PT  - JOURNAL ARTICLE
AU  - Hugh Ip
AU  - Chris Kosky
AU  - Anne Collett
TI  - Role of bronchoscopy in non-HIV immunocompromised patients
DP  - 2011 Sep 01
TA  - European Respiratory Journal
PG  - p2519
VI  - 38
IP  - Suppl 55
4099  - http://erj.ersjournals.com/content/38/Suppl_55/p2519.short
4100  - http://erj.ersjournals.com/content/38/Suppl_55/p2519.full
SO  - Eur Respir J2011 Sep 01; 38
AB  - Background: Non-HIV immuno-compromised patients with suspected chest sepsis commonly undergo bronchoscopy to aid microbiological diagnosis.Aims: Characterise non-HIV immuno-compromised patients undergoing bronchoscopy at a London hospital. To assess the diagnostic utility of this procedure with a focus on results leading to treatment changes.Methods: Clinical records of all patients undergoing bronchoscopies over a 14-month period were retrospectively reviewed. The diagnostic yield was calculated using the total number of bronchoscopies as the denominator. A positive result was defined as any positive culture (viral, fungal, bacterial) or diagnostic pathology. Using the clinical notes and prescription charts the percentage of treatment changes following bronchoscopy results was calculated.Results: 43 bronchoscopies were carried out in this group. Underlying diagnoses were haematological malignancy (27), organ transplant (10), solid organ tumour (2), rheumatological (4). The most common indication was suspected infection (76.7%); fungal infection was suspected in 46.5%.The overall diagnostic yield was 51.2% (95% CI, 36.1-66.3%). Many of the positive samples were for virus (10) followed by positive bacterial cultures (8). Few samples were positive for fungus (2). Therapy was changed after 18.6% (95% CI, 6.8-30.4%) of all bronchoscopies. The total complication rate was 4.6% (95% CI, -1.7-11.1%); both cases due to controllable bleeding.Conclusions: The most common underlying diagnoses were haematological malignancies. Fungal chest infection was frequently suspected, but the fungal organism yield was low; treatment decisions were based on clinical and radiological findings. Total diagnostic yield was similar to published data.