A large prospective audit of morbidity and mortality associated with feeding gastrostomies in the community.

BACKGROUND & AIMS: Morbidity after 30 days and morbidity after 1 year from gastrostomy placement is poorly characterised as patients are discharged into the community. We prospectively recorded morbidity and mortality associated with gastrostomy placement over a five year period. PATIENTS AND METHODS: Community dietitians regularly reviewed all patients with a gastrostomy after hospital discharge, prospectively recording morbidity and mortality between 2008 and 2012. In addition hospital databases and case notes were examined. Recorded morbidity included insertion site infection, leakage, over granulation, haemorrhage and buried bumper. RESULTS: The commonest indication for PEG placement was following an acute cerebral injury. There were no deaths and few complications directly related to gastrostomy insertion in 350 patients. We collected a total of 571 years of gastrostomy data. Mortality within 30 days was predominantly from a respiratory infection. 30 day, 3 and 12 month cumulative mortality (and morbidity) were 8% (2%), 16% (10%) and 35% (15%) respectively. 38% of patients required treatment for an insertion site infection with 70% of these having further infections. Overall there was a site infection every 2.1 years a gastrostomy was in situ. Complications such as buried bumpers (5(1.4%)), persistent fistulas (0) and over granulation (7(2%)) were rare. Few gastrostomies required replacement (11%). CONCLUSION: We have demonstrated reassuringly low rates of gastrostomy-associated morbidity and mortality. There was no direct mortality. The greatest morbidity resulted from gastrostomy-site infection.

Pain and Anxiety Experienced by Patients Following Placement of a Percutaneous Endoscopic Gastrostomy.

BACKGROUND: Abdominal pain following percutaneous endoscopic gastrostomy (PEG) placement is a recognized complication. However, the prevalence and degree of severity of pain are poorly characterized. We assessed abdominal pain and anxiety levels associated with PEG placement in communicative and noncommunicative patients. METHODS: A prospective questionnaire assessed patients' anxiety and abdominal pain 1 hour before, 1 hour after, and 24 hours after PEG placement using 11-point Likert-type scales. Patients were followed up until pain had resolved. Procedural data, analgesia requirements, and complications were recorded. For analysis, patients were divided into 2 groups: communicative (able to self-assess) and noncommunicative (clinician assessed). RESULTS: Seventy consecutive patients were assessed. Of the 49 self-assessed patients, 11 (22%) reported immediate pain, 32 (65%) reported pain at 1 hour (24 mild, 5 moderate, 3 severe), and 40 (82%) reported pain at 24 hours. Pain most commonly lasted between 24 and 48 hours (25 patients). Of the 21 clinician-assessed patients, only 1 was deemed to have pain, and this was at 24 hours. Four (6%) patients were admitted with pain. There was no relationship between preplacement anxiety scores and postplacement pain scores. DISCUSSION: Abdominal pain after PEG placement pain is common but resolved by 48 hours in most patients. In patients able to communicate, clinicians scored pain lower compared with patients' scores. It is likely that pain is not identified in patients unable to communicate. Patients need to be better informed about the possibility of postprocedural pain and routinely offered access to appropriate analgesia.

An examination of the relationship between the endoscopic appearance of duodenitis and the histological findings in patients with epigastric pain.

The endoscopic appearance of duodenitis is a common finding in patients undergoing endoscopy because of epigastric pain however, the relationship of the visual findings to histology is poorly defined. We set out to ascertain if there was a correlation between the endoscopic and histological appearances of the duodenal mucosa. Consecutive patients with epigastric pain referred for diagnostic gastroduodenoscopy were studied. The visual appearances of 'duodenitis' (erythema, erosions and sub-epithelial haemorrhage) were reported independently by two endoscopists. Duodenal biopsies were taken and assessed for: neutrophil infiltrate, mononuclear infiltrate, gastric metaplasia, villous atrophy and a breach in the mucosa. H pylori status was determined. Of the 93 patients with endoscopic features of duodenitis an increase in histological markers of inflammation was found in 75 (81%). However, histological inflammation was absent or minimal in 68 (73%). Conversely, biopsies from normal-looking mucosa revealed histological evidence of inflammation in 26 (27%). For patients with the endoscopic features of duodenitis the positive & negative predictive value for neutrophilic infiltrate was 39% and 98% respectively. Biopsies from erosions confirmed a breach in the mucosa in only 2 of 40 patients. Neutrophilic infiltrate occurred with NSAI ingestion and infection with H pylori. The endoscopic appearance of the duodenal mucosa is unreliable in determining the presence of histological inflammation. The endoscopic appearance of 'erosions' is not usually associated with a mucosal breach.