The relationship between disturbed gastric motor function and enteral nutrition in critically ill patients.

Abstract

Delayed gastric emptying, that manifests clinically as intolerance to enteral feeding, occurs in over 50% of critically ill patients and has a major impact on patient morbidity and mortality. Despite the recognition that the proximal stomach has a major role in gastric emptying of liquids, only the motor activity of the antro-pyloro-duodenal region has been evaluated in detail. In addition, many of the proposed risk factors for the gastric dysmotility, particularly a prior history of diabetes mellitus, have not been evaluated formally but have been extrapolated from data from non-critically ill patients. The currently available prokinetic drugs, erythromycin and metoclopramide, are considered to be the first line treatment for feed intolerance. However, neither data comparing the effectiveness of these agents nor the data on the effects of combination of therapy in the treatment of feed intolerance are available. The aims of this thesis were, therefore, to examine: (i) proximal gastric motor activity and the association between proximal and distal motility; (ii) the relationship between entero-gastric humoral responses to nutrients, gastric emptying and feed intolerance; (iii) the impact of admission diagnoses, choice of sedations, timing of initiation of feeding, and pre-existing history of diabetes mellitus on gastric emptying and feed intolerance; and (iv) the efficacy of erythromycin, metoclopramide and combination of these drugs in treatment of feed intolerance in critically ill patients. The current thesis indicates that motor activity is impaired in multiple regions of the stomach in the critically ill. When compared to healthy humans, proximal gastric relaxation was prolonged and fundic wave activity was educed during small intestinal nutrient infusion in critically ill patients. In addition, simultaneous assessment of proximal and distal gastric motility demonstrated a possible disruption of the motor integration between the proximal and distal stomach. In light of the recent data that suggested a significantly greater proportion of meal distributed proximally in critically ill patients with delayed gastric emptying (Nguyen, et al. 2006), the disruption of the gastric motor integration and the prolonged gastric relaxation in response to duodenal nutrients may play a significant role in the pathogenesis of slow gastric emptying during critical illness, especially as liquid formulae. The entero-gastric hormonal feedback responses were also disturbed during critical illness. Both fasting and duodenal nutrient-stimulated plasma CCK and PYY concentrations were significantly higher in critically ill patients, particularly those who did not tolerated gastric feeds. The rate of gastric emptying of a liquid meal was inversely related to both fasting and postprandial plasma CCK and PYY concentrations, supporting the potential role of plasma CCK and PYY in the pathogenesis of gastric dysmotility in critically ill patients. Admission diagnosis, choice of sedative drug and blood glucose control but not the timing of enteral feeds were important factors for delayed gastric emptying and feed intolerance in these patients. In particular, delaying enteral feeding by 4 days had no impact on the rate of gastric emptying, intra-gastric meal distribution, or plasma CCK and PYY concentrations. Contrary to traditional belief, critically ill patients with a pre-existing diagnosis of type 2 DM have only a minor disturbance to the proximal stomach, a relatively normal gastric emptying and are at no higher risk of feed intolerance than those without DM, suggesting the presence of pre-existing DM 2 in critically ill patients should not influence the standard practice of gastric feeding. Therapeutically, short-term treatment with low dose erythromycin was more effective than metoclopramide, but the effectiveness decreased rapidly overtime at similar rate as observed with metoclopramide. In patients who failed to response to either agent, treatment with both agents was highly effective in re-establishing feeding success. The use of combination therapy as the initial treatment for feed intolerance was also more effective than erythromycin alone and had less tachyphylaxis. Treatment with erythromycin and metoclopramide, either as a single agent or in combination did not associated with major cardiovascular adverse side effects. Although diarrhoea was a common side effect and was highest with combination therapy, it was not associated with Clostridium difficile infection and settled quickly after the cessation of the prokinetic therapy. In summary, the work performed in the current thesis has provided substantial insights into the understanding of the nature, risk factors, pathogenesis and treatment of disturbed gastric motor function in critically ill patients. Not only do these findings stimulate further research into the mechanisms responsible for gastric dysmotility in critical illness, they also lead to the development of new strategies for optimizing the management of feed intolerance.