GLUCOSE HOMEOSTASIS DURING PARENTERAL NUTRITION

Normally, the concentration of plasma glucose is kept within close limits. There is a constant requirement for glucose by glucose-dependent tissues (e.g., brain, erythrocytes). Equally, there are detrimental effects of extreme hyperglycemia (e.g., hyperosmolar coma).^{[16] [17]} The mechanisms responsible for the regulation of plasma glucose levels are of fundamental importance to the entire metabolic response to hyperalimentation.^[18]

2893

Figure 77-10 Effects of semistarvation on metabolic rate and hypoxic and hypercapnic ventilatory responses. The hypoxic response is quantitated by A, defined in the equation VE = V_o + A/(PaO₂ − 32), in which VE is minute ventilation, PaO₂ is alveolar oxygen tension, and V_o is the asymptote for ventilation obtained by extrapolation. The hypercapnic response is quantitated by S, defined in the equation VE = S(PaCO₂ − B), in which PaCO₂ is end-tidal CO₂ tension, and B is the extrapolated intercept on the abscissa (PaCO₂ axis). ψO₂ and ψCO₂ declined, reaching the lowest level by the 10th day, as did the hypoxic response. The hypercapnic response declined slightly but not significantly. The asterisk denotes a significant difference from the control (P < .05), and brackets indicate ± SEM. (Adapted from Doekel RC Jr, Zwillich CW, Scoggin CH, et al: Clinical semi-starvation: Depression of hypoxic ventilatory response. N Engl J Med 295:358, 1976.)

During intravenous infusion of glucose, two processes can minimize changes in plasma glucose concentration: reduction in the rate of glucose production and enhanced ability to clear glucose from the bloodstream. Wolfe and coworkers ^[19] demonstrated that when glucose was infused into normal volunteers at the rate of 1 mg/kg/min (84 mL/hour of 5% dextrose in water in a 70-kg patient), endogenous glucose production was suppressed. One of the principal reasons infused glucose is of nutritional benefit is that the suppression of glucose production resulting from glucose infusion spares amino acids that would otherwise be catabolized to provide gluconeogenic precursors. The rate of glucose infusion that provides the maximal benefit depends on the condition of the patient.^[20]

From a clinical point of view, the justification for administering glucose at rates in excess of 3 to 4 mg/kg/min must therefore reside in the ability of the body to oxidize the infused glucose. Table 77-3 shows that even during a 4-mg/kg/min infusion, less than one half of the infused glucose was directly oxidized, and that at an infusion rate of 9 mg/kg/min, approximately one third of the glucose was directly oxidized.^[21] At this high level of glucose infusion, however, only 61.6% of CO₂ was attributed to direct oxidation of glucose. Even at these levels of glucose supply, other substrates are being oxidized.

The percentage of CO₂ from glucose infusion also reaches a plateau at around 60% in burn patients, regardless of the infusion rate.^[22] If insulin is infused simultaneously with glucose, the glucose clearance rate increases, but the rate of glucose oxidation is unaltered. Perhaps 22% of carbon dioxide that is coming from fat oxidation can be attributed to the conversion of excess infused glucose to fat.

Hyperglycemia associated with insulin resistance is a common complication observed in critically ill patients, even in those patients without a history of diabetes.^[23] Van den Berghe and coworkers^{[24] [25]} examined the role of intensive insulin therapy (i.e., maintenance of blood glucose no higher than 110 mg/dL) to improve mortality in critically ill patients compared with conventional therapy (i.e., insulin treatment only when plasma glucose exceeds 215 mg/dL and maintenance of glycemia between 180 and 200 mg/dL). The mortality rate for the patients receiving intensive insulin therapy was significantly lower than for the group assigned to receive conventional therapy. The benefit was seen specifically in patients with stays of more than 5 days in the intensive care unit and was achieved principally through a reduction in the incidence of multiple organ failure with an identifiable septic focus.

Intensive glycemic control also reduced the morbidity associated with critical illness. Significantly fewer patients in the intensive-treatment group required prolonged ventilatory support and renal replacement therapy, whereas the proportion of patients needing inotropic or vasopressor therapy, or both, was the same in the two groups. Overall, intensive insulin therapy reduced the length of stay in the intensive care unit among patients requiring intensive care for more than 5 days. The reason for improvement in intensive insulintreated patients remains obscure until other investigators can replicate the improvement in mortality in critically ill patients.

2894