Food for Thought

Abstract & Commentary

Commentary by John J. Caronna, MD, Vice-Chairman, Department of Neurology, Cornell University Medical Center, Professor of Clinical Neurology, New York Hospital, Associate Editor, Neurology Alert.

Synopsis: NG feeding should be used for dysphagic patients early and PEG feeding reserved for those who do not tolerate NG feeding or who require long-term tube feeding.

Source: Dennis MS, et al. Routine Oral Nutritional Supplementation for Stroke Patients in Hospital (FOOD): A Multicenter, Randomized, Controlled Trial. Lancet. 2005;365:755-763; Dennis MS, et al. Effect of Timing and Method of Enteral Tube Feeding for Dysphagic Stroke Patients (FOOD): A Multicenter, Randomized, Controlled Trial. Lancet. 2005;365:764-772.

Nutritional support is an integral part of the management of neurologically ill patients. The FOOD trials were designed to assess the effect of feeding on outcome in adult stroke patients. In the first FOOD trial, 125 hospitals in 15 countries enrolled more than 4000 stroke patients. Those patients who could swallow were randomly assigned to a normal hospital diet or a normal hospital diet plus nutritional supplements equivalent to 360ml at 6.3 kJ/ml and 62.5 g/L in protein every day until hospital discharge. The primary outcome was death or poor outcome at 6 months after stroke.

Of the 2016 patients allocated to receive supplements, 79 (4%) did not receive any due to staff error, patients refusal, or clinical worsening. An additional 540 patients (28%) stopped receiving supplements before discharge, mainly because of patient refusal. The reasons for refusal were not liking the taste, unwanted weight gain, or feelings of nausea. The mean duration of hospital stay in the supplemented groups was 34 days. There was no significant difference between groups for any complications. Pneumonia and urinary tract infection were the most common in-hospital complications (6% and 7%, respectively).

The study did not show a significant effect of routine, oral nutritional supplements on the 6-month outcome of acute stroke patients, a not surprising result given that only 8% of enrolled patients were considered to be undernourished at baseline. Therefore, the study could not exclude a potential benefit from nutritional supplements in undernourished stroke patients.

The second report (pages 764-772) describes 2 food trials that included stroke patients unable to swallow. The first of these latter trials studied the effect of the timing of early feeding on post stroke outcome: Patients enrolled within the first week of admission were randomly allocated to early tube feeding via percutaneous endoscopic gastrostomy (PEG) or nasogastic (NG) feeding, or to no tube feeding for more than 7 days. In the other trial, the route of administration of early tube feeding was studied. Patients were allocated to PEG or NG feeding and primary outcome assessed at 6 months.

In the early (less than a week) vs late (after a week) tube feeding trial, almost 900 patients were enrolled by 83 hospitals. There was no significant difference in outcome between early tube feeding and late, although there was an absolute reduction in mortality of 5.8% that was not significant (P = 0.09). There was no excess of pneumonia associated with tube feeding. The rate of gastrointestinal (GI) hemorrhage was higher with early tube feeding rather than late (22 vs 11, P = 0.04). The trend favoring improved survival in the early tube feeding group was offset by an almost 5% increase in survivors with a poor outcome. Dennis and colleagues speculate that early tube feeding kept patients alive who would have otherwise died, but their improved survival was in a severely disabled state.

In the PEG vs NG tube trial, 321 patients were enrolled in 47 hospitals in 11 countries. PEG feeding was associated with a 1% absolute increase in risk of death that was not significant and a 7.8% increase in risk of death or poor outcome that was significant (P = 0.05). GI hemorrhage was more frequent with NG, rather than PEG tube feeding (18 vs 5, P = 0.005). Therefore, Dennis et al recommend that NG feeding should be used for dysphagic patients early and PEG feeding reserved for those who do not tolerate NG feeding or who require long-term tube feeding.


The 3 FOOD trials provide useful information on how and when to provide nutrition in acute stroke patients. The first trial found that oral food supplements are not necessary for patients who are not malnourished. The second observed that early tube feeding did not reduce stroke disability, and may have promoted the survival of severely disabled patients. The third study reported that NG tube feeding was superior to PEG tube feeding in the first weeks after stroke. Others have observed contradictory results. For example, Norton and colleagues (BMJ. 1996;312:13-16), in a small series of 30 patients, reported that PEG tube feeding, compared to NG tube feeding, significantly reduced mortality and aspiration pneumonia after stroke. The experience of clinicians is that NG tube feeding is more troublesome, if not more dangerous, than PEG tube feeding because NG tubes are subject to misplacement and obstructions; can produce ulceration of the nasal mucosa and reflux esophagitis, and allow aspiration of stomach contents. Nevertheless, the FOOD trial results underscore the need for clinicians to carefully consider the risks and benefits of the 2 routes of enteral nutrition in the first weeks after the onset of an acute stroke.