Key points from the evidence

1. Systematic reviews of RAB combined results from studies assessing different RAB platforms and did not adequately report patient characteristics or the appraisal of primary study quality. Many of the primary studies recruited small numbers of highly selected patients and few studies directly compared RAB with other diagnostic modalities.

2. A systematic review with meta-analysis (27 studies, n=2,315) reported a diagnostic yield (calculated based on intermediate criteria) of 86.6% (95% confidence interval [CI] 83.7% to 89.2%) for RAB in patients with peripheral lung nodules and a suspicion of cancer.² In the meta-analysis, pneumothorax occurred in 2.0% of patients having RAB (95% CI 1.3% to 2.7%).

3. A network meta-analysis (NMA; 37 studies, n=4,285) found no statistically significant differences in diagnostic yield (intermediate criteria) between RAB and CT-guided transthoracic bronchoscopy (CTTB), endobronchial ultrasound (EBUS), virtual bronchoscopic navigation (VBN) or electromagnetic navigation bronchoscopy (ENB) in patients with peripheral pulmonary lesions.³ Only three studies (n=242) contributed data on patients undergoing RAB in these comparisons.

4. Five primary studies compared RAB using the Ion™ platform with other diagnostic modalities. The patients in each group across these five studies may be highly selective because individuals eligible for RAB may be unsuitable for the comparator diagnostic modalities because of the size or anatomical location of their pulmonary nodules.

• A cluster randomised trial (n=411) comparing RAB with ENB in patients with
  peripheral pulmonary lesions found no statistically significant difference between the two modalities when diagnostic yield was calculated using strict criteria (odds ratio [OR] 1.23, 95% CI 0.73 to 2.07).⁴ Pneumothorax was reported in 1.97% of patients in the RAB group and 2.88% of patients in the ENB group (OR 0.39, 95% CI 0.12 to 1.25).
• A retrospective cohort study (n=92) comparing RAB with ENB in patients with
  pulmonary nodules reported a statistically significantly higher diagnostic yield (strict criteria) from RAB (89% versus 66%, p<0.001).⁵ Pneumothorax was reported in 1.8% of patients in the RAB group and 2.1% of patients in the ENB group.
• A second retrospective cohort study (n=116) comparing RAB with ENB in patients with pulmonary nodules also reported statistically significantly higher diagnostic yield (intermediate criteria) from RAB (86.1% versus 49.5%, p<0.0001).⁶ No complications were reported in either group in this study.
• A retrospective cohort study (n=59) comparing RAB with VBN in patients with
  peripheral pulmonary nodules reported that diagnostic yield (intermediate criteria)
  was 90% for RAB compared with 69% for VBN (p=0.045).⁷ Pneumothorax was reported in 6.9% of patients in the VBN group and no patients in the RAB group
  (p=0.237).
• A retrospective cohort study (n=296) comparing RAB with CTTB in patients with
  pulmonary nodules found no statistically significant difference in diagnostic yield
  between the two modalities using strict criteria (86.48% versus 89.18%, p=0.594).⁸ Pneumothorax was reported in 3.38% of patients in the RAB group compared with 44.45% in the CTTB group (p<0.001).

5. A retrospective cohort study (n=162) compared the diagnostic yield (strict criteria) between the Ion™ and Monarch™ RAB platforms in patients with pulmonary lesions.⁹ Diagnostic yield was statistically significantly higher with the Ion™ platform compared with the Monarch™ platform (84.2% versus 71.0%, p=0.003). There were no statistically significant differences in complication rates between the two platforms.

6. A patient organisation submission from the Roy Castle Lung Cancer Foundation reported that:

• lung cancer can have life-limiting symptoms with serious effects on patients’ physical and emotional well-being, quality of life and ability to perform daily tasks
• there is an unmet need among patients with small peripheral pulmonary lesions that are not suitable for biopsy using conventional techniques
• patients with lung lesions who are placed on a ‘watch and wait’ list, which may be for as long as 3 years, experience anxiety and may undergo unnecessary invasive
  treatments for lung cancer
• patients having RAB need a general anaesthetic rather than the sedation used for a
  standard bronchoscopy, but some patients prefer general anaesthesia
• RAB offers patients hope of a definitive diagnosis and improved outcomes.

7. Lung cancer incidence and mortality are approximately three times higher among people living in the most deprived areas of Scotland compared with people living in the least deprived areas.¹⁰ Ethnicity, sex, age and smoking status were each statistically significantly associated with people having late-stage (III or IV) lung cancer at diagnosis.¹¹

8. Organisational issues relating to the introduction of RAB include a recognised learning curve for clinicians, a need for training on using the RAB platforms and the need for infrastructure to manage cleaning and monitoring of reusable RAB bronchoscopes.^12-17

9. We performed an exploratory economic analysis comparing RAB with usual care, which suggested that RAB (Ion™ platform) is expected to be cost effective. The analysis assumes that a proportion of patients who would otherwise be on a ‘watch and wait’ list would be able to get an earlier biopsy-confirmed diagnosis with RAB, which would lead to earlier treatment and improved 5-year survival rates. The results of our analysis were consistent across sensitivity analyses including the time horizon and the assumed proportion of patients diagnosed with early-stage lung cancer in the RAB arm of the model. A lack of comparative clinical data means our analysis should be viewed as indicative.