Optimizing the Dose of Glutamine Dipeptides and Antioxidants in Critically Ill Patients: A Phase I Dose-Finding Study 
From the * Kingston General Hospital/Queen's
University, Kingston, Ontario, Canada; the Correspondence: Daren K. Heyland, Kingston General Hospital/Queen's University, 76 Stuart Street, Kingston, Ontario, Canada K7L 2V7. Electronic mail may be sent to dkh2{at}post.queensu.ca.
Background: Supplementation with glutamine and antioxidants may be
associated with an improvement in clinical outcomes, but the optimal dose of
these substrates is unknown. The purpose of this study was to determine the
safety of high doses of glutamine combined with antioxidants in critically ill
patients. Methods: We conducted a single-center, open-label,
dose-escalating clinical trial. Mechanically ventilated adult patients with
clinical evidence of hypoperfusion were sequentially enrolled to 1 of 5
groups. Group 1 (n = 30): no supplementation; group 2 (n = 7): 0.35 g/kg/d of
glutamine IV; group 3 (n = 7): same as group 2 plus 15 g/d of glutamine and
150 µg of selenium enterally; group 4 (n = 7): same as group 2 plus 30 g/d
of glutamine and 300 µg of selenium enterally; and group 5 (n = 7): same as
group 4 plus an additional 500 µg of selenium IV. After enrollment,
nutrients were started as soon as possible. All patients were fed enterally
according to clinical practice guidelines. Results: The primary
outcomes for this study were change in sequential organ function assessment
(SOFA) score and safety parameters. Secondary outcomes included whole blood
glutathione (GSH), thiobarbituric acid reactive substances (TBARS), and blood
cells' mitochondrial DNA/nuclear DNA ratio (RATIO). There were no adverse
events attributable to the study nutrients, and the maximum and In recent years, there have been numerous studies of the effect of various nutrients on clinical outcomes in critically ill patients. The individual examination of these studies often produces conflicting or inconclusive results.1–3 Furthermore, in the majority of studies, the sample sizes and event rates were too low to demonstrate significant differences in clinically important outcomes. We previously used meta-analytic techniques to determine the best overall estimate of treatment effects.1 From our review of >30 different nutrition-related topics, it would seem that glutamine and antioxidants have the greatest likelihood to positively affect clinical outcomes in critically ill patients.1 With respect to glutamine supplementation, when we aggregated across 16 studies in critical illness, we observed a significant reduction in mortality (risk ratio [RR], 0.75; 95% confidence interval [CI], 0.59–0.96; p = .02), in infectious complications (RR, 0.79; 95% CI, 0.63–0.98; p = .04), and in intensive care unit (ICU) length of stay (weighted mean difference in days, –4.50; 95% CI, –8.28 to –0.72; p = .02) in critically ill patients.4 With respect to antioxidant supplementation, we identified 13 trials that met the inclusion criteria, most of which studied the effects of selenium either alone or in combination with other trace elements and vitamins, whereas others looked at the effects of zinc and vitamins A, C, and E. When the results of all these trials were aggregated, overall antioxidants were associated with a significant reduction in mortality (RR, 0.70; 95% CI, 0.59–0.83; p < .0001) but had no effect on infectious complications (RR, 0.90; 95% CI, 0.65–1.24; p = .51).4 Despite the significant and large treatment effects observed in these meta-analyses, our overall results should be viewed more as hypothesis generating than hypothesis-confirming for the following reasons. First, compared with other meta-analyses, these reviews contained relatively few distinct trials, with even fewer observed clinical endpoints. Infrequent events in trials are more likely to produce unstable or erroneous estimates of treatment effect. Second, although we attempted to obtain data on an intent-to-treat basis, this was not possible in the majority of the cases. Third, the majority of trials in these meta-analyses provided the nutrient supplementation in association with parenteral nutrition (PN). The beneficial effect of supplementation in a critically ill population receiving enteral nutrition is less certain. It is not unusual for meta-analyses to disagree with subsequent large randomized clinical trials (RCTs), and such larger, confirmatory trials are clearly desirable.5 However, given that the upper 95% CI around the effect on mortality is <1.00 and that no individual study suggested harm, we can conclude with reasonable confidence that glutamine and antioxidant supplementation is safe (excludes harm). Therefore, we propose to move forward and test the hypothesis that glutamine and antioxidant supplementation improves the survival of critically ill patients. Before moving forward with such a large and expensive proposition, however, we felt it necessary to explore the optimal dose of these substrates. The reason some of the previous RCTs failed to demonstrate a treatment effect may relate to inadequate doses of these nutrients. At high doses, vitamin C, vitamin E, and selenium have been shown to have some pro-oxidant properties,6,7 and vitamin E is associated with excess mortality in patients with cardiovascular disease.8 Therefore, higher doses may not necessarily result in better outcomes. The purpose of this study was to determine the safety of high doses of glutamine combined with high doses of antioxidants in critically ill patients when administered separately from standard nutrition support.
This study was a single-center, open-label, phase I, dose-escalating, clinical trial conducted at the Kingston General Hospital, Kingston, Ontario, Canada. We enrolled consecutive eligible adult patients admitted to the ICU expected to stay >48 hours who required mechanical ventilation and had clinical evidence of hypoperfusion. We defined hypoperfusion as the need for vasopressor agents (norepinephrine, epinephrine, neosynephrine, vasopressin, or  5 mg/kg/min of dopamine) for >1 hour or a systolic blood pressure
 90 mm Hg or the mean arterial pressure <70 mm Hg for >1 hour despite
adequate fluid challenge. We excluded the following patients: not expected to
be in ICU for >24 hours (due to imminent death, withdrawing treatments, or
discharge), duration of stay in ICU >24 hours before enrollment, no
gastrointestinal (GI) tract access, severe head trauma (Glasgow Coma Score
[GCS] <8 or need for ventriculostomy), weight <50 kg, Child's class C
cirrhosis, pregnancy, and enrollment in other ICU interventional study. All
patients in both the control group and the prospectively enrolled study groups
met these eligibility criteria. These criteria were designed to include those patients who are likely to benefit from the therapeutic intervention tested in this study. We have included patients at highest risk for substrate deficiency9,10 and oxidative stress, whereas we excluded patients not likely to benefit from the intervention (not likely to survive beyond 24 hours and already in ICU >24 hours) and patients with no access to the GI tract. Given that glutamine is metabolized to glutamate and that there are concerns glutamate may be detrimental to the injured brain, we also excluded patients with severe brain injury. The amount of glutamine provided in the final groups (see below) may be excessive in small, malnourished patients or patients with cirrhosis so they were also excluded.
Dose-Escalating Study Groups Study nutrients were initiated as soon as possible after admission to ICU and provided continuously either parenterally (Dipeptiven and selenium) or enterally (Intestamin) until discharge, death, or up to a maximum of 28 days. Independent of study nutrients, all patients were fed according to the Canadian clinical practice guidelines that typically favored the use of standard enteral nutrition within the first 48 hours of ICU admission.1
Safety Assessments In addition to the above noted measurements, we tracked the development of renal failure requiring dialysis, adverse events related to study nutrients, and monitored routine blood work, including urea and liver function tests. In groups 2–5 only, we drew blood from study patients at baseline every Monday, Wednesday, and Friday while receiving the study protocol (for a maximum of 28 days), and 12 hours after discontinuation of the study nutrients. The tubes were spun to separate plasma and buffycoat-like cell pellet. The latter was stored frozen at –70°C until use. The plasma samples were processed, stored, and sent to a laboratory for measurement of plasma ammonia, plasma levels of provided substrates, and other markers, exploring how glutamine and antioxidants may exert their treatment effect. These latter tests included markers of antioxidant capacity (whole blood glutathione [GSH]), and markers of oxidative stress (thiobarbituric acid reactive substances [TBARS]). The buffycoat cellular pellet was used to measure an indirect marker of mitochondrial function, the levels of mitochondrial DNA (mtDNA) relative to nuclear DNA (nDNA) (RATIO). The relative mtDNA/nDNA ratios were determined by real-time polymerase chain reaction (PCR) with fluorescent probes, as described elsewhere.14 The nDNA level per cell can be considered constant; therefore, alterations of the ratio can be attributed to changes in mtDNA content. For reference, a concurrently assayed pool of DNA from 24 healthy male controls yielded a relative mtDNA/nDNA ratio of 1.09 ± 0.17. Finally, we followed study patients to evaluate tolerance of enteral nutrition, duration of mechanical ventilation, hospital length of stay, and short-term mortality. Of note, although samples were collected over 28 days, short-term survival status was extended to 31 days to include patients who were discharged from ICU alive but died within 72 hours.
Statistics
The primary endpoint of this study was the maximum For the mechanistic markers, we present plots of individual patient profiles, with average slopes overlaid by treatment group. The linear mixed model was used to estimate average slopes per group and to compare these slopes between groups. The mixed model included random intercepts, and random slopes were added when Akaike's information criterion, Schwarz's bayesian criterion, or the likelihood ratio test suggested they improved the fit.17 Consequently, a random intercepts model was used for the mtDNA/nDNA ratio, but the models for all other parameters included a random intercept and slope. The model is only suitable for detecting approximately linear trends over the first 14 days. All analyses were performed in SAS V9.1 (SAS Institute Inc, Cary, NC), and no adjustment was made for multiple comparisons.
Ethics
We enrolled 58 patients in this study; baseline characteristics of all patients by group assignment are shown in Table I. Overall, patients in groups 2–5 received 81% of study nutrients enterally and 87% of parenteral nutrients during the study period. In addition to study nutrients, 27 of 28 of these patients received standard enteral nutrition; no patient received PN. The average total amount of glutamine, protein, nitrogen, and calories received by study patients in each of the groups and their clinical outcomes is shown in Table II.
Primary Outcomes
Safety parameters. At baseline, the majority of study patients had a significant elevation in their serum creatinine. During the study period, the overall pattern of study nutrients on creatinine was consistent with resolution of renal dysfunction as the creatinine either improved or remained the same (see Table IV). One patient from group 2 began dialysis on the first study day and another patient on day 4. In group 4, one patient began dialysis on day 2 and another on day 3. No subjects from groups 3 or 5 underwent dialysis during the study period. These rates are similar to the control group, where 5 of 30 underwent dialysis during the study period, with 2 receiving dialysis by the first study day. However, as seen in Table IV, an increase in post-treatment urea was observed as the dose of glutamine increased across the 5 groups. There were no differences in plasma ammonia levels (see Table IV) or liver enzymes (data not shown) across groups. One patient in group 3 with a preexisting seizure disorder developed status epilepticus and was withdrawn from the study on day 10. The seizures persisted, life-sustaining treatments were withdrawn, and the patient died on day 17. Study nutrients were interrupted for 1–2 days in 3 patients from group 5 due to safety concerns about a high urea level (>50 µmol/L).
Secondary Outcomes
Mechanistic markers. Figure 1 plots GSH levels over time. In group 2, the p value is significant (p = .034), suggesting a decrease in GSH levels over time. With subsequent groups, the slopes straighten out and the p values are no longer significant, suggesting a greater preservation of GSH levels with escalating doses. The difference in slopes across all groups is not statistically different (p = .6). Figure 2 shows TBARS values in each group. In group 2, the slope is horizontal. With subsequent groups, the slopes decrease and by group 5, the p value is statistically significant (p = .028), suggesting a greater reduction in oxidative stress with the higher doses in group 5. The difference in slopes across all groups is not statistically significant (p = .25).
Figure 3 plots mitochondrial DNA/nuclear DNA ratio (RATIO). In group 2, the slope is horizontal (p = .99), suggesting no recovery of mitochondrial function at the lower doses. In both groups 3 and 5, the slopes increase and achieve statistical significance. The difference in slopes across all groups is statistically significant (p = .001), again suggesting that with higher doses there is increased mitochondrial function.
In a post hoc analysis, we evaluated the effect of antioxidants on the RATIO. Using group 2 levels as the control group, in Figure 4, we show that with increasing doses of antioxidants across the remaining groups, the RATIO significantly improves (p = .03).
Having systematically reviewed the literature on numerous nutrients and nutrition interventions, the use of glutamine supplementation and antioxidants seems most likely to improve the mortality of critically ill patients. We plan to move forward with a large-scale, multicenter trial of glutamine and antioxidants in the critical care setting,18 but before initiating this study, we needed to determine the optimal dose of glutamine and antioxidants. Our previous work suggests that high-dose therapy will afford the best opportunity to demonstrate a clinical effect. Therefore, the purpose of this project was to determine the safety of high-dose glutamine in combination with antioxidants when provided to critically ill patients receiving standard enteral nutrition support.
In this dose-escalating study, we found that the high doses in this study (approximately 40 g of glutamine and approximately 800 µg of selenium/d given via the combined IV and enteral route) seem to be safe. At baseline, the levels of all these measured nutrients were low, and glutamine, selenium, and vitamin C levels increased significantly, into the normal range, with treatment. There were no adverse events, either clinical or biochemical, attributable to the study nutrients. Overall, patterns of organ failure/resolution were similar across groups. In the groups that received high doses of glutamine, we did observe a higher plasma urea compared with controls or low-dose glutamine. This was likely due to the amount of nitrogen these patients received. This elevated urea was not associated with a decline in renal function and was not thought to be a significant safety hazard. Nevertheless, we did interrupt the glutamine doses for a few days in a few of the patients to minimize the continued rise of the urea levels. Although perhaps this study represented an unusual study design and small sample size, this approach is similar to that used in determining the maximal tolerable oral dose of glutamine in pediatric oncology.15 Beginning at 0.35 g/kg, investigators gave 4 groups of 1–6 patients (total n = 13) escalating doses of L-glutamine and concluded that 0.65 g/kg was the maximally tolerated glutamine dose in this patient population. They then plan to use this dose in subsequent randomized trials. Tjader and colleagues16 have also explored various doses of glutamine in critically ill patients. They randomized 40 ICU patients expected to stay for >5 days to receive 0, 0.28, 0.57, or 0.86 g/kg of L-glutamine administered concomitantly with PN. Similar to our study, they observed that all patients started with a low baseline plasma glutamine level, and as the dose of glutamine escalated, there was an increase in the plasma glutamine levels. However, their study examined the effect of glutamine supplementation on muscle glutamine concentration, muscle protein synthesis, and muscle protein content.16 Escalating doses of glutamine did not have an impact on these metabolic markers, suggesting that the treatment effect of glutamine may be due to other mechanisms.
Our findings also offer some insights into the potential mechanisms of benefit related to escalating doses of glutamine and antioxidant administration. Reactive oxygen species (ROS) and reactive nitrogen-oxygen species (RNOS) are generated during critical illness and are capable of damaging proteins, polysaccharides, nucleic acids, and polyunsaturated fatty acids, resulting in cellular damage and mitochondrial dysfunction.19 A recent study in critically ill patients demonstrates a relationship between increased inflammatory mediators, antioxidant depletion, reduced respiratory chain complex I activity in the mitochondria, and low cellular ATP levels.20 These endpoints correlated with severity of disease and clinical outcome and support the notion that mitochondrial dysfunction associated with bioenergetic failure may be an important factor in the underlying pathophysiology in critically ill patients with a systemic inflammatory response and multiorgan dysfunction.20 In critically ill patients, there are reduced stores of antioxidants, reduced plasma or intracellular concentrations of free electron scavengers or cofactors, and decreased activities of enzymatic systems involved in the detoxification of ROS.21 For the first time, we have shown that administration of high-dose glutamine and antioxidants may be associated with an improvement in mitochondrial function (or rather an increase in mitochondrial DNA). To the extent that mitochondrial DNA correlates with mitochondrial function, we can postulate that glutamine and antioxidant supplementation may be associated with greater antioxidant capacity (preservation of glutathione levels), less oxidative stress (less TBARS), and improved mitochondrial function (higher mtDNA content). This may explain the mechanism by which glutamine and antioxidants are associated with a reduction in mortality in critically ill patients.22,23 This study does not attempt to determine whether there is a different treatment effect between the enteral or parenteral delivery of these key nutrients. Previous studies have suggested that the majority of enteral glutamine is metabolized within the splanchnic area and may have minimal effect on systemic or blood levels of glutamine.24,25 Patients with low plasma glutamine levels have higher mortality rates compared with higher plasma levels.9 Therefore, additional parenteral supplementation may be necessary to achieve a systemic effect. Furthermore, the results of the glutamine and antioxidant meta-analysis seem to suggest that higher dose or the parenteral administration of the substrates may be associated with a greater treatment effect.22,23 The reason the parenteral route of administration appears more advantageous could be that it allows immediate delivery of the desired dose of these key nutrients that would enhance or support key immunologic functions.26,27 On the other hand, there are studies that support the notion that enterally provided glutamine supports the structure, function, and immunologic responsiveness of the GI tract, a desirable therapeutic effect in critically ill patients with hypoperfusion.28–31 For the purposes of the present study protocol and a subsequent phase III clinical trial, we achieved higher doses of study substrates by providing both parenteral and enteral nutrients. In doing so, to the extent that parenteral and enteral substrates have independent and different effects, we maximize the likelihood of demonstrating a treatment effect.
Combining the enteral nutrients with macronutrients in specialized commercial enteral formulas may limit the dose of the key substrates delivered because enteral nutrition is not adequately tolerated in the many critically ill patients, particularly in the early phase of their illness. We have overcome this deficiency by dissociating the nutrients from the nutrition support and administering them separately. Thus, this novel strategy involves parenteral nutrients, enteral nutrients, and enteral nutrition administered independently and simultaneously. This concept emphasizes that these individual nutrients, when given at high doses, have druglike effects and should be administered as separate pharmacologic agents. This new strategy has been termed pharmaconutrition, which is defined as using nutrients to modulate disease process to improve patient outcomes.32
Given the limitations of the nonrandomized design and small sample size, the inferences from this study are admittedly weak. The strongest inference we can make is that the high doses of glutamine (0.35 g/kg IV plus 30 g enterally daily) and antioxidants (500 µg IV plus 300 µg selenium, 20 mg of zinc, 10 mg of β-carotene, 500 mg of vitamin E, and 1500 mg of vitamin C provided enterally daily) tested in this study seem to be safe. They may even have positive effects on some physiologic or mechanistic parameters. A larger phase III clinical trial will be necessary to confirm their therapeutic effects on clinically important outcomes. Such a trial is currently under way in Canada under the auspices of the Canadian Critical Care Trials Group.33 We thank Kristina Stuerke and Ulrich Suchner of Fresenius Kabi for their assistance in providing thestudy nutrients. We also thank Dr James Reynolds, Kristen Singleton, Izabelle Gadawski, and Motoi Matsukura for their technical support. Received for publication July 13, 2006. Accepted for publication September 11, 2006.
Journal of Parenteral and Enteral Nutrition, Vol. 31, No. 2,
109-118 (2007) This article has been cited by other articles:
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
SOFA
did not differ across groups. In group 2, a significant decrease in GSH levels
was observed (p = .03). With subsequent groups, the slopes straighten
out and the p values are no longer significant, suggesting a greater
preservation of GSH levels with escalating doses. In group 2, the slope of the
line representing TBARS was horizontal. With subsequent groups, the slopes
decrease, and by group 5, this decrease reaches statistical significance
(p = .03), suggesting a greater reduction in oxidative stress with
the higher doses in group 5. The difference in slopes across all groups
describing the mitochondrial RATIO is statistically significant (p =
.001), again suggesting that, with higher doses, there is increased
mitochondrial function. Conclusions: The doses of glutamine and
antioxidants tested in this study seem to be safe and may have positive
effects on some mechanistic endpoints. A larger trial will be necessary to
confirm their therapeutic effects. 
Top
MATERIALS AND METHODS
RESULTS
