Lipid Oxidation and Nitrogen Balance in Critically Ill Obese PatientsDepartment of Physiology and Biophysics, College of Medicine, University of Illinois at Chicago Correspondence: Heather B. Breen, RD, CNSD, Department of Physiology and Biophysics, College of Medicine, University of Illinois at Chicago, 835 S. Wolcott Ave., Chicago, IL 60612-7342. Electronic mail may be sent to hbbreen{at}aol.com.
In 1991, the landmark Veterans Affairs Total Parenteral Nutrition Cooperative Study Group presented the unexpected finding that preoperative parenteral nutrition (PN) correlated with greater incidence of major complications in hospital patients with moderate or borderline malnutrition before surgery.1 At the same time, epidemiologists began to describe a burgeoning obesity epidemic in the population at large.2 Several multicenter studies including the Continuing Surveys of Food Intakes by Individuals conducted by the US Department of Agriculture revealed that food portion sizes started increasing in the mid-1980s,3 and television-watching was evaluated as an additional risk factor.4 The prevalence of obesity increased steadily through the past decades to its present status as one of the principal public health problems of the United States.5,6 In the 1990s, evaluation of altered metabolism in obese individuals was only beginning and not yet driven by the same urgency that is seen now. Several early 1990s review articles by prominent clinicians do not address nutrition support in this population7–12; however, it is now widely accepted that malnutrition can occur in most obese patients.13,14
The paper by Jeevanandam et al (Fig. 1),15 published in 1991, was presented in part at the 14th Clinical Congress of the American Society for Parenteral and Enteral Nutrition (January 1990). Using innovative techniques, the investigators evaluated 7 obese (BMI > 30 kg/m2) and 10 nonobese (BMI <30 kg/m2) multiple trauma patients admitted to the intensive care unit of a level 1 trauma center. Patients were enrolled in the protocol within 2–4 days of major injury when they were receiving maintenance fluid and electrolyte supplementation without calories or nitrogen. For convenience, this review will report results for obese patients first, followed by nonobese (control) patients in the discussion. Key findings have also been summarized in table form (Table 1).
Postresuscitation weights and blood samples were collected within 48 to 72 hours of injury. Mean BMI of obese patients (n = 7) was 36.2 kg/m2 (SEM ± 1.5); mean BMI of nonobese patients (n = 10) was 25.0 kg/m2 (SEM ± 1). Injury severity scores did not differ significantly between groups. To assess differences in basal metabolic rate, indirect calorimetry was conducted over 20 minutes after a 10-minute equilibration period. Measured resting energy expenditure (REE) did not differ significantly between the 2 groups (2550 ± 172 vs 2538 ± 162 kcal/d; NS). However, there was a significant difference when results were presented as a function of total body fat (56.7 ± 3.8 vs 198 ± 12.6 kcal/kg/d; p < .001), calculated according to body composition data reported by Wolfe et al.16 Mean respiratory quotient was 0.80 for both groups. Twenty-four-hour urine collections through a Foley catheter were initiated upon enrollment and continued until the conclusion of the study. Urinary creatinine was assessed as a reliable indicator of muscle mass in patients with normal renal function.17 Daily excretion was 2106 mg/d in obese and 1620 mg/d in nonobese patients. According to the calculation that 1 mg of creatinine represents 17.7 g of muscle tissue,17 results were 37.3 kg and 28.7 kg of muscle mass, respectively, which is consistent with 36% of body weight in both groups. Urinary urea nitrogen (UUN) and 3-methylhistidine (3MH) were measured in pooled aliquots representing 10% of the volume collected at each hour of study duration. Although UUN is to thought reflect total nitrogen metabolism,18 3MH is considered a reliable indicator of skeletal muscle proteolysis. Obese patients showed significantly greater nitrogen loss as urea (22.2 vs 14.3 g/d; p < .025) and 3MH (2.1 vs 1.6 g/d; p < .025). There were no significant differences in blood urea nitrogen (13.8 obese vs 12.1 nonobese; NS) during the collection period. Significant differences in UUN disappear when results are expressed as a function of body weight (211 vs 190 mg/kg; NS). To account for differences in body composition, the investigators also reported UUN as a function of lean body mass, again calculated from the data by Wolfe et al.16 Adjusted to milligram scale, results were 379 and 220 mg N/kg/day, respectively (p < .005). However, this difference, corresponding to <1 g of body protein per day, is not clinically significant. Adjustment data for 3MH are not presented. Whole body lipolysis rate (WBLR) was assessed by 2-stage primed infusion of glycerol (10% wt/wt). The carbon "backbone" of triglycerides, glycerol is not recycled in reesterification and serves as an index of endogenous lipid catabolism in fasting patients. Free fatty acids, in contrast, are recycled and reveal the balance between lipolysis and reesterification. In the first stage, the patient was primed with a 2-mmol dose of glycerol, followed by a 70-minute infusion of 250 µmol/min glycerol; in the second, priming was repeated and the dose advanced to 500 µmol/min. Blood samples were drawn sequentially in triplicate during the last 10 minutes of each stage. A significant increase in FFA was observed in obese patients (671 µM vs 392 µM; p < .005), whereas glycerol was significantly lower (69 µM vs 103 µM; p < .05). Ketone levels were 491 µM and 764 µM, respectively (NS). Following completion of WBLR testing, 24-hour infusion of [15N] glycine for assessment of whole body protein turnover (WBPT) was initiated. Isotope was administered through a venous catheter of the forearm. Priming dose was 2.26 mg [15N]/kg, followed by steady state infusion of 250 µg 15N/kg/min. Urine samples were assayed for isotope excretion as urea and ammonium at 0-, 12-, 16-, 18-, 20-, 22-, and 24-hour time points. Protein turnover rate, Q, and protein synthesis rate, S, did not differ significantly either absolutely or according to body weight. Obese individuals had higher rates of turnover (5.76 ± 0.56 vs 4.11 ± 0.25, p < .01) and synthesis (3.27 ± 0.40 vs 2.69 ± 0.16, NS) when adjusted by LBM. However, the same trend is seen when results are expressed as a function of ideal body weight (IBW). Because IBW is significantly correlated with height and gender,19,20 this introduces the possibility that protein turnover may actually be mediated by these variables, rather than obesity.
According to these findings, the investigators concluded that critically ill obese patients experience a relative block in lipolysis and impairment of protein synthesis. Evaluated absolutely, UUN, 3MH, and urinary creatinine differ significantly between the groups. These differences disappear when adjusted for actual body weight, although the authors caution that this does not consider differences in body composition between groups. However, while creatinine excretion suggests that muscle mass as percent body composition was equal between the groups, the authors subsequently use values from another study (Table 2) to present UUN and 3MH as a function of lean body mass. In that study, Wolfe et al16 studied 5 obese (mean BMI = 39.6 kg/m2) and five normal weight (mean BMI = 21.2 kg/m2) male volunteers of mean age 24 using the labeled water method. Review of the literature has shown significant differences in body composition across gender, age, and ethnic groups.21 In normal-weight men, the approximate range for muscle mass as percent of body composition is 12% to 20%; in women, it is 20% to 30%.22 Thus, the Wolfe sample differed significantly from the older, mixed gender sample of Jeevanandam et al. Therefore, although the body composition–adjusted data are of the greatest interest, extrapolation diminishes the confidence with which it can be interpreted. Indeed, the investigators' own data on body composition suggest that there is no difference between obese and nonobese patients on nitrogen balance parameters.
As indices of lipid metabolism, plasma FFA and glycerol are significantly different between groups. Lower glycerol concentrations suggest that lipolysis is lower in obese patients. However, the appearance of free fatty acids may be related to depressed oxidation or diversion to reesterification. It is possible, therefore, that lipid metabolism in stressed obese patients may be more efficient rather than impaired. Greater efficiency in energy metabolism has previously been observed in nonstressed obese individuals.23 Nordenstrom et al24 found that, among nonobese patients receiving 5% dextrose in IV fluids, plasma FFA accounted for less than one-half of total fat oxidation indicating that unlabeled sources that are not rapidly equilibrated in plasma accounted for the bulk of fat oxidation. Finally, the experimental design evaluated fasted trauma patients. Therefore, any putative impairment in lipid metabolism described by Jeevanandam et al15 may not be applicable to patients receiving nutrition support, possibly including D5W, or patients with conditions other than multiple trauma. Indeed, investigators evaluating hypocaloric PN regimens in obese patients report contradictory findings. Dickerson et al25 studied 13 severely stressed obese patients with postoperative complications including sepsis, anastomotic leaks, abscesses, fistulae, and wound dehiscence. Subjects averaged 208% IBW at 126.9 ± 60 kg. 1.5 g protein per kilogram of IBW was supplied with an average of 881 nonprotein kilocalories (NPK)/day or 51.5% of measured REE.25 REE dropped significantly over the course of the study, from 2205 ± 689 kcal to 1898 ± 498 kcal. Although not specifically a study goal, weight loss to 109.7 ± 32.5 kg was found, which is greater than predicted by energy deficits alone. UUN results showed positive nitrogen balance or equilibrium for all patients. Serum albumin and total iron binding capacity levels improved significantly. In addition, for 8 patients who did not receive parenteral lipid infusions, RQs obtained from indirect calorimetry were consistent with fat oxidation of 68% ± 19% of NPK REE.25 Burge et al26 replicated these findings in their protocol, which advanced protein to 2 g/kg IBW. Subsequently, Choban et al27 administered formulaic PN at 2.0 g protein/kg IBW and 14 ± 3 NPK kJ/kg IBW to a sample including critically ill and diabetic patients, also reporting positive nitrogen balance. Two interesting findings emerged with respect to glycemic control. First, although control patients required 61.1 ± 61.1 units of insulin per day, the treatment group required 36.1 ± 47.1 units per day (NS). Second, among individuals with non–insulin-dependent diabetes, the number of days where insulin therapy was required was decreased among treatment patients (3.2 ± 2.7 vs 8.0 ± 2.5 days; p < .05).27 These findings suggest that, in the context of modest doses of glucose, stressed obese patients are, in fact, able to oxidize lipid and accrue nitrogen.
The prevalence of obesity is increasing around the world, whereas comorbidities result in disproportionate presentation of overweight and obese patients in health care settings. However, management of nutrition support regimens in these patients remains controversial. Jeevanandam et al15 present an interesting preliminary study that has been extensively cited in the antecedent literature, but methodological issues and the frequent use of calculated values diminish the confidence with which their results can be interpreted. Future research should evaluate lipid oxidation, especially in patients receiving hypocaloric regimens. Assessment of β-hydroxybutyrate, a ketone intermelite of hepatic fatty acid oxidation, may clarify the relationship between free fatty acid oxidation, reesterification, and lipolysis.28 Contribution of other lipid sources such as tissue or plasma triglycerides, which are not in rapid equilibrium with plasma,24 to total fat oxidation is also of interest. Finally, results in total urinary nitrogen including ammonia losses (data collected but not reported in the present investigation) may shed further light on protein kinetics. Hypocaloric nutrition support regimens show promising preliminary evidence and merit further investigation, especially in view of the rising prevalence of obesity. In the meantime, it is inappropriate to extrapolate the findings of Jeevanandam et al15 in fasted, injured patients to patients receiving nutrition support regimens.
Pivotal Paper Review of Jeevanandam M, Young DH, Schiller WR. Obesity and the metabolic response to severe multiple trauma in man. J Clin Invest. 1991;87:262–269.
Nutrition in Clinical Practice, Vol. 20, No. 1,
98-102 (2005)
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Top
Unique Scientific Contribution
Validity