients entered the observation period after the first 2-week therapy. Additionally, using the calibrated dental probe to calculate lesional area (mm²) might yield some measurement bias, especially for lesions with irregular shapes. Therefore, we first divided an irregular lesion into several regular calculating parts and then combined their areas together to reduce measurement error. A software or scanner for automatically calculating oral lesional areas is clinically needed in the future.

Conclusion

We identified for the first time the superiority of short-course, low-concentration compound betamethasone mouthwash over dexamethasone for rapidly enhancing erosion healing, relieving pain and effectively prolonging the interval of erosion recurrence. Both treatments were well tolerated. This trial suggested that 0.137 mg/mL compound betamethasone mouthwash is a novel, rapid-acting, and safe topical agent for patients with severe EOLP.

Author contribution

Qingxiang Zeng: investigation, methodology, visualization, writing—original draft, writing—review and editing; Yangfan Liu: formal analysis, methodology, visualization, validation, software, writing—review and editing; Shimeng Wang: formal analysis, methodology, visualization, writing—original draft, writing—review and editing; Houshang Wang: investigation, validation, writing—review and editing; Shuang Yu: formal analysis, validation, writing—review and editing; Fanglong Wu: investigation, resources, funding acquisition, writing—review and editing; Jin Yang: data curation, funding acquisition, methodology, project administration, supervision, validation, visualization, writing—review and editing; Hongmei Zhou: conceptualization, data curation, funding acquisition, methodology, project administration, supervision, writing—review and editing, resources, investigation.

Funding

This work was financed by National Natural Science Foundation of China (No. 82071124, No. 82101028) and Natural Science Foundation of Sichuan Province (No. 2023NSFSC0552).

Data availability

The data that support the findings of this trial and the full trial protocol are available from the corresponding authors upon reasonable request.

Declarations

Ethics approval and consent to participate

The study was approved by the Ethical Committee of the West China Hospital of Stomatology, Sichuan University (grant number: WCHSIRB-D-2018-046-R). Written informed consent was obtained from all participants before the intervention.

Conflict of interest

The authors declare no competing interests.