Laryngopharyngeal reflux (LPR) disease is associated with gastroduodenal content reflux into the upper airway [1]. Typical symptoms include a globus sensation in the throat, hoarseness, excessive throat mucus levels, cough, and postnasal drip [1,2]. LPR accounts for approximately 10% of ear, nose, and throat (ENT) outpatient visits; however, insufficient reliable diagnostic technique remains a major problem [2,3]. LPR diagnosis is currently based on the self-report of symptoms, clinical findings from laryngoscopy, and pH analysis [3]. The reflux symptom index (RSI) and reflux finding score (RFS) are commonly used to assess symptom severity [3,4]. Lifestyle and diet modifications are the first steps in LPR treatment, and proton pump inhibitors (PPIs) are the preferred drugs to effectively reduce LPR symptoms. A recent meta-analysis revealedthat PPIs are useful in reducing throat symptoms, but not cough and hoarseness, in patients with LPR [5]. Hence, PPIs alone cannot manage all LPR symptoms, with > 90% of patients experiencing relapses after PPI discontinuation, there by requiring treatment with PPIs again. Additionally, adverse effects of the long-term use of PPIs have been reported, including calcium, iron, and vitamin malabsorption, atrophic gastritis, and renal failure [2,5,6]. Additional effective and safe treatments to reduce LPR symptoms are necessary, and acupuncture may be an option. Recent randomized controlled trials (RCTs) have evaluated the clinical benefits of acupuncture in reducing LPR symptoms; however, no relevant systematic reviews are available. Therefore, this systematic review and meta-analysis was conductedto compare the effectiveness of acupuncture with that of conventional treatments for LPR.

1. Study design

A systematic review and meta-analysis of RCTs was conducted under the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines. The study protocol was registered with PROSPERO (International Systematic Review Registry; CRD42023412151) [7].

2. Search strategy for identification of studies

We conducted extensive database searches in English, Japanese, Chinese, and Korean, and included relevant papers in our study to minimize language bias. 5 English language databases (MEDLINE [PubMed], Excerpta Medica Database [EMBASE], Cochrane Register of Controlled Trials [CENTRAL], EBSCOhost, and Springer HSS collection), 1 Chinese database (China Knowledge Infrastructure for Chinese studies [CNKI]), 1 Japanese database (Japan Science and Technology Information Aggregator Electronic database [J-STAGE]), and 3 Korean databases (Oriental Medicine Advanced Searching Integrated System [OASIS], Korean Studies Information Service System [KISS], and Research Information Sharing Service [RISS]) were searched for studies published up to March 2023. We applied a search strategy to identify all articles that addressed acupuncture and LPR (Table 1). Different search terms were utilized for each database to ensure comprehensive coverage of the relevant articles.

Table 1 . Search strategies for online search of PubMed.

PubMed

#1. Search: (laryngopharyngeal reflux[MeSH]) #2. Search: ((laryngopharyngeal[Title/Abstract] AND reflux[Title/Abstract]) OR laryngopharyngeal reflux[Title/Abstract]) #3. #1 OR #2 #4. Search: (acupuncture[Mesh]) #5. Search: (acupuncture[Title/Abstract] OR needle[Title/Abstract] OR needling[Title/Abstract] OR dry-needling[Title/Abstract] OR electroacupuncture[Title/Abstract] OR electro-acupuncture[Title/Abstract]) #6. #4 OR #5 #7. #3 AND #6

3. Eligibility criteria

The analysis only included RCTs, and other types of studies, such as case studies, observational studies, non-human studies, reviews, and protocols, were excluded, as well as studies with no access to the full text.

The study included patients diagnosed with LPR according to the case definition of the expert consensus on LPR disease diagnosis and treatment (2015) [8], with an RSI score of > 13 and an RFS of > 7, with no restrictions on symptom onset duration after disease morbidity as well as patients’ race, age, or sex.

We included studies assessing acupuncture as an intervention, either alone or in combination with other interventions, such as dietary modification, herbal medicines, and PPIs. We excluded studies that included acupuncture in both the experimental and control groups as this would not compare the effects of acupuncture alone. The type of acupuncture used, type of acupuncture stimulation, acupuncture points stimulated, treatment duration, or number of treatments were not restricted.

We included studies that reported at least one of the primary outcomes to assess treatment effectiveness. The primary outcomes included the RSI and RFS, which are widely used indicators of symptoms and clinical findings in patients with LPR. Secondary outcomes include lower esophageal sphincter pressure (LESP) and upper esophageal sphincter pressure (UESP), which are indicators of esophageal motility in patients with LPR.

4. Data extraction process

Two researchers independently searched and screened the articles following predetermined inclusion and exclusion criteria. We analyzed titles and abstracts for the first selection after excluding duplicates. Two independent researchers thoroughly evaluated the selected studies in full before making a final selection for inclusion in the study. We first attempted to resolve disagreements through discussion. We sought the opinion of a third researcher for final decision-making if a consensus was not reached. Finally, the following information was extracted from the included studies: sample size, illness duration, outcomes, and adverse events (AEs). Additionally, information on interventions, such as needling details, treatment duration, and control interventions was collected using the Standards for Reporting Interventions in Clinical Trials of Acupuncture (STRICTA) guidelines [9].

5. Risk of bias assessment

Two investigators independently used 7 questions from the Cochrane risk of bias (RoB) tool to evaluate the RoB in the included RCTs [10]. The questions covered random sequence generation, allocation concealment, participants and personnel blinding, outcome assessment blinding, incomplete outcome data, selective outcome reporting, and other potential biases. These were rated as high, low, or unclear. Any disagreements between the researchers were resolved through discussion with a third party, if necessary.

6. Data synthesis and analysis

Meta-analyses were conducted to compare the effectiveness of interventions among the included studies. RevMan version 5.4.1 for Windows was used for data synthesis. We used the mean difference (MD) with a 95% confidence interval (CI) because the outcomes were continuous and measured on the same scale. The Higgins I² heterogeneity test and the chi-square test were used for heterogeneity analysis. Arandom-effects model was used to synthesize the extracted data to elucidate the variability of studies since this review included RCTs using acupuncture interventions with various factors such as disease duration, acupoints, and treatment period. The quality of evidence from the studies included in the analysis was assessed using GRADEpro software, and a summary table was constructed [11]. Publication bias was not assessed when fewer than 10 studies were analyzed following the Cochrane Handbook for Systematic Reviews of Interventions [10].

1. Study selection

This study searched 10 domestic and international databases and revealed 9, 10, 5, 4, 16, 293, 5, and 5 articles in PubMed, EMBASE, CENTRAL, EBSCOhost, Springer HSS collection, CNKI, J-STAGE, and Korean databases, respectively. After excluding duplicates, 321 articles remained. Of these, 305 were excluded after a review of titles and abstracts, including 297 studies with no LPR or acupuncture involvement, 2 animal experiments, and 6 non-RCT studies. After reviewing the full text of the remaining 16 articles, 8 studies were excluded because they used interventions, such as TEAS, and 1 study usedacupuncture treatment in both the experimental and control groups. This study included and analyzed 7 articles (Fig. 1).

Figure 1. Flow diagram of the study selection process of this review. LPR, laryngopharyngeal reflux; RCTs, randomized controlled trials.

2. Study information

This study included and evaluated 695 patients with LPR in the 7 RCTs included [12-18], as shown in Table 2. The experimental group, receiving either combined acupuncture [12-17] or acupuncture alone [18], comprised 324 patients. The control group, which received the conventional treatment, comprised 371 patients. All 7 studies included in the analysis were conducted in China [12-17] between 2017 and 2022. All studies were conducted over the past 5 years except for 2 studies [15,18]. Six RCTs [12-16,18] were conducted as 2-arm parallel trials and only 1 RCT [17] as a 3-arm parallel trial. Six studies [12-17] used acupuncture and conventional treatment combination as an intervention and conventional treatment alone as a control. Another study [18] compared acupuncture alone with conventional treatments. Of the studies [12-17] that investigated the effectiveness of combined acupuncture, Pei et al. [17] used PPIs and herbal medicines as conventional treatment, 2 studies [12,13] used PPIs as the conventional treatment, and 2 others [14,15] used PPIs and diet modification as conventional treatments. Wang et al. [16] used conventional treatments, such as PPIs, diet modifications, and prokinetics. Yang et al. [18] compared acupuncture alone with PPIs as controls. Primary outcome measures in all studies included RSI and RFS. Secondary outcome measures, including UESP, LESP, the Short Form 36 health survey (SF-36), salivary pepsin level, salivary EGF level, and 24-hour pH monitoring (24-hesophageal pH < 4-hour occupancy), varied among studies. The participants included 342 males and 353 females. 2 studies [12,18] had missing values. Zhao et al. [12] reported 5 dropouts, including 2 in the experimental group due to workplace issues and 3 in the control group due to transport issues. Yang et al. [18] reported 6 dropouts, including 3 each in the intervention and control groups, with no reasons presented.

Table 2 . Summary of included studies.

Author (y)	Intervention (n)	Comparison (n)	Duration (wk)	Outcome measure	Result (mean ± standard deviation)	Adverse events
Acupuncture + PPI treatment vs. PPI treatment
Zhao (2021) [12]	Acupuncture + PPI (33)	PPI (32)	4	1) RSI 2) RFS 3) UESP (mmHg) 4) LESP (mmHg) 5) Salivary pepsin level (ug/L)	(Intervention/comparison, baseline, 4 weeks) 1) 22.39 ± 5.29→12.33 ± 4.05/21.41 ± 5.28→16.03 ± 4.48 2) 14.33 ± 2.79→8.67 ± 2.04/15.03 ± 2.53→11.00 ± 1.88 3) 54.94 ± 15.61→63.80 ± 15.87/50.34 ± 14.02→54.15 ± 13.72 4) 17.94 ± 3.65→23.37 ± 5.44/18.09 ± 4.45→18.35 ± 3.87 5) 122.51 ± 44.43→67.99 ± 28.28/116.40 ± 39.22→91.05 ± 37.42	NR
Li (2020) [13]	Acupuncture + PPI (34)	PPI (33)	8	1) RSI 2) RFS 3) UESP (mmHg) 4) LESP (mmHg) 5) Salivary EGF level (ng/mL)	(Intervention/comparison, baseline, 8 weeks) 1) 22.65 ± 5.98→10.65 ± 4.08/23.55 ± 4.94→17.09 ± 4.38 2) 14.82 ± 3.54→5.97 ± 1.87/15.33 ± 3.29→8.88 ± 1.90 3) 39.77 ± 9.80→49.17 ± 10.42/41.03 ± 9.45→42.92 ± 9.48 4) 9.69 ± 2.27→12.65 ± 2.78/10.09 ± 2.32→10.70 ± 2.28 5) 3.68 ± 0.41→4.97 ± 0.60/3.69 ± 0.38→4.43 ± 0.43	NR
Acupuncture + Conventional treatment (including PPI) vs. conventional treatment (including PPI)
Liu (2020) [14]	Acupuncture + PPI, Diet (24)	PPI + Diet (24)	8	1) RSI 2) RFS	(Intervention/comparison, baseline, 8 weeks) 1) 20.94 ± 1.34→14.45 ± 0.94/20.88 ± 1.40→17.33 ± 1.03 2) 9.19 ± 0.32→4.25 ± 0.35/9.21 ± 0.35→5.64 ± 0.39	NR
Zhang (2017) [15]	Acupuncture + PPI, Diet (102)	PPI + Diet (100)	8	1) RSI 2) RFS	(Intervention/comparison, baseline, 8 weeks) 1) 19.93 ± 0.27→14.84 ± 0.26/19.86 ±0.27→17.24 ± 0.27 2) 9.23 ± 0.22→4.23 ± 0.11/9.12 ± 0.17→5.63 ± 0.08	NR
Wang (2020) [16]	Acupuncture + PPI + Prokinetic + Diet (42)	PPI + Prokinetic + Diet (42)	2	1) RSI 2) RFS 3) 24 hours pH monitoring (%)	(Intervention/comparison, baseline, 2 weeks) 1) 21.43 ± 3.81→15.21 ± 2.88/20.21 ± 2.71→17.36 ± 2.75 2) 9.69 ± 1.72→6.43 ± 1.67/9.33 ± 1.46→7.95 ± 1.34 3) 12.57 ± 3.14→7.02 ± 1.85/12.69 ± 3.69→8.81 ± 2.47	NR
Pei (2022) [17]	Acupuncture + PPI + H-med (50)	CG1: PPI + H-med (50) CG2: PPI (50)	4	1) RSI 2) RFS 3) SF-36	(Intervention/comparison CG1/comparison CG2, baseline, 4 weeks, 8 weeks [follow-up]) 1) 26.74 ± 5.04→13.58 ± 2.49→7.82 ± 1.56/27.64 ± 4.73→ 16.32 ± 3.29→7.56 ± 1.49/26.41 ± 4.47→18.64 ± 3.17→ 10.88 ± 2.03 2) 17.54 ± 4.29→6.76 ± 1.72→4.82 ± 1.45/17.16 ± 4.38→7.94 ± 2.46→4.30 ± 1.16/16.74 ± 4.28→8.72 ± 2.67→7.98 ± 2.46 3) 39.88 ± 6.01→69.22 ± 9.73→88.08 ± 11.03/38.18 ± 6.42→ 65.32 ± 10.34→74.94 ± 10.90/39.92 ± 6.10→55.54 ± 8.47→ 66.50 ± 9.86	NR
Acupuncture vs. PPI treatment
Yang (2018) [18]	Acupuncture (36)	PPI (37)	4	1) RSI 2) RFS	(Intervention/comparison, baseline, 4 weeks) 1) 22.23 ± 5.31→11.46 ± 3.10/21.98 ± 4.22→14.98 ± 4.29 2) 15.18 ± 2.98→8.32 ± 1.55/15.46 ± 2.77→8.01 ± 1.21	NR

PPI, proton pump inhibitor; RSI, reflux symptom index; RFS, reflux finding score; UESP, upper esophageal sphincter pressure; LESP, lower esophageal sphincter pressure; NR, not reported; EGF, epidermal growth factor; H-med, herbal medicine; CG, control group; SF-36, Short Form 36 health survey..

3. Details of interventions

Table 3 shows the intervention-specific features of the 7 RCTs using the STRICTA guidelines [9].

Table 3 . Details of acupuncture interventions.

Author (y)	Treatment period	Details of needling
		Needle stimulation	Acupoints (uni/bilateral)	Needle type	Needle retention time
				De qi response and manipulation
	Treatment frequency
			Number of needle insertions
				Depth of insertion
Zhao (2021) [12 ]	4 weeks	Manual acupuncture	CV22 (U), CV13 (U), ST36 (NR), SP6 (NR), LR3 (NR)	25 or 40 mm length needle	30 min
	3x/wk		Not reported	Not reported/reinforcing-reducing manipulation
				CV22: 1 cun (along the posterior wall of sternum) CV13, ST36, SP6: 1 cun LR3: 0.5 cun
Li (2020) [13]	8 weeks	Manual acupuncture	CV22 (U), CV17 (U), CV12 (U), ST36 (B), LR3 (B)	25–40 mm length needle/diameter 0.25 mm (Huatuo brand)	30 min
	3x/wk		Not reported	Manipulation until de qi/manipulation once while needle retaining
				CV22: 0.8–1 cun (along the posterior wall of sternum) CV17: 0.5 cun (flatten downward) CV12, ST 36: 1 cun LR3: 0.5 cun
Liu (2020) [14]	8 weeks	Manual acupuncture	CV22 (U)	1.5 cun length needle	30 min
	2x/wk		Not reported	Manipulation until de qi/manipulation once after 15 minutes
				CV22: 1 cun (along the posterior wall of sternum)
Zhang (2017) [15]	8 weeks	Manual acupuncture	CV22 (U)	1.5 cun length needle	30 min
	2x/wk		Not reported	Manipulation until de qi/manipulation once after 15 minutes
				CV22: 1 cun (along the posterior wall of sternum)
Wang (2020) [16]	2 weeks	Manual acupuncture	CV22 (U), CV17 (U), CV12 (U), ST36 (NR), LR3 (NR), PC6 (NR) *and LI11 (NR), LI4 (NR) or ST40 (NR) or SP6 (NR), SP10 (NR)	0.30 mm × 40 mm, 0.30 mm × 25 mm (su food supervision equipment (quasi): 2012 No.2270864)	30 min
	5x/wk		Not reported	Not reported/supplementation and drainage manipulation
				CV22: 3–7 mm (along the posterior wall of sternum) CV17: 15–25 mm (flatten downward) CV12, ST36: 15–25 mm LR3: 7–15 mm PC6: 15–25 mm
Pei (2022) [17]	4 weeks	Manual acupuncture	CV23 (U), GV20 (U), PC6 (B), LR3 (B), LI4 (B), ST40 (B), ST36 (B), SP6 (B)	0.25 mm × 40 mm (Huatuo brand)	28 min
	2x/wk		Not reported	Not reported/reinforcing-reducing manipulation
				Not reported
Yang (2018) [18]	4 weeks	Manual acupuncture	GV20 (U), CV23 (U), LU7 (NR), CV12 (U), CV6 (U), KI6 (B), ST36 (B), GB34 (B), LR3 (B)	0.25 mm × 40 mm (Huatuo brand)	30 min
	3x/wk		Not reported	After sensation of qi arrival/supplementation and drainage manipulation 30 seconds
				Not reported

U, unilateral; B, bilateral; NR, not reported..

*At the discretion of diagnosis system of oriental medicine..

Three studies treated participants for 8 weeks, with 3 sessions per week in 1 study [13] and 2 sessions per week in 2 studies [14,15]. 3 other studies treated participants for 4 weeks, with 3 sessions per week in 2 studies [12,18] and 3 sessions per week in 1 study [17]. The remaining study [16] utilized 5 sessions per week for 2 weeks. All studies included a minimum of 8 and a maximum of 24 acupuncture sessions, with an average of 14 sessions.

The most commonly used acupoints were CV22 [12-16], ST36 [12,13,16-18], and LR3 [12,13,16-18], which were each used in 5 studies. 2 studies [14,15] selected only 1 acupoint: CV22. 2 studies [17,18] that did not use CV22 chose the acupoint CV23; thus, all studies included acupoints from the CV meridians. One study [16] did not use a uniform treatment for all patients but selected acupoints based on the patient diagnostic system of oriental medicine.

All studies used 25–40-mm long needles. Two studies [17,18] used a 40-mm long needle, 2 [14,15] used a 1.5-cun long needle, 2 [12,16] used a 25- or 40-mm long needle depending on the acupoint, and 1 [13] used a 25–40-mm long needle. 3 studies [13,17,18] used 0.25-mm and 1 study [16] used 0.30-mm diameter needles, the remaining studies reported no diameter data. Only 5 studies reported the depth and method of needling [12-16] which varied from 3 mm to 1 cun for each acupoint. 7 studies [12-18] presented needle manipulation. Two studies [12,17] used reinforcing-reducing manipulations, 2 other studies [16,18] used supplementation and drainage manipulations, and 3 studies [13-15] presented manipulation until de qi, of which 2 [14,15] presented intermediate manipulation with manipulation once after 15 minutes, and the remaining study [13] presented manipulation once during needle retention time. Another study [18] confirmed the de qi response by acupuncture needle insertion followed by supplementation and drainage manipulation for 30 seconds. The needle retention time was 30 minutes in 6 studies [12-16,18] and 28 minutes in 1 study [17].

4. Risk of bias

The Cochrane RoB Tool was used to assess the RoB in 7 RCTs. The results are presented in graphs and summarized using RevMan 5.4.1 (Figs. 2, 3).

Figure 2. Risk of bias (RoB) summary: judgments ofreview authors about each RoB item for each included study.

Figure 3. Risk of bias (RoB) graph: judgments ofreview authorsabout each RoB item presented as percentages across all included studies.

Five studies [12,13,15-17] were rated as “low” RoB because patients were randomized using a random number table. One study [18] was rated as “unclear” because the randomization methodology was not mentioned. One study [14] was rated as “high” RoB because it did not mention any randomization.

The RoB for this aspect was rated as “unclear” in all 7 studies because none described the concealment of assignment order to reduce bias [12-18].

The characteristics of acupuncture make it difficult for practitioners and patients to be blinded to the treatment. Additionally, all 7 studies [12-18] did not mention the use of blinding, and the RoB was rated as “high.”

All 7 studies [12-18] did not properly describe the method of rater blinding and were rated as “unclear.”

One study [12] reported an unbalanced number of missing values in the experimental and control groups, but with no significant difference, and the reasons for the missing values were explained; therefore, the RoB was rated as “low.” One study [18] reported a balanced number of missing values in the experimental and control groups; however, the RoB was rated as “unclear” because the reasons for missing values were not explained. The remaining 5 studies [13-17] had no missing data, and the RoB was rated as “low.”

All 7 studies [12-18] reported all outcomes expected from their experimental design and were rated as “low.”

All 7 studies [12-18] were rated as “low” because no additional biases were identified.

5. Effectiveness of intervention

Seven studies [12-18] were pooled and meta-analyzed using RSI, which is an indicator of LRP symptoms, as an outcome measure. Two studies [12,13] compared acupuncture in combination with PPI and PPI alone, and 4 studies [14-17] compared acupuncture in combination with conventional treatment (including PPI) and conventional treatment alone (including PPI). One study [18] compared acupuncture and PPI and was subgrouped, with the total results re-synthesized (Fig. 4). The MD was used to synthesize the same indicator, and a random-effects model was used, although the homogeneity between the studies could be high, because the studies could show substantial heterogeneity, such as in the experimental design and intervention details.

Figure 4. Reflux symptom index: comparison of treatments with and without acupuncture. SD, standard deviation; CI, confidence interval; PPI, proton pump inhibitors.

The subgroup analysis comparing acupuncture combined with PPI and PPI alone revealed that the experimental group demonstrated a greater reduction in RSI than the control group, with a statistically significant difference between the experimental and control groups and high heterogeneity (MD = −5.08; 95% CI, −7.76 to −2.39; p = 0.0002; I² = 71%). The comparison of acupuncture in combination with conventional treatment (including PPI) and conventional treatment alone (including PPI) revealed that the experimental group demonstrated a greater reduction in RSI than the control group. A statistically significant difference was found between the experimental and control groups and low heterogeneity (MD = −2.45; 95% CI, −2.63 to −2.27; p < 0.00001; I² = 9%). The experimental group demonstrated a greater improvement in RSI than the control group when comparing acupuncture with PPI (MD = −3.52; 95% CI, −5.23 to −1.81; p < 0.0001; I² = not applicable). We found that the intervention significantly reduced the RSI scores compared with conventional treatment in combining the subgroup analyses. The overall study (I² = 72%) and the subgroups were highly heterogeneous (I² = 61%). The quality of the meta-analysis was reduced because of the RoB and inconsistency in the quality assessment using the GRADE Tool (Table 4).

Table 4 . GRADE table of included studies.

No. of studies	Certainty assessment							No. of patients			Effect	Certainty
	Study design	Risk of bias	Inconsistency	Indirectness	Imprecision	Other considerations		Intervention	Control		Absolute (95% CI)
RSI: Comparison of treatments with and without acupuncture
7	Randomized trials	Serious*	Serious†	Not serious	Not serious	None		321	318		MD 3.00 lower (3.62 lower to 2.37 lower)	Low (⨁⨁◯◯)
RFS: Comparison of treatments with and without acupuncture
7	Randomized trials	Serious*	Very serious§	Not serious	Not serious	None		321	318		MD 1.39 lower (1.76 lower to 1.02 lower)	Very low (⨁◯◯◯)
UESP: Comparison of treatments with and without acupuncture
2	Randomized trials	Not serious	Not serious	Serious||	Serious‡	None		67	65		MD 7.29 higher (3.31 higher to 11.26 higher)	Low (⨁⨁◯◯)
LESP: Comparison of treatments with and without acupuncture
2	Randomized trials	Not serious	Very serious§	Serious||	Serious‡	None		67	65		MD 3.33 higher (0.33 higher to 6.32 higher)	Very low (⨁◯◯◯)

CI, confidence interval; MD, mean difference; RSI, reflux symptom index; RFS, reflux finding score; UESP, upper esophageal sphincter pressure; LESP, lower esophageal sphincter pressure..

*All studies had a high risk of performance bias, and 1 study had a high risk of selection bias. ^†I² was > 50% and > 75%. ^‡Total sample size was > 300, but the 95% CI of the MD did not cross zero. ^§I² was > 75%. ||Surrogate outcome..

All 7 studies [12-18] used RFS as an indicator of the laryngoscopic manifestation of LPR. Among these, 2 studies compared acupuncture in combination with PPI and PPI alone [12,13], whereas 4 studies compared acupuncture in combination with conventional treatment (including PPI) and conventional treatment alone (including PPI) [14-17]. One study [18] that compared acupuncture with PPI treatment was subgrouped, and the overall results were re-synthesized (Fig. 5). The MD was used to synthesize the same indicator, and a random-effects model was used because the studies could show substantial heterogeneity, such as the experimental design and intervention details.

Figure 5. Reflux finding score: comparison of treatments with and without acupuncture. SD, standard deviation; CI, confidence interval; PPI, proton pump inhibitors.

A subgroup analysis comparing acupuncture in combination with PPI treatment and PPI treatment alone revealed that the experimental group demonstrated a greater reduction in RFS than the control group, with a statistically significant difference between the intervention and control groups and low heterogeneity (MD = −2.64; 95% CI, −3.29 to −1.98; p < 0.00001; I² = 0%). The comparison of acupuncture in combination with conventional treatment (including PPI) and conventional treatment alone (including PPI) revealed that the experimental group demonstrated a greater reduction in RFS than the control group. A statistically significant difference was found between the experimental and control groups and low heterogeneity (MD = −1.40; 95% CI, −1.43 to −1.37; p < 0.00001; I² = 0%). The comparison of acupuncture with PPI treatment revealed that the experimental group demonstrated no greater improvement in RFS than the control group, with no statistically significant difference between the experimental and control groups (MD = 0.31; 95% CI, −0.33 to 0.95; p = 0.34; I² = not applicable). We found that the intervention significantly reduced RFS scores compared with conventional treatment alone by combining the subgroup analyses. The overall heterogeneity of the studies was high (I² = 86%), similar to the subgroups (I² = 95.1%). The quality of the meta-analysis was reduced because of the RoB and the inconsistency in the quality assessment using the GRADE Tool (Table 4).

We conducted a meta-analysis by combining 2 studies [12,13] that measured UESP as an indicator of esophageal motility (Fig. 6). Both studies compared acupuncture in combination with PPI and PPI alone. We used the MD as a summary statistic for the same outcome measure and random-effects model based on our judgment that differences in the experimental design and intervention details could contribute to variability in the study results.

Figure 6. Upper esophageal sphincter pressure: comparison of treatments with and without acupuncture. SD, standard deviation; CI, confidence interval.

This analysis revealed that the experimental group demonstrated higher UESP than the control group. A statistically significant difference was found between the intervention and control groups, and the heterogeneity was low (MD = 7.29; 95% CI, 3.31–11.26; p = 0.0003; I² = 0%). The quality of the meta-analyses was lowered because of indirect and imprecise factors in the quality assessment using the GRADE Tool (Table 4).

The same 2 studies [12,13] measured the LESP as an indicator of esophageal motility (Fig. 7). Both studies compared acupuncture in combination with PPI and PPI treatment alone. We used the MD as a summary statistic for the same outcome measure and random-effects model because differences in the experimental design and intervention details may contribute to variability in the study results.

Figure 7. Lower esophageal sphincter pressure: comparison of treatments with and without acupuncture. SD, standard deviation; CI, confidence interval.

This analysis revealed that the experimental group demonstrated higher LESP than the control group. A statistically significant difference was found between the intervention and control groups and the heterogeneity was high (MD = 3.33; 95% CI, 0.33–6.32; p = 0.03; I² = 81%). The quality of the meta-analysis was downgraded due to inconsistency, indirectness, and imprecision in the quality assessment using the GRADE Tool (Table 4).

6. Adverse events

All studies [12-18] did not report the occurrence of AEs.

7. Reporting bias assessment

The meta-analysis included 7 articles, which is less than the recommended minimum of 10 according to the Cochrane Handbook for Systematic Reviews of Interventions. Publication bias was not assessed [10].

The American Academy of Otolaryngology-Head and Neck Surgeons first adopted the term LPR in 2002, after Koufman reported it as an otolaryngological manifestation of gastroesophageal reflux disease. The number of publications and interest in LPR has gradually increased over the past 3 decades. LPR is a highly prevalent condition, with 1 study from Korea reporting that 25% of patients newly diagnosed with ENT conditions had LPR-related symptoms or findings [2,19]. However, objective and efficient diagnostic methods as well as accurate treatment standards for this condition remain lacking. Considerable disagreement exists regarding the drug treatment optimal dosage, administration, and duration for LPR [20]. PPIs are the preferred medications for managing LPR [6]. However, additional treatment options are needed due to the side effects of long-term use and the fact that PPIs do not resolve all LPR symptoms. Acupuncture and herbal treatments have been used in clinical practice to reduce LPR symptoms and have been consistently used in LPR treatment, as supported by numerous case studies. However, no systematic study has investigated the effects of acupuncture. Thus, this study aimed to systematically review RCTs investigating acupuncture for LPR to determine the effectiveness of acupuncture therapy and to provide evidence for its use.

We searched for clinical studies that included acupuncture as an intervention for patients with LPR up to March 2023 and selected 7 studies for the final analysis. The results indicated that the combination of acupuncture and conventional treatment was effective and statistically significant in reducing LPR symptoms compared with conventional treatment alone. However, the dependability of this meta-analysis was low according to the GRADE assessment. Therefore, further high-quality trials are needed to improve the evidence level and provide more robust conclusions regarding the effectiveness of acupuncture for LPR.

Following the STRICTA guidelines [9], we analyzed the treatments used in 7 RCTs and revealedthat CV22, ST36, and LR3 were the most commonly used acupoints, followed by CV12, CV17, and CV23. Additionally, most studies used CV22 or CV23 on the CV meridians near the throat and upper esophagus, where LRP symptoms occur, and acupuncture points that are effective for digestion, such as ST36, LR3, and CV12. 2 studies [14,15] added the stimulation of a single CV22 acupoint to the conventional treatment with PPIs and diet. This study design was selected to observe improvements in LRP symptoms in the posterior pharyngeal region, which had been disregarded in conventional treatment. Acupuncture treatment details, including the acupoint name, stimulation type, retention time, number, frequency, and duration of treatments, and evoked responses or manipulation, were presented in all studies. 5 studies reported needle insertion depth [12-16], and all studies reported the needle length, but only 4 reported the needle diameter [12,16-18]. Only 4 studies [13,16-18] reported production companies and materials. The number of acupuncture needles per patient, other treatment components, and practitioner background were not reported in any of the studies. These limitations restrict their reproducibility in the clinical setting. Therefore, future RCTs should prove detailed descriptions according to the STRICTA guidelines for reproduction in the clinical environment.

The evidence was insufficient to report the safety of the interventions because AEs were not reported. Further, future studies on LRP are recommended to include transparent reporting of AEs.

This study used systematic reviews and meta-analyses to generate high-level clinical evidence. However, this study has some limitations. First, all 7 studies were conducted in China, which limits the generalizability of the results globally. Second, the overall dependability of the results of the study was low due to possible biases in the included studies and the low quality of the meta-analysis, as evaluated by the Cochrane RoB and GRADE tools; therefore, the resultsshould be analyzed with caution. Third, the limited number of included studies and the large heterogeneity among them made the determination of the effect sizes and the identification of clear individual benefits of acupuncture difficult. Fourth, the lack of uniformity in the acupuncture interventions among the studies is a limitation. Acupuncture differed in each study in terms of acupuncture points, types of acupuncture needles, duration and frequency, and manual therapy methods used. Drawing firm conclusions that account for the clinical use of acupuncture in LRP is difficult because of the lack of consensus on acupuncture interventions.

This study indicates that acupuncture may effectively improve the symptoms and clinical signs of LPR. However, the evidence is limited due to various factors, such as inaccurate reporting of research methods in some RCTs, restriction to studies conducted only in China, and the small number of included studies. Therefore, the confidence level in the evidence was low, and further high-quality studies are required.

Conceptualization: JO, JL. Methodology: JO. Formal investigation: JO, Jaewoo Y, Jungmin Y, MK, SK, ML, JL. Data analysis: JO, Jaewoo Y, JL. Writing – original draft: JO. Writing – review and editing: All authors.

The authors have no conflicts of interest to declare.

None.

This research did not involve any human or animal experiment.