Introduction
Percutaneous endoscopic gastrostomy (PEG) insertion is a well-established and widely used intervention to maintain enteral nutrition in individuals with unsafe oral intake and a functionally intact gut. Since its introduction in 1980 it is now performed globally but is associated with significant mortality and morbidity.1,2 There has been recent helpful guidance around oral feeding difficulties and ethical dilemmas.3,4
Aims
The aims of this retrospective study were to identify markers predictive of increased mortality in patients having a PEG for the first time, and to assess if use of sedation with benzodiazepines and local anaesthetic throat spray was associated with higher mortality. We also aimed to determine if mortality rates have improved in our unit.
Methods
All procedures named, proposed and performed that contained the term PEG were identified from our endoscopy database in 2008. Demographic data, indication, blood results and mortality data were obtained from the Patient Administration System (PAS), Integrated Clinical Environment (ICE) reporting systems and Dietetic Continuum Database.
Results
In total, 74 patients had a PEG inserted for the first time in 2008. Of these, 45 (61%) were males and 29 (39%) females with a mean age of 64 years. The indication was stroke/neurological in 38%, nutritional/feeding problems 27%, neoplasm 19% and others 16%. The 30-day and three-month mortality was 8% and 26% respectively. An elevated platelet count was significantly associated with higher 30-day and three-month mortality (p<0.05) while albumin was only significantly associated with a higher three-month mortality (p<0.001) using the Fisher exact test. Age was significantly associated with higher mortality at 30 days and three months using the Mann–Whitney U test (p<0.05). Otherwise, gender, abnormal haemoglobin, white cell count, prothrombin time, creatinine and urea were not associated with mortality using the χ 2 or Fisher exact test. Higher than average doses of midazolam, fentanyl, combination midazolam and fentanyl or midazolam and local anaesthetic throat spray were not associated with increased mortality.
Conclusions
Mortality rates have improved from 2004 to 2008 from 24% to 8% at 30 days and 41% to 26% at three months.5 The mean age of patients having a PEG in 2004 was 75 compared to 64 in 2008. Factors predictive of poor outcome include age, platelet count and albumin. There was no increased mortality when benzodiazepines were used with local anaesthetic throat spray. This lack of association is likely due to improved patient selection and lower doses of sedation used. Ethical consideration, patient preference and factors predictive of poor outcome should all be taken into account when making decisions about PEGs.4
Footnotes
Letters not directly related to articles published in Clinical Medicine and presenting unpublished original data should be submitted for publication in this section. Clinical and scientific letters should not exceed 500 words and may include one table and up to five references.
- Royal College of Physicians
Related Articles
Cited By...
- No citing articles found.