Leahy 1985.
| Methods | RCT (two recruiting centres) of intrapleural Corynebacterium parvum and tetracycline for pleurodesis of malignant pleural effusion (UK) | |
| Participants | Inclusion: histologically or cytologically proven MPE Exclusion: participants on chemotherapy; participants receiving treatment with steroids 36 patients randomised. |
|
| Interventions | Effusion aspirated to dryness prior to administering study agent. After agent instilled, the participants moved from side to side for six hours. If the participant had symptomatic recurrence of the effusion within a month, the allocated treatment was repeated Tetracycline group: 500 mg in 20 ml saline given intrapleurally. The tetracycline was administered via an intercostal tube at one centre and with needle drainage at the other centre C. parvum group: 7 mg in 20 ml saline intrapleurally through a needle, after the effusion was drained to dryness |
|
| Outcomes | Symptomatic recurrence of pleural effusion one month after the last dose Side effects (pain, fever, nausea and vomiting, rash) |
|
| Notes | People with trapped lung eligible for trial entry The side effects were reported per procedure rather than per patient For this review, if participants had a successful pleurodesis after the second dose of study agent, these were included in the analysis as a success. For the tetracycline group, the results from the two administration methods were combined for the purposes of analysis Included in network meta‐analysis for pleurodesis efficacy, fever, pain and mortality. |
|
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Computer randomisation |
| Allocation concealment (selection bias) | Low risk | Computer randomisation |
| Blinding of participants and personnel (performance bias) All outcomes | Unclear risk | Blinding not mentioned in the paper. Both drugs reconstituted in 20 ml saline |
| Blinding of outcome assessment (detection bias) All outcomes | Unclear risk | If study was unblinded, reporting of side effects, symptomatic pleural fluid re‐accumulation could be biased |
| Incomplete outcome data (attrition bias) All outcomes | Low risk | Participants excluded from analysis if died prior to one month, but the numbers were small and fairly well balanced between the groups (1/17 in C. parvum group; 3/19 in tetracycline group ie 11% LTFU in total) |
| Selective reporting (reporting bias) | Low risk | Thorough reporting of toxicity |
| Other bias | Low risk | No other biases identified |