Article Text

Download PDFPDF

Clinical outcome in patients from a single region who were dependent on parenteral nutrition for 28 days or more
  1. J Köglmeier1,
  2. C Day2,
  3. J W L Puntis1
  1. 1
    The Children’s Centre, the General Infirmary at Leeds, Leeds, UK
  2. 2
    Bradford Royal Infirmary, Bradford, UK
  1. Dr J Köglmeier, Department of Paediatric Gastroenterology, Chelsea and Westminster Hospital, 369 Fulham Road, London SW10 9NH; jkoeg{at}


Background: The frequency and outcome of intestinal failure (IF) in children are not well defined in the UK. Long-term parenteral nutrition (PN) is an effective intervention, with intestine transplantation offering the possibility of survival should life-threatening complications arise in those with long-term dependency. The ideal model for service provision is a subject of debate.

Aims: We aimed to identify all new cases of IF (defined as PN dependency ⩾28 days) in West Yorkshire over a two-year period to determine the rate of serious complications, establish the outcome after two years and clarify the role of specialist referral.

Method: Pharmacists in all the West Yorkshire paediatric units were contacted to establish the number of children with IF during 2001–2002. Underlying diagnosis, complications and outcome at two years were obtained by case-note review for 93 of the 96 children identified.

Results: IF patients were exclusively managed in one or other of the three large teaching hospitals. At the two-year follow-up, six (6.4%) children had died (one while listed for a small bowel transplantation), but 85 (91%) had established full enteral feeding and were well. Two remained PN dependent and were assessed in the supra-regional intestinal transplantation unit (Birmingham); in neither case was small bowel transplantation thought to be appropriate. The most common complications were central venous catheter sepsis (69% of patients) and cholestasis (59%).

Conclusions: This study shows that a favourable outcome for IF can be achieved in a regional centre with appropriate multidisciplinary support. A single UK supra-regional unit undertaking small bowel transplantation is probably adequate for assessment of the most complex patients, although this should remain under review.

Statistics from

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.

Intestinal failure (IF) refers to a functionally impaired gastrointestinal tract that is unable to maintain biochemical homeostasis and support normal growth. Short bowel syndrome (SBS) is a common cause of IF. It is usually defined separately as a severe reduction in functional gut mass below the minimal amount necessary for digestion and absorption to satisfy the nutrient and fluid requirements for growth.1 Other causes of IF include mucosal abnormalities that give rise to protracted diarrhoea and neuromuscular disorders that result in chronic intestinal pseudo-obstruction.2 Bowel adaptation in SBS may take many months and, in cases of irreversible pathology, survival is only possible if patients are entered into a home parenteral nutrition (HPN) programme.3

Recent debate in the UK has focused on the proposal that a small number of supra-regional services should be established to act as referral centres for infants and children with complex IF.4 Currently, patients are looked after in approximately 14 regional gastroenterology units, with only one centre (Birmingham Children’s Hospital) being able to offer intestine transplants. Lack of data regarding the incidence and long-term outcome of IF make it difficult to plan services with confidence. Yorkshire is one of the largest counties in Britain, with West Yorkshire home to almost 3.5% of UK children. Our aim was to identify all new cases of IF over a two-year period in this well-defined geographical area that is representative of the UK population as a whole and to determine morbidity, mortality and outcome. We hoped that this data would help define the need for referral to a regional gastroenterology unit with multidisciplinary support services, the potential demand for assessment in a centre able to perform small bowel transplantation and, by extrapolation, give some indication of the national workload caused by IF. We chose the simple definition of IF adopted by the British Intestinal Failure Survey, as PN dependency for 28 days or more.5


Hospital pharmacists in all of West Yorkshire’s paediatric units were contacted by telephone and asked if there were any patients in their institution during 2001 and 2002 who received PN for 28 days or more. No hospital other than the three teaching hospitals had had such patients. Since the authors of this study had a duty of care to all these patients (either through the nutritional support team or as supervising consultant), formal ethical committee approval was not sought for accessing clinical notes; permission was asked from other supervising consultants. Patients were included in the review once it was confirmed that their home address was within West Yorkshire throughout a period of PN lasting at least 28 consecutive days during 2001 and 2002. Relevant clinical details including underlying diagnosis, PN-related complications and clinical status at 6, 12 and 24 months were recorded following note review by one of us (JK). This information was anonymised.


One hundred and ten patients with IF were identified, of whom 96 fulfilled the inclusion criteria. Two had left West Yorkshire and follow-up details were not available; one patient’s case notes were unobtainable. Hence, 93 sets of clinical records were examined. Patients fell into the following four categories: (1) premature newborns (n = 61; 65.6%), 37 with gastrointestinal immaturity, 23 with necrotising enterocolitis (NEC) and one with intestinal pseudo-obstruction; (2) term infants (n = 20; 21.5%) with congenital bowel disorders of whom 11 had gastroschisis, three had duodenal atresia, two had meconium ileus and cystic fibrosis, three had malrotation with volvulus and one had exomphalos; (3) oncology patients (n = 11; 12%) with gastrointestinal complications of chemotherapy; (4) one patient with liver disease and SBS. Among the 23 infants with NEC, 15 underwent surgery (leading to SBS in six) and eight were treated conservatively.

In total, 82/93 (88.1%) children had established full enteral feeds during the two-year follow-up; results of liver-function tests were normal in all children by six months, and all showed satisfactory growth at both the one- and two-year follow-up. Six patients (6.4%) died while receiving PN: five of these were premature newborns with NEC, and one was a child with gastroschisis. In three, life-prolonging care was withdrawn because of severe neurological problems associated with premature birth, whereas three others died from overwhelming Gram-negative sepsis, one while listed for combined liver and small bowel transplantation.

Five children were discharged on HPN at between 11 and 17 weeks of age; these three subsequently established full enteral feeding following a mean duration of PN of 13 months (range 6–22 months), and one underwent liver transplantation for IF-associated liver disease. The total number of patients who were successfully weaned from PN therefore rose to 85 (91.4%). At the two-year follow-up, two patients remained on HPN (both had gastroschisis that had been complicated by antenatal infarction of bowel) and were assessed in the supra-regional intestinal failure/small bowel transplant unit at Birmingham Children’s Hospital. Neither fulfilled the criteria for listing for transplantation.6

The most common complications of IF were central venous catheter-related blood-stream infection (defined as fever and positive blood cultures in a child with indwelling central venous catheter and no alternative focus of infection identified; n = 64 patients; 68.8%), IF-associated liver disease (defined as serum-conjugated bilirubin >50 μmol/l in the absence of any other identifiable cause; n = 55; 59.1%) and abnormality of serum sodium or potassium concentration judged clinically important enough to require either modification of PN prescription (n = 20; 21.5%) or a period of replacement therapy with “tailor-made” non-PN fluids (n = 8; 8.6%). Two patients needed correction of hypomagnesaemia, and three others suffered an extravasation injury. A further child with SBS and jejuno-colic anastomosis developed calcium-oxalate renal stones that required lithotripsy.


Congenital or acquired gastrointestinal disease resulting in IF requires prolonged PN to sustain life.7 An increasing number of underlying primary intestinal disorders are now recognised, including structural abnormalities of the enterocyte, inflammatory or allergic bowel diseases, autoimmune enteropathy and disorders of intestinal motility.1 2 Diagnosis and management present considerable challenges and require the resources of a regional gastroenterology unit. In recent years, increasing survival of very low birthweight infants and incidence of gastroschisis have both contributed to greater pressures on services.8 The clinical course of IF is variable and determined by the underlying disorder.2 Small intestine transplantation offers the possibility of survival in those children with life-threatening complications of PN.9 There are little published data relating to the frequency and outcome of IF in children, and this study provides some novel information. We are confident of a very high level of case ascertainment. Pharmacists from all relevant hospitals cooperated in reviewing data from the detailed records of their workload. Only one district general was unable to provide information regarding PN usage in its neonatal unit; however, at worst this has resulted in us missing only a tiny number of preterm infants.

Our data give an incidence of IF in West Yorkshire during the first year of life of 1.6 per 1000 live births. For cases of IF across the whole age range, if we tentatively extrapolate from our findings in West Yorkshire with its child population of 414 000 to that of the UK as a whole (11.86 million), around 1300 children each year probably receive PN for 28 days or more (93:414 137×11.85 million  = 2660 per two years; 1330 per year); approximately 70 would progress to HPN; fourty-five might merit assessment in a unit undertaking small bowel transplantation, of whom 15 might be listed for transplant surgery. The estimated figure for HPN agrees with the findings of the British Artificial Nutrition Survey.10

It is notable that almost half of all patients were premature newborns with IF that was related to intestinal immaturity or NEC that was managed medically. Most were unknown to the regional paediatric gastroenterology service. With the marked development of neonatal and paediatric intensive care services over the past 20 years, the number of survivors of massive small bowel resection has significantly increased and is likely to continue to do so in the future.7 11 We take the view that patients with complex issues, including those with severe SBS, intractable diarrhoea or pseudo-obstruction, should be managed in a regional paediatric gastroenterology centre where multidisciplinary expertise in diagnosis and management is available.4

We endorse the referral criteria shown below.

Criteria for consultation/referral to a tertiary gastroenterology unit providing intestinal failure services for infants and children

  • Children with massive small bowel resection.

  • All children with congenital enteropathies who are destined for life-long parenteral nutrition dependence.

  • Diagnostic/prognostic uncertainty (eg, uncharacterised protracted diarrhoea).

  • All children requiring parenteral nutrition for more than 28 days.

  • Hyperbilirubinaemia (greater than 50 μmol/l).

  • Vascular access problems in patients receiving long-term parenteral nutrition.

Criteria for consultation/referral for small bowel transplant assessment

  • Children with massive small bowel resection.

  • Children with severely diseased bowel and unacceptable morbidity.

  • Continuing prognostic/diagnostic uncertainty.

  • Microvillous inclusion disease.

  • Persistant hyperbilirubinaemia (>100 μmol/l).

  • Thrombosis of two out of four upper body central veins.

  • The request of the patient/family.

IF-associated liver disease is common (it was present in two thirds of our IF patients), but generally resolved as enteral feeding increased. One child with SBS and IF-associated liver disease died of overwhelming sepsis while on the waiting list for a combined liver and small bowel transplant. This shortage of donor organs highlights the need to bring the plight of such patients to public attention.12 Two other children who received HPN and were not thought likely to establish full enteral feeding were assessed but not listed for small bowel transplantation. Currently, the main indication for transplant is life-threatening complication of PN.1217 Our data suggest that one supra-regional intestinal transplant centre is probably adequate for assessment of such patients, although this provision should be kept under review. The findings of our study indicate that the overall outcome of IF (for the preterm, newborn surgical and oncology patients) is good and that care in a regional gastroenterology centre with multidisciplinary support services (level 3 neonatal intensive care unit for the premature newborn) is appropriate. Furthermore, in West Yorkshire it seems that patients with complex IF are being referred on to the regional gastroenterology unit. This situation may be different elsewhere in the country.

The UK model for HPN in adult patients, in contrast to that of paediatric services, involves two national IF centres.18 IF services for children in France are similarly centralised, with six centres undertaking HPN. Problems in the UK with this degree of centralisation include saturation of the supra-regional units and long journeys for patients. The National Service Commissioning Advisory Group of the Department of Health acknowledges that the needs of many patients can be met at a lower level.18 In general, we would endorse the model of regional-centre-based care for IF as outlined by the British Society of Paediatric Gastroenterology, Hepatology and Nutrition.4 A proviso would be that a centre managing HPN should have sufficient staff, facilities and experience to make this safe and to make it work well for the families involved. It is difficult to define the throughput of patients that this might mean, given the absence of outcome data from different units undertaking HPN. Clearly, surgical expertise in non-transplant surgery for IF is likely to be limited to a number of regional centres. Bowel-lengthening interventions such as serial tranverse enteroplasty (currently the subject of an evaluation by the National Institute for Health and Clinical Excellence) should be viewed critically and, ideally, be concentrated in the hands of surgeons working in a few larger centres with outcome data being shared. As in Scotland,3 all units undertaking HPN should agree on standards of care. The issue of underfunding for HPN services needs to be addressed urgently by the Department of Health.

What is already known on this topic

  • The incidence of intestinal failure in the UK and its outcome is currently being investigated through the British Intestinal Failure Survey, a patient registry supported by the British Society of Paediatric Gastroenterology, Hepatology and Nutrition, and the British Association of Paediatric Surgeons.

  • The frequency with which IF occurs in the UK, together with its outcome, is currently poorly documented.

What this study adds

  • We estimate that around 1300 children in the UK each year are likely to receive parenteral nutrition for 28 days or more; most have an excellent prognosis.

  • Complex cases can be successfully managed in regional gastroenterology units with multidisciplinary support; however, a relatively small number will require further assessment in a supra-regional unit that is able to undertake small bowel transplantation.


We would like to thank Helen Martin, Charlotte Gibb, Evelyn Ward and Kerstin Tremlett for their help with the patient ascertainment.



  • Competing interests: None.