Thirteen RCTs were included (n=3,473 patients randomised; range 84 to 509). Five trials reported no information on specific symptom outcomes.
Five trials scored 2 out of 5 points on the Jadad scale; the other trials scored 3 points.
There was no significant difference in complete response between high- and low-dose radiotherapy for haemoptysis (five RCTs), cough (five RCTs), or chest pain (five RCTs).
High-dose radiotherapy was associated with a statistically significant increase in the proportion of patients with some degree of improvement in overall symptom burden compared with low-dose radiotherapy (61% versus 56%; RR 0.86, 95% CI 0.78 to 0.95; four RCTs) but there was no significant difference between treatments in complete response (four RCTs).
Survival (13 studies)
High-dose radiotherapy was associated with a statistically significant increase in survival at one year (discrepancy in year of survival reported in text and figures, but appeared to be one-year survival from forest plot) compared to low-dose radiotherapy (26.5% versus 21.7%, p=0.002), but there was no significant difference between treatments at two year survival. Sensitivity analysis suggested that the probability of improved survival was greater with schedules of 35Gy10 biologically effective dose (BED) compared with lower-dose schedules (data presented graphically).
There was no significant heterogeneity for the above analyses, except for the chest pain analysis (I2=63%).
High-dose radiotherapy was associated with a statistically significant increase in physician-assessed dysphagia compared with low-dose radiotherapy (21% versus 15%; p=0.01). Results for patient-assessed dysphagia were not consistent and there was significant heterogeneity (I2=79%; six RCTs).
There was no significant difference between high-dose and low-dose radiotherapy in myelopathy (ten RCTs with three confirmed cases) or pneumonitis (six RCTs; I2=60.8%).
There was no significant difference between high-dose and low-dose radiotherapy in the re-irradiation rate (five RCTs; I2=54.3%).