Malnutrition can adversely affect physical and psychological function, influencing both morbidity and mortality. Despite the prevalence of malnutrition and its associated health and economic costs, malnutrition remains under-detected and under-treated in differing healthcare settings. For a subgroup of malnourished individuals, a gastrostomy (a feeding tube placed directly into the stomach) may be required to provide long-term nutritional support. In this review we explore the spectrum and consequences of malnutrition in differing healthcare settings. We then specifically review gastrostomies as a method of providing nutritional support. The review highlights the origins of gastrostomies, and discusses how endoscopic and radiological advances have culminated in an increased demand and placement of gastrostomy feeding tubes. Several studies have raised concerns about the benefits derived following this intervention and also about the patients selected to undergo this procedure. These studies are discussed in detail in this review, alongside suggestions for future research to help better delineate those who will benefit most from this intervention, and improve understanding about how gastrostomies influence nutritional outcomes.

Enteral feeding (or ‘tube feeding’) is a very common inpatient intervention to maintain nutritional status where the oral route is inadequate, unsafe or inaccessible. A proportion of patients will need to continue tube feeding in the community after their admission and will require a gastrostomy tube. Although gastrostomy insertion is relatively straightforward, it is not without complications in an often frail and vulnerable group of patients and a multidisciplinary approach is necessary to ensure that the procedure is appropriate. Some patients are better managed with careful assisted hand feeding or nasogastric tubes. Particular care needs to be taken in deciding whether patients with dementia should have a gastrostomy in view of data suggesting that this group of patients have a particularly poor prognosis after the procedure. Decisions regarding the provision of enteral nutrition at the end of life or where patients are not competent to make an informed judgement are particularly challenging and need to be made on a case-by-case basis.

The development of gastrostomy placement has been an important technological advance in the enteral-access field. However, its rapid growth in popularity could be viewed as problematic. The endoscopist or intervention radiologist can no longer act as a technician but requires to follow-up this group of patients in order to determine outcomes that will inform future practice. There has been emphasis on the importance of the multidisciplinary team in informing and assessing patients referred for gastrostomy insertion. Communication between all the professionals caring for a patient and between the carers and the patient allows information to be collated that will determine the benefits and burdens of long-term gastrostomy feeding. At present much of the published experience is limited to the acute care setting. The incidence of complications varies, depending on the investigator's definition of complication and the diagnosis of the patient group. Many reported studies are retrospective, which puts limitations on documentation. Increasing numbers of patients with diverse needs are now being discharged into the community with gastrostomy tubes in place. Whilst the hospital should ensure that written protocols are provided and that all carers involved should have adequate training, it is not unusual for patients to receive mixed messages from the different care teams responsible for their care. In South Glasgow NHS Hospital Division key members of all teams caring for these patients (acute care, community district nurses, learning disabilities team, physical disabilities team and commercial homecare companies) meet regularly to discuss equipment and protocols. The members of this group feel that this approach has improved communication, standardised practice and reduced complications by providing a service that delivers artificial nutrition support but is primarily suited to the patient's disease process.

A gastrostomy feeding tube has become the most acceptable method for long-term feeding support in patients with head and neck cancer. The aim of this study was to compare the complications of percutaneous endoscopically inserted gastrostomy (PEG) tubes, radiologically inserted gastrostomy (RIG) tubes and surgically inserted gastrostomy (open/laparoscopic) (SUR) tubes in head and neck cancer patients and also to compare the mortality rates of these patients.

Seventy-eight head and neck cancer patients underwent gastrostomy tube insertion (40 PEG, 28 RIG and 10 SUR) during the period February 2002 to February 2005. There were no significant demographic differences between the three groups. Thirty-six patients (46 per cent) developed complications, 32 minor and four major. All three groups were similar in their rate of minor complications, with the dislodgement and blockage rate being lowest in the PEG group (p > 0.05). The mortality rate was 4 per cent within 30 days of gastrostomy tube insertion. There were no deaths in the PEG group, two deaths in the RIG group and one in the SUR group. The PEG tube was considered superior to the RIG and SUR gastrostomy tubes, had fewer complications and was safer. Thus, PEG tube insertion is our first choice for head and neck cancer patients.

Since it was first described, the original percutaneous endoscopic gastrostomy (PEG) technique has proved to be a valuable adjunct in patients with head and neck tumours. This procedure is being increasingly utilized in the face of swallowing impairment related to head and neck carcinoma. Although generally well tolerated, it may be associated with complications. In this report, we document tumour implantation at the percutaneous endoscopic gastric site and review the report cases. It appears that implantation metastasis does alter prognosis.