The two primary goals of nutrition support in patients with critical illness are to prevent morbidity and mortality directly attributable to macronutrient or micronutrient deficiencies and to ameliorate loss of lean-body mass. Although nutrition support by enteral or parenteral may be lifesaving in patients who are malnourished (BMI < 15 kg/m2 and/or clinically significant weight loss), the role of nutrition support in critically ill, obese patients, and in particular patients with type 2 diabetes is more controversial. It is often not clear when nutrition support should be initiated in these patients given their caloric reserves. Accumulating evidence suggests that in obese patients, hypocaloric feeding (permissive underfeeding) during critical illness is not inferior to eucaloric feeding and may have clinical benefits. The purpose of this review is to discuss the recent data evaluating optimal nutrition support in obese, diabetic patients.
Metabolic effects of critical illness and obesity
The majority of Americans are now overweight or obese and overt macronutrient deficiency is uncommon in hospitalized patients . Almost 10% of adults in the USA now have type 2 diabetes, representing a tremendous health care burden . Illness, obesity and type 2 diabetes have synergistic, adverse effects on neurohormonal pathways worsening insulin sensitivity, muscle breakdown and hyperglycemia. Critical illness may blunt insulin sensitivity by increasing sympathetic nervous system activity , the plasma concentration of catecholamines, and the plasma concentration  and bioavailability of cortisol . Further, circulating levels of proinflammatory cytokines including IL-6 and tumor necrosis factor, rise dramatically following tissue injury from sepsis or trauma . Lipolytic rate is increased and the ability of insulin to suppress fatty acid release from adipose tissue is impaired, promoting elevated plasma free fatty acid (FFA) concentration, which is strongly associated with rapid development of insulin resistance . Hepatic gluconeogenesis is increased and clearance of glucose from plasma by skeletal muscle is impaired by these factors. Unlike healthy patients, administration of intravenous glucose fails to suppress hepatic glucose output in critical illness and indeed may exacerbate hyperglycemia . Obesity is also strongly associated with insulin resistance, high plasma FFA concentration and increases in sympathetic nervous system activity, providing a ‘second hit’; to insulin action in critical illness. Obese patients have defects in oxidation of circulating fatty acids during critical illness; further exacerbating elevated plasma FFA concentration and insulin sensitivity . Patients with type 2 diabetes are at particular risk for development of severe metabolic complications including diabetic ketoacidosis and severe hypertriglyceridemia because they are unable to increase insulin production further.
Initial control of hyperglycemia
Administration of exogenous insulin is often necessary in type 2 diabetics who are critically ill, even when not receiving exogenous glucose. Infusion of insulin will rapidly reduce plasma FFA, lower plasma triglyceride concentration and prevent development of ketoacidosis. Insulin should either be provided by intravenous drip or by the administration of short-acting subcutaneous insulin. Long-acting subcutaneous insulin should be used with caution in critically ill, diabetic patients because reductions of insulin clearance and renal gluconeogenesis increase the risk of hypoglycemia. Prolonged periods of hypoglycemia may go unnoticed and prove fatal. The precise level at which glycemic control should be maintained remains unknown (see later paragraph).
Importance of maintenance of muscle mass
Ultimately, the primary goal of nutrition support should be to minimize the loss of lean body mass. Critically ill patients lose up to 1% of their lean body mass per day , with the majority of these losses occurring in skeletal muscle . Loss of skeletal muscle hinders rehabilitation after recovery, and lowers the ability of insulin to stimulate glucose clearance. The factors responsible for these dramatic elevations of muscle protein breakdown are not fully understood but include increases in glucocorticoid availability. Suppression of growth hormone secretion is common in sedentary, obese patients and is further lowered by critical illness and infusion of vasopressors such as dopamine . Growth hormone and insulin have synergistic effects to increase amino acid uptake by muscle , however, the treatment of critically ill patients with growth hormone cannot be recommended . It is likely that interventions to increase insulin sensitivity, such as hypocaloric feeding, will improve the ability of insulin to suppress muscle protein breakdown.
Potential benefits of hypocaloric feeding
Although it may seem that meeting estimated caloric expenditure with nutrition support is optimal, data from prospective studies do no support this. In healthy obese humans, with or without type 2 diabetes, even brief periods of calorie restriction may markedly improve insulin sensitivity. Fasting may actually worsen insulin resistance in skeletal muscle and therefore cannot be recommended as an intervention to improve insulin sensitivity [14,15]. Although calorie restriction may have health benefits in well patients, is this true in critically ill, obese patients, with large ongoing metabolic stress?
Clinical outcomes with hypocaloric feeding
Several studies clearly show that hypocaloric feeding for a limited amount of time is not inferior to eucaloric feeding. Sandstrom et al. reported that matching measured energy expenditure of nonobese, postsurgical, using total parenteral nutrition (TPN) for up to 15 days did not confer clinical benefit over that of a glucose infusion alone (providing only approximately 600 kcal/day), however, the hypocaloric group had significantly lower nitrogen balance . Choban  examined the effects of hypocaloric TPN in a randomized, prospective, controlled study in a group of patients who were 163% of their ideal body weight (IBW) . Approximately half of the patients in each group had type 2 diabetes mellitus. By design, the hypocaloric group received fewer calories than the eucaloric group (22 kcal/kgIBW/d vs. 36 kcal/kg IBW/d) and both groups received 2 g/kg IBW to attempt to minimize negative nitrogen balance. Although objective clinical outcomes were not different between the two groups, reassuringly, nitrogen balance did not differ on the basis of caloric intake. Taken together, these two studies suggest that hypocaloric feeding is at least not inferior to eucaloric feeding with respect to clinical outcomes. Further, Choban's study  demonstrates that nitrogen losses can be minimized by providing 2 g/kg/day of protein in combination with a hypocaloric diet, even during critical illness. In a recent prospective study, McCowen et al. examined the effects of hypocaloric feeding with TPN on outcomes in critically ill adults with an average BMI of 27 kg/m2. Patients were randomized to either receive up to 1000 kcal/day or standard TPN providing approximately 25 kcal/kg. The authors found that hypocaloric feeding was again not inferior to eucaloric feeding with respect to clinical outcomes.
Although hypocaloric nutrition does not appear to be inferior to eucaloric feeding in critical illness, is there a clinical benefit over eucaloric feeding? Krishnan et al. in an observational study in 187 critically ill patients evaluated the relationship between caloric intake clinical outcomes. This study was aimed to determine the proportion of patients who received calories that met the guidelines of the American College of Chest Physicians (caloric intake of 25 kcal/kg/day). Patients were grouped into tertiles of caloric intake to examine the relationship between actual calories delivered and clinical outcomes in patients in the second tertile of caloric intake (33–67% of recommended calories; 9–18 kcal/kg/day) had a greater likelihood of hospital discharge alive than patients in the first tertile (lowest caloric intake), and a greater likelihood of achieving spontaneous ventilation prior to ICU discharge than the other groups. Indeed, patients in tertile III (highest calorie intake) had a lower likelihood of hospital discharge alive and spontaneous ventilation prior to hospital discharge than the other groups even after correcting for multiple risk factors. The outcomes of this study are supported by an earlier retrospective study by Dickerson et al., which examined the relationship between caloric intake and outcomes in 40 critically ill obese patients. Patients were stratified as having received 25–30 kcal/kg adjusted body weight per day (eucaloric) or less than 20 kcal/kg adjusted body weight per day (hypocaloric). Both groups received 2 g/kg ideal body weight/day of protein. The hypocalorically fed group had a shorter stay in the ICU (19 ± 10 days vs. 29 ± 16 days), a shorter duration of antibiotic use (17 ± 12 vs. 27 ± 17 days), a trend towards a shorter requirement of mechanical ventilation (16 ± 11 vs. 24 ± 17 days) but no difference in nitrogen balance. These results are more impressive given that the hypocaloric group was more obese and had more patients with multiple comorbidities.
Are these differences due to glycemic control?
Taken together, these data suggest that hypocaloric feeding may improve outcomes, due in part to a lower infection rate. Hyperglycemia is strongly associated with infection rates and blood sugars more than 220 mg/dl on the first postoperative day are associated with an almost six-fold increased risk of serious infection . The role of glycemic control in critically ill patients and outcomes remains controversial, however, particularly in patients in medical ICUs. Van den Berghe et al. found that in surgical ICU patients, glycemic control to achieve euglycemia (80–110 mg/dl) was associated with lower mortality rates, mainly due to lower rates of bloodstream infections. Curiously, outcomes of studies performed by the same group of investigators in medical ICU patients failed to show lowered mortality rates with tight glycemic control but again showed a lower rate of bacteremia . Conclusion that may be drawn from these studies have been complicated further by a study by Brunkhorst in which critically ill patients, with an average BMI of approximately 27 kg/m2 were randomized to either conventional (180–200 mg/dl) or tight (80–110 mg/dl) glycemic control. This study failed to find any benefit to tight glycemic control and had a high rate of severe hypoglycemia [24••]. These data have been further supported by a recently published meta-analysis [25•].
It is possible that the difference in outcomes between these large studies is that of the patients studied; patients in a medical ICU often already have underlying infection as opposed to surgical patients who are often recovering from major surgery. Clearly, a major goal of care in surgical patients is to prevent development of secondary infection or wound infection. In combination, these data suggest that in diabetic, critically ill patients, near-normal glycemia may reduce the rate of development of new infection but may not improve outcomes in current infection. Nutrition support teams place a major focus on prevention on development of secondary infections such as catheter-related infections. Therefore, measures to prevent hyperglycemia in patients receiving nutrition support should lower the rate of TPN or catheter-associated bacteremia. Surprisingly, it is not clear that hypocaloric feeding necessarily improves glycemic control in critically ill patients. In the studies by Dickerson et al., and McCowen et al. mean blood sugars did not differ between groups. McCowen's study  was performed under a protocol in which glycemic control was achieved prior to increasing the carbohydrate content of TPN. As a consequence both groups actually ended up receiving hypocaloric nutrition (hypocaloric received 12 kcal/kg/day and the control group received 18 kcal/kg/day) so that differences in glycemic control did not reach statistical significance. Patients who were fed hypocalorically did have a tendency towards lower insulin requirements. Choban et al. found that among patients with non insulin-requiring diabetes, insulin was required for less than half as much time in the hypocalorically fed group as the eucaloric group(3.2 ± 2.7 d vs. 8.0 ± 0.5 d) and insulin dose requirements tended to be lower with hypocaloric feeding (36.1 ± 47.1 U vs. 61.1 ± 61.1 U, P = NS).
Risks of hypocaloric feeding
Despite the theoretical advantages of hypocaloric feeding, this is not a risk-free intervention. The metabolic adaptations to underfed, critically ill patients have not been extensively studied in prospective, clinical trials. The strongest evidence that hypocaloric feeding may worsen outcomes comes from a study by Villet et al.. In this prospective, observational study, in consecutive patients admitted to an ICU, negative caloric balance was strongly associated with worse clinical outcomes. Rubinson et al. found that failure to receive more than 25% of estimated caloric needs was associated with a greater incidence of nosocomial infection suggesting that hypocaloric intake itself may increase the rate of nosocomial infection. However, if hypocaloric feeding per se was responsible for these findings, one would anticipate that achieving greater energy intake would progressively reduce the rate of nosocomial infection. Infection rates between patients receiving 25–50% of goal calories and those receiving more than 50% of goal calories did not differ in Rubinson's study. Further, both of these studies were observational and as such were not capable of determining cause and effect. It is far more likely that failure to tolerate goal rates of feeding is simply a marker for patients who will have worse outcomes; indeed failure to tolerate goal feeding has been associated with worse outcomes in other studies .
Obesity, diabetes and critical illness all contribute to the metabolic abnormalities seen in critically ill patients. Failing to tolerate goal-feeding rates is associated with, but probably not responsible for, poor clinical outcomes. Feeding below 25% of predicted energy expenditure appears to increase the rate of nosocomial infections, possibly by exacerbating underlying insulin resistance or increasing negative nitrogen balance. Conversely, feeding at more than 66% of estimate daily caloric expenditure in the initial 2 weeks after admission to an ICU may worsen outcomes. Hypocaloric nutrition at 33–66% of estimated caloric needs should be considered in most, well nourished, critically ill patients to improve insulin sensitivity, lower the risk of hyperglycemia and its associated risk of infection. If hypocaloric nutrition is used, protein should be provided at 2 g/kg ideal body weight per day to attenuate losses of lean body mass. The optimal duration of hypocaloric nutrition is unclear, however, and should be a focus of randomized, prospective clinical trials.
This publication was made possible by the American Society for Nutrition, Physician Nutrition Support Specialist Award and National Institutes of Health grant DK 56341 (Clinical Nutrition Research Unit).
There are no conflicts of interests.
References and recommended reading
Papers of particular interest, published within the annual period of review, have been highlighted as:
• of special interest
•• of outstanding interest
Additional references related to this topic can also be found in the Current World Literature section in this issue (p. 170).
1 Mokdad AH, Serdula MK, Dietz WH, et al
. The spread of the obesity
epidemic in the United States, 1991–1998. JAMA 1999; 282:1519–1522.
2 Mokdad AH, Ford ES, Bowman BA, et al
. Prevalence of obesity
, and obesity
-related health risk factors, 2001. JAMA 2003; 289:76–79.
3 Barton RN. The neuroendocrinology of physical injury. Bailliere's clinical endocrinology and metabolism 1987; 1:355–374.
4 van der Poll T, Barber AE, Coyle SM, Lowry SF. Hypercortisolemia increases plasma interleukin-10 concentrations during human endotoxemia: a clinical research center study. J Clin Endocrinol Metab 1996; 81:3604–3606.
5 Tsigos C, Kyrou I, Chrousos GP, Papanicolaou DA. Prolonged suppression of corticosteroid-binding globulin by recombinant human interleukin-6 in man. J Clin Endocrinol Metab 1998; 83:3379.
6 Boden G, Shulman GI. Free fatty acids in obesity
and type 2 diabetes
: defining their role in the development of insulin resistance and beta-cell dysfunction. Eur J Clin Invest 2002; 32(Suppl 3):14–23.
7 Wolfe RR, Martini WZ. Changes in intermediary metabolism in severe surgical illness. World J Surg 2000; 24:639–647.
8 Jeevanandam M, Young DH, Schiller WR. Obesity
and the metabolic response to severe multiple trauma in man. J Clin Invest 1991; 87:262–269.
9 Finn PJ, Plank LD, Clark MA, et al
. Progressive cellular dehydration and proteolysis in critically ill patients. Lancet 1996; 347:654–656.
10 Griffiths RD. Muscle mass, survival, and the elderly ICU patient. Nutrition (Burbank, Los Angeles County, California) 1996; 12:456–458.
11 Van den Berghe G, de Zegher F, Lauwers P, Veldhuis JD. Growth hormone secretion in critical illness
: effect of dopamine. J Clin Endocrinol Metab 1994; 79:1141–1146.
12 Reeds PJ, Munday KA, Turner MR. Action of insulin and growth hormone on protein synthesis in muscle from nonhypophysectomized rabbits. Biochem J 1971; 125:515–520.
13 Carroll PV. Treatment with growth hormone and insulin-like growth factor-I in critical illness
. Best Pract Res 2001; 15:435–451.
14 Duska F, Andel M, Kubena A, Macdonald IA. Effects of acute starvation on insulin resistance in obese patients with and without type 2 diabetes
mellitus. Clin Nutr 2005; 24:1056–1064.
15 Bergman BC, Cornier MA, Horton TJ, Bessesen DH. Effects of fasting on insulin action and glucose kinetics in lean and obese men and women. Am J Physiol 2007; 293:E1103–E1111.
16 Sandstrom R, Drott C, Hyltander A, et al
. The effect of postoperative intravenous feeding (TPN) on outcome following major surgery evaluated in a randomized study. Ann Surg 1993; 217:185–195.
17 Choban PS, Burge JC, Scales D, Flancbaum L. Hypoenergetic nutrition support
in hospitalized obese patients: a simplified method for clinical application. Am J Clin Nutr 1997; 66:546–550.
18 McCowen KC, Friel C, Sternberg J, et al
. Hypocaloric total parenteral nutrition: effectiveness in prevention of hyperglycemia and infectious complications–a randomized clinical trial. Crit Care Med 2000; 28:3606–36011.
19 Krishnan JA, Parce PB, Martinez A, et al
. Caloric intake in medical ICU patients: consistency of care with guidelines and relationship to clinical outcomes. Chest 2003; 124:297–305.
20 Dickerson RN, Boschert KJ, Kudsk KA, Brown RO. Hypocaloric enteral tube feeding in critically ill obese patients. Nutrition (Burbank, Los Angeles County, California) 2002; 18:241–246.
21 Pomposelli JJ, Baxter JK 3rd, Babineau TJ, et al
. Early postoperative glucose control predicts nosocomial infection rate in diabetic patients. JPEN 1998; 22:77–81.
22 van den Berghe G, Wouters P, Weekers F, et al
. Intensive insulin therapy in the critically ill patients. N Engl J Med 2001; 345:1359–1367.
23 van den Berghe G, Wilmer A, Hermans G, et al
. Intensive insulin therapy in the medical ICU. N Engl J Med 2006; 354:449–461.
24•• Brunkhorst FM, Engel C, Bloos F, et al
. Intensive insulin therapy and pentastarch resuscitation in severe sepsis. N Engl J Med 2008; 358:125–139. This paper provides very strong cautionary evidence that very tight glycemic control
may worsen outcomes in critically ill patients.
25• Wiener RS, Wiener DC, Larson RJ. Benefits and risks of tight glucose control in critically ill adults: a meta-analysis. JAMA 2008; 300:933–944. This meta-analysis summarizes the largest studies evaluating the effects of tight glycemic control
on clinical outcomes in critically ill patients.
26 Villet S, Chiolero RL, Bollmann MD, et al
. Negative impact of hypocaloric feeding
and energy balance on clinical outcome in ICU patients. Clin Nutr 2005; 24:502–509.
27 Rubinson L, Diette GB, Song X, et al
. Low caloric intake is associated with nosocomial bloodstream infections in patients in the medical intensive care unit. Crit care Med 2004; 32:350–357.