Effect of High/Low Dose N-Acetylcysteine on Chronic Obstructive Pulmonary Disease: A Systematic Review and Meta-analysis

Abstract

Background: Chronic obstructive pulmonary disease (COPD) is a leading cause of mortality and morbidity worldwide, charcterised by persistent airflow limitation, mucus hypersecretion, oxidative stress and airway inflammation. N-acetylcysteine (NAC) have anti-oxidant and anti-inflammatory properties, which have been shown an uncertain benefit in COPD patients. Method: Systematic searches were conducted in Cochrane, Medline and Embase electronic databases. A meta-analysis was performed to evaluate the different effect between high and low-dose NAC treatment on COPD exacerbation. Results: This review yielded 11 studies. The methodological quality of included studies were scored using the Jadad score, with a scale of 1 to 5 (score of 5 being the highest). Data showed high-dose NAC can reduce both the total number of exacerbations (RR = 0.59, 0.47 to 0.74, 95%CI, p < 0.001) and the proportion of patients with at least one exacerbation (RR = 0.76, 0.59 to 0.98, 95%CI, p = 0.03). In the low-dose group, subgroup with jadad ≤ 3 showed a significant decrease (RR = 0.69, 0.61 to 0.77, 95%CI, p < 0.001) in the proportion of patients with exacerbation, the other subgroup with Jadad score > 3 showed no significant decrease (RR = 0.98, 0.90 to 1.06, 95%CI, p = 0.59). And low-dose NAC showed no benifit in the total number of exacerbations (RR = 0.97, 0.68 to 1.37, 95%CI, p = 0.85). Neither high nor low-dose NAC treatment showed benifit in forced expiratory volume in one second(FEV₁)(WMD = 1.08, −9.97 to 12.13, 95%CI, p = 0.85). Conclusion: Long-term high-dose NAC treatment may lead to a lower rate of exacerbations. But the effect of low-dose NAC treatment remains uncertain. Further researches are needed to confirm this outcome and to clarify its mechanisms.

Keywords :

• exacerbation
• inflammation
• lung function
• oxidant stress

Abbreviations
COPD	=	Chronic obstructive pulmonary disease
FEV1	=	Forced expiratory volume in 1 second
WMD	=	Weighted mean difference
95%CI	=	95% confidence interval
RR	=	Rate Ratio
NAC	=	N-acetylcysteine.

Introduction

Chronic obstructive pulmonary disease (COPD) is characterised by persistent airflow limitation, mucus hypersecretion, oxidative stress and airway inflammation. It is a leading cause of mortality and morbidity worldwide, with rapid decline in pulmonary function, compromised quality of life and reduced life span. Although available treatment strategies for COPD are beneficial, they are far from satisfactory, and thus alternative treatments are urgently needed.

Oxidant/antioxidant imbalance and airway inflammation are important processes in the pathogenesis of COPD (1). Increased expression of markers of oxidative and nitrosative stress, such as 4-hydroxynonenal, ­thiobarbituric acid reactive substances and nitrotyrosine (2, 3) can aggravate the inflammation of the small airway, which is related to lung function (4), exacerbation and quality of life. Reducing oxidative stress and airway inflammation is supposed to be effective in the ­treatment of COPD.

N-acetylcysteine (NAC) is a derivative of amino acid L-cysteine, which is most commonly used as an ­apophlegmatisant, with direct/indirect anti-oxidant and anti-inflammatory properties. It has been confirmed to be beneficial in some pulmonary diseases such as ­pulmonary oxygen toxicity, idiopathic pulmonary fibrosis (5) and cystic fibrosis (6).

Recent studies (7, 8) showed that NAC helps to prevent COPD exacerbation and improve pulmonary function. However, results of these studies were conflicting, possibly due to different NAC doses or concomitantly used drugs. Studies (8, 9) with 600 mg NAC twice daily reported that NAC could reduce COPD exacerbations versus placebo.

Conversely, other studies (10, 11) with less than or equal to 600mg NAC daily failed to demonstrate the beneficial effect on exacerbation frequency. Meanwhile, some studies (12, 13) pointed out that the anti-oxidant effect of NAC was dose-dependent, which only exerts its anti-oxidant effect at high doses. Thus, we hypothesized that the effect of NAC in preventing COPD exacerbations only exists at high doses, which has never been specifically investigated in previous systematic reviews (14, 15).

This meta-analysis aimed to evaluate via a subgroup analysis (high dose and low dose NAC groups) the effect of long-term high/low dose NAC treatment in COPD exacerbations and pulmonary function.

Materials and methods

Search strategy and inclusion criteria

We conducted a comprehensive literature search for English language articles examining the effect of long-term high/low dose NAC treatment in stable patients with COPD. High-dose NAC treatment was defined as the use of more than 600 mg daily, and low dose NAC treatment was defined as the use of less than or equal to 600 mg daily. These definitions were chosen based on published studies (9, 16).

We searched the Cochrane database, Medline and Embase for randomized, controlled trials published up to August 1, 2013. We searched terms for disease (chronic obstructive pulmonary disease, chronic bronchitis, pulmonary emphysema, or COPD), and terms for drugs (acetylcysteine). We examined the bibliographies and reference lists of included articles and relevant reviews to identify additional relevant studies.

Studies were included if they met the following ­criteria: 1) Design: randomized controlled clinical ­trials, 2) intervention: orally-administered NAC plus standard therapy versus placebo plus standard therapy in subjects with stable COPD, more than 3 months; 3) outcome: total number of COPD exacerbations, proportion of patients with at least one COPD exacerbations, pulmonary function and adverse events. We excluded studies that were published as protocol, or written in non-­English language.

Types of outcome measures

Primary outcome

• The total number of exacerbations as a function of person-seasons.

Secondary outcomes

1. The number of patients with at least one exacerbation.

2. Spirometry: Forced expiratory volume in 1 second (FEV₁).

3. Adverse events: Gastrointestinal disorders (including diarrhea, reflux esophagitis, gastric complaint).

Data collection and analysis

Selection of studies

Two reviewers independently checked the relevant studies from the literature search. Trials were selected from identified studies, based on previously agreed inclusion criteria. Study characteristics and outcomes were collected by two reviewers. Two studies (17, 18) were included in this review, but we failed to get the full articles despite contacting the authors. The data of these studies were extracted from a published meta-analysis (14). Any difference in opinion about eligibility was resolved by consensus.

Assessment of risk of bias in included studies

Methodological quality of included studies was assessed independently by two reviewers. The Cochrane Allocation Concealment Scale and Jadad scores Jadad et al. (19) were used, with a scale of 1 to 5 (score of 5 being the highest). This scale is used to assess randomisation, double blinding and withdrawals/dropouts.

Data extraction and management

Data from included studies was extracted and rechecked independently by two reviewers. Of studies selected, a variety of specific trial characteristics were recorded including study design, Acetylcysteine dose/duration, age, smoking history, inhaled corticosteroids (ICS) treatment, COPD severity and Jadad score. The primary endpoint was the total number of COPD exacerbations as a function of person-seasons (one season include 12 weeks). Secondary endpoints included the ­proportion of patients with at least one COPD exacerbations, pulmonary function and adverse events.

One large trial (11) reported the exacerbation rate instead of the proportion of patients with exacerbation. The count data was estimated by halving the annualized exacerbation rate and multiplying the number of patients allocated to either placebo or NAC group by this 6-month exacerbation rate. Another study (20) reported number of patients with more than 2 exacerbations, so the total exacerbation number was estimated by multiplying the number of patients with more than 2 exacerbations by 2, and plus the number of patients with only 1 exacerbation.

For the forced expiratory volume in 1 second (FEV₁), the standard deviation of change from baseline was inferred from the standard deviation of baseline and final score (7). All formulations were converted based on the recommendations from the Cochrane handbook for systematic reviews of interventions. For dichotomous variables, withdrawals were taken as negative outcome. For continuous variables, withdrawals were not taken into calculation.

Measures of treatment effect

Trials were combined using the Review Manager. For dichotomous variables, a Mantel–Haenszel Rate Ratio with 95% confidence intervals was calculated. If significant heterogeneity was detected (p < 0.1), causes of heterogeneity were subsequently explored via subgroup analyses, otherwise a random effect model was selected. For continuous variables, we used a fixed effect weighted mean difference (WMD) for measurements and 95% confidence interval (95% CI) was calculated. Sensitivity analysis was made to assess the impact of excluding studies based on methodological quality. Statistical analysis was performed using ­RevMan software version 5.1.

Results

Study characteristics

We identified 11 trials (comprising 1281 patients who received NAC and 1306 patients who received placebo) that met our inclusion criteria, of which 2 used high dose NAC (600 mg twice daily) (8, 9), and 9 used low dose NAC (7, 10, 11, 17, 18, 20–23). A summary of the study selection process is shown in Figure 1.

Figure 1.  Flow chart of the literature search and selection (26–36).

All the studies were published between 1976 and 2013, and the relevant demographic data is summarized in Table 1. The treatment periods for the studies ranged from 4 months to 3 years. Ten studies were described as randomized, double-blind, placebo controlled studies, but methods of randomization or masking were not adequately described. Jadad score ranged from 2 to 5 points. One study was described as a randomized, single-blind, placebo controlled study, with a Jadad score of 1 point.

Table 1.  Major study characteristics from the 11 studies included in the meta-analysis

Study+year	TSE (2013) (9)	Schermer et al. 2009 (10)	Bachh et al.(2007) (20)	Decramer et al. (2005) (11)	Pela et al. (1999) (7)	Hansen et al. (1994) (8)	Rasmussen et al. (1988) (21)	McGavin et al. (1985) (22)	Boman et al. (1983) (17)	Multicenter Study (1980) (18)	Grassi et al. (1976) (23)
Number of patients (T:C)	58:62	96:96	50:50	256:267	85:84	75:78	59:57	85:96	127:132	371:373	40:40
Mean age (T:C)	71.0:70.8	59.2:59.6	62.6:60.1	62:62	66:66	51.1:51.7	58.8:58.9	64.3:62.6	53.2:51.0		60:62
Dose of NAC	600 mg twice daily	600 mg once daily	600 mg once daily	600 mg once daily	600 mg once daily	600 mg twice daily	300 mg twice daily	200 mg 3 times/day	200 mg twice daily	300 mg twice daily	300 mg twice daily, 3 times a week
Duration of study	1 year	6 months	4 months, followed for 1 year	3 years	6 months	22 weeks	6 months	5 month	6 months	6 months	6 months
COPD severity	Most of themModerate to severe	64% Moderate to severe	Moderate to severe	Moderate to severe	Moderate to severe	Mild or chronic bronchitis	chronic bronchitis	Severe to very severe	—	—	chronic bronchitis
Current smoking (T:C)	15:13	54:49	­ —	120:109	29:19	71:70	—	24:25	—	—	—
ICS treatment (T:C)	43:52	0:0	10:8	182:183	32:35	—	—	19:27	—	—	—
withdraw (T:C)	6:6	44:40	0:0	70:99	2:4	16:8	15:10	13:20	29:27	117:132	6:9
Jadad score	4	4	1	4	2	4	4	4	—	—	3

T:C: treatment group:control group; ICS: Inhaled corticosteroids.

Study results

Effect of NAC therapy on exacerbation

The primary endpoint was the total number of exacerbations as a function of person-seasons. High dose NAC treatment involving 2 studies and 273 patients showed a significant decrease (RR = 0.59, 0.47 to 0.74, 95%CI, ­Figure 2, p < 0.001), and low-dose NAC treatment involving 9 studies and 2314 patients showed no ­significant decrease (RR = 0.97, 0.68 to 1.37, 95%CI, ­Figure 2, p = 0.85). Overall, there was no statistical difference between NAC and ­placebo groups (RR = 0.86, 0.65 to 1.16, 95%CI, Figure 2, p = 0.33).

Figure 2.  Forest plot from meta-analysis, including 8 included RCTs, assessing relative risk (RR) of COPD exacerbations as a function of person-seasons.

Secondary endpoint was the number of patients with at least one exacerbation. High-dose NAC treatment showed a significant decrease in the number of exacerbation (RR = 0.76, 0.59 to 0.98, 95%CI, Figure 3, p = 0.03). In the analysis with low-dose NAC studies, statistical heterogeneity was detected, so subgroup was restricted to studies with Jadad score > 3. The subgroup (7, 17, 18, 20, 23) with jadad ≤ 3 showed a significant decrease in the risk of exacerbation (RR = 0.69, 0.61 to 0.77, 95%CI, Figure 3, p < 0.001). Subgroup analysis (10, 11, 21, 22) restricting to studies with Jadad score > 3 showed no significant difference in the risk of exacerbation (RR = 0.98, 0.90 to 1.06, 95%CI, Figure 3, p = 0.59). Overall, there was a ­statistical difference between high/low dose NAC treatment and placebo treatment (RR = 0.81, 0.76 to 0.87, 95%CI, Figure 3, p < 0.01).

Figure 3.  Forest plot from meta-analysis, including all 11 included RCTs, assessing relative risk (RR) of the number of patients with at least one exacerbation.

Effect of NAC on spirometry and adverse effect

A spirometry evaluation of patients between high/low dose NAC treatment and placebo treatment showed no significant increase in FEV1 (MD = 1.08, −9.97 to 12.13, 95%CI, Figure 4, p = 0.85). Gastrointestinal disorders including diarrhea, reflux esophagitis, and gastric complaint were reported in some studies, but NAC did not significantly increase the risk of such adverse reactions (RR = 1.30, 0.71 to 2.39, 95%CI, Figure 5, p = 0.39).

Figure 4.  Forest plot from meta-analysis, including 5 included RCTs, assessing mean difference (MD) of the forced expiratory volume in one second.

Figure 5.  Forest plot from meta-analysis, including 4 included RCTs, assessing relative risk (RR) of the patients with gastrointestinal disorders.

Discussion

NAC is an effective mucolytic agent with direct/indirect anti-oxidative and anti-inflammation properties that can reduce oxidative stress and inflammation in the lung tissue by replenishing glutathione levels.

Some clinical studies (9, 22, 23) have suggested a protective effect of NAC on the risk of exacerbations in COPD patients. However, conflicting results were reported, especially in low dose NAC treatment. Published meta-analyses (14, 24) indicated that NAC could reduce COPD exacerbations versus placebo, in which they pooled the high-dose and low-dose NAC treatment together and used the proportion of COPD patients with at least one exacerbation rather than the actual number of exacerbations as an endpoint.

This review is distinguished from others by making a subgroup analysis of the high-dose and low-dose NAC treatment. Furthermore, most of the included studies included patients with moderate to severe COPD, and they are more likely to have more than one exacerbation during the long-term treatment course. So the number of exacerbations was chosen as the primary end point, and the number of patients with at least one exacerbation was the secondary outcome.

The analysis involving high-dose NAC showed a statistically significant reduction in both the number of exacerbations and the proportion of patients with at least one exacerbation. In contrast, the subgroup ­analysis restricting to low-dose NAC showed a negative outcome in the number of exacerbations.

According to conclusions of some studies (12, 13), this can be explained by the fact that the anti-oxidant effect of NAC is dose-dependent. It is well known that NAC exerts its mucolytic effect at low dose while the anti-oxidant effect appears only at higher doses (1200–1800 mg daily). Another review (16) pointed out that NAC was able to inhibit oxidative stress at both high and low concentrations. The anti-inflammatory property of NAC is highly dose-dependent and can only be seen at high concentration as demonstrated in an in vitro experimental study. Another study (25) demonstrates that high dose NAC (1200mg daily) treatment shows no additional benefit on reducing oxidative stress, but it improves the symptom of COPD that is thought to be associated with its immunomodulating property.

Meanwhile, 9 included studies with low dose NAC reported the total number of patients with exacerbations as an important outcome of the treatment, but the results remain conflicting. Significant heterogeneity (p < 0.0001); I≤ = 77%) among component studies was identified and was addressed by using subgroup analysis. A Jadad score of 3 was employed as a boundary to define subgroups. The heterogeneity of each group became insignificant (I≤ = 0%,14%). The subgroup analysis (7, 17, 18, 20, 23) restricting to studies with Jadad ≤ 3 showed a significant decrease in the number of patients with exacerbations, but the other subgroup analysis (10, 11, 21, 22) restricting to ­studies with jadad score > 3 showed no significant decrease. Such results are inconsistent with most of the formerly published studies (7, 17, 18, 20) and reviews (14, 24). This maybe attributable to the fact that these studies were published fifteen years ago, and are not properly designed. These reviews did not separate the high-dose from low-dose NAC treatment, and used a random effect model when the heterogeneity was significant.

Our analysis has several limitations. First, publication bias was seen in this review, which is attributable to the exclusion of non-English language studies. Second, we failed to get full text articles of 2 studies published in the years 1983 and 1980. However, we tried to abstract data from another meta-analysis. Third, most of the included studies used inhaled corticosteroids as a standard treatment (7, 9, 11, 20, 22), and the heterogeneity was significant. Furthermore, in another 4 studies, the concomitant drugs were not explicitly described, so subgroup analysis restricting to studies without ICS was not made to confirm if ICS treatment would attenuate the effect of NAC treatment that has been reported in other studies (11, 14).

Although prior meta-analyses have evaluated the use of NAC in general for prevention of COPD exacerbations, none have specifically separated the high dose from low-dose NAC treatment. This review suggests substantial benefit from high-dose NAC treatment for COPD patients, but it is worth noting that only 2 studies were included in this subgroup analysis. More investigations with high/low dose NAC are needed to confirm this result.

Declaration of Interest Statement

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.