MedicineLakex

Cpccrn.org

The randomized comparative pediatric critical illness
stress-induced immune suppression (CRISIS) prevention trial
Joseph A. Carcillo, MD; J. Michael Dean, MD; Richard Holubkov, PhD; John Berger, MD; Kathleen L. Meert, MD;K. J. S. Anand, MBBS, DPhil; Jerry Zimmerman, MD, PhD; Christopher J. L. Newth, MB, ChB; Rick Harrison, MD;Jeri Burr, MS, RN-BC; C. C. R. C. Douglas F. Willson, MD; Carol Nicholson, MD Objectives: Nosocomial infection/sepsis occurs in up to
Measurements and Main Results: There were no differences by
40% of children requiring long-term intensive care. Zinc, se-
assigned treatment in the overall population with respect to time
lenium, glutamine, metoclopramide (a prolactin secretalogue),
until the first episode of nosocomial infection/sepsis (median
and/or whey protein supplementation have been effective in
whey protein 13.2 days vs. zinc, selenium, glutamine, and intra-
reducing infection and sepsis in other populations. We evalu-
venous metoclopramide 12.1 days; p ⴝ .29 by log-rank test) or
ated whether daily nutriceutical supplementation with zinc,
the rate of nosocomial infection/sepsis (4.83/100 days whey pro-
selenium, glutamine, and metoclopramide, compared to whey
tein vs. 4.99/100 days zinc, selenium, glutamine, and intravenous
protein, would reduce the occurrence of nosocomial infection/
metoclopramide; p ⴝ .81). Only 9% of the 293 subjects were
sepsis in this at-risk population.
immunocompromised and there was a reduction in rate of noso-
Design: Randomized, double-blinded, comparative effective-
comial infection/sepsis with zinc, selenium, glutamine, and intra-
ness trial.
venous metoclopramide in this immunocompromised group (6.09/
Setting: Eight pediatric intensive care units in the NICHD Col-
100 days whey protein vs. 1.57/100 days zinc, selenium,
laborative Pediatric Critical Care Research Network.
glutamine, and intravenous metoclopramide; p ⴝ .011).
Patients: Two hundred ninety-three long-term intensive care pa-
Conclusion: Compared with whey protein supplementation,
tients (age 1–17 yrs) expected to require >72 hrs of invasive care.
zinc, selenium, glutamine, and intravenous metoclopramide con-
Interventions: Patients were stratified according to immuno-
ferred no advantage in the immune-competent population. Further
compromised status and center and then were randomly assigned
evaluation of zinc, selenium, glutamine, and intravenous meto-
to receive daily enteral zinc, selenium, glutamine, and intravenous
clopramide supplementation is warranted in the immunocompro-
metoclopramide (n ⴝ 149), or daily enteral whey protein (n ⴝ
mised long-term pediatric intensive care unit patient. (Pediatr Crit
144) and intravenous saline for up to 28 days of intensive care
Care Med 2012; 13:000 – 000)
unit stay. The primary end point was time to development of
KEY WORDS: glutamine; nosocomial infection; prolactin; sele-
nosocomial sepsis/infection. The analysis was intention to treat.
nium; sepsis; whey protein; zinc
Centers for Disease Control care. Critical illness stress induces lym- tional supplementation is needed in this and Prevention recommen- phopenia and lymphocyte dysfunction as- population at risk for stress-induced lym- dations and bundled inter- sociated with hypoprolactinemia (1) and phocyte dysfunction and nosocomial in- ventions for preventing catheter-associ- also deficiencies in zinc and selenium (2, ated bloodstream infection, ventilator- 3) and amino acids (4, 5). Because lym- Metoclopramide, a prolactin secreta- associated pneumonia, and urinary phocyte integrity is important for the im- logue, administered at the dosage com- catheter-associated infections, nosoco- mune response to fight infection, stan- monly used for gastrointestinal prokinesis mial infection/sepsis remains a signifi- dard nutritional practice in critically ill maintains prolactin levels in the high- cant cause of morbidity in critically ill children includes zinc, selenium, and normal range in children. In mechanicallyventilated adults, metoclopramide delayedtime to onset of nosocomial pneumoniaby 50% but had no effect on the rate of From the Children's Hospital of Pittsburgh of UPMC of Health and Human Services (DHHS; U10HD050096, (JAC), Pittsburgh, PA; University of Utah (JMD, RH, JB); U10HD049981, U10HD500009, U10HD049945, nosocomial pneumonia or mortality (6).
Children's National Medical Center (JB); Children's U10HD049983, U10HD050012, and U01HD049934).
In malnourished children, zinc supple- Hospital of Michigan (KLM); Arkansas Children's Hos- Members of the Data Safety Monitoring Board mentation reduced morbidity and mor- pital (KJSA); Seattle Children's Hospital (JZ); Children's include Jeffrey R. Fineman, MD, Jeffrey Blumer, PhD, tality with severe pneumonia (7, 8) or Hospital Los Angeles (CJLN); University of California at MD, Thomas P. Green, MD, and David Glidden, PhD.
Los Angeles (RH), Los Angeles, CA; University of Vir- For information regarding this article, E-mail: diarrhea (9 –11) and reduced infectious ginia Children's Hospital (CDFW); and National Institute disease mortality in small-for-gestation- of Child Health and Human Development (CN).
Copyright 2011 by the Society of Critical Care al-age infants (12). Selenium supplemen- Supported, in part, by the following cooperative Medicine and the World Federation of Pediatric Inten- tation (13) or glutamine-enriched enteral agreements from the Eunice Kennedy Shriver National sive and Critical Care Societies nutrition (14) also reduced the risk of Institute of Child Health and Human Development (NICHD), National Institutes of Health (NIH), Department nosocomial sepsis in preterm neonates.
Pediatr Crit Care Med 2012 Vol. 13, No. 4

Enteral glutamine safely maintains T 1 lymphocyte function for bacterial killing(15, 16).
Essential amino acids are also impor- tant to overall immune function and lym-phocyte function in particular (5). Wheyprotein provides all the essential aminoacids needed to maintain immune func-tion in immune cells. Experimental ani-mal studies show that whey protein sup-plementation facilitates maturation ofthe immune system and is protectiveagainst rotavirus (17–21). Randomizedhuman clinical studies of whey proteinhave demonstrated improved lymphocytefunction and reduction in coinfection inhuman immunodeficiency virus-infectedchildren, reduction in infection in se-verely burned children, and improved im-munologic response to immunization inthe elderly (22–26).
The Eunice Kennedy Shriver National Institute of Child Health and Human De-velopment Collaborative Pediatric Criti-cal Care Research Network investigatorshypothesized that critical illness stress-induced immune suppression-relatednosocomial infection/sepsis would be Figure 1. Screening, enrollment, randomization, and study completion.
more effectively prevented by prophylac-tic supplementation of "standard" nutri- tor (J.M.D.) acted as the sponsor. The study was Safety Monitoring Board before authorizing tional practice with added zinc, selenium, registered with ClinicalTrials.gov (number enrollment of infants younger than 1 yr. After glutamine, and metoclopramide, (ZSGM) the first interim analysis, the Data Safety Mon- than by prophylactic supplementation Patients were eligible for enrollment if itoring Board deferred this authorization until with added amino acid (whey protein).
they: were older than 1 yr and younger than 18 a second interim analysis could be reviewed.
The Collaborative Pediatric Critical Care yrs of age; were within the first 48 hrs of PICU At the time of the second review, the trial was Research Network designed a random- admission; had an endotracheal tube, central terminated for futility. Therefore, no infants ized, double-blinded, comparative effec- venous catheter (new or old, tunneled or not younger than 1 yr were enrolled in the trial.
tiveness trial with the primary hypothesis tunneled), or urinary catheter; were antici- The study commenced in April 2007 and ter- that daily ZSGM would prolong the time pated to have an indwelling arterial or venous minated in November 2009.
to development of nosocomial infection/ catheter for blood sampling during the first 3 After informed consent was obtained from sepsis compared to daily whey protein. In days of study enrollment; and were anticipated parents, subjects were randomized by tele- this article we report the results of the to have venous access and an enteral feeding phone according to an a priori design using critical illness stress-induced immune tube for the administration of study drugs.
randomized blocks of variable length, strati- suppression prevention trial.
Patients were excluded from enrollment if fied according to center and immunocompro- they: had a known allergy to metoclopramide; mised status. Immunocompromised status MATERIALS AND METHODS
were expected to have planned removal of en- was defined by the known presence of acquired dotracheal tube, central venous, and urinary immune deficiency syndrome, cancer, trans- This randomized, double-blinded, compara- catheters within 72 hrs after study enroll- plantation, primary immune deficiency, or tive study was performed on two parallel groups ment; had suspected intestinal obstruction; chronic immune suppressant therapy. Chil- of children at eight pediatric intensive care units had intestinal surgery or bowel disruption; dren were randomized in a 1:1 ratio into the (PICUs) in the Collaborative Pediatric Critical had other contraindications to the enteral ad- two arms of the trial in these stratified groups.
Care Research Network. The Institutional Re- ministration of drugs or nutrients; received All patients, medical and nursing staffs, clini- view Boards of all Collaborative Pediatric Critical chronic metoclopramide therapy before en- cal site monitors, and most DCC staff re- Care Research Network centers approved the rollment; had a known allergy to whey (cow's mained blinded throughout the study period.
protocol and informed consent documents. Pa- milk) or soy-based products; had been dis- The DCC biostatistician (R.H.) prepared Data rental permission was provided for each subject.
charged from the PICU in the previous 28 Safety Monitoring Board reports and reviewed An independent Data Safety Monitoring Board days; had been previously enrolled in this results in the two arms but remained blinded was appointed by the NICHD, and two interim study; or had a positive pregnancy test. Pa- to actual group assignment throughout the safety and efficacy analyses were planned. The tients were also excluded if their parents indi- study period. Central and clinical site research study was performed under an Investigational cated a lack of commitment to aggressive in- pharmacists and the pharmacy site monitor New Drug application from the Food and Drug tensive care therapies.
were unblinded throughout the study.
Administration (investigational new drug The Food and Drug Administration re- Subjects were randomized to receive en- 74,500), for which the DCC Principal Investiga- quired an interim analysis review by the Data teral whey protein powder and intravenous Pediatr Crit Care Med 2012 Vol. 13, No. 4

Table 1. Epidemiologic and clinical characteristics at admission
diatric Critical Care Research Network inves-tigators adjudicated the presence or absence of Zinc, Selenium, Glutamine, a nosocomial clinical sepsis or infection event and Metoclopramide Group for every subject. Each case was reviewed in- Group (N ⫽ 144) dependently by two investigators and pre-sented in detail so that consensus for all out- Age (yrs), median (range) comes was attained. All participants in this Pediatric logistic organ dysfunction, median (range) adjudication process were blinded to treat- Pediatric risk of mortality, median (range) ment arm through the study period.
Organ failure index, median (range) Secondary outcome variables of this study Immunocompromised (%) included the rate of nosocomial infection/sepsis Postoperative pediatric intensive care unit per 100 PICU days, number of antibiotic-free Primary diagnosis (%) days, incidence of prolonged lymphopenia (ALC ⱕ1000/mm3 for ⱖ7 days), serum prolac- tin, zinc, and selenium levels before treatment and after 7 days of treatment, and the safety Cardiovascular disease indicator 28-day mortality and adverse events.
Serum zinc and selenium levels were classified as deficient if they were below the pediatric reference ranges of the core laboratory. Zinc deficiency was defined as a level ⬍0.60 ␮g/mL Human immunodeficiency virus in children aged 10 yrs or younger and ␮g/mL in children aged at least 11 yrs. Sele- nium deficiency was defined as a level ⬍70 ng/mL in children aged 10 yrs or younger and ⬍95 ng/mL in children aged at least 11 yrs.
Chronic diagnoses (%) Prolactin deficiency was defined as a level of Malnutrition (reported as primary or secondary ⬍3 ng/mL across all ages. Glutamine levels Infection status at entry (%) were not measured because they are not con- Existing infection sidered indicative of total body reserves.
The sample size was calculated to provide No infection or sepsis 90% power to detect an inverse hazard rate of Existing lymphopenia (ALC ⱕ1000/mm3) Baseline core laboratory data (%)
1.5 using a two-sided nonparametric test (log- rank test) with type I error (␣) of 0.05. This Prolactin deficiency required accrual of subjects until 263 had nos- Selenium deficiency ocomial events; the estimated total samplesize was 600 subjects based on initial eventrate estimates. A log-rank test with a two-sided saline (whey protein group) or ZSGM (ZSGM between admission to the PICU and occur- ␣ ⫽ 0.05, stratified by immune-compromised group). Subjects assigned to the whey protein rence of nosocomial infection/clinical sepsis in status, was used to compare the primary end group received 0.3 g/kg beneprotein each PICU patients who have endotracheal tubes, point of freedom from nosocomial infection or morning and intravenous saline every 12 hrs.
central venous catheters, or urinary catheters.
sepsis (from time of admission to the PICU Subjects assigned to the ZSGM group received Nosocomial events were defined as occurring until up to 5 days after discharge from the zinc (20 mg), selenium (40 ␮g age 1–3 yrs, at least 48 hrs after PICU admission during the PICU) between treatment arms. Outcome 100 ␮g age 3–5 yrs, 200 ␮g age 5–12 yrs, 400 hospital stay until 5 days after discharge from rates over time are presented using Kaplan- ␮g adolescent), and glutamine (0.3 g/kg) each the PICU; for children remaining in the PICU Meier freedom from event curves. In the sub- morning, and intravenous metoclopramide for ⬎28 days after randomization, events were group of immunocompromised whey protein (0.2 mg/kg, maximum 10 mg) every 12 hrs. All counted until day 33. The study protocol re- patients, median time to nosocomial infec- study drugs were shipped from a central phar- quired that patients be randomized within 48 tion/sepsis was derived at the 50.5% event-free macy (University of Utah) and dispensed by hrs of PICU admission and that study drug site research pharmacists. Subjects received administration begin within 72 hrs of PICU study drug until discharge from the PICU or admission. According to Center for Disease In prespecified analyses complementary to for 28 days from the time of randomization, Control and Prevention definitions, clinical the primary analysis, rates of infection were whichever occurred earlier. Enteral drugs sepsis occurs when patients older than 1 yr analyzed using Poisson regression analyses, were administered by feeding tube and discon- have development of fever (ⱖ38°C), hypoten- which count multiple events for a single sub- tinued if the feeding tube was removed or if sion (ⱕ90 mm Hg), or oliguria (ⱕ20 mL/hr) ject. Additionally, numbers and proportions of contraindications to enteral feeding developed and a clinician initiates antibiotic therapy with antibiotic-free days during the PICU stay were during the study. Intravenous drugs were dis- no positive microbiological evidence and no compared between study arms using the Wil- continued if the intravenous was removed.
other recognized cause. Nosocomial infection coxon rank-sum test. Incidence of prolonged The hypothesis of the critical illness stress- occurs when microbiologically (culture, anti- lymphopenia and all-cause mortality and ad- induced immune suppression Prevention trial gen, polymerase chain reaction, or antibody) verse events at 28 days after randomization was that daily prophylaxis with enteral zinc, proven infection is observed in a patient with were analyzed using the chi-square test or selenium, and glutamine, and intravenous fever, hypothermia, chills, or hypotension.
exact analogues when numbers of events were metoclopramide would delay the time (hours) The treatment arm blinded Collaborative Pe- small. Four ZSGM-assigned subjects (one of Pediatr Crit Care Med 2012 Vol. 13, No. 4

Figure 2. Top, Freedom from nosocomial sepsis according to assigned treatment for all randomized patients. Numbers above the horizontal time axis
denote number of patients remaining at risk at each time point. p ⫽ .29 for log-rank test comparing curves between study arms, stratified by immune
competent status. Middle, Freedom from nosocomial infection/sepsis according to assigned treatment for patients immunocompromised at study entry.
Numbers above the horizontal time axis denote number of patients remaining at risk at each time point. p ⫽ .24 for log-rank test comparing curves between
study arms. Lower, Freedom from nosocomial infection/sepsis according to assigned treatment for patients who were immune-competent at study entry.
Numbers above the horizontal time axis denote number of patients remaining at risk at each time point. p ⫽ .16 for log-rank test comparing curves between
study arms.
whom received no study treatment) had un- ment, randomization, and study comple- Treatment with ZSGM did not delay known 28-day survival status.
tion results.
the time until nosocomial infection/ Five factors (immunocompromised status, Table 1 shows the epidemiologic and sepsis compared to treatment with whey postoperative status, gender, race/ethnicity, clinical characteristics of the study pop- protein (median time, whey protein 13.2 and center) were prespecified for subgroup ulation at the time of enrollment in both days vs. ZSGM 12.1 days; log-rank p ⫽ analysis, and the Data Safety Monitoring treatment arms. The median age of chil- .29; Fig. 2 top). The median PICU stay Board subsequently added another factor (in- dren was 7 yrs and ⬍10% were immuno- was 10 days. Of subjects at risk for an fection or sepsis at study entry). The inten- compromised on entry. Baseline charac- event, approximately 50% in each treat- tion-to-treat approach was used for all study teristics were equally distributed between ment arm were event-free at 14 days after analyses of efficacy. Analysis by treatment re- the study arms. Among patients assigned PICU admission. The effect of immuno- ceived was performed for the safety outcomes to whey protein, 46.5% received paren- compromised status on time to nosoco- of mortality and occurrence of adverse events.
teral nutrition and 89.6% received en- mial infection/sepsis was not significant teral nutrition, compared to 43.0% and (median time in immunocompromised 90.6% of patients assigned to ZSGM. The patients, whey protein 10 days vs. ZSGM We enrolled 293 subjects (Fig. 1). En- proportion of patients who were allowed 32.4 days; median time in immune com- rollment was terminated for futility after nothing by mouth during one or more petent patients, whey protein 13.2 days the second interim analysis indicated that PICU study days was 55% in the whey vs. ZSGM 11.8 days; p ⫽ .12) for interac- the conditional power to determine a protein arm and 53% in the ZSGM arm, tion between treatment group and immu- beneficial effect by ZSGM, compared to with the average proportion of PICU nocompromised status in the time-to- the whey protein therapy, was ⬍10%.
study days being allowed nothing by event analysis (Fig. 2 middle, bottom).
Figure 1 shows the screening, enroll- mouth at 14% and 13%, respectively.
Other subgroup factors examined were Pediatr Crit Care Med 2012 Vol. 13, No. 4

Figure 2. Continued.
Pediatr Crit Care Med 2012 Vol. 13, No. 4

Table 2. Rates of nosocomial infection/sepsis, days of invasive lines, urinary catheterization, and
lenium (40% vs. 38%), glutamine (8% vs.
mechanical ventilation by treatment group 7%) primarily as part of routine TPN, andmetoclopramide (3% vs. 5%) for facilita- Zinc, Selenium, Glutamine, tion of nasoduodenal tube placement or and Metoclopramide gastroesophageal reflux, respectively.
Group (N ⫽ 144) Group (N ⫽ 149) Seven-day levels of all three measures Total events (infection or sepsis) were significantly higher in ZSGM pa- Total pediatric intensive care unit days tients than whey protein patients, and Total study daysa change from baseline was also higher Mean events/patient/100 study days (p ⬍ .001 for all six comparisons). At 7 95% confidence interval days, zinc deficiency was present in 19 of Therapeutic risk factorsDays in pediatric intensive care unit 83 (23%) ZSGM subjects vs. 36 of 80 (45%) whey protein subjects, selenium Ventilator days (mean/median) deficiency in 10 of 84 (12%) ZSGM sub- Central venous catheter days (mean/median) jects vs. 23 of 80 (29%) whey protein Endotracheal tube days (mean/median) subjects, and prolactin deficiency in 3 of Urinary catheter days (mean/median) Total ventilator days 84 (4%) ZSGM subjects vs. 14 of 81 (17%) Total respiratory infections whey protein subjects. Controlling for Mean respiratory infections/patient presence of baseline deficiencies, ZSGM Per 100 ventilator days (95% confidence interval) subjects with 7-day measures showed sig- Total urinary catheter days Total urinary tract infections nificantly lower 7-day deficiency rates Mean urinary tract infections/patient compared to whey protein subjects for Per 100 urinary catheter days (95% zinc (p ⫽ .001 by Cochran-Mantel- confidence interval) Haenszel test), selenium (p ⫽ .009), and Total central venous catheter days prolactin (p ⫽ .014).
Total bacteremia infections Mean bacteremia infections/patient Overall 28-day mortality was 8.1% Per 100 central venous catheter days (95% among the 284 children who received confidence interval) whey protein or ZSGM and had known28-day status. There was no significant aStudy days indicates days in pediatric intensive care unit plus additional days after pediatric difference in 28-day mortality by treat- intensive care unit discharge that patient was followed-up for events (5 days unless patient was ment received between whey protein and discharged from hospital earlier).
ZSGM (8/139 [5.8%] vs. 15/145 [10.3%];p ⫽ .16). Among the 287 children receiv- not significantly associated with the pri- treatment arms (Table 2). Distribution of ing treatment, there were 2624 adverse mary end point.
events, sites of infection, and infecting events, including 113 serious adverse There was no difference in the rate of organisms were also generally similar be- events with no significant differences by nosocomial infection or clinical sepsis tween the treatment groups (Table 3).
treatment received for whey protein and per 100 PICU days between the ZSGM and In the immunocompromised popula- ZSGM. Among 139 subjects receiving whey protein groups (Table 2; p ⫽ .81).
tion, the rate of nosocomial infection/ only whey protein treatment, adverse Examining study days in the PICU, me- sepsis was reduced in the ZSGM group events were reported in 126 (90.6%) and dian number of antibiotic-free days (2 vs.
compared with the whey protein group serious adverse events were reported in 1; p ⫽ .09) and proportion of days (17% (unadjusted p ⫽ .006 for interaction be- 37 (26.6%), whereas among 148 subjects vs. 10%; p ⫽ .19) did not differ between tween treatment group and immunocom- receiving ZSGM regimen adverse events subjects assigned to whey protein vs.
promised status; Table 4). The causes for were reported in 135 (91.2%) and serious ZSGM. There was no significant differ- immune compromise in whey protein adverse events were reported in 39 ence in the incidence of prolonged lym- compared to the ZSGM groups were bone (26.4%). There were also no differences phopenia (ALC ⱕ1000/mm3 for ⱖ7 days) marrow transplant two vs. three, other in specific adverse events, including diar- between subjects assigned to whey pro- organ transplant five vs. one, cancer one rhea (whey protein 12.2% vs. ZSGM tein (12/144 [8.3%] vs. ZSGM 5/149 vs. three, human immunodeficiency virus 12.2%), dysrhythmias (arrhythmia, extra- [3.4%]; p ⫽ .07). In the study population, zero vs. one, severe neutropenia one vs.
systole, nodal rhythm; whey protein 4.3% 41% receiving whey protein and 42% re- one, chronic high-dose steroids/immune vs. ZSGM 4.1%), and abnormal move- ceiving ZSGM had development of noso- suppressants one vs. three, congenital ment (akathisia, choreoathetosis, dyski- comial infection or sepsis. Approximately immunodeficiency one vs. one, and ther- nesia, dystonia; whey protein 2.9% vs.
one-third of patients had development of apeutic hypothermia zero vs. one.
nosocomial infection and one-fifth had Among subjects with available base- development of sepsis. Days of invasive line values, zinc deficiency was present at lines, urinary catheterization, and me- baseline in 235 of 280 (84%), selenium chanical ventilation and rates of site- deficiency in 156 of 278 (56%), and pro- Similar to previous literature, we ob- specific infections based on denominators lactin deficiency in 68 of 284 (24%) (Ta- served that nosocomial infection/sepsis of ventilator days, urinary catheter days, ble 1). Among whey protein and ZSGM occurred in ⬎40% of long-term PICU pa- and central venous catheter days were subjects, standard of care included the tients, with ⬍50% of these children be- not significantly different between the common use of zinc (44% vs. 39%), se- ing free from nosocomial infection or Pediatr Crit Care Med 2012 Vol. 13, No. 4

Table 3. Nosocomial infection/sepsis and sites of nosocomial infection by treatment group
needed, perhaps in a specialized PICUnetwork with a larger immunocompro- Zinc, Selenium, Glutamine, mised population, or in a PICU network and Metoclopramide with a larger number of centers to prop- Group (N ⫽ 144) Group (N ⫽ 149) erly assess the significance of this signal.
Patients with events With regard to limitations, several One or more events (%) study design and performance variables Nosocomial infection (%) require the reader's consideration. First, Nosocomial sepsis (%) this trial was performed in the "standard Site of infection practice" setting. Protein, zinc, and sele- Lower respiratory nium are an accepted part of "standard" Upper respiratory pediatric enteral and parenteral nutrition in the intensive care setting (29, 30) and Skin or soft tissue no effort was made to control this nutri- tional practice. The study results cannotbe applied to patients who are without any nutrition in the PICU. Second, ourtrial compared the effectiveness of two Total infecting organisms nutriceutical strategies to one another, Candida albicans rather than to placebo. The research planning committee wanted to follow previous study designs of glutamine sup- Candida glabrata Candida lusitanae plementation in newborns that used sin- gle amino acids as "placebos" to address Gram-negative bacilli potential criticism that an apparent effect in a glutamine arm could be an effect of protein nutrition rather than of glu- tamine per se. This rationale is problem- atic because all amino acids have specific immune cell effects and therefore are not Gram-positive bacilli true placebos (5). Whey protein was the Gram-negative cocci only Food and Drug Administration- approved amino acid supplement avail- Gram-positive cocci able to us. Because whey protein is mar- Staphylococcus coagulase negative keted as immune nutrition, we designed a comparative effectiveness trial rather than a true placebo-controlled trial. A true placebo arm without any zinc, sele- nium, or protein was considered outside of human subjects standards. Two ongoingadult trials comparing the use of a dopa- sepsis at 14 days and median time to The ZSGM supplement did not prevent mine-2 antagonist to placebo in mechani- nosocomial infection or sepsis being just persistent lymphopenia or nosocomial in- cal ventilation and zinc, selenium, and glu- less than 14 days (27). Similar to previous fection/sepsis compared with essential tamine supplements to placebo in severe reports, we also found a high incidence of amino acid supplementation from whey sepsis (NCT0013978, NCT00300391) will critical illness stress-related zinc and se- protein in the overall study population.
give information on the effect of these lenium deficiency, as well as prolactin We stratified the randomization of pa- supplements in the absence of concomi- deficiency in 24% and lymphopenia in tients to nutriceutical treatment arms ac- tant protein supplementation. Approxi- nearly 40% of the patients (1, 28). The cording to immunocompromised status mately ten patients in each treatment observation that nearly 95% of subjects because we thought it was biologically arm either did not receive the assigned had deficiencies at enrollment supports plausible that the T 2 phenotype-domi- treatment, or they had their treatments the study design that used the multi- nant immunocompromised group of pa- stopped prematurely on parental request.
modal ZSGM supplement strategy and tients would benefit differentially from Post hoc analysis excluding these patients analyzed the effects on both the pre hoc ZSGM supplementation. In this regard, did not change the overall findings of the stratified immune competent and immu- we did observe a reduction in the noso- study. Fourth, a low number of antibiot- nocompromised populations. We ob- comial infection/sepsis rate with use of ic-free days in the subjects enrolled in served more frequent resolution of zinc, ZSGM in this at-risk population. Because either arm of this study was discovered.
selenium, and prolactin deficiencies at 7 ⬍10% of our general PICU population This calls into question whether high an- days with ZSGM, but this pharmacoki- was immunocompromised, the small tibiotic use diminished any effects of the netic effect was not matched with the sample size leads us to view these find- nutriceuticals. However, post hoc analy- hypothesized pharmacodynamic effect.
ings rather cautiously. Repeated study is sis found no association between extent Pediatr Crit Care Med 2012 Vol. 13, No. 4 Table 4. Rates of nosocomial infection/sepsis per 100 days by treatment group and immunocompro-
immune system during zinc deficiency.
Annu Rev Nutr 2004; 24:277–298 3. Stone CA, Kawai K, Kupka R, et al: Role of Zinc, Selenium, Glutamine, selenium in HIV infection. Nutr Rev 2010; and Metoclopramide 4. Roth E: Immune and cell modulation by amino acids. Clin Nutr 2007; 26:535–544 5. Li P, Yin YL, Li D, et al: Amino acids and Total events (infection or sepsis) immune function. Br J Nutr 2007; 98: Total pediatric intensive care unit days 6. Yavagal DR, Karnad DR, Oak JL: Metoclopra- Mean events/patient/100 study days 6.09 (3.33–10.32) 1.57 (0.53–3.73) mide for preventing pneumonia in critically (95% confidence interval) ill patients receiving enteral tube feeding: Arandomized controlled trial. Crit Care Med Zinc, Selenium, Glutamine, 2000; 28:1408 –1411 and Metoclopramide 7. Brooks WA, Yunus M, Santosham M, et al: Group (N ⫽ 133) group (N ⫽ 135) Zinc for severe pneumonia in very young children: Double-blind placebo-controlled Total events (infection or sepsis) trial. Lancet 2004; 363:1683–1688 Total pediatric intensive care unit days 8. Fischer Walker C, Black RE: Zinc and the risk for infectious disease. Annu Rev Nutr Mean events/patient/100 study days 4.72 (3.87–5.69) 5.44 (4.47–6.55) 2004; 24:255–275 (95% confidence interval) 9. Baqui AH, Black RE, El Arifeen S, et al: Zinc therapy for diarrhoea increased the use oforal rehydration therapy and reduced the useof antibiotics in Bangladeshi children.
of antibiotic use and evidence of treat- (Data Coordinating Center), Salt Lake J Health Popul Nutr 2004; 22:440 – 442 ment effect.
City, UT: J. Michael Dean, MD, MBA, Jeri 10. Bhatnagar S, Bahl R, Sharma PK, et al: Zinc Burr, MS, RN-BC, CCRC, Amy Donald- with oral rehydration therapy reduces stool son, MS, Richard Holubkov, PhD, Angie output and duration of diarrhea in hospital- Webster, MStat, Stephanie Bisping, RN, ized children: A randomized controlled trial.
Nosocomial infection and sepsis re- Teresa Liu, MPH, Brandon Jorgenson, J Pediatr Gastroenterol Nutr 2004; 38:34 – 40 mains a prevalent public health problem BS, Rene Enriquez, BS, Jeff Yearley, BS; 11. Raqib R, Roy SK, Rahman MJ, et al: Effect of in critically ill children with long-term Children's National Medical Center, WA zinc supplementation on immune and in- stay in the PICU. Implementation of Cen- DC: John Berger, MD, Angela Wratney, flammatory responses in pediatric patients ters for Disease Control and Prevention- MD, Jean Reardon, BSN, RN; Children's with shigellosis. Am J Clin Nutr 2004; 79: recommended practices is the first step in Hospital of Michigan, Detroit, MI: Kath- prevention. Our study was performed on leen L. Meert, MD, Sabrina Heidemann, 12. Sazawal S, Black RE, Menon VP, et al: Zinc the premise that evaluation of the com- supplementation in infants born small for MD, Maureen Frey, PhD, RN; Arkansas parative effectiveness of prophylactic nu- gestational age reduces mortality: A prospec- Children's Hospital, Little Rock, AR: KJS tritional support strategies could inform tive, randomized, controlled trial. Pediatrics Anand, MBBS, DPhil, Parthak Prodhan, further improvement. The novel multi- 2001; 108:1280 –1286 MD, Glenda Hefley, MNSc, RN; Seattle 13. Darlow BA, Austin NC: Selenium supplemen- modal strategy designed and used to re- Children's Hospital, Seattle, WA: Jerry tation to prevent short-term morbidity in duce critical illness stress-induced zinc, Zimmerman, MD, PhD, David Jardine, preterm neonates. Cochrane Database Syst selenium, glutamine, and prolactin defi- MD, Ruth Barker, RRT; Children's Hospi- Rev 2003; 4:CD003312 ciencies was successful in part in revers- tal Los Angeles, Los Angeles, CA: Chris- 14. van den Berg A, van Elburg RM, Westerbeek ing these deficiencies. It conferred no ad- topher J. L. Newth, MB, ChB, J. Francisco EA, et al: Glutamine-enriched enteral nutri- vantage in nosocomial infection and Fajardo, CLS (ASCP), RN, MD; Mattel tion in very-low-birth-weight infants and ef- sepsis prevention in immune-competent Children's Hospital at University of Cali- fects on feeding tolerance and infectious children compared to whey-based amino fornia Los Angeles, Los Angeles, CA: Rick morbidity: A randomized controlled trial.
acid supplementation. Further study of Harrison, MD; University of Virginia Chil- Am J Clin Nutr 2005; 81:1397–1404 the ability of ZSGM supplementation to 15. Boelens PG, Houdijk AP, Fonk JC, et al: Glu- dren's Hospital, Charlottesville VA: Doug- prevent nosocomial infection and sepsis tamine-enriched enteral nutrition increases las F. Willson, MD; National Institute of in the immunocompromised PICU popu- in vitro interferon-gamma production but Child Health and Human Development, lation is warranted.
does not influence the in vivo specific anti- Bethesda, MD: Carol Nicholson, MD, body response to KLH after severe trauma. A Tammara Jenkins, MSN RN.
prospective, double blind, randomized clini- cal study. Clin Nutr 2004; 23:391– 400 16. Yalcin SS, Yurdakok K, Tezcan I, et al. Effect Members of the Collaborative Pediat- of glutamine supplementation on diarrhea, ric Critical Care Research Network par- 1. Felmet KA, Hall MW, Clark RS, et al: Pro- interleukin-8 and secretory immunoglobulin ticipating in this study: Children's Hospi- longed lymphopenia, lymphoid depletion, A in children with acute diarrhea. J Pediatr tal of Pittsburgh, Pittsburgh, PA: Joseph and hypoprolactinemia in children with nos- Gastroenterol Nutr 2004; 38:494 –501 Carcillo, MD, Michael Bell, MD, Alan ocomial sepsis and multiple organ failure.
17. Ha E, Zemel MB: Functional properties of Abraham, BA, Annette Seelhorst, RN, J Immunol 2005; 174:3765–3772 whey, whey components, and essential Jennifer Jones RN; University of Utah 2. Fraker PJ, King LE: Reprogramming of the amino acids: mechanisms underlying health Pediatr Crit Care Med 2012 Vol. 13, No. 4 benefits for active people (review). J Nutr effects of dietary whey proteins in mice.
26. Freeman SL, Fisher L, German JB, et al: Biochem 2003; 14:251–258 J Dairy Res 1995; 62:359 –368 Dairy proteins and the response to pneu- 18. Low PP, Rutherfurd KJ, Gill HS, et al: Effect 22. Micke P, Beeh KM, Buhl R: Effects of long- movax in senior citizens: A randomized, dou- of dietary whey protein concentrate on pri- term supplementation with whey proteins on ble-blind, placebo-controlled pilot study. Ann mary and secondary antibody responses in plasma glutathione levels of HIV-infected pa- N Y Acad Sci 2010; 1190:97–103 immunized BALB/c mice. Int Immunophar- tients. Eur J Nutr 2002; 41:12–18 27. Leclerc F, Leteutre S, Duhamel A, et al: Cu- macol 2003; 3:393– 401 23. Micke P, Beeh KM, Schlaak JF, et al: Oral mulative influence of organ dysfunctions and 19. Perez-Cano FJ, Marin-Gallen S, Castell M, et supplementation with whey proteins in- septic state on mortality of critically ill chil- al: Supplementing suckling rats with whey creases plasma glutathione levels of HIV- dren. Am J Resp Crit Care Med 2005; 171: protein concentrate modulates the immune infected patients. Eur J Clin Invest 2001; response and ameliorates rat rotavirus- 28. Skelton JA, Havens PL, Werlin JL: Nutrient induced diarrhea. J Nutr 2008; 138: 24. Moreno YF, Sgarbieri VC, da Silva MN, et al: deficiencies in tube fed children. Clin Pediatr Features of whey protein concentrate supple- 2006; 45:37– 41 20. Perez-Cano FJ, Marin-Gallen S, Castell M, et mentation in children with rapidly progressive 29. Jeejeebhoy K: Zinc essential trace element al: Bovine whey protein concentrate supple- HIV infection. J Trop Pediatr 2006; 52:34 –38 for parenteral nutrition. Gastroenterology mentation modulates maturation of immune 25. Alexander JW, MacMillan BG, Stinnett JD, et 2009; 137(5 Suppl):S7–S12 system in suckling rats. Br J Nutr 2007; al: Beneficial effects of aggressive protein 30. Shenkin A: Selenium in intravenous nutri- 98(Suppl 1):S80 –S84 feeding in severely burned children. Ann tion. Gastroenterology 2009; 137(Suppl 5): 21. Wong CW, Watson DL: Immunomodulatory Surg 1980; 192:505–517 Pediatr Crit Care Med 2012 Vol. 13, No. 4

Source: http://www.cpccrn.org/documents/PediatrCritCare2012Carcillo.pdf