Journal of Postgraduate Medicine Medknow Publications and Staff Society of Seth GS Medical College and KEM Hospital, Mumbai, India ISSN: 0022-3859 EISSN: 0972-2823 Vol. 51, Num. 2, 2005, pp. 152-152

Journal of Postgraduate Medicine, Vol. 51, No. 2, April-June, 2005, pp. 152

Letters

Mortality of percutaneous endoscopic gastrostomy in the UK

Leontiadis G.I., Moschos J., Cowper T., Kadis Sawas

Department of Gastroenterology, Queen Elizabeth Hospital, Sheriff Hill, Gateshead NE6 9SX

Correspondence Address: Department of Gastroenterology, Queen Elizabeth Hospital, Sheriff Hill, Gateshead NE6 9SX, skadis@doctors.org.uk

Code Number: jp05058

Sir,

We read with great interest the original article by Janes et al[1] and the accompanying Expert′s Comments by Lang[2] published in the March 2005 issue of J Postgrad Med . Janes et al found that 30-day mortality of percutaneous endoscopic gastrostomy (PEG) was 22% in a hospital in the UK during 2002 as opposed to a mortality of 10% 10 years earlier. This increase in mortality was attributed to a trend for less strict patient selection over the last few years.

We wish to support the findings of the above study by providing the results of a prospective audit we conducted in another hospital in the UK over the same period. All patients that received PEG in our hospital over a period of 38 months (May 1999 to June 2002) were followed up until reversion to oral feeding or until death. Seventy-three patients received PEG (mean age 71.7 years). The indications for PEG insertion were: cerebrovascular accident (CVA) 56.1%; non-CVA dementia 30.1%; other organic neurological diseases 11%; malignancy-associated anorexia 1.4%; neck cancer 1.4%.

Overall mortality over the follow-up period was 64.4%. In specific, early mortality (within 4 weeks) was 23.3%, mid-term mortality (4-8 weeks) 11% and late mortality (>8 weeks) 30.1%. Causes of death were: chest infection 83%; myocardial infarction 4%; congestive cardiac failure 6.4%; pulmonary embolism 2.1%; progression of pre-existing malignancy 2.1%. Mortality was 82% for patients with non-CVA dementia and 54% for patients with CVA, P = 0.05. No differences were found in survival and in complication rate between patients discharged home (11 patients) and those referred to nursing homes or long-term hospital care (26 patients).

In conclusion, 30-day mortality following PEG insertion is high in British hospitals. Our findings are remarkably similar to the findings of Janes et al .[1] Local and national guidelines on indications for PEG insertion need to be developed. Based on these guidelines a local multidisciplinary team should assess individually each request for PEG insertion.

REFERENCES

Copyright 2005 - Journal of Postgraduate Medicine