- Email: [email protected]
Pelargonium sidoides for acute bronchitis: A systematic review and meta-analysis
ARTICLE IN PRESS
Phytomedicine 15 (2008) 378–385 www.elsevier.de/phymed
Pelargonium sidoides for acute bronchitis: A systematic review and meta-analysis Taofikat B. Agbabiaka, Ruoling Guo, Edzard Ernst Complementary Medicine, Peninsula Medical School, Universities of Exeter and Plymouth, 25 Victoria Park Road, Exeter EX2 4NT, UK
Abstract Objective: To critically assess the efficacy of Pelargonium sidoides for treating acute bronchitis. Data sources: Systematic literature searches were performed in 5 electronic databases: (Medline (1950 – July 2007), Amed (1985 – July 2007), Embase (1974 – July 2007), CINAHL (1982 – July 2007), and The Cochrane Library (Issue 3, 2007) without language restrictions. Reference lists of retrieved articles were searched, and manufacturers contacted for published and unpublished materials. Review methods: Study selection was done according to predefined criteria. All randomized clinical trials (RCTs) testing P. sidoides extracts (mono preparations) against placebo or standard treatment in patients with acute bronchitis and assessing clinically relevant outcomes were included. Two reviewers independently selected studies, extracted and validated relevant data. Methodological quality was evaluated using the Jadad score. Meta-analysis was performed using a fixed-effect model for continuous data, reported as weighted mean difference with 95% confidence intervals. Results: Six RCTs met the inclusion criteria, of which 4 were suitable for statistical pooling. Methodological quality of most trials was good. One study compared an extract of P. sidoides, EPss7630, against conventional non-antibiotic treatment (acetylcysteine); the other five studies tested EPss7630 against placebo. All RCTs reported findings suggesting the effectiveness of P. sidoides in treating acute bronchitis. Meta-analysis of the four placebo-controlled RCTs suggested that EPss7630 significantly reduced bronchitis symptom scores in patients with acute bronchitis by day 7. No serious adverse events were reported. Conclusion: There is encouraging evidence from currently available data that P. sidoides is effective compared to placebo for patients with acute bronchitis. r 2007 Elsevier GmbH. All rights reserved. Keywords: Acute bronchitis; Bronchitis Symptom Score; EPss7630; Pelargonium sidoides; Umckaloabo
Introduction Acute bronchitis refers to the acute or sub-acute onset of productive cough with no history of chronic pulmonary disease and without evidence of pneumonia Corresponding author. Tel.: +44 1392 424 942; fax: +44 1392 427 562. E-mail address: taofi[email protected] (T.B. Agbabiaka).
0944-7113/$ - see front matter r 2007 Elsevier GmbH. All rights reserved. doi:10.1016/j.phymed.2007.11.023
or sinusitis (Gonzales and Sande, 1995). It is one of the most common diagnoses in primary care (Knutson and Braun, 2002), affecting 44 adults in every 1000 (416 years old) each year in the UK, with 82% of the episodes occurring in autumn or winter (Macfarlane et al., 2001). Hueston and Mainous (1998) reported that viruses are the most common cause of this condition. However, non-viral agents (bacteria and allergens) play a role in its etiology in less than 10% of cases (Niroumand and
ARTICLE IN PRESS T.B. Agbabiaka et al. / Phytomedicine 15 (2008) 378–385
Grossman, 1998). Surprisingly, prescription of antibiotics for acute bronchitis is still wide-spread even though their efficacy is limited to bacterial infections (Bent et al., 1999; Hueston and Mainous, 1998; Orr et al., 1993). Perhaps due to the limited success of antibiotics, substantial cost, associated adverse effects and antibiotic resistance as a result of overuse and misuse (Bent et al., 1999) more attention is now focused on alternative treatments for acute bronchitis and other upper respiratory tract infections; hence the search for effective herbal options (Kligler et al., 2006). Pelargonium sidoides is an herbaceous perennial that is popular in South African traditional medicine for the treatment of infectious respiratory diseases (Watt and Breyer-Brandwyk, 1962). Pelargonium containing phytopharmaceuticals are currently widely used in Europe to treat respiratory tract infections (Matthys et al., 2003). The aim of this systematic review was to assess available evidence from rigorous clinical trials on the efficacy and/or effectiveness of Pelargonium sidoides for the treatment of acute bronchitis.
379
Data extraction and assessment of methodological quality All titles and abstracts identified by the literature search were initially assessed against the inclusion criteria to decide if the full text article should be obtained (TBA). The retrieved full text papers were independently assessed by two reviewers (TBA and RG) to decide on inclusion. Disagreements were resolved by discussion between the two reviewers and by seeking the opinion of a third reviewer (EE) when necessary. Data concerning the details of study design, quality of the study, participants, intervention, outcomes and adverse events were independently extracted by two reviewers (TBA and RG) using a pre-designed data extraction sheet. The methodological quality of all included studies was assessed independently by two reviewers (TBJ and RG) using the 5-point Jadad score (Jadad et al., 1996). We also assessed allocation concealment and the use of intention to treat analysis. Discrepancies were resolved by discussion between the two authors.
Data analysis
Methods Search strategy Systematic literature searches were conducted in 5 electronic databases: (Medline (1950 – July 2007), Amed (1985 – July 2007), Embase (1974 – July 2007), CINAHL (1982 – July 2007), and The Cochrane Library (Issue 3, 2007) without language restrictions. The search strategy was structured as ‘‘search terms for condition’’ AND ‘‘search terms for intervention’’. Both MeSH terms and text terms were searched for the condition as well as the intervention. The MeSH term for the condition was acute bronchitis and for the intervention were Pelargonium OR sidoides OR Umckaloabo OR EPss7630. We also searched the reference lists of all retrieved articles. We contacted 4 manufacturers of commercial preparations of Pelargonium sidoides for further information, particularly unpublished materials.
Included trials were categorised by the type of control interventions and the following comparisons were made: (a) herbal medicine versus placebo and (b) herbal medicine versus conventional therapy. Meta-analysis was carried out using Review Manager Software (version 4.2.10 2007, Cochrane Collaboration and Updated Software). We quantitatively combined the results of four studies that were suitable for statistical pooling. We calculated weighted mean differences (WMD) and 95% confidence intervals (CI) for continuous data using fixed-effect model. The chi-square test for homogeneity was performed to determine whether the distribution of results was compatible with the assumption that differences between trials were due to chance variation alone. To evaluate potential publication bias we constructed a funnel plot for the primary outcome of change in Bronchitis Severity Scores (BSS) score, using effect size as a measure, and visually examined it for asymmetry.
Inclusion and exclusion criteria
Results Only randomized clinical trials (RCTs) testing monopreparations of P. sidoides as sole or adjunctive treatment administered orally against a control intervention (placebo or conventional therapy), in patients of any age or sex with acute bronchitis and assessing clinically relevant outcomes were included. Trials including patients with pre-existing chronic bronchitis or other infectious diseases were excluded.
The literature searches identified 295 potentially relevant titles and abstracts. We also identified two unpublished trials (Kieser, 2007a, b). Initial screening of the titles and abstracts identified 16 relevant references for which full texts were retrieved for further evaluation. Ten references were subsequently excluded for the following reasons: eight references as they were
ARTICLE IN PRESS 380
T.B. Agbabiaka et al. / Phytomedicine 15 (2008) 378–385
non-controlled trials or reviews (Haidvogl and Heger, 2007; Matthys and Heger, 2007a; Matthys et al., 2007; Schulz, 2006; Kolodziej and Kayser, 1998; Haidvogl et al., 1996; Konig, 1995; Kolodziej et al., 1995), 1 RCT was excluded because the condition treated was not acute bronchitis (Bereznoy et al., 2003). A duplicate paper in German was also excluded (Golovatiouk and Chuchalin, 2002). Six RCTs (Matthys et al., 2003; Kieser, 2007a, b; Blochin et al., 1999; Chuchalin et al., 2005; Matthys and Heger, 2007b) met the inclusion criteria and were included in the review. Four trials (Matthys et al., 2003; Kieser, 2007a; Chuchalin et al., 2005; Matthys and Heger, 2007b) with total number of subjects n ¼ 1012 were suitable for statistical pooling and included in a meta-analysis. Agreement between reviewers on inclusion of RCTs was 100%. Fig. 1 is a flow chart of the trial selection process. A summary of the main characteristics of these studies is presented in Table 1. The studies were published between 1999 and 2007, originating from two countries (Russia, n ¼ 3; Ukraine, n ¼ 3) and in two languages (English, n ¼ 5; German, n ¼ 1). A total of 1681 subjects were randomised and 1647 analysed by the original investigators. The sample sizes ranged from 60 to 476, all patients were diagnosed with acute bronchitis,
Total number of citations identified: Electronic searches n= 295 Manufacturers n=2 Citations excluded after removal of duplicates, screening of titles and abstracts: n= 281 Full manuscript retrieved for detailed evaluation: From electronic search From manufacturer From reference list
n= 16 n= 13 n= 2 n= 1
Excluded with reasons: n=9 Not RCT n= 8 Condition treated not acute bronchitis n=1
Potential relevant articles identified
with BSS of 5 points or more, and duration of complaint less than 48 h prior to enrolment. Four trials were carried out in adult patients aged over 18 years (Matthys et al., 2003; Kieser, 2007a; Chuchalin et al., 2005; Matthys and Heger, 2007b), two RCTs involved children aged between 6 and 18 years (Kieser, 2007b; Blochin et al., 1999). Five RCTs compared EPss7630 (Dr. Willmar Schwabe Pharmaceuticals, Germany) against placebo (Matthys et al., 2003; Kieser, 2007a, b; Chuchalin et al., 2005; Matthys and Heger, 2007b). Another RCT tested EPss7630 as sole treatment against a mucolytic commonly used for this condition (Blochin et al., 1999). EPss7630 was administered as solution (drops) in all trials except two (Kieser, 2007a, b), where tablets were administered. EPss7630 is an ethanolic extract of the roots of Pelargonium sidoides which is marketed in Germany, in the Commonwealth of Independent States (CIS), in the Baltic States, and in Mexico, for the treatment of ear, nose and throat (ENT) and respiratory infections (Matthys et al., 2003).The formulation of EPss7630 solution and its botanical origin are identical to the finished product Umckaloabos registered by ISO Pharmaceuticals, Ettlingen, Germany. Methodological quality of the RCTs reviewed was generally good; Jadad score was 5 points (maximum 5) for three trials (Matthys et al., 2003; Chuchalin et al., 2005; Matthys and Heger, 2007b), the unpublished trials were assessed but not scored for methodological quality (Table 1). Five of the RCTs (Matthys et al., 2003; Kieser, 2007a, b; Chuchalin et al., 2005; Matthys and Heger, 2007b) described methods of randomisation and double blinding, allocation concealment, follow up and intention to treat analysis. Three published (Matthys et al., 2003; Chuchalin et al., 2005; Matthys and Heger, 2007b) and the two unpublished studies (Kieser, 2007a, b) reported details of dropouts and withdrawals. The 6 trials were categorised into two groups according to the type of control interventions. Detailed results are summarised in Table 1 and described below.
n= 7
Herbal medicine versus conventional treatment Articles excluded with reasons: Duplicate of included paper n= 1
RCTs included in the systematic review
n=6
RCTs excluded from meta-analysis with reasons: Not placebo controlled n=1 Different population n= 1
RCTs included in the meta-analysis
n=4
Fig. 1. Flow chart of the study selection process.
Only one RCT (Blochin et al., 1999) compared EPss7630 with a conventional therapy (Acetylcysteine). In this trial, 60 children aged between 6 and 12 yrs were randomised into two groups to receive either EPss7630 liquid formula (20 drops every hour up to 12 times on day 1 and 2; 20 drops 3/daily from day 3–7) or Acetylcystein granules (2 mg 200 mg daily for 7 days). Outcome measures were changes in typical symptoms of bronchitis (cough, sputum, rattling noise, chest pain and dyspnoea). Bronchitis specific symptoms were reduced more effectively and faster with EPss7630 than acetylcysteine at the end of 7 days treatment, intergroup
Table 1.
Characteristics of included randomised clinical trials of Pelargonium sidoides for the treatment of acute bronchitis
First author
Participants Design and quality (JS)
R/A (T, C)
Intervention and daily dose Control group
Treatment duration/follow-up period (days)
Main outcome measure
Main results, mean difference (95% CI)
Adverse events (AEs)
Herbal medicine versus conventional medicine Blochin et al. Open parallel 60/60 (30,30) (1999) (2)
6–12
EPss7630 solution, 20 drops/ h up to 12 for days 1& 2 and 20 drops 3/d on days 3–7
Acetylsteine granules (200 mg 2/d)
7
Bronchitis typical symptoms
7.0 (0.02–0.02)
Not reported
Herbal medicine versus placebo Matthys DB parallel 476/468 et al. (2003) (5) (237,239)
X18
EPss7630 liquid extract, 30 drops 3/d EPss7630 liquid extract, 30 drops 3/d EPss7630 solution, 30 drops 3/d
placebo
7
2.7 (2.1–3.3)
No serious AEs (test, n ¼ 20; control, n ¼ 16)
Placebo
7
2.3 (1.3–3.3)
Placebo
7
Change of total score of BSS Change of total score of BSS Change of total score of BSS
EPss7630 tablet
Placebo
7
Change of total score of BSS
4.3 (0.4–1.2)
Mild to moderate AEs (test, n ¼ 15; control, n ¼ 10) No serious AEs; blood and lymphatic disorders and laboratory abnormalities (test, n ¼ 37; control, n ¼ 27); No intergroup differences No serious AEs in any group; minor to moderate AEs (gastrointestinal and nervous system disorders, test: LD ¼ 22, MD ¼ 25 and HD ¼ 31; control, n ¼ 14) No serious AEs in any group; minor to moderate AEs (gastrointestinal disorders infections and infestations, test: LD ¼ 23, MD ¼ 20 and HD ¼ 19; control, n ¼ 18)
Chuchalin et al. (2005)
DB parallel (5)
124/117 (64, 60)
X18
Matthys and Heger (2007b)
DB parallel (5)
217/198 (108,109)
18–66
Kieser (2007a)*
DB
405/405 (303,102)
418
Parallel#
Kieser (2007b)*
DB Parallel#
LD – 3 10 mg/d MD – 3 20 mg/d HD – 3 30 mg/d
399/399 (298,101)
6–18
EPss7630 tablet LD – 3 10 mg/d MD – 3 0 mg/d HD – 3 30 mg/d
Placebo
7
Change of total score of BSS
2.3(1.6–3.1)
6.1 (0.8–1.6) 6.3 (0.9–1.6) for the 30, 60 and 90 mg/d doses respectively 3.6 (0.4–1.3) 4.4 (0.6–1.5) 5.0 (0.6–1.3) for the 30, 60 and 90 mg/d doses respectively
JS ¼ Jadad Score (maximum of five points); DB ¼ double blind; x/d ¼ times per day; AEs ¼ adverse events; R ¼ number of patients randomised; A ¼ number of patients analysed; T ¼ number of participants randomised to treatment group; C ¼ number of participants randomised to control group (given in parenthesis); BSS ¼ Bronchitis Symptom Score; LD ¼ low dose; MD ¼ medium dose; HD ¼ high dose; # ¼ unpublished trial, not scored for methodological quality, * ¼ unpublished RCT.
ARTICLE IN PRESS
Test group
T.B. Agbabiaka et al. / Phytomedicine 15 (2008) 378–385
Age range (yrs)
381
ARTICLE IN PRESS 382
T.B. Agbabiaka et al. / Phytomedicine 15 (2008) 378–385
difference was however not significant (p ¼ 0.285). The authors reported no information on drop outs or withdrawals.
Kieser (2007a) due to the similarity of the medium dose to the doses administered in the other studies (i.e. a 20mg-tablet is equivalent to 30 drops). A meta-analysis of the four trials (n ¼ 1012) showed a significant decrease of BSS score compared to placebo (WMD 2.80 points, 95% CI 2.44–3.15). The chi-square test for homogeneity signifies heterogeneity among the four trials (p ¼ 0.09; I2 ¼ 54.4%), which indicates that the observed differences in results between trials could be caused by factors other than chance (Fig. 2). However, the small number of trials n ¼ 4 was inadequate to produce a meaningful funnel plot. A subgroup analysis was conducted to test the robustness of the results of the overall analysis and also test for publication bias. We excluded the unpublished trial (Kieser, 2007a) from the analysis, retaining only the three published trials with similar methodologic features (Matthys et al., 2003; Chuchalin et al., 2005; Matthys and Heger, 2007b). These three RCTs used the same preparation-EPss7630 (a liquid herbal drug preparation from the roots of Pelargonium sidoides (1:8–10), extraction solvent: ethanol 11% (w/w) produced by Dr. Willmar Schwabe Pharmaceuticals, Germany) at the same dose of 30 drops thrice daily over a treatment period of 7 days. Meta-analysis of these data (Fig. 1) revealed a WMD of 2.48 points (CI, 2.05–2.92), still indicating a significant change in BSS score in patients treated with EPss7630 compared with placebo but homogeneity was absolute this time (p ¼ 0.67; I2 ¼ 0%). Some other secondary outcomes were also assessed in these RCTs. In one RCT (Matthys et al., 2003), the
Herbal medicine versus placebo A total of 5 RCTs tested EPss7630 against placebo, three (Matthys et al., 2003; Chuchalin et al., 2005; Matthys and Heger, 2007b) of which tested EPss7630 solution (30 drops 3/d) and two (Kieser, 2007a, b) tested EPss7630 tablets. According to the manufacturer’s data, these two dosage forms are equivalents. In both trials testing EPss7630 tablets (Kieser, 2007a, b), patients were randomised into four groups to receive EPss7630 tablets: low dose ¼ 1 10 mg, 3/d; medium dose ¼ 1 20 mg, 3/d and high dose ¼ 1 30 mg, 3/d respectively, or placebo for 7 days. One of the RCTs was carried out amongst children aged between 6 and 18 years (Kieser, 2007b). Outcome measures were changes in the bronchitis symptom score (BSS) between day 0 and 7, and changes in individual components of BSS. All five trials (Matthys et al., 2003; Kieser, 2007a, b; Chuchalin et al., 2005; Matthys and Heger, 2007b) reported effects in favour of EPss7630 compared to placebo at the end of 7 days treatment. The four RCTs carried out in adults (Matthys et al., 2003; Kieser, 2007a; Chuchalin et al., 2005; Matthys and Heger, 2007b) were considered suitable to be combined in a meta-analysis, the trial in children (Kieser, 2007b) was excluded due to the difference in population. We included in the meta-analysis data from the RCT by Review: Pelargonium sidoides for the treatment of acute bronchitis Comparison: 01 EPs7630 vs. placebo Outcome: 01 BSS score
Study
or sub-category
N
01 All included studies Matthys 2003 Chuchalin 2005 Kiesser 2007 Matthys 2007
233 64 101 108
Treatment Mean (SD)
N
5.90(2.90) 7.20(3.10) 6.10(2.10) 7.60(2.20)
235 60 102 109
Control Mean (SD)
WMD (fixed) 95% CI
3.20(4.10) 4.90(2.70) 2.70(2.30) 5.30(3.20)
506 506 Total (95% CI) Test for heterogeneity: Chi² = 6.58, df = 3 (P = 0.09), I² = 54.4% Test for overall effect: Z = 15.48 (P < 0.00001) 02 Published studies only Matthys 2003 Chuchalin 2005 Matthys 2007
233 64
5.90(2.90) 7.20(3.10)
235 60
3.20(4.10) 4.90(2.70)
108
7.60(2.20)
109
5.30(3.20)
Weight %
WMD (fixed) 95% CI
30.32 12.01 34.16 23.51
2.70 [2.06, 3.34] 2.30 [1.28, 3.32] 3.40[2.79, 4.01] 2.30 [1.57, 3.03]
100.00
2.80 [2.44, 3.15]
46.04 18.24 35.71
Total (95% CI) 405 404 Test for heterogeneity: Chi² = 0.80, df = 2 (P = 0.67), I² = 0% Test for overall effect: Z = 11.16 (P < 0.00001)
100.00
-10
-5
Favours treatment
0
5
2.70 [2.06, 3.34] 2.30 [1.28, 3.32] 2.30 [1.57, 3.03]
2.48 [2.05, 2.92]
10
Favours control
Fig. 2. Effect of EPss7630 on BSS score in patients with acute bronchitis. Mean difference of the change from baseline given with their 95% confidence intervals. The vertical line represents no difference between EPss7630 and placebo.
ARTICLE IN PRESS T.B. Agbabiaka et al. / Phytomedicine 15 (2008) 378–385
patient’s health status showed a positive tendency in favour of EPss7630 at the end of the trial. Onset of treatment effect was recognised within 4 days in 53.6% of subjects in the treatment group compared to 36.2% in the control group. Duration of illness was also significantly shorter for subjects treated with EPss7630 compared to placebo (po0.001). In another RCT (Chuchalin et al., 2005) an overall improvement was observed in all individual components of BSS (cough, chest pain on coughing, sputum, rales/ronchi and dyspnoea) in the EPss7630 treatment group compared to the placebo group. Better quality of life was reported by 61.9% of patients treated with EPss7630 compared with 41.4% in placebo patients (p ¼ 0.012). The third RCT (Matthys and Heger, 2007b) reported individual symptoms and patients satisfaction over 7 days, complete remission in cough was observed in 51.9% for the treatment group compared to 11.9% in placebo; 68.3% improvement in sputum production among the controls and 40.0% in the placebo group. Complete recovery by day 7 was observed by the physician in 45.4% of patients taking active treatment compared to 6.4% of patients on placebo.
Adverse events Five RCTs (Matthys et al., 2003; Kieser, 2007a, b; Chuchalin et al., 2005; Matthys and Heger, 2007b) reported information on adverse events (AE). No serious AEs were reported in any of these RCTs. Mild to moderate AEs were observed in all RCTs, but there was no significant difference in the number of AEs reported between the two treatment groups (Matthys and Heger, 2007b). Some of the AEs reported include gastrointestinal disorders (Matthys et al., 2003; Kieser, 2007a, b), nervous system disorders (Matthys et al., 2003; Kieser, 2007a), ear and labyrinth disorders (Matthys et al., 2003) which was slightly higher with EPss7630 than with placebo, while gastrointestinal disorders were slightly more frequent in the placebo group. Infections and infestations were also reported in the RCT involving children (Kieser, 2007b).
Discussion This systematic review identified 6 studies, 4 of which are published (Matthys et al., 2003; Blochin et al., 1999; Chuchalin et al., 2005; Matthys and Heger, 2007b) and two unpublished (Kieser, 2007a, b), all of which suggest that preparations of P. sidoides are effective in the treatment of acute bronchitis. When compared to placebo, EPss7630 was found to significantly decrease BSS scores within 7 days of treatment (WMD 2.80 points, 95% CI 2.44–3.15) (Fig. 1) with improvement in individual components of BSS.
383
Only randomized, double-blind, placebo-controlled trials were included in this review. However, methodological quality varied among studies. Three trials scored a maximum of 5 points on the quality assessment (Matthys et al., 2003; Chuchalin et al., 2005; Matthys and Heger, 2007b), one trial had a low score of 2 points (Blochin et al., 1999). Five studies described randomization methods and allocation concealment with all analyses based on intention to treat (Matthys et al., 2003; Kieser, 2007a, b; Chuchalin et al., 2005; Matthys and Heger, 2007b). Three RCTs (Matthys et al., 2003; Chuchalin et al., 2005; Matthys and Heger, 2007b) were of similar design, two of them (Chuchalin et al., 2005; Matthys and Heger, 2007b) were conducted in the same country. One researcher participated in all three studies. Another two RCTs (Kieser, 2007a, b) were of the same design although each was conducted in different populations. Five of the RCTs were funded, at least in part, by the manufacturer of the product. Independent replication of the results would be desirable despite the difficulties encountered in getting funding from independent sources for research of commercial products. In view of the fact that the licence for EPss7630 was only acquired in 2005, we anticipate that independent replications, which are desirable for confirming these findings, will soon be forthcoming. We found substantial heterogeneity among the 4 trials included in the meta-analysis, as indicated by a chisquare test (w2 ¼ 6.58, p ¼ 0.09 and I2 ¼ 54.4%). To explore the possible reasons for this, we performed a sensitivity analysis. We felt that the heterogeneity could be attributed to the single unpublished trial by Kieser (2007a).When this trial was excluded from the analysis, effect size for EPss7630 was slightly decreased but remained superior to placebo (WMD of 2.48; 95% CI 2.05–2.92). We also observed no inconsistency among these trials (I2 ¼ 0%). Compared to N-acetylcysteine (NAC), a non-antibiotic treatment, Blochin et al. (1999) reported that EPss7630 showed a reduction in bronchitis-typical symptoms in both treatment groups but no significant intergroup differences (p ¼ 0.285). NAC is a precursor of glutathione, an important internal antioxidant which has been in clinical use for more than 30 years as a mucolytic drug to boost antioxidant levels and dissolve mucus in people suffering from respiratory ailments (Kupczyk and Kuna, 2002). However, the study by Blochin et al. (1999) was not blinded and, even though randomization was mentioned, it was not adequately described. There was also no description of dropouts and withdrawals, all this contributed to a low Jadad score (2 out of 5). The evidence from this study is therefore weak, particularly as it also had a small sample size and no allocation concealment, which may overestimate intervention benefits.
ARTICLE IN PRESS 384
T.B. Agbabiaka et al. / Phytomedicine 15 (2008) 378–385
The characteristic and therapeutically most important natural substances of EPss7630 are polymeric polyphenolic compounds. Simple phenolic compounds and coumarins are also mentioned in the literature (Kolodziej et al., 2003). The mechanism of action is ascribed to the ability of EPss7630 to antagonize bacterial adhesion and/or invasion to intact epithelia, thus protecting the upper respiratory tract from bacteria colonisation and infection (Conrad et al., 2007). Available safety data for EPss7630 indicate mild to moderate adverse events comprising gastrointestinal complaints which are likely due to the phenols contained in the liquid herbal drug preparation from Pelargonium roots, no serious risks or detrimental effects on quality of life was reported. The use of EPss7630 could also prove useful by minimising the use of antibiotics for this condition. This is particularly important when the over-prescription of antibiotics for acute bronchitis, and increased bacterial resistance is considered (Wise et al., 1998). Our findings may be subject to the general limitations of systematic review, pertaining to potential incompleteness of the reviewed evidence. Although we aimed to identify all RCTs on this topic, the distorting effects on systematic reviews arising from publication bias and location bias cannot be completely ruled out including the tendency for negative trials to remain unpublished (Easterbrook et al., 1991). We searched all available databases with a focus on American and European literature. Our search was not restricted in terms of language of publication. Therefore, we are confident that our search had located all relevant data on the subject. However, whether other unpublished trials were not located remains uncertain as contact to manufacturers produced only two relevant studies (Kieser, 2007a, b). The relatively small number of studies (n ¼ 6) available for this review may be seen as a further limitation, but we agree with Derry et al. (2006) that four trials and/or a minimum of 200 patients is sufficient to calculate statistical significance and make a conclusion on the level of evidence from the trials.
Conclusions Currently available data from 6 high quality randomised clinical trials suggests there is encouraging evidence that P. sidoides is effective compared to placebo for patients with acute bronchitis.
Acknowledgements The authors are grateful to Barbara Wider for translating the German article. TBA’s research fellow-
ship is supported by Dr. Willmar Schwabe Pharmaceuticals, Germany.
References Bent, S., Saint, S., Vittinghoff, E., Grady, D., 1999. Antibiotics in acute bronchitis: a meta-analysis. Am. J. Med. 107, 62–67. Bereznoy, V.V., Riley, D.S., Wassmer, G., Heger, M., 2003. Efficacy of extract of Pelargonium sidoides in children with acute non-group A beta-hemolytic streptococcus tonsillopharyngitis: a randomized, double-blind, placebo-controlled trial. Altern. Ther. Health Med. 9, 68–79. Blochin, B., Haidvogl, M., Heger, M., 1999. Umckaloabo im Vergleich zu Acetylcystein bei Kindern mit akuter Bronchitis: prospektive, randomisierte, kontrollierte, offene Studie zur Wirksamkeit und Vertra¨glichkeit. Der Kassenarzt. 49, 46–50. Chuchalin, A.G., Berman, B., Lehmacher, W., 2005. Treatment of acute bronchitis in adults with a Pelargonium sidoides preparation (EPss7630): a randomized, doubleblind, placebo-controlled trial. Explore (NY) 1, 437–445. Conrad, A., Jung, I., Tioua, D., Lallemand, C., Carrapatoso, F., Engels, I., et al., 2007. Extract of Pelargonium sidoides (EPss7630) inhibits the interactions of Group A – streptococci and host epithelia in vitro. Phytomedicine 14, 52–59. Derry, C.J., Derry, S., McQuay, H.J., Moore, R.A., 2006. Systematic review of systematic reviews of acupuncture published 1996–2005. Clin. Med. 6 (4), 381–386. Easterbrook, P.J., Berlin, J.A., Gopalan, R., Matthews, D.R., 1991. Publication bias in clinical research. Lancet 337, 867–872. Golovatiouk, A., Chuchalin, A.G., 2002. Wirksamkeit eines extraktes aus Pelargonium sidoides (EPss7630) im vergleich zu placebo bei patienten mit akuter bronchitis. Phytopharmaka VII, 3–12. Gonzales, R., Sande, M., 1995. What will it take to stop physicians from prescribing antibiotics in acute bronchitis? Lancet 345 (8951), 665–666. Haidvogl, M., Schuster, R., Heger, M., 1996. Bronchitis im kindesalter- multizenter-studie zur wirksamkeit und vertraglichkeit des phytotherapeutikums Umckaloabo. Z. Phytother. 17, 300–313. Haidvogl, M., Heger, M., 2007. Treatment effect and safety of EPss 7630 solution in acute bronchitis in childhood: report of a multicentre observational study. Phytomedicine 14, 60–64. Hueston, W.J., Mainous, A.G., 1998. Acute bronchitis. Am. Fam. Physician 57, 1270–1272. Jadad, A.R., Moore, R.A., Carroll, D., Jenkinson, C., Reynolds, D.J., Gavaghan, D.J., 1996. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin. Trials 17, 1–12. Kligler, B., Ulbricht, C., Basch, E., Kirkwood, C.D., Abrams, T.R., Miranda, M., et al., 2006. Andrographis paniculata for the treatment of upper respiratory infection: a systematic review by the Natural Standard Research Collaboration. Explore 2, 25–29.
ARTICLE IN PRESS T.B. Agbabiaka et al. / Phytomedicine 15 (2008) 378–385
Kieser, M., 2007a. Efficacy and tolerability of Pelargonium sidoides extract EPss7630 in patients (older than 18 years) with acute bronchitis, forthcoming. Kieser, M., 2007b. Efficacy and tolerability of Pelargonium sidoides extract EPss7630 in patients (6–18 years old) with acute bronchitis, forthcoming. Kolodziej, H., Kayser, O., Gutmann, M., 1995. Arzneilich verwendete Pelargonien aus Su¨dafrika. Dtsch. Apoth. Ztg. 135, 853–864. Knutson, D., Braun, C., 2002. Diagnosis and management of acute bronchitis. Am. Fam. Physician 65, 2039–2044. Kolodziej, H., Kayser, O., 1998. Pelargonium sidoides DC. – Neuste Erkenntnisse Zum Phytotherapeutikums Umckaloabo. Z. Phytother. 19, 141–151. Kolodziej, H., Kayser, O., Radtke, O.A., Kiderlen, A.F., Koch, E., 2003. Pharmacological profile of extracts of Pelargonium sidoides and their constituents. Phytomedicine 10, 18–24. Konig, I., 1995. Natural respiratory tract therapy: from Umckaloabo drug to therapy of respiratory tract infections. Therapiewoche 45, 1123–1126. Kupczyk, M., Kuna, P., 2002. Mucolytics in acute and chronic respiratory tract disorders. I. Pathophysiology and mechanisms of action. Pol. Merkur. Lekarski. 12, 245–247. Macfarlane, J., Holmes, W., Gard, P., Macfarlane, R., Rose, D., Weston, V., et al., 2001. Prospective study of the incidence, aetiology and outcome of adult lower respiratory tract illness in the community. Thorax 56, 109–114. Matthys, H., Eisebitt, R., Seith, B., Heger, M., 2003. Efficacy and safety of an extract of Pelargonium sidoides (EPss7630)
385
in adults with acute bronchitis: a randomised, double-blind, placebo-controlled trial. Phytomedicine 10, 7–17. Matthys, H., Heger, M., 2007a. EPss7630 solution – an effective therapeutic option in acute and exacerbating bronchitis. Phytomedicine 14, 65–68. Matthys, H., Heger, M., 2007b. Treatment of acute bronchitis with a liquid herbal drug preparation from Pelargonium sidoides preparation (EPss7630): a randomized doubleblind, placebo-controlled, multicentre study. Curr. Med. Res. Opin. 23, 323–331. Matthys, H., Kamin, W., Funk, P., Heger, M., 2007. Pelargonium sidoides preparation (EPss 7630) in the treatment of acute bronchitis in adults and children. Phytomedicine 14, 69–73. Niroumand, M., Grossman, R.F., 1998. Airway infection. Infect. Dis. Clin. North Am. 12, 671–688. Orr, P.H., Scherer, K., Macdonald, A., Moffatt, M.E., 1993. Randomized placebo-controlled trials of antibiotics for acute bronchitis: a critical review of the literature. J. Fam. Pract. 36, 507–512. Schulz, V., 2006. Extract from Pelargonium sidoides root is effective in acute bronchitis: result of a doubleblind study with 124 patients – comment. Z. Phytother. 27, 237–238. Watt, C., Breyer-Brandwyk, M.G., 1962. Medicinal and Poisonous Plants of Southern and Eastern Africa. Livingstone, Edinburgh/London, pp. 449–455. Wise, R., Hart, T., Cars, O., Streulens, M., Helmuth, R., Huovinen, P., et al., 1998. Antimicrobial resistance is a major threat to public health. Br. Med. J. 317, 609–610.
Phytomedicine 15 (2008) 378–385 www.elsevier.de/phymed
Pelargonium sidoides for acute bronchitis: A systematic review and meta-analysis Taofikat B. Agbabiaka, Ruoling Guo, Edzard Ernst Complementary Medicine, Peninsula Medical School, Universities of Exeter and Plymouth, 25 Victoria Park Road, Exeter EX2 4NT, UK
Abstract Objective: To critically assess the efficacy of Pelargonium sidoides for treating acute bronchitis. Data sources: Systematic literature searches were performed in 5 electronic databases: (Medline (1950 – July 2007), Amed (1985 – July 2007), Embase (1974 – July 2007), CINAHL (1982 – July 2007), and The Cochrane Library (Issue 3, 2007) without language restrictions. Reference lists of retrieved articles were searched, and manufacturers contacted for published and unpublished materials. Review methods: Study selection was done according to predefined criteria. All randomized clinical trials (RCTs) testing P. sidoides extracts (mono preparations) against placebo or standard treatment in patients with acute bronchitis and assessing clinically relevant outcomes were included. Two reviewers independently selected studies, extracted and validated relevant data. Methodological quality was evaluated using the Jadad score. Meta-analysis was performed using a fixed-effect model for continuous data, reported as weighted mean difference with 95% confidence intervals. Results: Six RCTs met the inclusion criteria, of which 4 were suitable for statistical pooling. Methodological quality of most trials was good. One study compared an extract of P. sidoides, EPss7630, against conventional non-antibiotic treatment (acetylcysteine); the other five studies tested EPss7630 against placebo. All RCTs reported findings suggesting the effectiveness of P. sidoides in treating acute bronchitis. Meta-analysis of the four placebo-controlled RCTs suggested that EPss7630 significantly reduced bronchitis symptom scores in patients with acute bronchitis by day 7. No serious adverse events were reported. Conclusion: There is encouraging evidence from currently available data that P. sidoides is effective compared to placebo for patients with acute bronchitis. r 2007 Elsevier GmbH. All rights reserved. Keywords: Acute bronchitis; Bronchitis Symptom Score; EPss7630; Pelargonium sidoides; Umckaloabo
Introduction Acute bronchitis refers to the acute or sub-acute onset of productive cough with no history of chronic pulmonary disease and without evidence of pneumonia Corresponding author. Tel.: +44 1392 424 942; fax: +44 1392 427 562. E-mail address: taofi[email protected] (T.B. Agbabiaka).
0944-7113/$ - see front matter r 2007 Elsevier GmbH. All rights reserved. doi:10.1016/j.phymed.2007.11.023
or sinusitis (Gonzales and Sande, 1995). It is one of the most common diagnoses in primary care (Knutson and Braun, 2002), affecting 44 adults in every 1000 (416 years old) each year in the UK, with 82% of the episodes occurring in autumn or winter (Macfarlane et al., 2001). Hueston and Mainous (1998) reported that viruses are the most common cause of this condition. However, non-viral agents (bacteria and allergens) play a role in its etiology in less than 10% of cases (Niroumand and
ARTICLE IN PRESS T.B. Agbabiaka et al. / Phytomedicine 15 (2008) 378–385
Grossman, 1998). Surprisingly, prescription of antibiotics for acute bronchitis is still wide-spread even though their efficacy is limited to bacterial infections (Bent et al., 1999; Hueston and Mainous, 1998; Orr et al., 1993). Perhaps due to the limited success of antibiotics, substantial cost, associated adverse effects and antibiotic resistance as a result of overuse and misuse (Bent et al., 1999) more attention is now focused on alternative treatments for acute bronchitis and other upper respiratory tract infections; hence the search for effective herbal options (Kligler et al., 2006). Pelargonium sidoides is an herbaceous perennial that is popular in South African traditional medicine for the treatment of infectious respiratory diseases (Watt and Breyer-Brandwyk, 1962). Pelargonium containing phytopharmaceuticals are currently widely used in Europe to treat respiratory tract infections (Matthys et al., 2003). The aim of this systematic review was to assess available evidence from rigorous clinical trials on the efficacy and/or effectiveness of Pelargonium sidoides for the treatment of acute bronchitis.
379
Data extraction and assessment of methodological quality All titles and abstracts identified by the literature search were initially assessed against the inclusion criteria to decide if the full text article should be obtained (TBA). The retrieved full text papers were independently assessed by two reviewers (TBA and RG) to decide on inclusion. Disagreements were resolved by discussion between the two reviewers and by seeking the opinion of a third reviewer (EE) when necessary. Data concerning the details of study design, quality of the study, participants, intervention, outcomes and adverse events were independently extracted by two reviewers (TBA and RG) using a pre-designed data extraction sheet. The methodological quality of all included studies was assessed independently by two reviewers (TBJ and RG) using the 5-point Jadad score (Jadad et al., 1996). We also assessed allocation concealment and the use of intention to treat analysis. Discrepancies were resolved by discussion between the two authors.
Data analysis
Methods Search strategy Systematic literature searches were conducted in 5 electronic databases: (Medline (1950 – July 2007), Amed (1985 – July 2007), Embase (1974 – July 2007), CINAHL (1982 – July 2007), and The Cochrane Library (Issue 3, 2007) without language restrictions. The search strategy was structured as ‘‘search terms for condition’’ AND ‘‘search terms for intervention’’. Both MeSH terms and text terms were searched for the condition as well as the intervention. The MeSH term for the condition was acute bronchitis and for the intervention were Pelargonium OR sidoides OR Umckaloabo OR EPss7630. We also searched the reference lists of all retrieved articles. We contacted 4 manufacturers of commercial preparations of Pelargonium sidoides for further information, particularly unpublished materials.
Included trials were categorised by the type of control interventions and the following comparisons were made: (a) herbal medicine versus placebo and (b) herbal medicine versus conventional therapy. Meta-analysis was carried out using Review Manager Software (version 4.2.10 2007, Cochrane Collaboration and Updated Software). We quantitatively combined the results of four studies that were suitable for statistical pooling. We calculated weighted mean differences (WMD) and 95% confidence intervals (CI) for continuous data using fixed-effect model. The chi-square test for homogeneity was performed to determine whether the distribution of results was compatible with the assumption that differences between trials were due to chance variation alone. To evaluate potential publication bias we constructed a funnel plot for the primary outcome of change in Bronchitis Severity Scores (BSS) score, using effect size as a measure, and visually examined it for asymmetry.
Inclusion and exclusion criteria
Results Only randomized clinical trials (RCTs) testing monopreparations of P. sidoides as sole or adjunctive treatment administered orally against a control intervention (placebo or conventional therapy), in patients of any age or sex with acute bronchitis and assessing clinically relevant outcomes were included. Trials including patients with pre-existing chronic bronchitis or other infectious diseases were excluded.
The literature searches identified 295 potentially relevant titles and abstracts. We also identified two unpublished trials (Kieser, 2007a, b). Initial screening of the titles and abstracts identified 16 relevant references for which full texts were retrieved for further evaluation. Ten references were subsequently excluded for the following reasons: eight references as they were
ARTICLE IN PRESS 380
T.B. Agbabiaka et al. / Phytomedicine 15 (2008) 378–385
non-controlled trials or reviews (Haidvogl and Heger, 2007; Matthys and Heger, 2007a; Matthys et al., 2007; Schulz, 2006; Kolodziej and Kayser, 1998; Haidvogl et al., 1996; Konig, 1995; Kolodziej et al., 1995), 1 RCT was excluded because the condition treated was not acute bronchitis (Bereznoy et al., 2003). A duplicate paper in German was also excluded (Golovatiouk and Chuchalin, 2002). Six RCTs (Matthys et al., 2003; Kieser, 2007a, b; Blochin et al., 1999; Chuchalin et al., 2005; Matthys and Heger, 2007b) met the inclusion criteria and were included in the review. Four trials (Matthys et al., 2003; Kieser, 2007a; Chuchalin et al., 2005; Matthys and Heger, 2007b) with total number of subjects n ¼ 1012 were suitable for statistical pooling and included in a meta-analysis. Agreement between reviewers on inclusion of RCTs was 100%. Fig. 1 is a flow chart of the trial selection process. A summary of the main characteristics of these studies is presented in Table 1. The studies were published between 1999 and 2007, originating from two countries (Russia, n ¼ 3; Ukraine, n ¼ 3) and in two languages (English, n ¼ 5; German, n ¼ 1). A total of 1681 subjects were randomised and 1647 analysed by the original investigators. The sample sizes ranged from 60 to 476, all patients were diagnosed with acute bronchitis,
Total number of citations identified: Electronic searches n= 295 Manufacturers n=2 Citations excluded after removal of duplicates, screening of titles and abstracts: n= 281 Full manuscript retrieved for detailed evaluation: From electronic search From manufacturer From reference list
n= 16 n= 13 n= 2 n= 1
Excluded with reasons: n=9 Not RCT n= 8 Condition treated not acute bronchitis n=1
Potential relevant articles identified
with BSS of 5 points or more, and duration of complaint less than 48 h prior to enrolment. Four trials were carried out in adult patients aged over 18 years (Matthys et al., 2003; Kieser, 2007a; Chuchalin et al., 2005; Matthys and Heger, 2007b), two RCTs involved children aged between 6 and 18 years (Kieser, 2007b; Blochin et al., 1999). Five RCTs compared EPss7630 (Dr. Willmar Schwabe Pharmaceuticals, Germany) against placebo (Matthys et al., 2003; Kieser, 2007a, b; Chuchalin et al., 2005; Matthys and Heger, 2007b). Another RCT tested EPss7630 as sole treatment against a mucolytic commonly used for this condition (Blochin et al., 1999). EPss7630 was administered as solution (drops) in all trials except two (Kieser, 2007a, b), where tablets were administered. EPss7630 is an ethanolic extract of the roots of Pelargonium sidoides which is marketed in Germany, in the Commonwealth of Independent States (CIS), in the Baltic States, and in Mexico, for the treatment of ear, nose and throat (ENT) and respiratory infections (Matthys et al., 2003).The formulation of EPss7630 solution and its botanical origin are identical to the finished product Umckaloabos registered by ISO Pharmaceuticals, Ettlingen, Germany. Methodological quality of the RCTs reviewed was generally good; Jadad score was 5 points (maximum 5) for three trials (Matthys et al., 2003; Chuchalin et al., 2005; Matthys and Heger, 2007b), the unpublished trials were assessed but not scored for methodological quality (Table 1). Five of the RCTs (Matthys et al., 2003; Kieser, 2007a, b; Chuchalin et al., 2005; Matthys and Heger, 2007b) described methods of randomisation and double blinding, allocation concealment, follow up and intention to treat analysis. Three published (Matthys et al., 2003; Chuchalin et al., 2005; Matthys and Heger, 2007b) and the two unpublished studies (Kieser, 2007a, b) reported details of dropouts and withdrawals. The 6 trials were categorised into two groups according to the type of control interventions. Detailed results are summarised in Table 1 and described below.
n= 7
Herbal medicine versus conventional treatment Articles excluded with reasons: Duplicate of included paper n= 1
RCTs included in the systematic review
n=6
RCTs excluded from meta-analysis with reasons: Not placebo controlled n=1 Different population n= 1
RCTs included in the meta-analysis
n=4
Fig. 1. Flow chart of the study selection process.
Only one RCT (Blochin et al., 1999) compared EPss7630 with a conventional therapy (Acetylcysteine). In this trial, 60 children aged between 6 and 12 yrs were randomised into two groups to receive either EPss7630 liquid formula (20 drops every hour up to 12 times on day 1 and 2; 20 drops 3/daily from day 3–7) or Acetylcystein granules (2 mg 200 mg daily for 7 days). Outcome measures were changes in typical symptoms of bronchitis (cough, sputum, rattling noise, chest pain and dyspnoea). Bronchitis specific symptoms were reduced more effectively and faster with EPss7630 than acetylcysteine at the end of 7 days treatment, intergroup
Table 1.
Characteristics of included randomised clinical trials of Pelargonium sidoides for the treatment of acute bronchitis
First author
Participants Design and quality (JS)
R/A (T, C)
Intervention and daily dose Control group
Treatment duration/follow-up period (days)
Main outcome measure
Main results, mean difference (95% CI)
Adverse events (AEs)
Herbal medicine versus conventional medicine Blochin et al. Open parallel 60/60 (30,30) (1999) (2)
6–12
EPss7630 solution, 20 drops/ h up to 12 for days 1& 2 and 20 drops 3/d on days 3–7
Acetylsteine granules (200 mg 2/d)
7
Bronchitis typical symptoms
7.0 (0.02–0.02)
Not reported
Herbal medicine versus placebo Matthys DB parallel 476/468 et al. (2003) (5) (237,239)
X18
EPss7630 liquid extract, 30 drops 3/d EPss7630 liquid extract, 30 drops 3/d EPss7630 solution, 30 drops 3/d
placebo
7
2.7 (2.1–3.3)
No serious AEs (test, n ¼ 20; control, n ¼ 16)
Placebo
7
2.3 (1.3–3.3)
Placebo
7
Change of total score of BSS Change of total score of BSS Change of total score of BSS
EPss7630 tablet
Placebo
7
Change of total score of BSS
4.3 (0.4–1.2)
Mild to moderate AEs (test, n ¼ 15; control, n ¼ 10) No serious AEs; blood and lymphatic disorders and laboratory abnormalities (test, n ¼ 37; control, n ¼ 27); No intergroup differences No serious AEs in any group; minor to moderate AEs (gastrointestinal and nervous system disorders, test: LD ¼ 22, MD ¼ 25 and HD ¼ 31; control, n ¼ 14) No serious AEs in any group; minor to moderate AEs (gastrointestinal disorders infections and infestations, test: LD ¼ 23, MD ¼ 20 and HD ¼ 19; control, n ¼ 18)
Chuchalin et al. (2005)
DB parallel (5)
124/117 (64, 60)
X18
Matthys and Heger (2007b)
DB parallel (5)
217/198 (108,109)
18–66
Kieser (2007a)*
DB
405/405 (303,102)
418
Parallel#
Kieser (2007b)*
DB Parallel#
LD – 3 10 mg/d MD – 3 20 mg/d HD – 3 30 mg/d
399/399 (298,101)
6–18
EPss7630 tablet LD – 3 10 mg/d MD – 3 0 mg/d HD – 3 30 mg/d
Placebo
7
Change of total score of BSS
2.3(1.6–3.1)
6.1 (0.8–1.6) 6.3 (0.9–1.6) for the 30, 60 and 90 mg/d doses respectively 3.6 (0.4–1.3) 4.4 (0.6–1.5) 5.0 (0.6–1.3) for the 30, 60 and 90 mg/d doses respectively
JS ¼ Jadad Score (maximum of five points); DB ¼ double blind; x/d ¼ times per day; AEs ¼ adverse events; R ¼ number of patients randomised; A ¼ number of patients analysed; T ¼ number of participants randomised to treatment group; C ¼ number of participants randomised to control group (given in parenthesis); BSS ¼ Bronchitis Symptom Score; LD ¼ low dose; MD ¼ medium dose; HD ¼ high dose; # ¼ unpublished trial, not scored for methodological quality, * ¼ unpublished RCT.
ARTICLE IN PRESS
Test group
T.B. Agbabiaka et al. / Phytomedicine 15 (2008) 378–385
Age range (yrs)
381
ARTICLE IN PRESS 382
T.B. Agbabiaka et al. / Phytomedicine 15 (2008) 378–385
difference was however not significant (p ¼ 0.285). The authors reported no information on drop outs or withdrawals.
Kieser (2007a) due to the similarity of the medium dose to the doses administered in the other studies (i.e. a 20mg-tablet is equivalent to 30 drops). A meta-analysis of the four trials (n ¼ 1012) showed a significant decrease of BSS score compared to placebo (WMD 2.80 points, 95% CI 2.44–3.15). The chi-square test for homogeneity signifies heterogeneity among the four trials (p ¼ 0.09; I2 ¼ 54.4%), which indicates that the observed differences in results between trials could be caused by factors other than chance (Fig. 2). However, the small number of trials n ¼ 4 was inadequate to produce a meaningful funnel plot. A subgroup analysis was conducted to test the robustness of the results of the overall analysis and also test for publication bias. We excluded the unpublished trial (Kieser, 2007a) from the analysis, retaining only the three published trials with similar methodologic features (Matthys et al., 2003; Chuchalin et al., 2005; Matthys and Heger, 2007b). These three RCTs used the same preparation-EPss7630 (a liquid herbal drug preparation from the roots of Pelargonium sidoides (1:8–10), extraction solvent: ethanol 11% (w/w) produced by Dr. Willmar Schwabe Pharmaceuticals, Germany) at the same dose of 30 drops thrice daily over a treatment period of 7 days. Meta-analysis of these data (Fig. 1) revealed a WMD of 2.48 points (CI, 2.05–2.92), still indicating a significant change in BSS score in patients treated with EPss7630 compared with placebo but homogeneity was absolute this time (p ¼ 0.67; I2 ¼ 0%). Some other secondary outcomes were also assessed in these RCTs. In one RCT (Matthys et al., 2003), the
Herbal medicine versus placebo A total of 5 RCTs tested EPss7630 against placebo, three (Matthys et al., 2003; Chuchalin et al., 2005; Matthys and Heger, 2007b) of which tested EPss7630 solution (30 drops 3/d) and two (Kieser, 2007a, b) tested EPss7630 tablets. According to the manufacturer’s data, these two dosage forms are equivalents. In both trials testing EPss7630 tablets (Kieser, 2007a, b), patients were randomised into four groups to receive EPss7630 tablets: low dose ¼ 1 10 mg, 3/d; medium dose ¼ 1 20 mg, 3/d and high dose ¼ 1 30 mg, 3/d respectively, or placebo for 7 days. One of the RCTs was carried out amongst children aged between 6 and 18 years (Kieser, 2007b). Outcome measures were changes in the bronchitis symptom score (BSS) between day 0 and 7, and changes in individual components of BSS. All five trials (Matthys et al., 2003; Kieser, 2007a, b; Chuchalin et al., 2005; Matthys and Heger, 2007b) reported effects in favour of EPss7630 compared to placebo at the end of 7 days treatment. The four RCTs carried out in adults (Matthys et al., 2003; Kieser, 2007a; Chuchalin et al., 2005; Matthys and Heger, 2007b) were considered suitable to be combined in a meta-analysis, the trial in children (Kieser, 2007b) was excluded due to the difference in population. We included in the meta-analysis data from the RCT by Review: Pelargonium sidoides for the treatment of acute bronchitis Comparison: 01 EPs7630 vs. placebo Outcome: 01 BSS score
Study
or sub-category
N
01 All included studies Matthys 2003 Chuchalin 2005 Kiesser 2007 Matthys 2007
233 64 101 108
Treatment Mean (SD)
N
5.90(2.90) 7.20(3.10) 6.10(2.10) 7.60(2.20)
235 60 102 109
Control Mean (SD)
WMD (fixed) 95% CI
3.20(4.10) 4.90(2.70) 2.70(2.30) 5.30(3.20)
506 506 Total (95% CI) Test for heterogeneity: Chi² = 6.58, df = 3 (P = 0.09), I² = 54.4% Test for overall effect: Z = 15.48 (P < 0.00001) 02 Published studies only Matthys 2003 Chuchalin 2005 Matthys 2007
233 64
5.90(2.90) 7.20(3.10)
235 60
3.20(4.10) 4.90(2.70)
108
7.60(2.20)
109
5.30(3.20)
Weight %
WMD (fixed) 95% CI
30.32 12.01 34.16 23.51
2.70 [2.06, 3.34] 2.30 [1.28, 3.32] 3.40[2.79, 4.01] 2.30 [1.57, 3.03]
100.00
2.80 [2.44, 3.15]
46.04 18.24 35.71
Total (95% CI) 405 404 Test for heterogeneity: Chi² = 0.80, df = 2 (P = 0.67), I² = 0% Test for overall effect: Z = 11.16 (P < 0.00001)
100.00
-10
-5
Favours treatment
0
5
2.70 [2.06, 3.34] 2.30 [1.28, 3.32] 2.30 [1.57, 3.03]
2.48 [2.05, 2.92]
10
Favours control
Fig. 2. Effect of EPss7630 on BSS score in patients with acute bronchitis. Mean difference of the change from baseline given with their 95% confidence intervals. The vertical line represents no difference between EPss7630 and placebo.
ARTICLE IN PRESS T.B. Agbabiaka et al. / Phytomedicine 15 (2008) 378–385
patient’s health status showed a positive tendency in favour of EPss7630 at the end of the trial. Onset of treatment effect was recognised within 4 days in 53.6% of subjects in the treatment group compared to 36.2% in the control group. Duration of illness was also significantly shorter for subjects treated with EPss7630 compared to placebo (po0.001). In another RCT (Chuchalin et al., 2005) an overall improvement was observed in all individual components of BSS (cough, chest pain on coughing, sputum, rales/ronchi and dyspnoea) in the EPss7630 treatment group compared to the placebo group. Better quality of life was reported by 61.9% of patients treated with EPss7630 compared with 41.4% in placebo patients (p ¼ 0.012). The third RCT (Matthys and Heger, 2007b) reported individual symptoms and patients satisfaction over 7 days, complete remission in cough was observed in 51.9% for the treatment group compared to 11.9% in placebo; 68.3% improvement in sputum production among the controls and 40.0% in the placebo group. Complete recovery by day 7 was observed by the physician in 45.4% of patients taking active treatment compared to 6.4% of patients on placebo.
Adverse events Five RCTs (Matthys et al., 2003; Kieser, 2007a, b; Chuchalin et al., 2005; Matthys and Heger, 2007b) reported information on adverse events (AE). No serious AEs were reported in any of these RCTs. Mild to moderate AEs were observed in all RCTs, but there was no significant difference in the number of AEs reported between the two treatment groups (Matthys and Heger, 2007b). Some of the AEs reported include gastrointestinal disorders (Matthys et al., 2003; Kieser, 2007a, b), nervous system disorders (Matthys et al., 2003; Kieser, 2007a), ear and labyrinth disorders (Matthys et al., 2003) which was slightly higher with EPss7630 than with placebo, while gastrointestinal disorders were slightly more frequent in the placebo group. Infections and infestations were also reported in the RCT involving children (Kieser, 2007b).
Discussion This systematic review identified 6 studies, 4 of which are published (Matthys et al., 2003; Blochin et al., 1999; Chuchalin et al., 2005; Matthys and Heger, 2007b) and two unpublished (Kieser, 2007a, b), all of which suggest that preparations of P. sidoides are effective in the treatment of acute bronchitis. When compared to placebo, EPss7630 was found to significantly decrease BSS scores within 7 days of treatment (WMD 2.80 points, 95% CI 2.44–3.15) (Fig. 1) with improvement in individual components of BSS.
383
Only randomized, double-blind, placebo-controlled trials were included in this review. However, methodological quality varied among studies. Three trials scored a maximum of 5 points on the quality assessment (Matthys et al., 2003; Chuchalin et al., 2005; Matthys and Heger, 2007b), one trial had a low score of 2 points (Blochin et al., 1999). Five studies described randomization methods and allocation concealment with all analyses based on intention to treat (Matthys et al., 2003; Kieser, 2007a, b; Chuchalin et al., 2005; Matthys and Heger, 2007b). Three RCTs (Matthys et al., 2003; Chuchalin et al., 2005; Matthys and Heger, 2007b) were of similar design, two of them (Chuchalin et al., 2005; Matthys and Heger, 2007b) were conducted in the same country. One researcher participated in all three studies. Another two RCTs (Kieser, 2007a, b) were of the same design although each was conducted in different populations. Five of the RCTs were funded, at least in part, by the manufacturer of the product. Independent replication of the results would be desirable despite the difficulties encountered in getting funding from independent sources for research of commercial products. In view of the fact that the licence for EPss7630 was only acquired in 2005, we anticipate that independent replications, which are desirable for confirming these findings, will soon be forthcoming. We found substantial heterogeneity among the 4 trials included in the meta-analysis, as indicated by a chisquare test (w2 ¼ 6.58, p ¼ 0.09 and I2 ¼ 54.4%). To explore the possible reasons for this, we performed a sensitivity analysis. We felt that the heterogeneity could be attributed to the single unpublished trial by Kieser (2007a).When this trial was excluded from the analysis, effect size for EPss7630 was slightly decreased but remained superior to placebo (WMD of 2.48; 95% CI 2.05–2.92). We also observed no inconsistency among these trials (I2 ¼ 0%). Compared to N-acetylcysteine (NAC), a non-antibiotic treatment, Blochin et al. (1999) reported that EPss7630 showed a reduction in bronchitis-typical symptoms in both treatment groups but no significant intergroup differences (p ¼ 0.285). NAC is a precursor of glutathione, an important internal antioxidant which has been in clinical use for more than 30 years as a mucolytic drug to boost antioxidant levels and dissolve mucus in people suffering from respiratory ailments (Kupczyk and Kuna, 2002). However, the study by Blochin et al. (1999) was not blinded and, even though randomization was mentioned, it was not adequately described. There was also no description of dropouts and withdrawals, all this contributed to a low Jadad score (2 out of 5). The evidence from this study is therefore weak, particularly as it also had a small sample size and no allocation concealment, which may overestimate intervention benefits.
ARTICLE IN PRESS 384
T.B. Agbabiaka et al. / Phytomedicine 15 (2008) 378–385
The characteristic and therapeutically most important natural substances of EPss7630 are polymeric polyphenolic compounds. Simple phenolic compounds and coumarins are also mentioned in the literature (Kolodziej et al., 2003). The mechanism of action is ascribed to the ability of EPss7630 to antagonize bacterial adhesion and/or invasion to intact epithelia, thus protecting the upper respiratory tract from bacteria colonisation and infection (Conrad et al., 2007). Available safety data for EPss7630 indicate mild to moderate adverse events comprising gastrointestinal complaints which are likely due to the phenols contained in the liquid herbal drug preparation from Pelargonium roots, no serious risks or detrimental effects on quality of life was reported. The use of EPss7630 could also prove useful by minimising the use of antibiotics for this condition. This is particularly important when the over-prescription of antibiotics for acute bronchitis, and increased bacterial resistance is considered (Wise et al., 1998). Our findings may be subject to the general limitations of systematic review, pertaining to potential incompleteness of the reviewed evidence. Although we aimed to identify all RCTs on this topic, the distorting effects on systematic reviews arising from publication bias and location bias cannot be completely ruled out including the tendency for negative trials to remain unpublished (Easterbrook et al., 1991). We searched all available databases with a focus on American and European literature. Our search was not restricted in terms of language of publication. Therefore, we are confident that our search had located all relevant data on the subject. However, whether other unpublished trials were not located remains uncertain as contact to manufacturers produced only two relevant studies (Kieser, 2007a, b). The relatively small number of studies (n ¼ 6) available for this review may be seen as a further limitation, but we agree with Derry et al. (2006) that four trials and/or a minimum of 200 patients is sufficient to calculate statistical significance and make a conclusion on the level of evidence from the trials.
Conclusions Currently available data from 6 high quality randomised clinical trials suggests there is encouraging evidence that P. sidoides is effective compared to placebo for patients with acute bronchitis.
Acknowledgements The authors are grateful to Barbara Wider for translating the German article. TBA’s research fellow-
ship is supported by Dr. Willmar Schwabe Pharmaceuticals, Germany.
References Bent, S., Saint, S., Vittinghoff, E., Grady, D., 1999. Antibiotics in acute bronchitis: a meta-analysis. Am. J. Med. 107, 62–67. Bereznoy, V.V., Riley, D.S., Wassmer, G., Heger, M., 2003. Efficacy of extract of Pelargonium sidoides in children with acute non-group A beta-hemolytic streptococcus tonsillopharyngitis: a randomized, double-blind, placebo-controlled trial. Altern. Ther. Health Med. 9, 68–79. Blochin, B., Haidvogl, M., Heger, M., 1999. Umckaloabo im Vergleich zu Acetylcystein bei Kindern mit akuter Bronchitis: prospektive, randomisierte, kontrollierte, offene Studie zur Wirksamkeit und Vertra¨glichkeit. Der Kassenarzt. 49, 46–50. Chuchalin, A.G., Berman, B., Lehmacher, W., 2005. Treatment of acute bronchitis in adults with a Pelargonium sidoides preparation (EPss7630): a randomized, doubleblind, placebo-controlled trial. Explore (NY) 1, 437–445. Conrad, A., Jung, I., Tioua, D., Lallemand, C., Carrapatoso, F., Engels, I., et al., 2007. Extract of Pelargonium sidoides (EPss7630) inhibits the interactions of Group A – streptococci and host epithelia in vitro. Phytomedicine 14, 52–59. Derry, C.J., Derry, S., McQuay, H.J., Moore, R.A., 2006. Systematic review of systematic reviews of acupuncture published 1996–2005. Clin. Med. 6 (4), 381–386. Easterbrook, P.J., Berlin, J.A., Gopalan, R., Matthews, D.R., 1991. Publication bias in clinical research. Lancet 337, 867–872. Golovatiouk, A., Chuchalin, A.G., 2002. Wirksamkeit eines extraktes aus Pelargonium sidoides (EPss7630) im vergleich zu placebo bei patienten mit akuter bronchitis. Phytopharmaka VII, 3–12. Gonzales, R., Sande, M., 1995. What will it take to stop physicians from prescribing antibiotics in acute bronchitis? Lancet 345 (8951), 665–666. Haidvogl, M., Schuster, R., Heger, M., 1996. Bronchitis im kindesalter- multizenter-studie zur wirksamkeit und vertraglichkeit des phytotherapeutikums Umckaloabo. Z. Phytother. 17, 300–313. Haidvogl, M., Heger, M., 2007. Treatment effect and safety of EPss 7630 solution in acute bronchitis in childhood: report of a multicentre observational study. Phytomedicine 14, 60–64. Hueston, W.J., Mainous, A.G., 1998. Acute bronchitis. Am. Fam. Physician 57, 1270–1272. Jadad, A.R., Moore, R.A., Carroll, D., Jenkinson, C., Reynolds, D.J., Gavaghan, D.J., 1996. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin. Trials 17, 1–12. Kligler, B., Ulbricht, C., Basch, E., Kirkwood, C.D., Abrams, T.R., Miranda, M., et al., 2006. Andrographis paniculata for the treatment of upper respiratory infection: a systematic review by the Natural Standard Research Collaboration. Explore 2, 25–29.
ARTICLE IN PRESS T.B. Agbabiaka et al. / Phytomedicine 15 (2008) 378–385
Kieser, M., 2007a. Efficacy and tolerability of Pelargonium sidoides extract EPss7630 in patients (older than 18 years) with acute bronchitis, forthcoming. Kieser, M., 2007b. Efficacy and tolerability of Pelargonium sidoides extract EPss7630 in patients (6–18 years old) with acute bronchitis, forthcoming. Kolodziej, H., Kayser, O., Gutmann, M., 1995. Arzneilich verwendete Pelargonien aus Su¨dafrika. Dtsch. Apoth. Ztg. 135, 853–864. Knutson, D., Braun, C., 2002. Diagnosis and management of acute bronchitis. Am. Fam. Physician 65, 2039–2044. Kolodziej, H., Kayser, O., 1998. Pelargonium sidoides DC. – Neuste Erkenntnisse Zum Phytotherapeutikums Umckaloabo. Z. Phytother. 19, 141–151. Kolodziej, H., Kayser, O., Radtke, O.A., Kiderlen, A.F., Koch, E., 2003. Pharmacological profile of extracts of Pelargonium sidoides and their constituents. Phytomedicine 10, 18–24. Konig, I., 1995. Natural respiratory tract therapy: from Umckaloabo drug to therapy of respiratory tract infections. Therapiewoche 45, 1123–1126. Kupczyk, M., Kuna, P., 2002. Mucolytics in acute and chronic respiratory tract disorders. I. Pathophysiology and mechanisms of action. Pol. Merkur. Lekarski. 12, 245–247. Macfarlane, J., Holmes, W., Gard, P., Macfarlane, R., Rose, D., Weston, V., et al., 2001. Prospective study of the incidence, aetiology and outcome of adult lower respiratory tract illness in the community. Thorax 56, 109–114. Matthys, H., Eisebitt, R., Seith, B., Heger, M., 2003. Efficacy and safety of an extract of Pelargonium sidoides (EPss7630)
385
in adults with acute bronchitis: a randomised, double-blind, placebo-controlled trial. Phytomedicine 10, 7–17. Matthys, H., Heger, M., 2007a. EPss7630 solution – an effective therapeutic option in acute and exacerbating bronchitis. Phytomedicine 14, 65–68. Matthys, H., Heger, M., 2007b. Treatment of acute bronchitis with a liquid herbal drug preparation from Pelargonium sidoides preparation (EPss7630): a randomized doubleblind, placebo-controlled, multicentre study. Curr. Med. Res. Opin. 23, 323–331. Matthys, H., Kamin, W., Funk, P., Heger, M., 2007. Pelargonium sidoides preparation (EPss 7630) in the treatment of acute bronchitis in adults and children. Phytomedicine 14, 69–73. Niroumand, M., Grossman, R.F., 1998. Airway infection. Infect. Dis. Clin. North Am. 12, 671–688. Orr, P.H., Scherer, K., Macdonald, A., Moffatt, M.E., 1993. Randomized placebo-controlled trials of antibiotics for acute bronchitis: a critical review of the literature. J. Fam. Pract. 36, 507–512. Schulz, V., 2006. Extract from Pelargonium sidoides root is effective in acute bronchitis: result of a doubleblind study with 124 patients – comment. Z. Phytother. 27, 237–238. Watt, C., Breyer-Brandwyk, M.G., 1962. Medicinal and Poisonous Plants of Southern and Eastern Africa. Livingstone, Edinburgh/London, pp. 449–455. Wise, R., Hart, T., Cars, O., Streulens, M., Helmuth, R., Huovinen, P., et al., 1998. Antimicrobial resistance is a major threat to public health. Br. Med. J. 317, 609–610.