Table 2.
Key Findings of a Meta-Analysis of Inhaled Antibiotics for the Treatment of Adult Patients With Bronchiectasis
| Outcomes | Anticipated Absolute Effects (95% CI)a |
Relative Effect (95% CI) | No. of Participants (Studies) | Certainty of Evidenceb | Comments | |
|---|---|---|---|---|---|---|
| Risk With Placebo | Risk With Inhaled Antibiotics | |||||
| Frequency of exacerbations | 93 per 100 | 73 per 100 (63-84) | RR, 0.78 (0.68-0.90) | 2,930 (12 RCTs) | ⨁⨁⨁⨁ highc | Inhaled antibiotics result in a slight reduction in frequency of exacerbations. |
| Frequency of severe exacerbations | 29 per 100 | 14 per 100 (9-21) | RR, 0.48 (0.31-0.74) | 828 (7 RCTs) | ⨁⨁⨁◯ moderated | Inhaled antibiotics likely reduce frequency of severe exacerbations. |
| Time to first exacerbation | 0 per 100 | NaN per 100 (— to —) | HR, 0.80 (0.68-0.94) | 2,725 (12 RCTs) | ⨁⨁◯◯ lowef | Inhaled antibiotics may increase time to first exacerbation slightly. |
| Change from baseline QOL-B RSS score (scale, 0-100) | Mean change from baseline QoL-B RSS score was 4.9 pointsg | MD 2.37 points higher (0.44 higher-4.31 higher) | NA | 2,315 (11 RCTs) | ⨁⨁⨁⨁ highch | Inhaled antibiotics probably result in a slight increase in change from baseline QOL-B RSS score, without reaching the minimal clinical important difference of 8 points. |
| Change from baseline SGRQ score (scale, 0-100) | Mean change from baseline SGRQ score was –0.62 pointsi | MD 3.13 points lower (5.93 lower-0.32 lower) | NA | 1,338 (10 RCTs) | ⨁◯◯◯ very lowjk1mn | Inhaled antibiotics may reduce or have little to no effect on change from baseline SGRQ score, but the evidence is uncertain. |
| Isolates with resistant MIC at the end of treatment | 9 per 100 | 17 per 100 (14-21) | Risk ratio, 1.86 (1.51-2.30) | 2,619 (17 RCTs) | ⨁⨁⨁◯ moderaten | Inhaled antibiotics likely result in an increase in isolates with resistant MIC at the end of treatment. |
| No. of patients reporting TEAE | 80 per 100 | 80 per 100 (75-84) | OR, 0.99 (0.75-1.30) | 3,207 (15 RCTs) | ⨁⨁⨁◯ moderateao | Inhaled antibiotics likely result in little to no difference in number of patients reporting TEAEs. |
HR = hazard ratio; MD = mean difference; MIC = minimum inhibitory concentration; NA = not applicable; NaN = not a number; QOL-B = Quality of Life Questionnaire-Bronchiectasis; RCT = randomized controlled trial; RR = rate ratio; RSS = respiratory symptoms scale; SGRQ = St. George Respiratory Questionnaire; TEAE = treatment-emergent adverse event.
The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
Grading of Recommendations, Assessment, Development, and Evaluations Working Group grades of evidence: high certainty = we are very confident that the true effect lies close to that of the estimate of the effect; moderate certainty = we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different; low certainty = our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect; very low certainty = we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.
The heterogeneity can be attributed to the AIR-BX trials. Explanation is provided in the text. P value for heterogeneity remains > .05.
Weight average of studies with high risk of bias is 42.4%, with the rest of the studies being low risk of bias.
Although most studies are low risk of bias, no significant benefit exists after excluding the studies with high risk of bias.
Lack of mean or median values in some of the studies because of low no. of events, making it difficult to estimate effect size correctly.
Based on change from baseline (placebo) in studies of > 48-wk duration.
Small CI. Although it does not reach the minimal clinical difference (which was defined using the AIR-BX trials), the results are solid with good studies of low risk of bias. For this reason, we decided not to downgrade, because 2.73 points can be relevant in our population.
Based on change from baseline (placebo) studies with SGRQ as an outcome.
Unreliable methods in one study.
Concern in patient and personnel masking and incomplete outcome data.
Nonoverlapping CI; studies of low risk of bias with opposite results; P < .05 for heterogeneity.
Wide CI overlapping with the minimal clinical difference of 4 points.
Assymetrical funnel plot, probably exacerbating effects.
CI includes benefit and harm.