With the use of a meticulous, daily record of the ICU feeding practice we evaluated the feasibility of prescribed enteral feeding for a 1-year period. The prescribed nutritional volume turns out to be hardly feasible in the patients involved in our study. When actual intake is compared with ideal energy and protein needs, protein shows the largest overall deficit. Current feeding practice (including the 5-day build-up schedule for enteral nutrition) fails to provide ICU patients with adequate nutrition.
Other studies found comparably bad results. A prospective cohort study among 99 ICU patients found that only the half of patients achieved tolerance of the feeding regime (90% of estimated energy for more than 48 hours) [12
Better results were found in a multicenter prospective study that followed 193 patients during 1,929 patient days. An average of 76% of the prescribed feed was delivered to the patient. They also concluded that using well-defined protocols significantly improved the intake [13
A prospective study in ICUs and coronary care units revealed that barely one-half of the 44 patients studied met their caloric requirements because of underordering by physicians and reduced delivery arising from frequent and inappropriate cessation of feeding [14
Another prospective study found also a low caloric intake in 51 enterally fed ICU patients for whom 78% of the mean caloric amount required was prescribed and 71% was actually delivered [15
An audit of 40 ICU patients for which the ideal feeding target was calculated by the Harris–Benedict equation. Patients received only 51% of these energy requirements during the 7-day study period [16
A cross-sectional survey of 66 responding dieticians of ICUs revealed that among patients receiving enteral nutrition only 58% met their prescribed energy and protein needs [17
Although we were aware of these studies, we did not expect this result until we kept these records. Despite having at our disposal an enteral feeding protocol and despite daily bedside consultations with the intensivist, nurse and dietician, only 50% of the enterally fed patients achieved a successful intake at the end of a 5-day feeding build-up scheme. Although a further improvement in intake occurred as the ICU stay was prolonged, the overall success per feeding day remained low during the ICU stay. Apparently, implementation of a protocol, once it has been set out and accepted, is difficult and needs more attention [13
The feeding with a NCJ resulted in odds ratios that favor this enteral route over the gastric tube. In addition semi-elemental formula seemed to be three times better than standard formula (Table ). In part, this might have been confounded by the use of either duodenal tubes or NCJ, because the NCJ showed the fewest problems in use. Because of this and because it concerned a small group of patients, we cannot unambiguously recommend semi-elemental formula although others have done so [20
Disordered upper gastrointestinal tract motility frequently occurs in ICU patients [22
], yet the gastric tube remains the first and simplest choice and the easiest way of starting enteral nutrition. This does not detract from the significant number of patients who have to be switched to a duodenal tube because of persisting gastric retention. We also found that nurses tended to overestimate gastric retention as a risk factor and, more importantly, violated the protocol by discarding a gastric retention volume of less than 200 ml over 6 hours. This behavior might be the result of a misplaced ambition to achieve safer care. Although the measurement of gastric retention is an important tool for guaranteeing safe enteral feeding, no difference is reported between gastric tube and duodenal tube use among ICU patients in terms of aspiration and nosocomial pneumonia. Moreover, the insignificant role of gastric retention levels of up to 250 ml has been reported [23
Using the feeding tube to administer contrast for a CT scan precludes the use of the tube for administering nutrition. In general, a high therapy intensity reflected by a high TISS score indicated a more difficult feeding practice because the subject was more critically ill. This might also reflect the lower priority given in the care routine for optimal continuation of the feeding process in comparison with the efforts taken to support patients in need of ventilation and assisted circulation.
Improvement of nutritional intake can be achieved by implementing simple rules, such as limiting the interruption of enteral nutrition because of diagnostic or therapeutic interventions, a quick replacement of accidentally removed tubes, and giving back gastric retention of less than 250 ml [14
Whereas a high TISS score did seem to interfere with the administration of enteral nutrition, the severity of illness did not. It took several days for 50% of the patients to achieve an optimal intake, which to some extent might reflect the unstable physical condition of the ICU patient. This is also shown by the relationship between success of feeding and prolonged ICU stay.
A limitation of this study is that we did not collect or analyze a nutritional anamnesis or patient outcome data. We have focused on measurable aspects of feeding practice. It will be worthwhile to expand the continuous recording to include a (nutritional) anamnesis of the patient. Improving the information load of this record would also require more information about outcome.