User login
Micronutrients’ role against sepsis
The potential role of micronutrients in sepsis has created much recent interest and may offer novel therapies. Oxidant stress and inflammation are important components of sepsis pathobiology. Several micronutrients, notably zinc and selenium, are known to have antioxidant and anti-inflammatory properties and are, therefore, plausible candidates for adjunctive therapies in sepsis.
Selenium
Selenium plays a major role in the intra-cellular antioxidant system as a structural component of the selenoprotein enzymes, including glutathione peroxidase. These enzymes catalyze the reduction of hydroperoxidases to less toxic products, thereby protecting cells against oxidative stress. Selenium can also modulate the inflammatory cascade by inhibiting pro-inflammatory gene expression.
Selenium plasma concentrations are significantly lower in critically ill patients compared with age-matched healthy control subjects. Current evidence suggests that selenium concentrations are lowest in patients with sepsis and that low plasma selenium concentrations correlate with poor outcomes and organ failure (Alhazzani et al. Crit Care Med. 2013;41[6]:1555). Whether these associations represent pathological responses or epiphenomena is not entirely known. Nonetheless, because selenium is critical for the selenoprotein enzymes that function in antioxidant defense, it is biologically plausible that selenium supplementation may improve the outcomes of patients with sepsis.
A recent clinical trial (Heyland et al. N Engl J Med. 2013;368[16]:1489) showed that early administration of glutamine and/or a combination of antioxidants (selenium, zinc, beta-carotene, vitamin C, and vitamin E) did not improve mortality in critically ill adults with multiorgan failure. Glutamine administration was actually associated with increased mortality in this study.
This trial offers strong evidence that selenium administration, in combination with other antioxidants, does not improve clinical outcomes in critically ill patients, although it is difficult to assign effect to any one micronutrient, as selenium was administered in combination with other agents. Furthermore, this study enrolled patients with heterogeneous forms of critical illness, thereby making it difficult to isolate any beneficial effect in more specific conditions, such as sepsis.
A recent meta-analysis by (Alhazzani and colleagues et al. (Crit Care Med. 2013;41[6]:1555) examined the effect of isolated selenium supplementation on mortality in critically ill patients with sepsis. They reported a trend toward reduced mortality (odds ratio for mortality = 0.73, P =.03) in patients receiving selenium at higher than the daily recommended dose. Selenium supplementation had no effect on secondary outcomes, including a reduction of nosocomial pneumonia or length of stay. The quality of this evidence was graded as "low" based on the high risk of bias and imprecision, but the authors encouraged further well-designed trials addressing the efficacy of selenium supplementation in sepsis.
A trial of IV selenium supplementation in sepsis was recently completed by the German Sepsis Network with an estimated total enrollment of 1,180 subjects (clinicaltrials.gov: NCT00832039). The primary outcome of this study was all-cause mortality at 28 days. This study, when published, may provide more conclusive evidence to support or oppose selenium supplementation in sepsis.
Zinc
Zinc supplementation may also have a therapeutic role in sepsis. Zinc is involved in both innate and adaptive immune function. Notably, zinc-deficient states produce lymphopenia, impaired natural killer and phagocytic cell function, and impaired cytokine production. Zinc directly regulates signal transduction mechanisms involved in immunity and inflammation. Zinc also serves important roles in oxidative stress responses, neurocognitive function, growth, and development.
Metallothioneins are metal-binding proteins involved in zinc homeostasis. They serve roles in the scavenging of free radicals, detoxification of heavy metals, and participate in the inflammatory response to stress. It has been estimated that approximately 10% of the human proteome potentially interacts with zinc (Andreini et al. J Proteome Res. 2006;5[1]:196).
Transcriptomic studies demonstrated that within 1 day of admission for septic shock, children’s genomes demonstrate decreased expression of a large number of genes involved in zinc homeostasis (Wong et al. Physiol Genomics. 2007;30[2]:146). When comparing pediatric survivors with nonsurvivors of septic shock, nonsurvivors had significantly lower serum zinc concentrations. Furthermore, metallothionein expression was increased in the nonsurvivors, suggesting a functional consequence of altered zinc homeostasis.
The role of metallothionein expression and zinc homeostasis was further investigated in a study involving a heterogeneous cohort of critically ill children (Cvijanovich et al. Pediatr Crit Care Med. 2009:10[1]:29). All patients in this study had low serum zinc concentrations on days 1 and 3 of illness. Furthermore, on day 1 of illness, there was a positive correlation between zinc levels and metallothionein protein expression.
Several murine models have demonstrated the efficacy of zinc supplementation in sepsis. Zinc deficiency increased the bacterial burden and enhanced lung NF-kappa-B activity in a mouse model of sepsis, and short-term zinc supplementation reversed these effects (Bao et al. Am J Physiol Lung Cell Mol Physiol. 2010;298[6]:L744). Mouse mortality from polymicrobial sepsis was significantly increased with zinc deficiency and correlates with higher plasma cytokines, increased oxidative tissue damage, and cell death in the lungs and spleen of zinc-deficient animals (Knoell et al. Crit Care Med. 2009;37[4]:1380). These biological effects were diminished, and mortality improved in deficient mice that received zinc supplementation following septic injury. Another study demonstrated decreased mortality and bacterial load in septic mice receiving zinc supplementation for 3 days prior to septic injury compared with unsupplemented control mice (Nowak et al. Pediatr Crit Care Med. 2012;13[5]:e323).
In contrast to the murine studies, the trial by Heyland and colleagues did not demonstrate improved mortality in critically ill adults with multiorgan failure who received zinc supplementation in combination with other micronutrients. In another recent clinical trial, critically ill children were randomized to receive either whey protein supplementation or a cocktail of zinc, selenium, glutamine, and metoclopramide (Carcillo et al. Pediatr Crit Care Med. 2012;13[2]:165). There was no difference in the primary endpoint—time to nosocomial infection—between the two groups. However, in a secondary analysis, there was a reduced nosocomial infection rate in the immunocompromised subgroup members who received the antioxidant cocktail, suggesting that zinc supplementation may be useful in patients who are particularly infection-vulnerable.
Although zinc supplementation in combination with other antioxidants has been trialed in critically ill patients, the effect of zinc alone has not been fully explored. Investigators at Children’s Hospital and Research Center Oakland (clinicaltrials.gov: NCT01062009) recently completed a phase 1 safety and dose-finding trial of IV zinc supplementation in critically ill pediatric patients. Investigators at the University of Vermont (clinicaltrials.gov: NCT01162109) are exploring the safety and pharmacokinetics of IV zinc supplementation in critically ill adults with severe sepsis. The results of these trials will help to inform the future design of studies to test the efficacy of zinc supplementation in pediatric and adult sepsis.
Summary
Although selenium and zinc levels are decreased in pediatric and adult patients with sepsis, it is not clear whether correcting these alterations is of clinical benefit. Recent human trials investigating selenium supplementation are difficult to interpret in the context of sepsis as many have been conducted in patient populations that include heterogeneous forms of critical illnesses and have often included selenium in combination with other micronutrients.
The recent meta-analysis by Alhazzani and colleagues, however, provides evidence that isolated selenium supplementation may be effective in reducing mortality in patients with sepsis. The results of the recently completed clinical trial by the German Sepsis Network may provide conclusive evidence to confirm this trend. While animal studies supporting zinc supplementation in sepsis abound, human trials have investigated zinc in combination with other antioxidants, and phase 1 trials of zinc alone are only currently underway. Further large, randomized trials will be needed before zinc supplementation can be considered efficacious in reducing mortality in sepsis.
Dr. Sandquist is a pediatric critical care medicine fellow, Department of Pediatrics, University of Cincinnati College of Medicine; Dr. Wong is a Professor of Pediatrics and the Director of the Division of Critical Care Medicine, Cincinnati Children’s Hospital Medical Center and Cincinnati Children’s Hospital Research Foundation, Cincinnati, Ohio.
 |
|
Dr. Peter Spiro, FCCP, comments: Drs. Wong and Sandquist and Wong present a significant and up-to-date review of potentially promising and, what would seem, common sense therapies for sepsis. Despite this, the literature has not yet shown significant benefit to antioxidants and elemental replacement; in fact, glutamine has shown questionable harm. There is a number of large trials underway and, hopefully, these results might help clear the haze. Selenium and zinc supplementation may prove to be significant primary and adjunctive therapies but to which patients remains truly unclear. Sepsis being such a heterogeneous entity, that it is still not surprising that the "one size fits all approach" does not give us clear answers, but we continue to try to find these potential solutions.
|
|
Dr. Peter Spiro, FCCP, comments: Drs. Wong and Sandquist and Wong present a significant and up-to-date review of potentially promising and, what would seem, common sense therapies for sepsis. Despite this, the literature has not yet shown significant benefit to antioxidants and elemental replacement; in fact, glutamine has shown questionable harm. There is a number of large trials underway and, hopefully, these results might help clear the haze. Selenium and zinc supplementation may prove to be significant primary and adjunctive therapies but to which patients remains truly unclear. Sepsis being such a heterogeneous entity, that it is still not surprising that the "one size fits all approach" does not give us clear answers, but we continue to try to find these potential solutions.
|
|
Dr. Peter Spiro, FCCP, comments: Drs. Wong and Sandquist and Wong present a significant and up-to-date review of potentially promising and, what would seem, common sense therapies for sepsis. Despite this, the literature has not yet shown significant benefit to antioxidants and elemental replacement; in fact, glutamine has shown questionable harm. There is a number of large trials underway and, hopefully, these results might help clear the haze. Selenium and zinc supplementation may prove to be significant primary and adjunctive therapies but to which patients remains truly unclear. Sepsis being such a heterogeneous entity, that it is still not surprising that the "one size fits all approach" does not give us clear answers, but we continue to try to find these potential solutions.
The potential role of micronutrients in sepsis has created much recent interest and may offer novel therapies. Oxidant stress and inflammation are important components of sepsis pathobiology. Several micronutrients, notably zinc and selenium, are known to have antioxidant and anti-inflammatory properties and are, therefore, plausible candidates for adjunctive therapies in sepsis.
Selenium
Selenium plays a major role in the intra-cellular antioxidant system as a structural component of the selenoprotein enzymes, including glutathione peroxidase. These enzymes catalyze the reduction of hydroperoxidases to less toxic products, thereby protecting cells against oxidative stress. Selenium can also modulate the inflammatory cascade by inhibiting pro-inflammatory gene expression.
Selenium plasma concentrations are significantly lower in critically ill patients compared with age-matched healthy control subjects. Current evidence suggests that selenium concentrations are lowest in patients with sepsis and that low plasma selenium concentrations correlate with poor outcomes and organ failure (Alhazzani et al. Crit Care Med. 2013;41[6]:1555). Whether these associations represent pathological responses or epiphenomena is not entirely known. Nonetheless, because selenium is critical for the selenoprotein enzymes that function in antioxidant defense, it is biologically plausible that selenium supplementation may improve the outcomes of patients with sepsis.
A recent clinical trial (Heyland et al. N Engl J Med. 2013;368[16]:1489) showed that early administration of glutamine and/or a combination of antioxidants (selenium, zinc, beta-carotene, vitamin C, and vitamin E) did not improve mortality in critically ill adults with multiorgan failure. Glutamine administration was actually associated with increased mortality in this study.
This trial offers strong evidence that selenium administration, in combination with other antioxidants, does not improve clinical outcomes in critically ill patients, although it is difficult to assign effect to any one micronutrient, as selenium was administered in combination with other agents. Furthermore, this study enrolled patients with heterogeneous forms of critical illness, thereby making it difficult to isolate any beneficial effect in more specific conditions, such as sepsis.
A recent meta-analysis by (Alhazzani and colleagues et al. (Crit Care Med. 2013;41[6]:1555) examined the effect of isolated selenium supplementation on mortality in critically ill patients with sepsis. They reported a trend toward reduced mortality (odds ratio for mortality = 0.73, P =.03) in patients receiving selenium at higher than the daily recommended dose. Selenium supplementation had no effect on secondary outcomes, including a reduction of nosocomial pneumonia or length of stay. The quality of this evidence was graded as "low" based on the high risk of bias and imprecision, but the authors encouraged further well-designed trials addressing the efficacy of selenium supplementation in sepsis.
A trial of IV selenium supplementation in sepsis was recently completed by the German Sepsis Network with an estimated total enrollment of 1,180 subjects (clinicaltrials.gov: NCT00832039). The primary outcome of this study was all-cause mortality at 28 days. This study, when published, may provide more conclusive evidence to support or oppose selenium supplementation in sepsis.
Zinc
Zinc supplementation may also have a therapeutic role in sepsis. Zinc is involved in both innate and adaptive immune function. Notably, zinc-deficient states produce lymphopenia, impaired natural killer and phagocytic cell function, and impaired cytokine production. Zinc directly regulates signal transduction mechanisms involved in immunity and inflammation. Zinc also serves important roles in oxidative stress responses, neurocognitive function, growth, and development.
Metallothioneins are metal-binding proteins involved in zinc homeostasis. They serve roles in the scavenging of free radicals, detoxification of heavy metals, and participate in the inflammatory response to stress. It has been estimated that approximately 10% of the human proteome potentially interacts with zinc (Andreini et al. J Proteome Res. 2006;5[1]:196).
Transcriptomic studies demonstrated that within 1 day of admission for septic shock, children’s genomes demonstrate decreased expression of a large number of genes involved in zinc homeostasis (Wong et al. Physiol Genomics. 2007;30[2]:146). When comparing pediatric survivors with nonsurvivors of septic shock, nonsurvivors had significantly lower serum zinc concentrations. Furthermore, metallothionein expression was increased in the nonsurvivors, suggesting a functional consequence of altered zinc homeostasis.
The role of metallothionein expression and zinc homeostasis was further investigated in a study involving a heterogeneous cohort of critically ill children (Cvijanovich et al. Pediatr Crit Care Med. 2009:10[1]:29). All patients in this study had low serum zinc concentrations on days 1 and 3 of illness. Furthermore, on day 1 of illness, there was a positive correlation between zinc levels and metallothionein protein expression.
Several murine models have demonstrated the efficacy of zinc supplementation in sepsis. Zinc deficiency increased the bacterial burden and enhanced lung NF-kappa-B activity in a mouse model of sepsis, and short-term zinc supplementation reversed these effects (Bao et al. Am J Physiol Lung Cell Mol Physiol. 2010;298[6]:L744). Mouse mortality from polymicrobial sepsis was significantly increased with zinc deficiency and correlates with higher plasma cytokines, increased oxidative tissue damage, and cell death in the lungs and spleen of zinc-deficient animals (Knoell et al. Crit Care Med. 2009;37[4]:1380). These biological effects were diminished, and mortality improved in deficient mice that received zinc supplementation following septic injury. Another study demonstrated decreased mortality and bacterial load in septic mice receiving zinc supplementation for 3 days prior to septic injury compared with unsupplemented control mice (Nowak et al. Pediatr Crit Care Med. 2012;13[5]:e323).
In contrast to the murine studies, the trial by Heyland and colleagues did not demonstrate improved mortality in critically ill adults with multiorgan failure who received zinc supplementation in combination with other micronutrients. In another recent clinical trial, critically ill children were randomized to receive either whey protein supplementation or a cocktail of zinc, selenium, glutamine, and metoclopramide (Carcillo et al. Pediatr Crit Care Med. 2012;13[2]:165). There was no difference in the primary endpoint—time to nosocomial infection—between the two groups. However, in a secondary analysis, there was a reduced nosocomial infection rate in the immunocompromised subgroup members who received the antioxidant cocktail, suggesting that zinc supplementation may be useful in patients who are particularly infection-vulnerable.
Although zinc supplementation in combination with other antioxidants has been trialed in critically ill patients, the effect of zinc alone has not been fully explored. Investigators at Children’s Hospital and Research Center Oakland (clinicaltrials.gov: NCT01062009) recently completed a phase 1 safety and dose-finding trial of IV zinc supplementation in critically ill pediatric patients. Investigators at the University of Vermont (clinicaltrials.gov: NCT01162109) are exploring the safety and pharmacokinetics of IV zinc supplementation in critically ill adults with severe sepsis. The results of these trials will help to inform the future design of studies to test the efficacy of zinc supplementation in pediatric and adult sepsis.
Summary
Although selenium and zinc levels are decreased in pediatric and adult patients with sepsis, it is not clear whether correcting these alterations is of clinical benefit. Recent human trials investigating selenium supplementation are difficult to interpret in the context of sepsis as many have been conducted in patient populations that include heterogeneous forms of critical illnesses and have often included selenium in combination with other micronutrients.
The recent meta-analysis by Alhazzani and colleagues, however, provides evidence that isolated selenium supplementation may be effective in reducing mortality in patients with sepsis. The results of the recently completed clinical trial by the German Sepsis Network may provide conclusive evidence to confirm this trend. While animal studies supporting zinc supplementation in sepsis abound, human trials have investigated zinc in combination with other antioxidants, and phase 1 trials of zinc alone are only currently underway. Further large, randomized trials will be needed before zinc supplementation can be considered efficacious in reducing mortality in sepsis.
Dr. Sandquist is a pediatric critical care medicine fellow, Department of Pediatrics, University of Cincinnati College of Medicine; Dr. Wong is a Professor of Pediatrics and the Director of the Division of Critical Care Medicine, Cincinnati Children’s Hospital Medical Center and Cincinnati Children’s Hospital Research Foundation, Cincinnati, Ohio.
The potential role of micronutrients in sepsis has created much recent interest and may offer novel therapies. Oxidant stress and inflammation are important components of sepsis pathobiology. Several micronutrients, notably zinc and selenium, are known to have antioxidant and anti-inflammatory properties and are, therefore, plausible candidates for adjunctive therapies in sepsis.
Selenium
Selenium plays a major role in the intra-cellular antioxidant system as a structural component of the selenoprotein enzymes, including glutathione peroxidase. These enzymes catalyze the reduction of hydroperoxidases to less toxic products, thereby protecting cells against oxidative stress. Selenium can also modulate the inflammatory cascade by inhibiting pro-inflammatory gene expression.
Selenium plasma concentrations are significantly lower in critically ill patients compared with age-matched healthy control subjects. Current evidence suggests that selenium concentrations are lowest in patients with sepsis and that low plasma selenium concentrations correlate with poor outcomes and organ failure (Alhazzani et al. Crit Care Med. 2013;41[6]:1555). Whether these associations represent pathological responses or epiphenomena is not entirely known. Nonetheless, because selenium is critical for the selenoprotein enzymes that function in antioxidant defense, it is biologically plausible that selenium supplementation may improve the outcomes of patients with sepsis.
A recent clinical trial (Heyland et al. N Engl J Med. 2013;368[16]:1489) showed that early administration of glutamine and/or a combination of antioxidants (selenium, zinc, beta-carotene, vitamin C, and vitamin E) did not improve mortality in critically ill adults with multiorgan failure. Glutamine administration was actually associated with increased mortality in this study.
This trial offers strong evidence that selenium administration, in combination with other antioxidants, does not improve clinical outcomes in critically ill patients, although it is difficult to assign effect to any one micronutrient, as selenium was administered in combination with other agents. Furthermore, this study enrolled patients with heterogeneous forms of critical illness, thereby making it difficult to isolate any beneficial effect in more specific conditions, such as sepsis.
A recent meta-analysis by (Alhazzani and colleagues et al. (Crit Care Med. 2013;41[6]:1555) examined the effect of isolated selenium supplementation on mortality in critically ill patients with sepsis. They reported a trend toward reduced mortality (odds ratio for mortality = 0.73, P =.03) in patients receiving selenium at higher than the daily recommended dose. Selenium supplementation had no effect on secondary outcomes, including a reduction of nosocomial pneumonia or length of stay. The quality of this evidence was graded as "low" based on the high risk of bias and imprecision, but the authors encouraged further well-designed trials addressing the efficacy of selenium supplementation in sepsis.
A trial of IV selenium supplementation in sepsis was recently completed by the German Sepsis Network with an estimated total enrollment of 1,180 subjects (clinicaltrials.gov: NCT00832039). The primary outcome of this study was all-cause mortality at 28 days. This study, when published, may provide more conclusive evidence to support or oppose selenium supplementation in sepsis.
Zinc
Zinc supplementation may also have a therapeutic role in sepsis. Zinc is involved in both innate and adaptive immune function. Notably, zinc-deficient states produce lymphopenia, impaired natural killer and phagocytic cell function, and impaired cytokine production. Zinc directly regulates signal transduction mechanisms involved in immunity and inflammation. Zinc also serves important roles in oxidative stress responses, neurocognitive function, growth, and development.
Metallothioneins are metal-binding proteins involved in zinc homeostasis. They serve roles in the scavenging of free radicals, detoxification of heavy metals, and participate in the inflammatory response to stress. It has been estimated that approximately 10% of the human proteome potentially interacts with zinc (Andreini et al. J Proteome Res. 2006;5[1]:196).
Transcriptomic studies demonstrated that within 1 day of admission for septic shock, children’s genomes demonstrate decreased expression of a large number of genes involved in zinc homeostasis (Wong et al. Physiol Genomics. 2007;30[2]:146). When comparing pediatric survivors with nonsurvivors of septic shock, nonsurvivors had significantly lower serum zinc concentrations. Furthermore, metallothionein expression was increased in the nonsurvivors, suggesting a functional consequence of altered zinc homeostasis.
The role of metallothionein expression and zinc homeostasis was further investigated in a study involving a heterogeneous cohort of critically ill children (Cvijanovich et al. Pediatr Crit Care Med. 2009:10[1]:29). All patients in this study had low serum zinc concentrations on days 1 and 3 of illness. Furthermore, on day 1 of illness, there was a positive correlation between zinc levels and metallothionein protein expression.
Several murine models have demonstrated the efficacy of zinc supplementation in sepsis. Zinc deficiency increased the bacterial burden and enhanced lung NF-kappa-B activity in a mouse model of sepsis, and short-term zinc supplementation reversed these effects (Bao et al. Am J Physiol Lung Cell Mol Physiol. 2010;298[6]:L744). Mouse mortality from polymicrobial sepsis was significantly increased with zinc deficiency and correlates with higher plasma cytokines, increased oxidative tissue damage, and cell death in the lungs and spleen of zinc-deficient animals (Knoell et al. Crit Care Med. 2009;37[4]:1380). These biological effects were diminished, and mortality improved in deficient mice that received zinc supplementation following septic injury. Another study demonstrated decreased mortality and bacterial load in septic mice receiving zinc supplementation for 3 days prior to septic injury compared with unsupplemented control mice (Nowak et al. Pediatr Crit Care Med. 2012;13[5]:e323).
In contrast to the murine studies, the trial by Heyland and colleagues did not demonstrate improved mortality in critically ill adults with multiorgan failure who received zinc supplementation in combination with other micronutrients. In another recent clinical trial, critically ill children were randomized to receive either whey protein supplementation or a cocktail of zinc, selenium, glutamine, and metoclopramide (Carcillo et al. Pediatr Crit Care Med. 2012;13[2]:165). There was no difference in the primary endpoint—time to nosocomial infection—between the two groups. However, in a secondary analysis, there was a reduced nosocomial infection rate in the immunocompromised subgroup members who received the antioxidant cocktail, suggesting that zinc supplementation may be useful in patients who are particularly infection-vulnerable.
Although zinc supplementation in combination with other antioxidants has been trialed in critically ill patients, the effect of zinc alone has not been fully explored. Investigators at Children’s Hospital and Research Center Oakland (clinicaltrials.gov: NCT01062009) recently completed a phase 1 safety and dose-finding trial of IV zinc supplementation in critically ill pediatric patients. Investigators at the University of Vermont (clinicaltrials.gov: NCT01162109) are exploring the safety and pharmacokinetics of IV zinc supplementation in critically ill adults with severe sepsis. The results of these trials will help to inform the future design of studies to test the efficacy of zinc supplementation in pediatric and adult sepsis.
Summary
Although selenium and zinc levels are decreased in pediatric and adult patients with sepsis, it is not clear whether correcting these alterations is of clinical benefit. Recent human trials investigating selenium supplementation are difficult to interpret in the context of sepsis as many have been conducted in patient populations that include heterogeneous forms of critical illnesses and have often included selenium in combination with other micronutrients.
The recent meta-analysis by Alhazzani and colleagues, however, provides evidence that isolated selenium supplementation may be effective in reducing mortality in patients with sepsis. The results of the recently completed clinical trial by the German Sepsis Network may provide conclusive evidence to confirm this trend. While animal studies supporting zinc supplementation in sepsis abound, human trials have investigated zinc in combination with other antioxidants, and phase 1 trials of zinc alone are only currently underway. Further large, randomized trials will be needed before zinc supplementation can be considered efficacious in reducing mortality in sepsis.
Dr. Sandquist is a pediatric critical care medicine fellow, Department of Pediatrics, University of Cincinnati College of Medicine; Dr. Wong is a Professor of Pediatrics and the Director of the Division of Critical Care Medicine, Cincinnati Children’s Hospital Medical Center and Cincinnati Children’s Hospital Research Foundation, Cincinnati, Ohio.
