Article Text


Nutrition posters 313–320

Statistics from


V. Christopher, S.D. Hearing, J. Smithson, S. Hughes. Dept Gastroenterology, Southmead Hospital, Bristol, UK

Percutaneous gastrostomy (PEG)insertion is associated with a significant morbidity. Recent studies suggest that antibiotics given at the time of PEG insertion may reduce complications, in particular soft tissue infection. A previous audit of all PEGs from Feb 1998 to July 1999 performed in our unit revealed a high rate of PEG site infection, although most cases were not severe (14.2%). A policy of prophylactic antibiotics (Ampicillin 1g IV, Gentamicin 80mg IV) administered at the time of PEG insertion was instituted and complication rate was reaudited.

Methods: Retrospective case note review from Jan 2000 to April 2001.

Results: 79 PEGs performed, 51 (65%) notes were available for review. Mean age 70.8yrs (SD 15.3±, range 36–98), M:F =27:25. Indications for PEG: CVA/CNS disorders =27, pre-op ENT malignancy =12, nutrition support =12. In both audits patient characteristics were similar and differed only in antibiotic use. Audit 1: 4/42 (9%) concurrent antibiotics, 38/42 (91%) none given. Audit 2: 37/51(73%) antibiotics given, 14/51 (27%) none. See table.

Abstract 313

Conclusions: On basis of this data, there was no statistically significant reduction wound sepsis, or mortality from the routine use of antibiotics at the time of PEG insertion. However, not all case notes were available and this reduced the number of subjects included in this audit. This reduction in numbers may mean that non-significance does not exclude a clinically important difference in PEG site, wound infection. Audit of complications following PEG insertion is continuing.


A. Hurst, A. Bassi, R. Nicholson, P. O'Toole. University Hospital Aintree (UHA), Liverpool, UK

Introduction: A growing number of patients are being discharged into the community with gastrostomy feeding tubes. Minor complications related to the tube are not uncommon in this group but availability of expert advice in the community is limited. As a result many patients contact the Endoscopy unit where the tube was inserted for help in managing minor complications.

Objectives: To identify the reasons that prompted review of patients with gastrostomy in the endoscopy unit. The advice given/action taken was also recorded to assess whether hospital attendance could be avoided if appropriate support care was available in the community.

Methods: Data was collected retrospectively over 2 years (01/99 -12/00) of patients attending the endoscopy unit at UHA with PEG tube related problems.

Results: Seventy patients attended our endoscopy unit with problems relating to their gastrostomy tube. The mean age of these patients was 64.5 years (range 32 – 85 years). The reasons for attendance were: red/inflamed and sore areas around the PEG stoma site (35%), broken/missing attachment (14%), blocked tubes (14%), leaking (12%), tube displacement (12%), hypergranulation (10%) and split tube (2%). The management of these problems included: trimming or attachment replacement (30%), cleaning and swabbing for infection around stoma sites (25%), advice regarding dressing (13%), change of PEG via endoscopy (15%) or without the aid of endoscopy (7%) and miscellaneous (10%). Eighty five percent of the presenting complaints could have been solved in the community by an appropriately trained staff.

Discussion: Our experience shows that more than three quarter of the problems could have been dealt in the home environment and would have required only one visit by the liaison nurse. This audit provides further evidence for the recommendation by the BAPEN that creating a suitable support system to continue treatment at home is desirable and would avoid unnecessary hospital attendance.


B. Amadi1, M. Mwiya1, P. Kelly2, S. Sinyangwe1, B. Simpito1, M. Shisholeka1, M. Katubulushi1, M. Thomson3, J. Walker-Smith3. 1Dept Paediatrics & Child Health, University Teaching Hospital, Lusaka, Zambia; 2Dept Adult & Paediatric Gastroenterology, St Bart's & Royal London School of Medicine & Dentistry, London, UK; 3University Dept Paediatric Gastroenterology, Royal Free Hospital School of Medicine, London, UK

Introduction: The persistent diarrhoea-malnutrition syndrome (PDM) remains a leading cause of morbidity and mortality in many resource-poor countries, but even in hospitals treatment is unsatisfactory. We report a randomised controlled trial of an elemental diet compared to standard nutritional rehabilitation for PDM in the University Teaching Hospital, Lusaka.

Study design: 200 children (106 HIV seropositive, 90 HIV seronegative) were randomised to an elemental diet with Neocate (SHS International) or to a skimmed milk-based followed by soy-based diet. Treatment was given for 4 weeks in hospital, and intestinal and systemic infection treated with routine therapies.

Results: 155 children completed 4 weeks of therapy, 39 died and 6 were lost. They were severely malnourished with median baseline weight-for-age z scores around –4.0; 9% were underweight, 23% had marasmus, 47% had kwashiorkor, and 21% marasmic-kwashiorkor. Weight gain was greater in the Neocate group (median gain in weight-for-age z score 1.23, interquartile range 0.89 - 1.57) compared to Control (0.87, 0.47 - 1.25; p=0.002), despite greater calorie intakes in the Control group. Increase in haemoglobin concentration was also greater in the Neocate group (0.8g/dl, 0 - 1.8) than the Control group (0.3, -0.6 - 1.6; p=0.04). Diarrhoea frequency and global recovery scores improved equally in both treatment groups. Mortality was higher in HIV seropositive children and those with cryptosporidiosis, but did not differ between treatment groups.

Conclusions: Exclusive use of an elemental diet for 4 weeks was associated with significantly improved nutritional recovery in children with severe PDM, irrespective of HIV infection.


A.I. Thuraisingam, S.A. Cairns. Royal Preston Hospital, Preston PR2 9HT, UK

Aims: Percutaneous endoscopic gastrostomies (PEG) are an increasingly frequent method for providing nutritional support. We set out to assess current practice surrounding referrals for PEG insertion in a district general hospital and their outcomes.

Methods: All patients who were referred for PEG insertion over a six-month period were assessed. Data were prospectively collected for each referral with regards to; duration between dates of admission, referral and insertion of PEG; indications; appropriateness of referral; prior clinical assessment by dietician or speech therapists and ability of patient to give informed consent for the procedure. Six-month follow up of all patients referred was then performed by case record examination.

Results: 50 patients were referred for consideration of PEG insertion. 24% of patients were deemed inappropriate for PEG insertion. This was either because the patient was currently unfit/had a poor prognosis or was able to swallow or because it would be technically difficult. There was a 44% 30-day mortality rate in the PEG insertion group compared to a 50% 30-day mortality rate in the PEG not inserted group (a 70% rate if patients who did not have a PEG inserted because they had an intact swallow are excluded from consideration). 60% of patients were unable to give informed consent for the procedure and these patients had a much higher 30-day mortality rate, 60% compared to 21% in those able to give informed consent. Mortality rates also varied appreciably according to the indication for PEG insertion. The overall six-month mortality rate for all patients referred was 70%.

Conclusions: The 30-day mortality rate following PEG insertion was higher than in published series, however this is offset by an even higher rate in patients in whom PEG insertion was declined. Clinician review appropriately reduced the number of PEG insertions, thereby reducing workload and preventing patients from undergoing an unnecessary procedure. Simple clinical predictors such as reviewing the indication for PEG insertion and the ability of patients to give informed consent may further reduce inappropriate PEG insertions. Special care needs to be taken when assessing patients unable to give informed consent as this vulnerable group has a much higher mortality rate.


J.R. Boulton-Jones, J. Lewis, J.C. Jobling, K. Teahon. Nottingham City Hospital, Hucknall Road, Nottingham NG5 1PB, UK

Introduction: Enteral nutrition is regarded as superior to and is cheaper than parenteral nutrition in critically ill patients. Oral and nasogastric tube (NGT) feeding may not always be possible or successful in these patients due to vomiting or gastric stasis. Nasojejunal tube (NJT) feeding is an alternative method of providing enteral nutrition. The use of NJT feeding in 100 patients is presented.

Methods: Patients who had NJT feeding were identified from the dietetic records. Their records were reviewed and data on the indication for and outcome of NJT feeding was recorded.

Results: 102 separate instances of NJT feeding were attempted in 100 patients. 36 patients had failed NGT feeding and 23 patients had parenteral feeding prior to NJT feeding. The commonest indications for the use of NJT feeding were reduced oral intake during chemotherapy (24 patients), gastro-oesophageal malignancy (14), prolonged or complicated post-operative recovery (14), acute pancreatitis (12), benign gastro-intestinal disease (10) and severe burns (10). 96 patients had an NJT successfully placed and in 88 nutritional requirements were successfully met. NJT feeding was continued for a median of 11 days (range 0–180 days). 65 patients were discharged home, 30 died as an in-patient and 7 were discharged to another hospital. 11 patients had NJ feeding continued at home. In 37 patients, medication was given by the NJT. There was one serious complication of a significant small bowel bleed possibly caused by NSAIDs given down the NJT. Minor complications included NJT displacement necessitating replacement (13), diarrhoea (6) and blocked NJT (4). 1 patient had small bowel obstruction that appeared to be related to their underlying malignancy.

Conclusions: NJT feeding can successfully and safely meet nutritional requirements of many critically ill patients. Patients with gastric stasis or vomiting, such as patients undergoing chemotherapy, with severe burns, prolonged or complicated post-operative recovery or with acute pancreatitis, can be successfully fed nasojejunally. There may be cost savings related to reducing the demand for parenteral nutrition.


T. Trebble, S.A. Wootton, M.A. Stroud, A. Ballinger, P.C. Calder. Institute of Human Nutrition, Southampton University Hospitals, Southampton SO16 6YD, UK

Prostaglandin E2 (PGE2) is an eicosanoid synthesised by monocytes from the n-6 polyunsaturated fatty acid (PUFA) arachidonic acid (ArA) with proposed effects on CD4+ T-helper cells (Th1 and Th2). Increased intake of the n-3 PUFA eicosapentaenoic acid (EPA) (in fish oil) may inhibit PGE2 production by substrate competition or enzyme inhibition.

We investigated the effect of increasing dietary n-3 PUFA intake, with or without antioxidant co-supplementation, on plasma and erythrocyte phospholipid composition and ex vivo LPS stimulated monocyte PGE2 synthesis, and Con A stimulated T-cell proliferation (L. prolif) and synthesis of interferon-γ (IFN-γ) (Th1) and interleukin-4 (IL-4)(Th2).

Sixteen healthy male subjects were randomised to 12 weeks of antioxidant co-supplementation (vitamins A, C and E and selenium) (n8) or placebo (n8). All subjects simultaneously received identical regimens of fish oil equivalent to 4 weeks of 0.3 g/d, 1 g/d and 2 g/d n-3 PUFA consecutively. Venous blood was taken at baseline, 4, 8 and 12 weeks for phospholipid composition, cell isolation and culture. EPA incorporation increased incrementally in all phospholipid pools with increasing n-3 PUFA intake. Only T-cell proliferation was modulated by antioxidant cosupplementation (augmented response to n-3 PUFA), and therefore for cytokine/PGE2 production the two groups were pooled (n16). See table.

Abstract 318

n-3 PUFA dietary supplementation is associated with inhibition of PGE2 synthesis by monocytes and parallel increases in Th1 and Th2 cytokine synthesis and T-cell proliferation, in a dose responsive manner. Therapeutic efficacy of n-3 PUFA may relate to the relative importance of monocyte or T-cell activation and PGE2 in the inflammatory response.


M.R. Anderson, M. O'Connor, J. Woodward, D. Mayer, K. Kane. Selly Oak Hospital, University Hospital Birmingham, UK

Background: The nasal loop (NL) is a novel method of securing nasogastric tubes (NGT) in dysphagic stroke patients that can be performed under light or no sedation. A loop of tape is passed round the nasal septum and this secures the NGT in place.

Aims: To demonstrate that NL in dysphagic stroke patients: (1) improves nutrition; (2) may offer an alternative to PEG in high risk patients; (3) may avoid premature PEG placement for NGT displacement.

Methods: A six month prospective audit of dysphagic stroke patients referred for PEG. Patients referred to one gastroenterologist were offered NL. Others proceeded direct to PEG. NL patients had feed intake monitored prior to and post NL. Complications and outcome at 3 months were recorded for all patients.

Results: Group 1) 14 patients had NL for a median of 15 days (range 1 – 46). Median prescribed feed intake before NL was 0% (range 0 – 47%), after NL was 100% (range 67 – 100%). 4 patients recovered normal swallowing and 4 patients died. 6 proceeded to PEG of whom 4 died within 3 months, 1 recovered normal swallowing and 1 continued to be PEG fed.

Group 2) 7 patients proceeded direct to PEG, 1 died and 6 were alive and PEG fed at 3 months. There were 6 complications from PEG insertion (including 1 peritonitis, 2 aspiration pneumonias, 2 wound infections and 1 severe pain) and no patients recovered normal swallowing.

Conclusions: NL improves nutritional intake and allows some patients to recover swallow without PEG. NL may avoid PEG intervention in patients who have a poor prognosis. Mortality was higher in NL patients which may be due to referral bias.


S. Chattophadhyay1, P.M. Irving1, A. Cunliffe1, C. Archer1, D. Murphy2, F. King2, K. Durman3, J. Powell-Tuck1. 1Human Nutrition, St Bartholomews and Royal London School of Medicine and Dentistry, 2Pharmacy, 3Dietetics, Barts and the London NHS Trust, London, UK

Aim: We wanted to know if the feeling that some patients report of feeling better soon after starting TPN was real or due to placebo.

Method: 10 patients (4F,6M), mean (range) age 51 (19–80), mean BMI (sd) 22.4(6.9), were selected. TPN was clinically indicated in all and none had received food for at least 5 days. Patients were randomised to receive either TPN or a dummy feed containing hypocaloric dextrose, water, sodium and potassium for the first 24 hours. Central and peripheral fatigue were assessed using validated measures: visual flicker fusion frequency (FFF), grip strength (GS), reaction time (RT) and a visual analogue scale (VAS). Both subjects and observers were blind to the contents of the feed for the first 24 hours. Measurements were taken before and 2,4,6 & 24 hours after starting the feed or placebo. Subsequently both groups received a complete feed and further measurements were taken 24 hours later. Dummy bags had calculated Na, K and water requirements with volume being made up by 5% dextrose. TPN contained calculated requirements of amino acids, Na, K, Ca, PO4, Mg, vitamins, trace elements, energy and water.

Results: Between 2–6 hours the change in fatigue was significantly different between the two groups for RT (p<0.000) and GS (p=0.02). Over the period 2–24 hours there were significant differences between the two groups for RT (p=0.01) and GS (p<0.02). By 24 h the differences between the two groups had diminished. At 48h the results of the two groups were similar.

Conclusions: Within 6 hours of starting TPN, fatigue is improved. Water, Na, and K infusion, and placebo do not account for this effect

View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.