Article Text

Download PDFPDF

Risk disclosure prior to bronchoscopy
  1. Stephen M Bianchi,
  2. Shironjit Saha,
  3. Paul Anderson
  1. Northern General Hospital, Sheffield, UK
  1. Correspondence to Dr Stephen M Bianchi, Chest Clinic, Northern General Hospital, Sheffield S5 7AU, UK; stephen.bianchi{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

We read with interest the article by Uzbeck et al1 demonstrating increased patient anxiety upon receiving detailed information regarding complications of bronchoscopy, in addition to the letter by Echavarria et al2 documenting the consenting practices of 33 respiratory physicians in the north east of England. A wide variation in practice is identified. The General Medical Council guidance for doctors relating to consent3 indicates that a physician ‘must tell patients if an investigation… might result in a serious adverse outcome, even if the likelihood is very small’. An adverse outcome is defined as one ‘resulting in death, permanent or long term physical disability or disfigurement, medium or long term pain, or admission to hospital’. The guidance also indicates that less serious side-effects or complications should be communicated if they occur frequently.

In reviewing the Uzbeck paper and their local practice, Echavarria et al2 feel that the appropriate balance between the disclosure of relevant risks and patient anxiety is one in which a high risk disclosure is advisable. However, it can be argued that this balance can only be struck in the knowledge of local and even personal bronchoscopic practice and performance, and that it is unethical to advise patients of risks that are neither serious nor common.

Many hospitals now utilise computer software to record and analyse bronchoscopic findings and outcomes. These software packages frequently allow the recording of relevant complications or side-effects with free text areas for the documentation of less frequently encountered, but clinically relevant, events. In a review of 1261 fibre-optic bronchoscopies, recorded on InfoFlex5 software (CIMS, Hertfordshire, UK) at Sheffield Teaching Hospitals NHS Foundation Trust over a 24-month period (1 December 2007 to 1 December 2009) 86.5% of patients did not encounter complications of sufficient severity for a record to be created. Data were unavailable for 4.2% of patients; 9.4% had documented bleeding and 2% were noted to have undergone desaturation requiring premature termination of the procedure or considered to be clinically relevant or unexpected. This latter group included those developing pneumothorax following transbronchial biopsy. No deaths were encountered in the patient cohort despite approximately 10% of the patient group undergoing interventional bronchoscopic procedures including laser therapy or stenting.

The quality of statistical output from any database is dependent on the quality of data entry and the consistency between clinicians in identifying and recording relevant complications. For instance, two clinicians may differ in their assessment of a ‘clinically relevant’ desaturation or bleeding event, and may therefore enter different datasets for a similar clinical experience, thus confounding analysis. However, for major complications, such as intraprocedural death or large volume haemorrhage, this is less likely to occur.

Our data would suggest that, in an appropriately selected patient group, administered to by experienced medical staff with appropriate training and expertise, major risks for bronchoscopy are infrequent and rarely life-threatening. On this basis the more limited information disclosure outlined by Uzbeck et al,1 with resultant lower levels of anxiety for patients may be more appropriate.



  • Linked articles 151761.

  • Competing interests None.

  • Provenance and peer review Not commissioned; not externally peer reviewed.

Linked Articles