A Clinical Risk Case Study – Failure to recognise a Fistula in the Oesophagus

A Clinical Risk Case Study – Failure to recognise a Fistula in the Oesophagus


A complaint was raised that there was a failure to recognise a fistula in the patient’s oesophagus, resulting in food particles spilling over into the chest cavity causing the patient’s death.

The Scene

Following a diagnosis of adenocarcinoma of the oesophagus, the patient underwent a course of chemotherapy, radiotherapy and the insertion of an oesophageal stent.

A few months after starting radiotherapy, the Deceased attended A&E with difficulty in eating. A CT scan demonstrated a narrowing within the oesophagus above the stent which was presumed to be secondary to oedema caused by the radiotherapy. The Deceased was commenced on steroid therapy, later diagnosed with aspiration pneumonia and then discharged.

The patient re-attended A&E with ongoing difficulty eating. A scan was undertaken which demonstrated a perforation of the distal oesophagus and leakage of oesophageal contents into the left pleural space, however, this was not shared or followed up with the clinical team and the patient was discharged with a course of antibiotics for a suspected chest infection.

A few days later the patient again re-attended where it was now suspected that he was suffering from a blocked stent and aspiration. Following this, a decision was made to carry out an endoscopy.

An endoscopy was later carried out which found a large fistula and it was only now that earlier CT scan results from the previous admission to A&E were reviewed and perforation suspected, resulting in the patient being nil by mouth and the fistula being sealed with a stent. Unfortunately, the Deceased’s condition deteriorated and he sadly passed away.

Our Independent Clinical Findings

Our independent review identified a clear missed opportunity by the treating team to review earlier scans which would have led to an earlier diagnosis, earlier nil-by-mouth, earlier stenting and avoidance of deterioration in the patient’s condition, terminal aspiration failure and death.

The independent review found that when making a diagnosis of pneumonia following readmission, the clinical team failed to review results from recent scans (which were available) resulting in an unnecessary delay of 16 days during which the Deceased had not been kept nil by mouth and the opportunity for more food to transfer from the oesophagus into the chest increased.

Furthermore, had the team looking after the Deceased from his re-admission onwards been made aware of the perforation then the diagnostic gastroscopy may not have been undertaken. In turn, had this not been undertaken then the perforation may not have enlarged. Instead with the knowledge from the CT scan the team should have arranged early repeat stenting to seal the new perforation.

How to prevent recurrence and improve patient safety

Where a patient with cancer is admitted increasingly unwell than all steps necessary to gain relevant information upon which to base treatment decisions should have been taken, this, unfortunately, did not occur in this case.

Obtaining a copy of the initial CT report would have taken a competent doctor or medical secretary a few minutes by phoning the radiology department. Had this been done then the following would have occurred:

  1. The claimant would have been placed on nil by mouth immediately

  2. There would have been a referral for immediate stenting

To Conclude

When considering what may be aggravating an existing diagnosis of carcinoma of the oesophagus which has been recently treated with complex chemotherapy and radiotherapy then earlier CT imaging reports should be reviewed. In this case, CT results would have immediately changed the management that was instituted at the Hospital preventing the patient’s deterioration and death.

TMLEP’s independent review found that it was difficult to understand why in this case the medical team did not request or review the result of that scan despite being prompted to do so by the family of the Deceased.

The failing in this case had a material effect in causing suffering and on the balance of probabilities prematurely ending the life of the Deceased who was terminally ill.


  • TMLEP Clinical Risk and Patient Safety Publishing Group and Dr Roger Prudham MBBS FRCP(UK) Dip Health Informatics. (2017). A Clinical Risk Case Study – Failure to recognise a Fistula in the Oesophagus. TMLEP Clinical Risk Case Studies. 1 (3), 1.

Important Note

This article is intended to raise awareness to clinical risk issues in an effort to reduce incidence recurrence and improve patient safety. This is not intended to be relied upon as advice. Facts have been altered to ensure this case is non-identifiable, albeit clinical learning points remain applicable. To request an independent clinical review, please contact admin@tmlep.com or call +44 (0) 203 355 9796.