JRSM >
Volume 97, Number 12 >
Pp. 588-589
1 Department of General Surgery, Northampton General Hospital, Northampton NN1
5B•, UK
2 Department of Histopathology, Northampton General Hospital, Northampton NN1
5B•, UK
Correspondence to: Charles Bailey, c/o Mr Hunter's Secretary, Northampton General Hospital E-mail: emhbailey{at}doctors.org.uk
In an elderly patient the most likely cause of recurrent laryngeal nerve palsy and dysphagia is a malignant tumour. Among the other conditions to be considered is tuberculosis.
CASE HISTORY
A woman aged 89 consulted her general practitioner after four weeks of hoarseness, sore throat and difficulty with swallowing. She also reported pain in the right side of the neck and recent breathlessness. Her sister had died from laryngeal cancer at age 93 and she was concerned that she might have the same condition. On referral to the ear, nose and throat department she was found to have a right recurrent laryngeal nerve palsy. Subsequent oesophagogastroduodenoscopy revealed a 3 cm ulcerated lesion in the upper oesophagus with what was considered to be an associated tracheo-oesophageal fistula (Figure 1). The provisional diagnosis was oesophageal squamous carcinoma and her dysphagia was relieved by a 12 cm stent placed under endoscopic and X-ray guidance. Biopsies of the ulcer showed fragments of oesophageal mucosa with a squamous epithelial covering. Focal ulceration was associated with granulation tissue but there was no evidence of malignancy. A chest X-ray showed no abnormality apart from slight cardiomegaly. One week later she developed severe back pain and died. At necropsy the cause of death was found to be a massive gastrointestinal bleed from an aorto-oesophageal fistula. Although the macroscopic appearance was typical of a squamous carcinoma, histological examination revealed the fistula to be caused by oesophageal tuberculosis with erosion into the aorta. The patient had lived in England all her life and had not recently travelled abroad.
|
COMMENT
Tuberculosis can affect the oesophagus either indirectly by ulceration from infected mediastinal or hilar lymph nodes or directly as part of a miliary spread. Oesophageal tuberculosis poses a considerable diagnostic challenge since the presentation and macroscopic appearances are very similar to those of carcinoma and only 20% have evidence of tuberculosis on chest X-ray.1 In the present case, even when malignancy had been excluded by biopsy, there were other possibilities such as peptic ulceration and Crohn's disease. In a review of 54 cases only 12 had the diagnosis made on mucosal biopsy alone;1 in others diagnosis depended on lymph node biopsies or the finding of acid-fast bacilli in sputum. We suspect that our patient had contracted tuberculosis as a child and it had been dormant for much of her life.
We have found only four previous reports of aortooesophageal fistula resulting from tuberculosis—two cases from Africa2 and one each from Singapore3 and England4. In all four, as here, the diagnosis was made post mortem.
REFERENCES
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
