acupuntura e asma

8/12/2019 Acupuntura e Asma

http://slidepdf.com/reader/full/acupuntura-e-asma 1/7



to ascertain and quantify the different sources of biasof the meta-analysis. Finally, although the studyeligibility criteria of the Cochrane9s Database over-view are similar to the ones used in this study, thestudies included are not the same. The authors give afull account of the methodology used and compare theuse of unstandardised and standardised difference of means for the overall effect.

Methods

Eligibility of trials

The authors formulated two eligibility criteria.First, the study had to be a randomised clinical trialcomparing real and placebo acupuncture in subjectswith asthma. Second, the study had to measure atleast one of the objective end points: peak expiratoryflow rate (PEFR), forced expiratory volume in onesecond (FEV1) and forced vital capacity (FVC).

Retrieving the literature

Initially, references from published papers [1, 6] andnarrative reviews about the topic [3, 8] were searched.A computer-assisted search examining the followingliterature databases was also performed: Medline,Biological Abstracts and Dissertation Abstracts. Thekeywords used were: "acupuncture", "asthma", "pulmo-nary disease", "clinical trial", "alternative medicine","randomised controlled trial" and "complementarymedicine". Successive searches were performed duringthe period of study, modifying the initial keywords inorder to incorporate any new information. The search

spanned the period 1970–2000. Finally, the authors of all eligible reports were contacted and asked if theywere aware of any further published or unpublishedwork. Retrieval spanned the period December 1994– December 2000.

Characterisation of the studies

From each eligible report the following were noted:year and source of publication, number of patients,type of randomisation, blindness, number of acu-puncture points, number of excluded patients, medi-cation, and outcomes and statistical summaries of

sex, age and duration of disease and treatment. Aquestionnaire was designed to assess other subjectivefeatures related to the quality of the reports [14]. Thequestionnaire had three sections to assess, separately,study design (10 items), statistical analysis (threeitems) and presentation of results (two items). A scorewas assigned to each item to give a maximum achiev-able score of 39 points (27 for study design, 10 forstatistical analysis section and 2 for presentation of results). Four experienced biostatisticians indepen-dently assessed the papers. Author and publishedsource were unknown to the different assessors. Thereliability of this assessment was evaluated by theeffective reliability [15].

Statistical analysis

Every trial reported on at least one of the twooutcome measures PEFR (six trials) and FEV1 (seventrials), and no single outcome was reported for all thetrials. Consequently, two meta-analyses were per-formed; one based on PEFR (using FEV1 when PEFRwas not available) and the other one based on FEV1

(using PEFR when FEV1 was not available). Sincedifferent outcomes had to be combined, the standar-dised differences between means were used instead of the difference between means, despite the fact that thissummary limits the comparability and interpretationof the analysis [16]. In crossover studies the corre-lation coefficient was used to compute the variance of the mean difference. A weighted-combination of thecorrelation coefficients was used in those studies thatdid not provide enough information for the corre-lation coefficient to be recovered [17].

If available, the authors planned to use the indivi-dual patient data. Otherwise, only summaries wereused. A scanner and a technical drawing program

were used to estimate means and standard errors whenresults were displayed graphically. For those trialswith a series of repeated measurements, the maximummean change was chosen as a criterion to obtain asingle effect size for each trial.

To combine results from trials and estimate anoverall effect, the authors used the fixed-effects model[18], weighting each trial with the inverse of thevariance of the effect size. To assess whether there wasany evidence of statistical disparity in results acrosstrials a test of heterogeneity was performed [19].Given the low power of this test [20], possible sourcesof heterogeneity were also investigated. To explore thecontribution of each study, the authors partitioned the

sum of squares (of deviations between individualeffects and overall) from the test of hetereogeneity(QH) into two parts, one related to the between-subgroup (QB) differences and the other related to thewithin-subgroup differences (QW). When any hetero-geneity was not explained by any identifiable cause,the robustness of the overall treatment effect wasassessed using the random-effects model [21].

Results

Description of the clinical trials

Over 200 possible trials were identified but only 12satisfied the inclusion criteria [1, 2, 6, 22–30]. One trialpublished in a Chinese journal [22] could not berecovered as the authors were only able to retrieve apoor translation of the abstract summarising theresults, which did not provide the information neededto extract any useful summary data. The main authorwas contacted by mail but all attempts were unsuc-cessful. Consequently, only 11 studies were includedfor further analysis. The descriptive information foreach of the 11 trials is shown in table 1. Some trialshad missing information. No trial stated the form of randomisation used and how it was performed.

In table 2, the trials are classified by type of design,

847ACUPUNCTURE IN ASTHMA: META-ANALYSIS OF RCTS



either crossover or parallel groups, and by threesubcategories, according to whether the analysis hadbeen: unadjusted by baseline measures, adjusted bybaseline measures or adjusted by baseline measuresand reported as percentage change from the baselinevalue. Only two of the crossover trials considered thepossibility of a period effect [23, 24] and none tookaccount of the possibility of a carry-over effect.

Consequently, in the present analysis the authors didnot assume that either period or carry-over effectswere important in any of the trials. Another featurewhich varied across trials was whether or not asthmahad been induced; asthma was induced by means of exercise or some sort of bronchospasm in five out of eleven trials. Reports were poorly presented. In manycases p-values were not precisely stated and, instead,their relationship to a conventional significance levelwas given. In addition, means were often stated with-out an indication of variability (such as a standarderror or a plot of the means).

Estimation of the treatment effect

Figure 1 and table 3 show the standardised differ-ence between means from nine studies, choosing asingle outcome from each report. The overall treat-ment effect, estimated by the unbiased standardiseddifference between means (using the pooled corre-lation coefficient to estimate its variance), was d=0.12and its 95% confidence interval (CI) was (-0.07–0.31).This corresponds to an approximate difference inFEV1 means of 1.7 (95% CI -1.3–4.7). Figure 1 showsthese results with the corresponding plot for indivi-dual studies and the overall combined result.

The test of hetereogeneity was not statisticallysignificant at conventional levels (QH=12.54 witheight degress of freedom (df); p=0.13). However,when the contributions were examined, the study byDIAS et al. [6] presented the greatest contribution tothe hetereogeneity statistic. After removing the studyfrom the analysis, the test of heterogeneity showed aconsiderably lower value (QH=5.41 with seven df;p=0.61). The overall-effect size estimator without thistrial was 0.167, (95% CI -0.02–0.359). Under therandom-effects model this estimator was 0.12 (95%CI -0.14–0.38). This result was similar to the oneobtained in the fixed-effects model, although the CIwere slightly wider.

Studies with induced (provoked) and noninducedbronchoconstriction were analysed separately. Thesubset of studies in which bronchoconstriction wasprovoked gave an estimated effect of 0.3 (95% CI0.04–0.56). In addition, there was very little evidenceof hetereogeneity of results across these trials. Incontrast, the estimated effect for the studies wherebronchoconstriction was not provoked was -0.08 (95%CI -0.28–0.20). The test of heterogeneity for thesetrials approached conventional significance (QH=7.49with four df, p=0.11). This hetereogeneity was mainlydue to the study by DIAS et al. [6]. For this subset theeffect-size estimator under the random-effects modelwas -0.08 (95% CI -0.45–0.29). T

a b l e 1 . – D e s c r i p t i v e i n f o r m a t i o n f

o r t h e 1 1 t r i a l s

F i r s t a u t h o r

[ r e f n o . ]

S u b j e c t s n

F : M

A g e m e a n ¡

S D ( r a n g e )

D u r a t i o n o f a s t h m a

y r s m e

a n ( r a n g e )

B l i n d n e s s

D u r a t i o n o f

t r e a t m e n

t d a y s

E x c l u d e d

p a t i e n t s n

M e d i c a t i o n

O u t c o m e

m e a s u r e s r e p o r t e d

C H O W

[ 2 5 ]

1 6

1 1 ( 8 – 1 3 )

S B

2 1

0

Y e s

F E V 1

C H R I S T E N S E N [ 2 ]

1 7

1 1 : 6

3 2 . 2 ( 1 9 – 4 8 )

D B

3 5

0

N o t c l e a r

P E F R

D I A S [ 6 ]

2 0

9 : 1 1

4 1 . 6

¡ 1 5 ( 1 8 – 7 3 )

D B

2 1 . 7

0

Y e s

P E F R

F U N G [ 2 6 ]

1 9

( 9 – 1 3 . 5 )

S B

0

Y e s

F E V 1 / F V C / P E F R

L U U [ 2 7 ]

1 7

1 1 : 6

3 1 ¡ 1 1 ( 1 9 – 6 1 )

1 6

( 2 – 5 1 )

D B

1 1

0

N o t c l e a r

V C / P E F R

M I T C H E L L [ 1 ]

3 1

1 3 : 1 8

2 9 ( 1 5 – 5 5 )

2 3

( 5 – 3 8 )

S B

2 9

5

Y e s

P E F R

M O R T O N [ 2 8 ]

1 3

1 3 : 0

( 1 7 – 2 7 )

D B

3 5

0

Y e s

F E V 1

T A N D O N [ 2 4 ]

1 5

6 : 9

4 0 . 5 ( 1 9 – 5 7 )

( 5 – 3 4 )

D B

3 5

0

Y e s

P E F R / F E V 1 / F V C /

T A N D O N [ 2 9 ]

1 6

6 : 1 0

3 9 . 1

¡ 1 6 ( 1 1 – 6 0 )

1 9 . 6

2 ( 2 – 4 9 )

D B

3

0

Y e s

F E V 1

T A S H K I N [ 3 0 ]

1 2

1 0 : 2

4 3 ¡ 1 1 ( 1 6 – 6 4 )

2 1 . 8

3 ( 1 – 4 5 )

D B

3

0

Y e s

F V C / F E V 1 / F E F

T A S H K I N [ 2 3 ]

2 6

1 6 : 1 0

3 4 ¡ 1 8 . 5 9 ( 8 – 7 3 )

2 1 . 2

8 ( 2 – 5 9 )

D B

2 8

1

Y e s

F V C / F E V 1

M : m a l e ; F : f e m a l e ; S B : s i n g l e - b l i n d

; D B : d o u b l e b l i n d ; F E V 1 : f o r c e d e x p i r a t o r y v o l u m e i n o n e s e c o n d ; F E F : f o r c e d e x p i r a t o r y fl o w ; P E F R : p e a k e x p i r a t o r y fl o w r a t e ; V C :

v i t a l c a p a c i t y ; F V C : f o r c e d v i t a l c a p a c i t y . I n a l l 1 1 t r i a l s t h e r a n d o m i s a t i o

n m e t h o d w a s u n k n o w n .

848 J. MARTIN ET AL.



Evidence for publication bias

The small number of randomised trials and theirrelatively small size meant that there was little powerto assess the evidence for trials in favour of acupunc-ture to be preferentially reported. The funnel plotshown in figure 3 is difficult to assess. The authorsobserved that all the effects corresponding to samplesizes between 11 and 19 were positive and the twolargest trials had negative effects. One of the latter, the

study by DIAS et al. [6], is seen as a clear outlier. Thiscould suggest a small publication bias towards posi-tive results but both negative and positive results werepublished, reflecting the controversy of the subject(the Chinese trial for which data was unavailable[22], reported a positive effect of the acupuncture onasthma with 184 patients).

Discussion

Complementary therapies are of growing interest inhealthcare. Acupuncture is one of the most popular of the alternative therapies. Some of the attractions stem

from its long standing use in Chinese medicine and the

avoidance of the side-effects of more conventionaltreatments for asthma such as corticosteroids andb2-agonist sympathomimetics [8]. Overall effect sizes(standardised differences between means) of magni-tudes between 0.07 and 0.13 (95% CI -0.07–0.31) wereobtained. This corresponds to a largest plausibleincrease of FEV1 of y1.7, which may suggest thatacupuncture for asthma has little effect on the objec-tive outcomes considered. However, interestingly, asmall effect may have been observed for experimen-tally-induced bronchoconstriction. The differentaetiology may have resulted in this effect. Whethersubjective outcomes (quality of life in general orperceived breathlessness or anxiety) were affected

cannot reliably be assessed and may need testing inthe future.

It appears that there is no clear agreement onthe best method of conducting controlled trials of acupuncture in asthma in relation to the type of design, selected end-points and data analysis. Inexploring hetereogeneity of effects across studies, thedifferences between studies with varying experimental

Detrimental effect

-3 -2

Overall

3

MORTON [28]

TASKIN [30]

FUNG [26]

TASKIN [23]

CHOW [25]

210-1

Beneficial effect

b)

-40 -30

Overall

40

MORTON [28]

TASKIN [30]

FUNG [26]

TASKIN [23]

CHOW [25]

30100-10

a)

20-20

Fig. 2.– Forest plot for a) the mean difference and b) thestandardised mean difference of the five trials integrating theforced expiratory volume in one second. —: noninduced; – –:induced. Data are presented as standardised mean¡95% confi-

dence intervals; the size of the square indicates sample size of thestudy; diamond height represents the combined effects and widththe 95% confidence interval.

Table 4. – Standardised mean difference with FEV1 asoutcome measure

First author[ref no.]

Effect size

d Sd

CHOW [25] 4.425 4.008FUNG [26] 8.850 3.506MORTON [28] 1.710 3.376TASHKIN [23] -7.800 10.933TASHKIN [30] 2.100 2.482Overall 3.535 1.577CHOW [25] 0.257 0.254FUNG [26] 0.553 0.247MORTON [28] 0.130 0.279TASHKIN [23] -0.138 0.201TASHKIN [30] 0.225 0.292Overall 0.172 0.111

FEV1: forced expiratory volume in one second; d: standar-dised mean difference; Sd: standard error.

Sample size

0 5-1.5

30

0.0

-1.0

0.5

-0.5

1.0

25201510

1.5

E f f e c t s i z e

Fig. 3. – Funnel plot of the nine trials integrated.




designs (crossover, unpaired comparison, pairedcomparison) and with varying quality in the presenta-tion of the reports were assessed. Nevertheless, thesesources of heterogeneity did not seem to affect thefinal conclusion. First, the test of hetereogeneitygenerally gave low values although the results mightindicate a modest effect of acupuncture in cases wherethe asthma was provoked and it would be wise to be

aware of this hypothesis in future investigations.Secondly, both the fixed- and random-effects modelshave delivered similar results. The authors considerthat the fixed-effects model is appropriate in thismeta-analysis since the degree of heterogeneity is nottoo large and, as seen previously, it may be explained.Moreover, the comparative study using both thestandardised and nonstandardised mean differenceshows no contradictory conclusions, except for theend-point FEV1. For the five studies that measuredFEV1, the standardised mean difference was 0.17 (95%CI of -0.05–0.39) while the mean FEV1 difference was3.5 (95% CI 0.3–9.5). The discrepancy for this end-point may have resulted from the inclusion of the

study by TASHKIN et al. [23]: it contributes negativelywith a large weight in the estimation of the stand-ardised differences between means, whereas its con-tribution to the pooled estimator of the differencesbetween means is small. The reason for this is that thestudy by TASHKIN et al. [23] shows the largest samplesize with the largest variance. This is obviously anextraordinary situation and suggests that there may bea mistake in this paper with regard to the calculationof the estimate.

Given the difficulty of assessing the impact of mostbiases on the overall result, the limitations that mayhave affected the reliability of this meta-analysisshould be considered. There were factors that may

have introduced a bias against acupunture. First,there was no evidence to suggest that any of the trialsestimated the sample size a priori and all of them weretoo small to detect a modest effect of acupuncture. Anaim of meta-analysis is to increase the number of patients in order to detect such moderate effects withclinical significance [11] but the integrated number of patients in the present meta-analysis was still belowthe size given by a conventional power requirement.Secondly, placebo points used in asthma trials seem tobe active in pulmonary disease [8]. Thirdly, missinginformation was a considerable limitation. None of the papers presented sufficient information necessaryto estimate the effect size and indirect methods had

to be used that may have made the results moreconservative. The use of the approach by FOLLMAN

et al. [17] yields conservative results. Furthermore,some basic statistics were obtained by scanner fromgraphs and this would have added a measurementerror. However, the authors believe that this wasunlikely to be systematic in one direction. The effectof other factors such as missing information and poorreport writing are likely to be important but theirimpact is difficult to estimate. In relation to presenta-tion of reports, ROSENBERGER [31] presents a list of recommendations which will be essential informationif an updated and more powerful meta-analysis of acupuncture is to be performed in the future.

The assessment of the quality of the studies by theindependent assessors in this study is in generalagreement with other studies [3]. It shows that thereare several shortcomings in the studies of acupunctureon asthma, above all, in terms of sample size, effectsof prognostic variables, missing information and thebias against acupuncture introduced by the use of placebo points that may not be completely inactive. It

may still be possible to obtain the patient data toavoid the problem of missing information and to havethe option of using more complex analyses [32]. Itis important to locate the chinese report [22] bycontacting different libraries and through the internet.The current meta-analysis did not find evidence of theefficacy of acupuncture in the treatment of patientswith asthma, in agreement with the result presented inthe Cochrane Database of Systematic Reviews [13].However, it is important to mention that theintegrated sample size in both studies was still belowthe sample size given by a conventional powerrequirement. Hence, there is an obvious need todesign a large randomised clinical trial in which the

above limitations are addressed.

References

1. Mitchell P, Wells JE, Box PO. Acupuncture forchronic asthma: controlled trial with six monthsfollow-up. Am J Acupunct 1989; 17: 5–13.

2. Christensen PA, Laursen LC, Taudorf SC, SorensenC, Weeke B. Acupuncture and bronchial Asthma.Allergy 1984; 39: 379–385.

3. Kleijnen J, ter Riet G, Knipschild P. Acupuncture andasthma: a review of controlled trials. Thorax 1991; 46:799–802.

4. Takishima T, Mue S, Tamura G, Ishihara T,Watanabe K. The bronchodilating effect of acupunc-ture in patients with acute asthma. Ann Allergy 1982;48: 44–49.

5. Virsik K, Kristufek P, Bangha O, Urban S. The effectof acupuncture on pulmonary function in bronchialasthma. Prog Respir Res 1980; 14: 271–275.

6. Dias PLR, Subramaniam S, Lionel ND. Effects of acupuncture in bronchial asthma: preliminary com-munication. J R Sco Med 1982; 75: 245–248.

7. Ernst-E. Acupuncture–how effective is it really?Fortschiritte-der-Medizin 1998; 116: 20–26.

8. Jobst K A. A Critical analysis of acupuncture inpulmonary disease: efficacy and safety of the acu-puncture needle. J Altern Complement Med 1995; 1:

57–85.9. Lancet editorial. Acupuncture, asthma and breath-

lessness (editorial). Lancet 1986; 19: 1427–1428.10. Glass GV. Primary, secondary and meta-analysis of

research. Educ Res 1976; 5: 3–8.11. Peto R. Why do we need systematic overviews of

randomized trials? Stat Med 1987; 6: 233–240.12. Hennekens CH, Burning JE, Herbert PR. Implications

of overview of randomized trials. Stat Medicine 1987;6: 397–402.

13. Linde K, Jobst K, Panton J. Acupuncture for chronicasthma (Cochrane Review). The Cochrane Library,Issue 2. Oxford, Update Software, 2002.

14. Chalmers TC, Smith HJR, Blackburn B, et al. A

851ACUPUNCTURE IN ASTHMA: META-ANALYSIS OF RCTS



method for assessing the quality of a randomizedcontrol trial. Controlled Clin Trials 1981; 2: 31–49.

15. Rosenthal R. Meta-analytic procedures for socialresearch. Beverly Hills, Sage, 1984.

16. Hedges LV. "Directions for future methodology". In:Watcher K W, Straf ML, eds. The Future of Meta-analysis. New York, Russell Sage Foundation, 1990;pp. 3–10.

17. Follmann D, Elliott P, Suh I, Cutler J. Varianceimputation for overviews of clinical trials withcontinuous response. J Clin Epidemiol 1992; 4: 769– 773.

18. Berlin JA, Laird NM, Sacks HS, Chalmers T. Acomparison of statistical methods for combining eventrates from clinical trials. Stat Med 1989; 8: 141– 151.

19. Hedges LV, Olkin J. Statistical methods for meta-analysis. Orlando, Academic Press, 1985.

20. Thompson SG. Controversial in meta-anlysis: the caseof the trials of seum cholesterol reduction. StatMethods Med Res 1993; 2: 173–191.

21. Petitti DB. Meta-analysis, decision analysis and cost-effectiveness analysis. Methods for quantitative synth-

esis in medicine. New York, Oxford University Press,1994.

22. You YZ. Influence of application-impulse stimulationtreatment on airway9s reactivity with asthma. Chung-Kuo-Chung-Hsi-I-Chieh-Ho-Tsa-Chih 1992; 2: 86–87.

23. Tashkin DP, Kroening RJ, Bresler DE, Simmons M,Coulson AH, Kershnar H. A controlled trial of realand simulated acupuncture in the management of

chronic asthma. J Allergy Clin Inmunol 1985; 76: 855– 863.

24. Tandon MK, Soh PFT, Wood AT. Acupuncture forbronchial asthma? A double-blind crossover study.Med J Aust 1991; 154: 409–412.

25. Chow OKW, So SY, Lam WK, Yu DYC, Yeung CY.Effect of acupuncture on exercise-enduced asthma.Lung 1983; 161: 321–326.

26. Fung KP, Chow OKW, So SY. Attenuation of exercise-induced asthma by acupuncture. Lancet1986; 20: 1419–1421.

27. Luu M, Maillard D, Pradalier A, Boureau F. Controlespiromtrique dans la maladie asthmatique des effets dela puncture de points douloureux thoraciquesı́.Respiration 1985; 48: 340–345.

28. Morton AR, Fazio SM, Miller D. Efficacy of laser-acupuncture in the prevention of exercise-inducedasthma. Ann Allergy 1993; 70: 295–298.

29. Tandon MK, Soh PFT. Comparision of real andplacebo acupunture in histamine-induced asthma. Adouble-blind crossover study. Chest 1989; 96: 102–105.

30. Tashkin DP, Bresler DE, Kroening RJ, KerschnerH, Katz RL, Coulson A. Comparision of real

and simulated acupuncture and isoproterenol inmethacholine-induced asthma. Ann Allergy 1977; 39:379–387.

31. Rosenberger JL. Design of experiments for evaluatingfrozen sperm. J Androl 1990; 11: 89–96.

32. Stewart LA, Parmar MKB. Meta-analysis of theliterature or of individual patient data: is there adifference? Lancet 1993; 341: 418–422.


acupuntura e asma

Documents