Acupuncture-related therapy for chronic cough: A systematic review and meta-analysis

Abstract

Background

Despite the high prevalence and socioeconomic burden of chronic cough, there has been an unmet medical need. Acupuncture may be promising for treating chronic cough with various pathophysiologies involving several neurotransmission mechanisms. We aimed to systematically compile evidence on the effect and safety of acupuncture-related therapy for chronic cough.

Methods

Through a search of 11 databases, randomized controlled trials (RCTs) published until February 15, 2024 that evaluated the effect of acupuncture-related therapy at acupuncture points (including acupuncture, acupoint herbal patching, and moxibustion) for patients with chronic cough were identified and analyzed without restrictions on the cause of cough or age. When there were two or more studies that measured the same outcome measures, a meta-analysis was performed, and the certainty of evidence was evaluated based on the GRADE methodology.

Results

A total of 30 RCTs with 2835 participants were included. When acupuncture-related therapy was implemented in addition to conventional treatment including medicine and lifestyle guidance according to symptoms and causes, cough severity, cough-related quality of life, and the total effective rate (TER) were significantly improved with no difference in the incidence of adverse events. Furthermore, compared with conventional treatment, acupuncture-related therapy significantly improved cough severity and the TER. The certainty of the evidence was generally moderate due to the risk of bias. The most frequently used acupuncture points in the included studies were BL13, GV14, CV17, and EX-B1.

Conclusion

Acupuncture-related therapy might be an effective and safe treatment for patients with chronic cough of various causes with complex pathophysiologies.

Protocol registration

PROSPERO, CRD42024518115. PROSPERO (CRD42024518115).

1. Introduction

Chronic cough refers to a cough that lasts for >4 weeks in children and 8 weeks in adults.^1,2 The prevalence of chronic cough has been reported to vary between 3 % and 24 % depending on the country.³ Although chronic cough is generally not life-threatening, it can have severe physical and psychological effects and result in the substantial utilization of medical resources.⁴ As chronic cough does not improve well with cough suppressants or expectorants, it is important to identify the cause and treat it accordingly. However, chronic cough has a variety of causes including upper airway cough syndrome (UACS), cough-variant asthma (CVA), gastroesophageal reflux disease (GERD), and respiratory tract infections, can be caused by more than one disease, and often has no specific symptoms or definitive testing methods; thus, empirical treatment is performed first in many cases.5, 6, 7, 8 Consequently, patients who visited clinics with chronic cough have reported unmet medical needs such as treatment ineffectiveness and unclear diagnosis, and the cough is not controlled despite treatment in >60 % of cases.⁹ Therefore, there is a need for an effective treatment for the chronic cough itself, which is not simply a symptom resulting from diseases but shows complex pathophysiologies with various causes.

East Asian traditional medicine (EATM) treatment, such as acupuncture and moxibustion, has been used to treat various diseases, including respiratory symptoms and diseases. As a non-pharmacological EATM therapy, acupuncture has attracted attention from clinicians and researchers not only for its local effects but also for its systemic effects. The effect of acupuncture on oxidative stress and systemic immunity in animal models highlights various potential indications for acupuncture.^10,11 Chronic cough of various causes can aggravate the course of cough with cough hypersensitivity, and airway inflammation and systemic inflammation can contribute to plastic changes in airway mucosal innervation, an increased inflow of inflammatory cells, and the central sensitization of cough.^12,13 Notably, some clinical and experimental studies showed an improvement in chronic cough after acupuncture-related therapy.^14,15

An analysis of the status of clinical research involving acupuncture in the treatment of chronic cough and a systematic review of acupuncture for common causes of chronic cough (e.g., CVA) have been conducted^16,17; however, to the best of our knowledge, the evidence on acupuncture-related therapy for chronic cough symptoms with complex pathophysiologies and various causes has not been systematically summarized. Therefore, in this study, we sought to systematically summarize and synthesize the evidence on the effectiveness and safety of acupuncture-related therapy for chronic cough symptoms by comprehensively collecting relevant studies regardless of the underlying cause of the cough. We examined the complex pathophysiology and various etiologies in our study, which included subgroup analyses based on the underlying causes.

2. Methods

2.1. Protocol registration

The protocol of this systematic review was registered in PROSPERO (CRD42024518115).

2.2. Eligibility criteria

• (1)Population: Studies involving participants with chronic cough (cough lasting >4 weeks in children and >8 weeks in adults^1,2) were included, without restriction on age, sex, race, or cause of cough. However, studies that did not report the cough duration of participants were excluded due to uncertainty as to whether they had chronic cough.
• (2)Intervention: All types of acupuncture-related therapy to acupuncture points, including acupuncture, acupressure, acupoint herbal patching (AHP; stimulation of acupuncture points with a herbal medicine patch), pharmacopuncture (combination of acupuncture with the injection of herbal medicine), bloodletting, and moxibustion (use of ignited material (usually moxa) to apply heat to certain points or areas of the body surface for curing disease through regulation of the function of meridians/channels and visceral organs¹⁸) as monotherapy or adjunctive therapy, were included.
• (3)Comparator: Studies with sham intervention, no treatment, and conventional treatment including medicine and lifestyle guidance according to symptoms and causes as control group interventions were included. However, studies with EATM interventions such as herbal medicine, acupuncture, and moxibustion as a control group intervention were excluded.
• (4)Outcomes: The primary outcome was post-treatment cough severity, measured by scales such as the cough symptom score, simplified cough score (SCS), and visual analog scale (VAS).¹⁹ Secondary outcomes included post-treatment cough-related quality of life, measured by measures such as the Leicester Cough Questionnaire (LCQ) and Cough Quality-of-Life Questionnaire (CQLQ),¹⁹ post-treatment cough frequency, total effective rate (TER) based on cough symptom improvement, recurrence rate, and incidence of adverse events during the trial period. Only studies reporting at least one of the above outcomes of interest were included.
• (5)Study design: Only parallel-group randomized controlled trials (RCTs) were included, and crossover RCTs were excluded to reduce the risk of potential bias.

2.3. Information sources and search strategy

A total of 11 bibliographic databases were searched on February 15, 2024 without language restrictions: MEDLINE, Embase, Cochrane Central Register of Controlled Trials (CENTRAL), Allied and Complementary Medicine Database (AMED), Oriental Medicine Advanced Searching Integrated System (OASIS), Korean Medical Database (KMbase), ScienceON, Research Information Sharing Service (RISS), China National Knowledge Infrastructure (CNKI), Wanfang Data, and CiNii. The reference lists of the included articles and relevant review articles were searched to identify eligible studies. Search strategies were set through discussions with experts on pulmonology in Korean medicine and systematic reviews, and detailed search strategies are described in Supplement 1.

2.4. Study selection and data collection process

Studies retrieved from the database were imported into Endnote 20 (Clarivate Analytics, Philadelphia, PA, USA). Relevant studies were initially included following a review of titles and abstracts, and the final included studies were selected through a full-text review. The following items were extracted from the included studies using a pilot-tested Excel sheet: basic study characteristics (publication year, country, sample size, study setting, and funding sources), population (age, cause of cough, and pattern identification), intervention, comparator, outcomes of interest, results, and information for risk of bias assessment. In particular, details of acupuncture-related therapy were extracted according to the STandards for Reporting Interventions in Clinical Trials of Acupuncture (STRICTA).²⁰ Two researchers independently conducted study selection (BL and CYK) and data extraction (BL and YKJ), and in cases of disagreement, consensus was reached through discussion.

2.5. Risk of bias assessment

The risk of bias was assessed using the Cochrane Collaboration's risk of bias tool.²¹ The following items were assessed as “low”, “unclear”, or “high” risk of bias for the included studies: random sequence generation method, allocation concealment, blinding of participants, personnel and outcome assessors, completeness of outcome data, selective reporting, and other bias (focusing on imbalances of baseline characteristics between the two groups). Two researchers independently assessed the risk of bias, and consensus was reached through discussion if there was disagreement.

2.6. Data analysis and synthesis

Qualitative analysis was performed on all included studies. If there were two or more studies that used the same type of intervention and comparator and measured the same outcomes of interest, a meta-analysis was performed using Review Manager software, version 5.4 (Cochrane, London, UK). Meta-analysis results for continuous variables are expressed as the mean difference (MD) and 95 % confidence interval (CI), and binary outcomes are expressed as the risk ratio (RR) and 95 % CI. In cases where only the sub-item scores of each questionnaire corresponding to the outcomes of interest were given in the included studies, the total score was calculated using a formula. Considering the inevitable clinical heterogeneity in the details of acupuncture-related therapy in the included studies, a random-effects meta-analysis model was used. The statistical heterogeneity of meta-analysis results was measured using I² statistics, and if I² was 50 % or more, substantial heterogeneity was considered to exist.

Subgroup analysis was performed to identify the cause of heterogeneity and determine whether the effect estimates were different according to the¹ type of acupuncture-related therapy (acupuncture, AHP, moxibustion, etc.),² age of participants (children, adults, and both), and³ cause of cough. We planned for sensitivity analysis by excluding¹ studies with a high risk of bias and² outliers to determine the robustness of the synthesized results. When 10 or more studies were included in the meta-analysis, the risk of publication bias was assessed using the symmetry of the funnel plot and Egger's test.

2.7. Certainty of evidence

The certainty of evidence for individual outcomes of interest was evaluated as “high”, “moderate”, “low”, and “very low” considering the risk of bias, indirectness, inconsistency, imprecision, and publication bias according to the GRADE methodology.²²

3. Results

3.1. Study selection and characteristics

A total of 1763 articles were identified from databases, and there were no additional studies from other sources. After excluding 130 duplicate articles and additionally excluding 1518 studies following the screening of titles and abstracts, the full texts of 115 articles were reviewed. Subsequently, 30 studies with 2835 participants that met the eligibility criteria were finally included in the analysis23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52 (Fig. 1 and Supplement 2).

Fig. 1.

A flow diagram of the study selection process.

EATM, East Asian traditional medicine; RCT, randomized controlled trial.

All studies were conducted in China, and except for one study jointly conducted at a hospital and a community health service center,²⁶ the settings of the remaining studies were all clinics/hospitals. A total of 8 studies reported funding sources,24, 25, 26^{,35,42,46,47,52} all of which were academic funding from governments or universities. There were 6 studies targeting children^{29,30,38,45,46,48} and 3 studies targeting both children and adults,^37,49,51 and the remaining studies targeted adults. Among 14 studies that reported the causes of chronic cough, 2 studies targeted CVA,^41,48 1 study targeted cough due to GERD,³⁵ 1 study targeted unexplained chronic cough,⁴⁷ and 3 studies included participants with various causes of cough.^29,32,34 In addition, 3 studies included participants after surgical treatment of lung cancer,^25,26,52 2 studies included participants with pulmonary tuberculosis,^40,44 1 study targeted cough due to chronic obstructive pulmonary disease (COPD),²⁷ and 1 study targeted cough due to vocal cord dysfunction.⁴² There were 8 studies using acupuncture,^25,26,31,41, 42, 43^,45,52 12 studies using AHP,^23,29,32, 33, 34^{,37,38,40,46,48,49,51} 3 studies using moxibustion,^28,35,36 3 studies using AHP plus acupressure,^30,44,50 2 studies using acupoint hot compress,^24,47 1 study using bloodletting,³⁹ and 1 study using pharmacopuncture²⁷ as types of intervention. All control group interventions were conventional treatment, except for 1 study with no treatment.⁵² A total of 4 studies evaluated cough severity, of which 3 studies used the SCS^26,35,43 and 1 study used the cough VAS.³¹ A total of 7 studies evaluated cough-related quality of life, of which 6 studies used the LCQ24, 25, 26^,35,36,52 and 1 study used the CQLQ.⁴⁷ No study evaluated cough frequency, and the TER based on cough symptom improvement was evaluated in 23 studies.^23,25,27, 28, 29^,31, 32, 33, 34, 35^,37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47^,49,50 The recurrence rate was evaluated in 1 study,⁴⁸ and 8 studies reported adverse events during the treatment period^{30,34,38,39,42,45,49,51} (Table 1). In the included studies, acupuncture-related therapy involved a total of 39 acupuncture points. The most frequently used acupuncture point was BL13 (25 studies), followed by GV14 (12 studies), CV17 (9 studies), EX-B1 (8 studies), BL20 (7 studies), ST36 (7 studies), and CV22 (7 studies) (Supplement 3).

Table 1.

Clinical characteristics of the included studies.

Study ID (Ref.)	Age (Year)	Cause of cough	(A) Treatment intervention (No. of participants)	(B) Control intervention (No. of participants)	Outcome of interest	Detailed results	Adverse event
Chen 202425	18-81	After surgical treatment of lung cancer (thoracoscopic lobectomy or sublobar resection)	Acupuncture + (B) (30)	Compound methoxynamine (2 capsules) (3 times/d) (30)	1) TER 2) LCQ	1) (A)>(B)* 2) (A)>(B)+	Not reported
Ding 202326	(A) 58.7; (B) 56.7	After complete radical surgical treatment of lung cancer	Acupuncture + (B) (30)	Ambroxol hydrochloride oral solution (10 ml) (3 times/d), terbutaline sulfate atomized inhalation solution (5 mg) (2 times/d), budesonide suspension for inhalation (2 mg) (2 times/d) (30)	1) SCS 2) LCQ	1) daytime (A)<(B)*; nighttime (A)<(B)+ 2) (A)>(B)+	Not reported
Huang 201831	(A) 36.5; (B) 38.1	Not reported	Acupuncture (46)	Compound methoxynamine (2 capsules) (3 times/d), montelukast sodium (Singulair) (10 mg) (1 time/d) (46)	1) Cough VAS 2) TER	1) (A)<(B)* 2) (A)>(B)*	Not reported
Lu 201141	(A) 38.4; (B) 36.2	CVA	Acupuncture (60)	Budesonide inhaler (100 μg) (2 times/d), albuterol aerosol (2 sprays) (2 times/d), aminophylline tablets (0.1 g) (3 times/d), montelukast sodium (Singulair) tablets (10 mg) (1 time/d) (60)	TER	N.S	Not reported
Luo 201642	(A) 18-65; (B) 19-65	Vocal cord dysfunction	Acupuncture + (B) (15)	Pulmicort (2 ml spray), psychological counseling, pronunciation treatment (15)	1) TER 2) Adverse events	1) (A)>(B)* 2) NA	varying degrees of sore throat in a short period, and the symptoms disappeared after 2 weeks (A: number of patients was not reported)
Meng 2023#43	(A) 40.0; (B) 41.0	Not reported	Acupuncture + (B) (30)	Intravenous infusion of ambroxol injection, oral methoxyphenamine capsules, aerosolized acetylcysteine, budesonide suspension/ipratropium bromide (30)	1) SCS 2) TER	1) (A)<(B)+ 2) N.S	Not reported
Wang 201945	(A) 7.2; (B) 7.7	Not reported	Acupuncture + (B) (60)	montelukast sodium (Singulair) (<6 yr, 4 mg; 6∼14 yr, 5 mg; 1 time/d) (60)	1) TER 2) Adverse events	1) (A)>(B)* 2) NA	None
Zhu 202252	(A) 56.3; (B) 58.8	Postoperative patients with non-small cell lung cancer	Acupuncture (20)	No treatment 20	LCQ	(A)>(B)+	Not reported
Chen 201824	NR	Not reported	Acupoint hot compress + (B) (30)	Antitussive, expectorant, anti-inflammatory, antiviral treatments (30)	LCQ	(A)>(B)+	Not reported
Xie 202147	(A) 43.6; (B) 42.6	Unexplained	Acupoint hot compress + (B) (40)	Ambroxol hydrochloride and clenbuterol hydrochloride oral solution (20 ml), amoxicillin and clavulanate potassium chewable tablets (3 tablets) (2 times/d) (39)	1) TER 2) CQLQ	1) (A)>(B)* 2) (A)<(B)*	Not reported
Chen 201523	(A) 35.7; (B) 36.8	Not reported	AHP + (B) (56)	Procaterol hydrochloride (25 μg) (2 times/d) (56)	TER	(A)>(B)*	Not reported
He 201429	(A) 5.5; (B) 5.0	UACS, CVA, after respiratory infection, unexplained	AHP + (B) (65)	Expectorant, antihistamine, anti-asthma, anti-inflammatory treatments -CVA: bronchodilators, glucocorticoids -Respiratory tract infection: antibiotics -UACS: pseudoephedrine hydrochloride (90 mg), chlorpheniramine (4.0 mg), nasal spray (70)	TER	(A)>(B)*	Not reported
Huo 201932	(A) 50.4; (B) 50.3	UACS, pulmonary interstitial disease, COPD	AHP (75)	Cough suppressant, anti-infection (74)	TER	(A)>(B)+	Not reported
Lai 201833	(A) 45.3; (B) 47.1	Not reported	AHP + (B) (30)	Loratadine tablets (10 mg), aminophylline sustained-release tablets (0.1 g), ambroxol tablets (30 mg) (2 times/d) (30)	TER	(A)>(B)*	Not reported
Lei 201534	(A) 42.7; (B) 43.8	UACS, CVA, pulmonary interstitial disease, COPD	AHP + (B) (50)	Routine clinical anti-infection, cough relief, and asthma treatment methods (50)	1) TER 2) Adverse events	1) (A)>(B)* 2) N.S	allergic reaction (A:1) (withdraw from the trial)
Li 2019a37	(A) 27.5; (B) 26.3	Not reported	AHP + (B) (78)	Antibiotics, antitussives, nursing based on pattern identification (78)	TER	(A)>(B)*	Not reported
Li 202038	(A) 6.4 mo; (B) 7.0 mo	Not reported	AHP + (B) (110)	Antitussive, expectorant treatments (110)	1) TER 2) Adverse events	1) (A)>(B)* 2) NA	None
Lin 201940	(A) 60.3; (B) 61.0	Pulmonary tuberculosis	AHP (43)	Oxygen therapy, bronchodilators, antibiotics, glucocorticoids, expectorants (43)	TER	(A)>(B)+	Not reported
Wang 201646	(A) 7.2; (B) 7.1	Not reported	AHP + (B) (25)	Compound pseudoephedrine hydrochloride oral solution, terbutaline, montelukast sodium (Singulair) (1 time/d) (25)	TER	(A)>(B)*	Not reported
Yang 201548	(A) 11.1; (B) 12.1	CVA	AHP + (B) 35	Theophylline, beta-receptor agonists, montelukast sodium (Singulair) tablet (5 mg) (1 time/d) 35	Recurrence rate	(A)<(B)*	Not reported
Yang 201649	(A) 28; (B) 28	Not reported	AHP + (B) (81)	Antibiotic, bronchodilator, expectorant, antitussive treatments (81)	1) TER 2) Adverse events	1) (A)>(B)* 2) N.S	itching (A:1), nausea and vomiting (B:2), others (B:1)
Zhao 202051	(A) 42.5; (B) 42.3	Not reported	AHP + (B) (38)	Antibiotics and additional treatments such as cough relieving, vasodilation, phlegm reduction (38)	Adverse events	(A)<(B)*	nausea and vomiting (A:1; B:3), dizziness and headache (B:2), skin rash and itching (B:2)
He 202230	(A) 8.1; (B) 8.5	Not reported	AHP + acupressure + (B) (64)	Provide timely diagnosis and treatment (64)	Adverse events	N.S	skin itching (A:1; B:2), nausea and vomiting (A:1; B:2), others (A:1; B:1)
Shen 202144	(A) 59.4; (B) 57.5	Pulmonary tuberculosis	AHP + acupressure + (B) (45)	Routine nursing measures (lifestyle guidance, dietary guidance, emotional care, etc.) (45)	TER	(A)>(B)+	Not reported
Yao 201950	(A) 3.1; (B) 3.3	Not reported	AHP + acupressure (63)	Roxithromycin dispersible tablets (3–4 mg/kg*d), ambroxol tablets (2 times/d) (63)	TER	(A)>(B)*	Not reported
Li 202139	(A) 44.5; (B) 44.5	Not reported	Bloodletting + (B) (35)	montelukast sodium (Singulair) tablets (10 mg), inhaled budesonide suspension (1 mg) (1 time/d) (35)	1) TER 2) Adverse events	1) (A)>(B)* 2) N.S	drowsiness (A:1; B:1), skin rash (A:1; B:1), loss of appetite (B:1), diarrhea (B:1)
Han 201728	(A) 8.8; (B) 8.7	Not reported	Moxibustion (51)	Ambroxol (1.2∼1.6 mg/kg*d), roxithromycin (3–4 mg/kg*d), pulmicort (1 mg) (2 times/d) (51)	TER	(A)>(B)*	Not reported
Li 201236	42.1	Not reported	Moxibustion + (B) (40)	Loratadine (10 mg), triamcinolone acetonide nasal spray (1 time/d) (40)	LCQ	(A)>(B)*	Not reported
Li 2019b35	(A) 37; (B) 37	GERD	Moxibustion + (B) (39)	Omeprazole enteric-coated capsules 20 mg (1 time/d), mosapride dispersible tablets 5 mg (3 times/d) (39)	1) SCS 2) LCQ 3) TER	1) (A)<(B)* 2) (A)>(B)* 3) (A)>(B)*	Not reported
Gu 201927	(A) 63.6; (B) 64.1	COPD	Pharmacoacupuncture + (B) (32)	-Allergic inflammation: oral antihistamines, beta-receptor agonists, inhaled corticosteroids, etc. -Infectious cough: antibiotics, supplemented by measures such as drainage or sputum -Gastroesophageal reflux cough: pay attention to their diet and sleeping position at night (32)	TER	(A)>(B)*	Not reported

Abbreviations. AHP, acupoint herbal patching; COPD, chronic obstructive pulmonary disease; CQLQ, Cough Quality-of-Life Questionnaire; CVA, cough variant asthma; GERD, gastroesophageal reflux disease; LCQ, Leicester Cough Questionnaire; NA, not applicable; NR: not reported; SCS, simplified cough score; SD, standard deviation; TER, total effective rate; UACS, upper airway cough syndrome; VAS, visual analog scale.

* p < 0.05; + p < 0.01; N.S, no significant difference between the two groups.

# Only information on groups that meet the eligibility criteria for intervention and comparison was extracted.

3.2. Risk of bias assessment

A total of 19 studies were evaluated as having a low risk of bias in the random sequence generation domain because they appropriately generated random numbers using computer software or random number tables,26, 27, 28, 29, 30^,34,35,37, 38, 39, 40^,44, 45, 46, 47^,49, 50, 51, 52 and the remaining studies did not report detailed methods. All studies did not report allocation concealment before randomization and the blinding of outcome assessors. Additionally, all studies did not report the blinding of participants and personnel. However, blinding was not possible due to the nature of the comparison groups, and all studies were evaluated as having a high risk of performance bias. One study had missing values, performed a per-protocol analysis, and did not report the number of participants included in the final analysis; thus, it was assessed as having a high risk of attrition bias.⁵² Two studies that did not report all of the observed outcomes were assessed as having a high risk of reporting bias.^30,52 One study that did not report the homogeneity of baseline clinical characteristics between groups was assessed as having an unclear risk of other bias²⁴ (Fig. 2).

Fig. 2.

Risk of bias of the included studies.

Low, unclear, and high risks are represented by the respective symbols: “+”, “?”, and “-”.

3.3. Acupuncture-related therapy plus conventional treatment versus conventional treatment alone

A total of 23 studies compared acupuncture-related therapy plus conventional treatment with conventional treatment alone.23, 24, 25, 26, 27^,29,30,33, 34, 35, 36, 37, 38, 39^,42, 43, 44, 45, 46, 47, 48, 49^,51

3.3.1. Primary outcome

In comparison with the conventional treatment alone group, the combined treatment group showed significantly improved cough severity measured by the SCS (3 RCTs, MD −0.85, 95 % CI −1.25 to −0.45, I² = 86 %) (Fig. 3A). Daytime and nighttime SCS was also significantly improved in the combined treatment group compared with the conventional treatment alone group.

Fig. 3.

Forest plots of the comparison between acupuncture-related therapy plus conventional treatment and conventional treatment alone: (A) simplified cough score, (B) Leicester Cough Questionnaire, and (C) total effective rate.

ART, acupuncture-related therapy; CT, conventional treatment.

3.3.2. Secondary outcomes

Cough-related quality of life measured by the LCQ was significantly improved in the combined treatment group compared with the conventional treatment alone group (4 RCTs, MD 0.88, 95 % CI 0.35 to 1.41, I² = 76 %) (Fig. 3B). Among the sub-items of the LCQ, the physical and psychological domains were significantly improved in the combined treatment group; however, there was no difference between the groups in the social domain. In a study,³⁶ only the sum of 19 questions in the LCQ was presented, making meta-analysis with other studies impossible; however, the sum of 19 questions in the LCQ was significantly improved in the combined treatment group compared with the control group (p < 0.05). One study evaluated cough-related quality of life using the CQLQ, and quality of life was significantly improved in the acupoint hot compress plus conventional treatment group compared with the conventional treatment alone group (p < 0.05).⁴⁷ The TER based on cough symptom improvement was significantly higher in the combined treatment group (17 RCTs, RR 1.16, 95 % CI 1.11 to 1.22, I² = 33 %) (Fig. 3C). Funnel plots of the meta-analysis results for the TER showed asymmetry, and Egger's test showed a p value < 0.001, suggesting a potential risk of publication bias (Supplement 4). One study compared the recurrence rate of CVA between the two groups, and a significantly lower CVA recurrence rate after 1, 2, and 3 years of treatment was observed in the AHP plus conventional treatment group compared with the conventional treatment alone group (p < 0.05).⁴⁸ There was no difference in the incidence of adverse events between the two groups (7 RCTs, RR 0.46, 95 % CI 0.20 to 1.07, I² = 0 %). Subgroup analysis according to the intervention type, age, and cause of cough did not generally change the directionality of effect estimates (Supplement 5).

3.4. Acupuncture-related therapy versus conventional treatment

A total of 6 studies compared acupuncture-related therapy with conventional treatment.^{28,31,32,40,41,50}

3.4.1. Primary outcome

In one study,³¹ cough severity was evaluated using the cough VAS, which was significantly improved after acupuncture compared with conventional treatment (p < 0.05).

3.4.2. Secondary outcomes

In comparison with the conventional treatment group, the acupuncture-related therapy group had a significantly higher TER based on cough symptom improvement (6 RCTs, RR 1.14, 95 % CI 1.03 to 1.26, I² = 66 %) (Fig. 4). Overall, subgroup analysis reduced statistical heterogeneity and did not change the directionality of effect estimates (Supplement 5).

Fig. 4.

Forest plot of the comparison between acupuncture-related therapy and conventional treatment: total effective rate.

ART, acupuncture-related therapy; CT, conventional treatment.

3.5. Acupuncture-related therapy versus no treatment

3.5.1. Secondary outcomes

One study compared electroacupuncture with no treatment for postoperative patients with non-small cell lung cancer with chronic cough.⁵² In comparison with no treatment, electroacupuncture significantly increased the LCQ total score (p < 0.01) and physical domain score (p < 0.001); however, there was no significant difference between the two groups in psychological and social domain scores.

3.6. Certainty of evidence

The certainty of evidence for outcomes of interest was generally moderate due to the risk of bias in the included studies (Table 2).

Table 2.

Summary of findings.

Outcome	No. of participants (studies)	Effect estimate [95 % CI]	Certainty of evidence	Reason for downgrading
ART plus CT vs. CT
Total SCS	195 (3 RCTs)	MD −0.85 [−1.25, −0.45]	Moderate	Risk of bias1
Daytime SCS	120 (2 RCTs)	MD −0.54 [−0.65, −0.43]	Moderate	Risk of bias1
Nighttime SCS	120 (2 RCTs)	MD −0.53 [−0.77, −0.30]	Moderate	Risk of bias1
Total LCQ	255 (4 RCTs)	MD 0.88 [0.35, 1.41]	Moderate	Risk of bias1
LCQ physical domain	195 (3 RCTs)	MD 0.50 [0.01, 0.98]	Moderate	Risk of bias1
LCQ psychological domain	195 (3 RCTs)	MD 0.80 [0.02, 1.58]	Moderate	Risk of bias1
LCQ social domain	195 (3 RCTs)	MD 0.16 [−0.23, 0.56]	Very low	Risk of bias1, Inconsistency2, Imprecision3
TER	1642 (17 RCTs)	RR 1.16 [1.11, 1.22]	Low	Risk of bias1, Publication bias4
Adverse events	876 (7 RCTs)	RR 0.46 [0.20, 1.07]	Low	Risk of bias1, Imprecision3
ART vs. CT
TER	675 (6 RCTs)	RR 1.14 [1.03, 1.26]	Low	Risk of bias1, Inconsistency2

ART, acupuncture-related therapy; CI, confidence interval; CT, conventional treatment; LCQ, Leicester Cough Questionnaire; MD, mean difference; RCT, randomized controlled trial; RR, risk ratio; SCS, simplified cough score; TER, total effective rate.

¹High risk of performance bias.

²The direction of the effect size of the included studies is different, and the I² value is greater than 50 %.

³The 95 % confidence interval for the mean difference or risk ratio value is wide and passes the null value.

⁴The funnel plot shows visual asymmetry, and the p value in Egger's test is <0.05.

4. Discussion

4.1. Findings of this review

By analyzing 30 studies with 2835 participants that applied acupuncture-related therapy to patients with chronic cough and evaluated its effectiveness, several findings were derived. When acupuncture-related therapy was implemented in addition to conventional treatment including medicine and lifestyle guidance according to symptoms and causes (combined treatment), cough severity, cough-related quality of life (especially in the physical and psychological domains), and the TER based on symptom improvement were significantly improved. The recurrence rate was also significantly lower in the acupuncture-related therapy group than in the conventional treatment group, and there was no difference in the incidence of adverse events between the two groups. In addition, compared with conventional treatment, acupuncture-related therapy significantly improved cough severity and the TER based on cough symptom improvement. Overall, the included studies were at risk of selection and performance bias. The certainty of the evidence of effect estimates was generally moderate due to the risk of bias.

4.2. Clinical interpretation

Chronic cough can be caused by a combination of etiologies, and nonspecific chronic cough (cause is unknown even after initial examination and evaluation) and unexplained chronic cough (cause cannot be determined even after comprehensive investigation and appropriate therapeutic trials) are also common.^8,53 In general, conventional medications are administered for specific coughs depending on the cause. However, due to the complexity and uncertainty of the cause, empirical treatment and antitussives/expectorants may be administered; thus, the effect could remain insufficient.^8,9 In addition, with the recent increased interest in the whole health concept,⁵⁴ acupuncture can be a promising option for the treatment of chronic cough with complex pathophysiologies and various causes.^10,11 In particular, EATM therapies including acupuncture and herbal medicine have been actively used to treat cough in East Asian clinical settings.^55,56 However, a previous study has only summarized the evidence on the effectiveness of acupuncture for specific causes of cough, such as CVA.¹⁷ Therefore, to accumulate evidence on EATM treatment for chronic cough symptoms, our research team first summarized the status of clinical research on EATM treatment for chronic cough regardless of the cause.¹⁶ As a result, a significant number of clinical studies were identified, which applied acupuncture-related therapies such as acupuncture, moxibustion, and AHP to patients with chronic cough of various causes.¹⁶ In addition, the eligibility criteria for this systematic review were set using the population, comparator, and outcome information confirmed through a scoping review.¹⁶ Therefore, this study is significant in that it is the first to organize and analyze various RCTs that applied acupuncture-related therapy at acupuncture points for patients with chronic cough with complex pathophysiologies and various causes.

The frequently used acupuncture points in the included trials were BL13, GV14, CV17, EX-B1, BL20, ST36, and CV22. From the perspective of EATM, acupuncture points can be selected based on pattern identification for individual patients, and these acupuncture points have been extensively studied for various respiratory diseases in modern medical research. Positioned near the third thoracic vertebra, BL13 plays a crucial role in modulating cytokine signaling within the immune system, particularly involving interleukins (ILs).⁵⁷ An animal study has indicated that moxibustion at BL13 could regulate lung immune responses, potentially preventing and alleviating asthma symptoms by affecting inflammatory cells, eosinophil levels, and various cytokines.⁵⁸ Moreover, electroacupuncture stimulation at BL13 and ST36 could enhance its anti-inflammatory effects by increasing the levels of nuclear factor erythroid-2-related factor-2 and heme oxygenase-1 in rabbits.⁵⁹ This process may be facilitated by the activation of p38 mitogen-activated protein kinase signaling, protecting the lungs from oxidative stress and damage caused by the excessive discharge of oxygen free radicals and inflammatory agents.⁵⁹ Electroacupuncture at ST36 and BL13 in a rat model has been found to regulate the inflammatory responses of COPD via the cholinergic anti-inflammatory pathway.⁶⁰ Stimulation at BL13 and CV22 could improve pulmonary function and reduce inflammatory infiltration in lung tissues.⁶¹ Moreover, acupuncture at BL13, BL20, ST36, and other acupoints could improve the nutritional state of COPD patients, as demonstrated by enhancements in body weight, inflammatory biomarkers, and quality of life indicators.⁶² These findings underscore the multifaceted and complex mechanisms by which acupuncture-related therapy exerts its therapeutic effects under various respiratory conditions through the targeted stimulation of acupuncture points.

The socioeconomic burden associated with chronic cough is high,^4,63 and the evidence on available non-pharmacological treatments is limited.⁶⁴ Considering that it can be difficult to identify a treatable cause of chronic cough even after a thorough clinical evaluation,⁶ there is a need for non-pharmacological therapies that have fewer side effects, are more economical, and have multifaceted mechanisms. The current findings show that acupuncture-related therapy may be considered as one of the candidates. The multifaceted effects of acupuncture-related therapy have the potential to improve not only the cough severity but also the quality of life of individuals with chronic cough, as demonstrated by our findings. However, the economic efficiency of this treatment for patients with chronic cough needs to be further investigated.

4.3. Suggestions for further studies

None of the included studies reported cough frequency after the intervention, and among EATM clinical studies on chronic cough, only <2 % of studies reported cough frequency as the outcome.¹⁶ Cough frequency is an ideal primary outcome measure in clinical studies of patients with cough; however, its application has been limited by its high cost and reliance on aural validation.⁶⁵ Recently, research has been conducted on cough detection technology for the development of smartphone apps and wearable devices.⁶⁵ Therefore, programs that can measure cough frequency may be widely available for future research and clinical practice,⁶⁵ which should also be applied in future clinical studies related to acupuncture-related therapy. In addition, less than half of the included studies reported the causes of chronic cough, and there was a study on unexplained chronic cough.⁵³ Recently, the definitions of nonspecific and unexplained chronic cough have been established through guidelines.^8,53 Furthermore, the potential of EATM treatments including acupuncture and herbal medicine for both chronic cough with a specific cause and nonspecific chronic cough has been reported by several studies.^16,66,67 Therefore, in future chronic cough research, it is necessary to mention the cause of cough in the case of a specific cough and to report the corresponding terminology in the case of a nonspecific or unexplained cough. This will help establish clear evidence of the effect of acupuncture-related therapy for the treatment of chronic cough. In addition, although the safety of acupuncture performed by certified acupuncture practitioners has been reported,⁶⁸ only 7 studies included in this study reported whether adverse events occurred following the intervention. Therefore, future clinical studies should report safety-related information on acupuncture-related therapy.

4.4. Limitations

This study has some limitations. First, statistical heterogeneity remained unresolved despite attempts at subgroup analysis according to the intervention type, age, and cause of cough. In particular, the limited number of studies evaluating individual outcomes resulted in very small subgroup sizes, restricting the exploration of heterogeneity and the interpretation of results. Consequently, factors such as intervention type, age, and cause of cough—insufficiently addressed through subgroup analysis—might have contributed to substantial heterogeneity. In addition, even though we performed subgroup analyses according to the type of acupuncture-related therapy, it was difficult to distinguish differences in effectiveness according to the application and potential mechanism of the interventions due to the small number of included studies. Furthermore, the diversity of acupuncture points and treatment duration in the included studies might be another source of clinical heterogeneity. Second, we planned for sensitivity analysis to evaluate the robustness of the meta-analysis results; however, we could not perform this because all studies were judged to have a high risk of performance bias. Additionally, this high risk of bias affected the certainty of the evidence for the main findings. Although the risk of performance bias can be reduced, the physiological inertness of sham acupuncture continues to raise questions about whether it is an appropriate control group for assessing the efficacy of acupuncture.^69,70 Therefore, if the blinding of participants and acupuncturists is difficult due to the comparison of the intervention with conventional treatment, the addition of objective evaluation indicators, including smartphone apps and wearable devices that measure cough frequency, can be considered.⁶⁵ Furthermore, the risk of detection bias should be lowered by blinding the evaluator. Lastly, all of the included studies were published in China, which limits the generalizability of our findings. Therefore, research on acupuncture-related therapy for chronic cough should be conducted in other countries such as South Korea and Japan, which actively apply acupuncture in clinical settings.

4.5. Conclusions

In conclusion, acupuncture-related therapy, when provided in addition to or as an alternative to conventional treatment, may significantly improve the cough severity and cough-related quality of life of patients with chronic cough of various causes with complex pathophysiologies without affecting the incidence of adverse events. The most frequently used acupuncture points were BL13, GV14, CV17, and EX-B1. Further well-designed RCTs for evaluating the efficacy of acupuncture-related therapy should be conducted in the future with the use of objective assessment tools.

CRediT authorship contribution statement

Boram Lee: Conceptualization, Methodology, Formal analysis, Writing – original draft, Writing – review & editing. Chan-Young Kwon: Methodology, Formal analysis, Writing – original draft, Writing – review & editing. Yoon Kyoung Jeong: Formal analysis, Writing – review & editing. Na-Yeon Ha: Writing – original draft, Writing – review & editing. Kwan-Il Kim: Writing – review & editing. Beom-Joon Lee: Writing – review & editing. Jun-Hwan Lee: Writing – review & editing, Supervision, Funding acquisition.

Declaration of competing interest

The authors have no conflict of interest to declare involving this study.