- Email: [email protected]
acute respiratory distress syndrome
Journal of Critical Care (2007) 22, 314–318
Steroids in Critical Illness
Adrenal insufficiency in early phase of pediatric acute lung injury/acute respiratory distress syndrome Rujipat Samransamruajkit MD a,⁎, Siriwan Jitchaiwat MD b , Jitladda Deerojanawong MD a , Suchada Sritippayawan MD a , Nuanchan Praphal MD a a
Respiratory and Critical Care unit, Faculty of Medicine, Department of Pediatrics, Chulalongkorn University, Bangkok 10330, Thailand b Faculty of Medicine, Department of Pediatrics, Chulalongkorn University, Bangkok 10330, Thailand
Keywords: Adrenal insufficiency; Acute lung injury/ARDS; Prevalence
Abstract Introduction: Adequate adrenal function is essential to survive critical illness. Several recent articles have reported the significant effect of adrenal insufficiency (AI) in patients with sepsis. However, the prevalence of AI in pediatric acute lung injury (ALI)/acute respiratory distress syndrome (ARDS) is so far still scanty. Thus, we elected to study its prevalence and its clinical outcome. Methods: This is a cross-sectional observational study. We enrolled eligible infants and children aged between 1 month and 15 years who were admitted to our tertiary pediatric intensive care unit from February 1, 2005, to December 31, 2005, with ALI or ARDS diagnosed by the American-European Consensus criteria. A short corticotropin stimulation test (250 μg) was done within 24 hours of enrollment, and all clinical data were also recorded. Cortisol levels were measured at baseline, 30 minutes, and 60 minutes posttest. Adrenal insufficiency was defined as a baseline cortisol level of less than 15.1 μg/dL or an increment of cortisol level of less than 9 μg/dL after the adrenocorticotropic hormone stimulation test. Results: Of 507 patients admitted to the pediatric intensive care unit, there were 20 diagnosed with ALI/ ARDS. Of 20 children, 16 met the inclusion criteria and had none of the exclusion criteria. Of 16, there were 9 (56%) with ARDS, and 7 (44%) of 12 had ALI. The prevalence of AI was observed in 37.5% (6/ 16), diagnosed by baseline level criteria in 25% (4/16) and by incremental criteria in 12.5% (2/16). The Baseline level of the adrenocorticotropic hormone was 7.8 ± 5 (nmol/L). The median age in the AI group was 2 months. Of 6 children, 5 (83.3%) were in the ARDS group. Pediatric Risk of Mortality III score was significantly higher in the AI group compared with that in the non-AI (P b .05). Initial PaO2/fraction
Abbreviations: ALI/ARDS, acute lung injury/acute respiratory distress syndrome; AI, adrenal insufficiency; Non-AI, non–adrenal insufficiency; PRISM, Pediatric Risk of Mortality; ACTH, adrenocorticotropic hormone. Part of this work was presented at the American Thoracic Society meeting, May 2006, San Diego, Calif. * Corresponding author. Pediatric pulmonary & Critical Care Division, Department of Pediatrics, Faculty of Medicine, Chulalongkorn University, Bangkok, 10330 Thailand. Tel.: +66 2 256 4996x129, 123; fax: +66 2 256 4911. E-mail address: [email protected] (R. Samransamruajkit). 0883-9441/$ – see front matter © 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.jcrc.2007.03.003
Adrenal insufficiency in early phase of pediatric ALI/ARDS
315
of inspired oxygen ratio tended to be lower in the AI group (123.2 ± 62.2) compared with that in the nonAI group (183.8 ± 79.1), although not statistically significant (P = .1). The mortality was also not statistically different between the AI (1/6, 16.7%) and the non-AI groups (1/10, 10%). Conclusions: Our study demonstrated that the prevalence of AI was common in pediatric ALI/ARDS. These results would be an initial step to further study the impact of AI on clinical outcomes of these children in a larger scale. © 2007 Elsevier Inc. All rights reserved.
1. Introduction
2. Materials and methods
The hypothalamic-pituitary adrenal axis (HPA) is an important component of compensatory anti-inflammatory of critically illness. Patients with critical illnesses have elevated glucocorticoid secretions marked by an increase in the serum total cortisol concentration [1,2]. Cortisol supports vascular tone and endothelial integrity, modulates a large number of proinflammatory cytokines, and suppresses phospholipase A 2, cyclooxygenase, and nitric oxide synthase [3]. It has been reported to be important in maintaining endothelial integrity and vascular permeability [4]. With severe infection, trauma, burns, illness or surgery, there is an increase in cortisol production proportional to the severity of illness [5]. Acute lung injury (ALI)/acute respiratory distress syndrome (ARDS) are the major cause of acute respiratory failure in critically ill patients diagnosed in the pediatric intensive care unit (PICU). It is characterized by increased vascular permeability, extravasations of plasma, and recruitment of inflammatory cells [6]. The HPA response may influence both the initial inflammatory response after ALI/ARDS and subsequent repair process, including the rate and degree of fibroproliferation [7]. In general, adrenal responsiveness to exogenous corticotrophin is maintained during acute illness [2,5]. Although during severe illness, many factors can impair the normal corticosteroid response. Therefore, adrenal insufficiency (AI) can potentially develop during ALI/ARDS [8]. The extensive tissue inflammation observed in ALI/ARDS would appear to be a promising target for corticosteroid therapy. A recent multicenter trial using corticosteroids in the acute phase of ARDS, not for AI but to reduce inflammation, failed to find benefit with this therapy and found some suggestion of potential harm [9]. Based on this study and others, the role of corticosteroids in the management of ALI/ARDS remains unclear. Several recent articles have reported the significant effect of AI in critically ill patients and the role this may play in hemodynamic stability in these patients. Pizarro et al [10] have recently described absolute and relative AI in children with septic shock that is actually common. In addition, the benefits of using corticosteroid replacement in several sepsis trials has been reported [8,10-12]. However, the prevalence of AI in children with ALI/ARDS is so far still obscure. Thus, we elected to investigate the prevalence of AI in pediatric ALI/ARDS and its relation to clinical outcome.
A cross-sectional observational study was performed in a PICU of a tertiary care referral center. We enrolled consecutive eligible infants and children aged between 1 month to 15 years who were admitted to our tertiary PICU from February 1, 2005, to December 31, 2005, with diagnosis of ALI/ARDS. Adrenal function was assessed by the adrenocorticotropic hormone (ACTH) stimulation test within 24 hours in patients who met the inclusion criteria and had none of the exclusion criteria. A short corticotropin test (250 μg of cosyntropin, Organon; Bangkok, Thailand) was performed, and serum cortisol levels were measured at baseline, 30 minutes, and 60 minutes posttest. Clinical parameters were also recorded. Definition of ALI/ARDS by American-European Consensus Conference [13]: 1. Impaired oxygenation shown by a PaO2/fraction of inspired oxygen (FIO2) ratio of less than 300 for ALI and less than 200 for ARDS 2. Chest x-ray showing bilateral pulmonary infiltrates 3. Pulmonary artery occlusion pressure (Wedge pressure) of less than 18 mm Hg or absence of clinical evidence of left atrial hypertension Definition of AI [8,14]: 1. Basal cortisol level of less than 15.2 μg/dL or less than 415 nmol/L with clinical of shock in spite of adequately volume load and remain dopamine resistant or 2. Incremental cortisol level of less than 9 μg/dL or less than 250 nmol/L (increment at 30 minutes and 60 minutes = peak level − basal level) Exclusion criteria: 1. Children with underlying adrenal disease 2. Children who received corticosteroids within the past 3 months 3. Children who recently received medication that may affect adrenal function within the month of diagnosis of ALI/ARDS, for example, etomidate The protocol was approved by our institutional review board and informed consent was obtained from the patient's legal guardian.
316
R. Samransamruajkit et al.
3. Results Twenty children with diagnosis of ALI/ARDS were admitted to the PICU during the study period. Sixteen of 20 children met the inclusion criteria and were recruited to our study. Of 20 children, 4 were excluded because of history of recent steroid use. Demographic and clinical characteristics of patients with ALI/ARDS compared between AI and non-AI groups in the study are shown in Table 1. The prevalence of AI was observed in 37.5% (6/16), diagnosed by basal cortisol level criteria in 67% (4/6) and by incremental criteria in 33% (2/6) (Fig. 1). The number of patients with ALI/ARDS having AI was equal for both sexes (3 males, 3 females). The median age of the AI group was less than that of the non-AI group (3.5 and 24 months, respectively). The most common causes of ALI/ARDS in patients with AI were severe pneumonia and septic shock (Table 1). Eighty percent of patients with AI were in the ARDS group, and 66% (4/6) of them were diagnosed with septic shock and multiple organ dysfunction.
Fig. 1 Comparison of the basal cortisol levels and levels of cortisol at 30 minutes and at 60 minutes after ACTH stimulation in patients with AI (n = 6). * Indicate statistically significant (P b.05).
3.2. Adrenocorticotropic hormone and cortisol levels
3.1. Clinical and laboratory parameters The Pediatric Risk of Mortality (PRISM) score was significantly higher in the AI group (8.3 ± 2.5) compared with that in the non-AI group (5.6 ± 3.7, P b .05). The initial serum sodium level within 24 hours of enrollment of the 2 groups was almost equal, but the serum potassium level was significantly lower in the AI group compared with that in the non-AI group (3.6 ± 1 vs 4.6 ± 0.6, P = .03, Table 2). In addition, PaO2/FIO2 ratio tended to be lower in the AI group compared with that in non-AI group (123.2 ± 62.2 vs 183.8 ± 79.1, P = .1).
Of 6 patients with AI, 5 (83%) had low initial ACTH level, and 4 (67%) of 6 had low basal cortisol level. The remaining 2 of the patients with AI had low initial ACTH and normal basal cortisol levels but showed inadequate adrenal response after ACTH stimulation (Fig. 1). The median of cortisol levels at baseline, 30 minutes, and 60 minutes after ACTH stimulation test was lower in the AI group compared with that in the non-AI group, as shown in Fig. 2. Of our 16 children, 2 who had ALI/ARDS died. There was 1 from each group. The causes of death were severe malnutrition (nonAI) and multiple organ failure (AI).
Table 1 Comparison of clinical characteristics between AI and non-AI groups
4. Discussion
Clinical characteristic AI (n = 6)
Our study demonstrated that the prevalence of AI was at 37.5% in pediatric ALI/ARDS. This finding may be common in children with ALI/ARDS, although the prevalence of AI
Sex Male Female Median age (mo) Causes of ALI/ARDS Severe pneumonia Sepsis Chest trauma PRISM III score Oxygen parameters PaO2/FIO2 ARDS Septic shock MODS
Non-AI (n = 10) P
3 (50%) 3 (60%) 3.5
8 (80%) 2 (20%) 24
2 (33%) 3 (50%) 1 (17%) 8.3 ± 2.5
7 (70%) 3 (30%) None 5.6 ± 3.7
b.05
123.2 ± 62.2 4 (66.7) 4 (66.7) 3 (50%)
183.8 ± 79.1 5 (50%) 3 (30%) 3 (30%)
.1 ns ns ns
MODS indicates multiple organ dysfunction syndrome; ns, not significant. PaO2/FIO2 = mmHg.
b.05
Table 2 Comparison of initial laboratory results among patients in the AI group Patient
Na
K
CO2 content
Blood glucose
ACTH (25-50 nmol/L)
1 2 3 4 5 6⁎
135 139 141 140 138 133
3.9 2.0 3.4 3.5 4.0 5.2
18 14 22 31 34 27
109 221 83 126 105 116
7.2 16.2 5.5 7.4 2.9 7.7
ACTH = nmol/L; Na+; K+ = meg/L; Blood sugar = mg/dl; Cortisol level = microgram/dl; nmol/L. ⁎ Died.
Adrenal insufficiency in early phase of pediatric ALI/ARDS
Fig. 2 Comparison of median cortisol levels (μg/dL) at baseline and cortisol levels after ACTH stimulation test at 30 minutes and at 60 minutes between AI and non-AI patients.
may vary according to the previous published criteria. The most reliable method is based on the assessment of plasma cortisol levels before and after exogenous cosyntropin (ACTH) stimulation, which is currently a gold standard. There are no established and accepted criteria to define AI in critically ill children. The cutoff levels that we used in this study were widely accepted [10,12,15]. Patients with AI had significantly higher PRISM score at enrollment and tended to be younger than those in the non-AI group. The mortality was also higher in AI (16.7%; odds ratio, 1.8) compared with that in the non-AI (10%), again not statistically significant. This may be due to our study not having power to detect the difference in mortality. In this study, our patients with AI had no hyponatremia or hyperkalemia, which is classically found in an individual with AI. In fact, the AI group had significantly lower potassium level than the non-AI group (P = .03). This may be explained by several reasons such as concurrent initial fluid replacement therapy, secondary AI (their mineralocorticoid secretions may remain intact) or the preexisting underlying conditions of the patients such as diarrhea/vomiting, or diuretic therapy [8]. Our patients in the AI group had undergone many fluid resuscitation and several days of diarrhea/vomiting before admission. Of the 6 patients with AI, 4 had abnormally low initial ACTH and low basal cortisol levels, although they had a good response to ACTH stimulation. This indicated that they actually had an adequate adrenal response, or they had secondary AI. Two of the patients with AI had low initial ACTH and high normal basal cortisol levels but had inadequate adrenal response to ACTH stimulation. This suggested that it was actually relative AI [10]. From our study, most of the children with AI had normal response to exogenous ACTH. Transient AI (functional AI) has been reported in critically ill patients, suggesting that reassessing adrenal function at 24 to 48 hours later should be done to confirm the diagnosis [5]. But it is important to be
317 aware that even in patients who have normal test results, AI can develop later in the illness. Therefore, it is unclear how often one should perform such test [8]. Adequate adrenal response in acute stress may be one factor contributing to the survival of the patient demonstrated by adequate production of cortisol; however, very high basal cortisol level may be associated with poor prognosis and high mortality [14,16,17]. One of our patients in the AI group who finally died showed very high baseline cortisol level and had inadequate response after ACTH stimulation. This may also suggest tissue resistance to glucocorticoid [18] or adrenal dysfunction. Almost all of the patients had good response of adrenals to ACTH stimulation during acute stress, suggesting that they had adequate adrenal reserve during the acute phase of ALI/ARDS. Thus, the treatment with glucocorticoid in the early phase of ALI/ARDS (in relation to adrenal dysfunction) may not be beneficial in most. However, our findings should be interpreted with awareness that the sample size was limited to those who presented with ALI/ARDS during the 1-year study period. Thus, the sample size may have led us to underestimate the clinical impact of AI in pediatric ALI/ ARDS. In previous clinical trials, high doses of methylprednisolone for 1 to 2 days early in the course of severe sepsis or established ARDS do not prevent ARDS development or reduce mortality rate, and they may be even harmful [19,20]. In contrast, the lower doses of methylprednisolone in late, unresolving ARDS may be beneficial [21], but again this is not confirmed by recent published data [9]. In conclusion, these data indicate that AI may be common among children with ALI/ARDS. The HPA axis appears to be an important component of the stress response and modulates the systemic cardiovascular and cellular response to injury. No study in pediatrics has yet examined HPA axis in patients with ALI/ARDS. These results would be an initial step to further study the impact of AI on clinical outcome of these children in a larger scale.
Acknowledgment This work was supported by grants (TRG 4780011) from Thailand Research Fund (Bangkok, Thailand) and Ratchadapiseksomphod Fund (Bangkok, Thailand). We thank the staff and fellows of the Pediatric Pulmonology and Critical Care Division of Chulalongkorn University (Bangkok, Thailand) and the PICU team and Prof Suthipong Wacharasindhu, MD (Chief, Pediatric Endocrinology) of King Chulalongkorn Memorial Hospital (Bangkok, Thailand). We are indebted to the Central Laboratory of King Chulalongkorn Memorial Hospital who assisted in cortisol level measurement and to Mettanando Bhikkhu for editing the manuscript.
References [1] Quiney N, Durkin M. Adrenocortical failure in intensive care. BMJ 1995;310:1253-4.
318 [2] Lamberts S, Bruining H, De Jong F. Corticosteroids therapy in severe illness. N Engl J Med 1997;337:1285-92. [3] Thompson BT. Glucocorticoids and acute lung injury. Crit Care Med 2003;31(4 Suppl):S253-7. [4] Chirousos G. Seminars in medicine of the Beth Israel Hospital, The hypothalamic-pituitary-adrenal axis and immune mediated inflammation. N Engl J Med 1995;332:1351-62. [5] Cooper MS, Stewart PM. Corticosteroid insufficiency in acutely ill patients. N Engl J Med 2003;348(8):727-34. [6] Suter PM. Lung inflammation in ARDS—friend or foe? N Engl J Med 2006;354:1739-42. [7] Meduri G, Kantangat S. Glucocorticoid treatment of sepsis and acute respiratory distress syndrome: time for a critical reappraisal. Crit Care Med 1998;26:630-3. [8] Gonzalez H, Nardi O, Annane D. Relative adrenal failure in the ICU: an identifiable problem requiring treatment. Crit Care Clin 2006;22 (1):105-18, vii. [9] Steinberg KP, Hudson LD, Goodman RB, Hough CL, Lanker PN, et al.. The National Heart, Lung, and Blood Institute Acute Respiratory Distress Syndrome (ARDS) Clinical Trials Network. Efficacy and safety of corticosteroids for persistent acute respiratory distress syndrome. N Engl J Med 2006;354(16):1671-84. [10] Pizarro CF, Troster EJ, Damiani D, et al. Absolute and relative adrenal insufficiency in children with septic shock. Crit Care Med 2005;33 (4):855-9.
R. Samransamruajkit et al. [11] Agus M. One step forward: an advance in understanding of adrenal insufficiency in the pediatric critically ill. Crit Care Med 2005;33(4):911-2. [12] Annane D. Glucocorticoids in the treatment of severe sepsis and septic shock. Curr Opin Crit Care 2005;11(5):449-53. [13] Ware L, Matthay MA. The acute respiratory distress syndrome. N Engl J Med 2000;342:1334-49. [14] Annane D, Sebille V, Troche G, et al. A 3 level prognostic classification in septic shock based on cortisol levels and cortisol response to corticotropin. JAMA 2000;283:1038-45. [15] Hatherill M, Tibby S, Hillard T, et al. Adrenal insufficiency in septic shock. Arch Dis Child 1999;80:51-5. [16] Annane D, Sebille V, Charpentier C, et al. Effect of treatment with low doses of hydrocortisone and fludrocortisone on mortality in patients with septic shock. JAMA 2002;288:862-71. [17] Yildiz O, Doganay M, Aygen B, et al. Physiologic dose steroid therapy in sepsis. Crit Care 2002;6:251-8. [18] Prigent H, Maxime V, Annane D. Clinical review: corticotherapy in sepsis. Crit Care 2004;8:122-9. [19] Sprung C, Caralis P, Marcial E, et al. The effect of high dose corticosteroids in patients with septic shock. N Engl J Med 1984;311:1137-43. [20] Bernard GR, Luuce J, Sprung C. High dose corticosteroids in patients with the ARDS. N Engl J Med 1987;317:1565-70. [21] Meduri G, Chinn A, Leeper K, et al. Corticosteroids rescue treatment of progressive fibroproliferation in late ARDS: patterns of response and predictors of outcome. Chest 1994;105:1516-27.
Steroids in Critical Illness
Adrenal insufficiency in early phase of pediatric acute lung injury/acute respiratory distress syndrome Rujipat Samransamruajkit MD a,⁎, Siriwan Jitchaiwat MD b , Jitladda Deerojanawong MD a , Suchada Sritippayawan MD a , Nuanchan Praphal MD a a
Respiratory and Critical Care unit, Faculty of Medicine, Department of Pediatrics, Chulalongkorn University, Bangkok 10330, Thailand b Faculty of Medicine, Department of Pediatrics, Chulalongkorn University, Bangkok 10330, Thailand
Keywords: Adrenal insufficiency; Acute lung injury/ARDS; Prevalence
Abstract Introduction: Adequate adrenal function is essential to survive critical illness. Several recent articles have reported the significant effect of adrenal insufficiency (AI) in patients with sepsis. However, the prevalence of AI in pediatric acute lung injury (ALI)/acute respiratory distress syndrome (ARDS) is so far still scanty. Thus, we elected to study its prevalence and its clinical outcome. Methods: This is a cross-sectional observational study. We enrolled eligible infants and children aged between 1 month and 15 years who were admitted to our tertiary pediatric intensive care unit from February 1, 2005, to December 31, 2005, with ALI or ARDS diagnosed by the American-European Consensus criteria. A short corticotropin stimulation test (250 μg) was done within 24 hours of enrollment, and all clinical data were also recorded. Cortisol levels were measured at baseline, 30 minutes, and 60 minutes posttest. Adrenal insufficiency was defined as a baseline cortisol level of less than 15.1 μg/dL or an increment of cortisol level of less than 9 μg/dL after the adrenocorticotropic hormone stimulation test. Results: Of 507 patients admitted to the pediatric intensive care unit, there were 20 diagnosed with ALI/ ARDS. Of 20 children, 16 met the inclusion criteria and had none of the exclusion criteria. Of 16, there were 9 (56%) with ARDS, and 7 (44%) of 12 had ALI. The prevalence of AI was observed in 37.5% (6/ 16), diagnosed by baseline level criteria in 25% (4/16) and by incremental criteria in 12.5% (2/16). The Baseline level of the adrenocorticotropic hormone was 7.8 ± 5 (nmol/L). The median age in the AI group was 2 months. Of 6 children, 5 (83.3%) were in the ARDS group. Pediatric Risk of Mortality III score was significantly higher in the AI group compared with that in the non-AI (P b .05). Initial PaO2/fraction
Abbreviations: ALI/ARDS, acute lung injury/acute respiratory distress syndrome; AI, adrenal insufficiency; Non-AI, non–adrenal insufficiency; PRISM, Pediatric Risk of Mortality; ACTH, adrenocorticotropic hormone. Part of this work was presented at the American Thoracic Society meeting, May 2006, San Diego, Calif. * Corresponding author. Pediatric pulmonary & Critical Care Division, Department of Pediatrics, Faculty of Medicine, Chulalongkorn University, Bangkok, 10330 Thailand. Tel.: +66 2 256 4996x129, 123; fax: +66 2 256 4911. E-mail address: [email protected] (R. Samransamruajkit). 0883-9441/$ – see front matter © 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.jcrc.2007.03.003
Adrenal insufficiency in early phase of pediatric ALI/ARDS
315
of inspired oxygen ratio tended to be lower in the AI group (123.2 ± 62.2) compared with that in the nonAI group (183.8 ± 79.1), although not statistically significant (P = .1). The mortality was also not statistically different between the AI (1/6, 16.7%) and the non-AI groups (1/10, 10%). Conclusions: Our study demonstrated that the prevalence of AI was common in pediatric ALI/ARDS. These results would be an initial step to further study the impact of AI on clinical outcomes of these children in a larger scale. © 2007 Elsevier Inc. All rights reserved.
1. Introduction
2. Materials and methods
The hypothalamic-pituitary adrenal axis (HPA) is an important component of compensatory anti-inflammatory of critically illness. Patients with critical illnesses have elevated glucocorticoid secretions marked by an increase in the serum total cortisol concentration [1,2]. Cortisol supports vascular tone and endothelial integrity, modulates a large number of proinflammatory cytokines, and suppresses phospholipase A 2, cyclooxygenase, and nitric oxide synthase [3]. It has been reported to be important in maintaining endothelial integrity and vascular permeability [4]. With severe infection, trauma, burns, illness or surgery, there is an increase in cortisol production proportional to the severity of illness [5]. Acute lung injury (ALI)/acute respiratory distress syndrome (ARDS) are the major cause of acute respiratory failure in critically ill patients diagnosed in the pediatric intensive care unit (PICU). It is characterized by increased vascular permeability, extravasations of plasma, and recruitment of inflammatory cells [6]. The HPA response may influence both the initial inflammatory response after ALI/ARDS and subsequent repair process, including the rate and degree of fibroproliferation [7]. In general, adrenal responsiveness to exogenous corticotrophin is maintained during acute illness [2,5]. Although during severe illness, many factors can impair the normal corticosteroid response. Therefore, adrenal insufficiency (AI) can potentially develop during ALI/ARDS [8]. The extensive tissue inflammation observed in ALI/ARDS would appear to be a promising target for corticosteroid therapy. A recent multicenter trial using corticosteroids in the acute phase of ARDS, not for AI but to reduce inflammation, failed to find benefit with this therapy and found some suggestion of potential harm [9]. Based on this study and others, the role of corticosteroids in the management of ALI/ARDS remains unclear. Several recent articles have reported the significant effect of AI in critically ill patients and the role this may play in hemodynamic stability in these patients. Pizarro et al [10] have recently described absolute and relative AI in children with septic shock that is actually common. In addition, the benefits of using corticosteroid replacement in several sepsis trials has been reported [8,10-12]. However, the prevalence of AI in children with ALI/ARDS is so far still obscure. Thus, we elected to investigate the prevalence of AI in pediatric ALI/ARDS and its relation to clinical outcome.
A cross-sectional observational study was performed in a PICU of a tertiary care referral center. We enrolled consecutive eligible infants and children aged between 1 month to 15 years who were admitted to our tertiary PICU from February 1, 2005, to December 31, 2005, with diagnosis of ALI/ARDS. Adrenal function was assessed by the adrenocorticotropic hormone (ACTH) stimulation test within 24 hours in patients who met the inclusion criteria and had none of the exclusion criteria. A short corticotropin test (250 μg of cosyntropin, Organon; Bangkok, Thailand) was performed, and serum cortisol levels were measured at baseline, 30 minutes, and 60 minutes posttest. Clinical parameters were also recorded. Definition of ALI/ARDS by American-European Consensus Conference [13]: 1. Impaired oxygenation shown by a PaO2/fraction of inspired oxygen (FIO2) ratio of less than 300 for ALI and less than 200 for ARDS 2. Chest x-ray showing bilateral pulmonary infiltrates 3. Pulmonary artery occlusion pressure (Wedge pressure) of less than 18 mm Hg or absence of clinical evidence of left atrial hypertension Definition of AI [8,14]: 1. Basal cortisol level of less than 15.2 μg/dL or less than 415 nmol/L with clinical of shock in spite of adequately volume load and remain dopamine resistant or 2. Incremental cortisol level of less than 9 μg/dL or less than 250 nmol/L (increment at 30 minutes and 60 minutes = peak level − basal level) Exclusion criteria: 1. Children with underlying adrenal disease 2. Children who received corticosteroids within the past 3 months 3. Children who recently received medication that may affect adrenal function within the month of diagnosis of ALI/ARDS, for example, etomidate The protocol was approved by our institutional review board and informed consent was obtained from the patient's legal guardian.
316
R. Samransamruajkit et al.
3. Results Twenty children with diagnosis of ALI/ARDS were admitted to the PICU during the study period. Sixteen of 20 children met the inclusion criteria and were recruited to our study. Of 20 children, 4 were excluded because of history of recent steroid use. Demographic and clinical characteristics of patients with ALI/ARDS compared between AI and non-AI groups in the study are shown in Table 1. The prevalence of AI was observed in 37.5% (6/16), diagnosed by basal cortisol level criteria in 67% (4/6) and by incremental criteria in 33% (2/6) (Fig. 1). The number of patients with ALI/ARDS having AI was equal for both sexes (3 males, 3 females). The median age of the AI group was less than that of the non-AI group (3.5 and 24 months, respectively). The most common causes of ALI/ARDS in patients with AI were severe pneumonia and septic shock (Table 1). Eighty percent of patients with AI were in the ARDS group, and 66% (4/6) of them were diagnosed with septic shock and multiple organ dysfunction.
Fig. 1 Comparison of the basal cortisol levels and levels of cortisol at 30 minutes and at 60 minutes after ACTH stimulation in patients with AI (n = 6). * Indicate statistically significant (P b.05).
3.2. Adrenocorticotropic hormone and cortisol levels
3.1. Clinical and laboratory parameters The Pediatric Risk of Mortality (PRISM) score was significantly higher in the AI group (8.3 ± 2.5) compared with that in the non-AI group (5.6 ± 3.7, P b .05). The initial serum sodium level within 24 hours of enrollment of the 2 groups was almost equal, but the serum potassium level was significantly lower in the AI group compared with that in the non-AI group (3.6 ± 1 vs 4.6 ± 0.6, P = .03, Table 2). In addition, PaO2/FIO2 ratio tended to be lower in the AI group compared with that in non-AI group (123.2 ± 62.2 vs 183.8 ± 79.1, P = .1).
Of 6 patients with AI, 5 (83%) had low initial ACTH level, and 4 (67%) of 6 had low basal cortisol level. The remaining 2 of the patients with AI had low initial ACTH and normal basal cortisol levels but showed inadequate adrenal response after ACTH stimulation (Fig. 1). The median of cortisol levels at baseline, 30 minutes, and 60 minutes after ACTH stimulation test was lower in the AI group compared with that in the non-AI group, as shown in Fig. 2. Of our 16 children, 2 who had ALI/ARDS died. There was 1 from each group. The causes of death were severe malnutrition (nonAI) and multiple organ failure (AI).
Table 1 Comparison of clinical characteristics between AI and non-AI groups
4. Discussion
Clinical characteristic AI (n = 6)
Our study demonstrated that the prevalence of AI was at 37.5% in pediatric ALI/ARDS. This finding may be common in children with ALI/ARDS, although the prevalence of AI
Sex Male Female Median age (mo) Causes of ALI/ARDS Severe pneumonia Sepsis Chest trauma PRISM III score Oxygen parameters PaO2/FIO2 ARDS Septic shock MODS
Non-AI (n = 10) P
3 (50%) 3 (60%) 3.5
8 (80%) 2 (20%) 24
2 (33%) 3 (50%) 1 (17%) 8.3 ± 2.5
7 (70%) 3 (30%) None 5.6 ± 3.7
b.05
123.2 ± 62.2 4 (66.7) 4 (66.7) 3 (50%)
183.8 ± 79.1 5 (50%) 3 (30%) 3 (30%)
.1 ns ns ns
MODS indicates multiple organ dysfunction syndrome; ns, not significant. PaO2/FIO2 = mmHg.
b.05
Table 2 Comparison of initial laboratory results among patients in the AI group Patient
Na
K
CO2 content
Blood glucose
ACTH (25-50 nmol/L)
1 2 3 4 5 6⁎
135 139 141 140 138 133
3.9 2.0 3.4 3.5 4.0 5.2
18 14 22 31 34 27
109 221 83 126 105 116
7.2 16.2 5.5 7.4 2.9 7.7
ACTH = nmol/L; Na+; K+ = meg/L; Blood sugar = mg/dl; Cortisol level = microgram/dl; nmol/L. ⁎ Died.
Adrenal insufficiency in early phase of pediatric ALI/ARDS
Fig. 2 Comparison of median cortisol levels (μg/dL) at baseline and cortisol levels after ACTH stimulation test at 30 minutes and at 60 minutes between AI and non-AI patients.
may vary according to the previous published criteria. The most reliable method is based on the assessment of plasma cortisol levels before and after exogenous cosyntropin (ACTH) stimulation, which is currently a gold standard. There are no established and accepted criteria to define AI in critically ill children. The cutoff levels that we used in this study were widely accepted [10,12,15]. Patients with AI had significantly higher PRISM score at enrollment and tended to be younger than those in the non-AI group. The mortality was also higher in AI (16.7%; odds ratio, 1.8) compared with that in the non-AI (10%), again not statistically significant. This may be due to our study not having power to detect the difference in mortality. In this study, our patients with AI had no hyponatremia or hyperkalemia, which is classically found in an individual with AI. In fact, the AI group had significantly lower potassium level than the non-AI group (P = .03). This may be explained by several reasons such as concurrent initial fluid replacement therapy, secondary AI (their mineralocorticoid secretions may remain intact) or the preexisting underlying conditions of the patients such as diarrhea/vomiting, or diuretic therapy [8]. Our patients in the AI group had undergone many fluid resuscitation and several days of diarrhea/vomiting before admission. Of the 6 patients with AI, 4 had abnormally low initial ACTH and low basal cortisol levels, although they had a good response to ACTH stimulation. This indicated that they actually had an adequate adrenal response, or they had secondary AI. Two of the patients with AI had low initial ACTH and high normal basal cortisol levels but had inadequate adrenal response to ACTH stimulation. This suggested that it was actually relative AI [10]. From our study, most of the children with AI had normal response to exogenous ACTH. Transient AI (functional AI) has been reported in critically ill patients, suggesting that reassessing adrenal function at 24 to 48 hours later should be done to confirm the diagnosis [5]. But it is important to be
317 aware that even in patients who have normal test results, AI can develop later in the illness. Therefore, it is unclear how often one should perform such test [8]. Adequate adrenal response in acute stress may be one factor contributing to the survival of the patient demonstrated by adequate production of cortisol; however, very high basal cortisol level may be associated with poor prognosis and high mortality [14,16,17]. One of our patients in the AI group who finally died showed very high baseline cortisol level and had inadequate response after ACTH stimulation. This may also suggest tissue resistance to glucocorticoid [18] or adrenal dysfunction. Almost all of the patients had good response of adrenals to ACTH stimulation during acute stress, suggesting that they had adequate adrenal reserve during the acute phase of ALI/ARDS. Thus, the treatment with glucocorticoid in the early phase of ALI/ARDS (in relation to adrenal dysfunction) may not be beneficial in most. However, our findings should be interpreted with awareness that the sample size was limited to those who presented with ALI/ARDS during the 1-year study period. Thus, the sample size may have led us to underestimate the clinical impact of AI in pediatric ALI/ ARDS. In previous clinical trials, high doses of methylprednisolone for 1 to 2 days early in the course of severe sepsis or established ARDS do not prevent ARDS development or reduce mortality rate, and they may be even harmful [19,20]. In contrast, the lower doses of methylprednisolone in late, unresolving ARDS may be beneficial [21], but again this is not confirmed by recent published data [9]. In conclusion, these data indicate that AI may be common among children with ALI/ARDS. The HPA axis appears to be an important component of the stress response and modulates the systemic cardiovascular and cellular response to injury. No study in pediatrics has yet examined HPA axis in patients with ALI/ARDS. These results would be an initial step to further study the impact of AI on clinical outcome of these children in a larger scale.
Acknowledgment This work was supported by grants (TRG 4780011) from Thailand Research Fund (Bangkok, Thailand) and Ratchadapiseksomphod Fund (Bangkok, Thailand). We thank the staff and fellows of the Pediatric Pulmonology and Critical Care Division of Chulalongkorn University (Bangkok, Thailand) and the PICU team and Prof Suthipong Wacharasindhu, MD (Chief, Pediatric Endocrinology) of King Chulalongkorn Memorial Hospital (Bangkok, Thailand). We are indebted to the Central Laboratory of King Chulalongkorn Memorial Hospital who assisted in cortisol level measurement and to Mettanando Bhikkhu for editing the manuscript.
References [1] Quiney N, Durkin M. Adrenocortical failure in intensive care. BMJ 1995;310:1253-4.
318 [2] Lamberts S, Bruining H, De Jong F. Corticosteroids therapy in severe illness. N Engl J Med 1997;337:1285-92. [3] Thompson BT. Glucocorticoids and acute lung injury. Crit Care Med 2003;31(4 Suppl):S253-7. [4] Chirousos G. Seminars in medicine of the Beth Israel Hospital, The hypothalamic-pituitary-adrenal axis and immune mediated inflammation. N Engl J Med 1995;332:1351-62. [5] Cooper MS, Stewart PM. Corticosteroid insufficiency in acutely ill patients. N Engl J Med 2003;348(8):727-34. [6] Suter PM. Lung inflammation in ARDS—friend or foe? N Engl J Med 2006;354:1739-42. [7] Meduri G, Kantangat S. Glucocorticoid treatment of sepsis and acute respiratory distress syndrome: time for a critical reappraisal. Crit Care Med 1998;26:630-3. [8] Gonzalez H, Nardi O, Annane D. Relative adrenal failure in the ICU: an identifiable problem requiring treatment. Crit Care Clin 2006;22 (1):105-18, vii. [9] Steinberg KP, Hudson LD, Goodman RB, Hough CL, Lanker PN, et al.. The National Heart, Lung, and Blood Institute Acute Respiratory Distress Syndrome (ARDS) Clinical Trials Network. Efficacy and safety of corticosteroids for persistent acute respiratory distress syndrome. N Engl J Med 2006;354(16):1671-84. [10] Pizarro CF, Troster EJ, Damiani D, et al. Absolute and relative adrenal insufficiency in children with septic shock. Crit Care Med 2005;33 (4):855-9.
R. Samransamruajkit et al. [11] Agus M. One step forward: an advance in understanding of adrenal insufficiency in the pediatric critically ill. Crit Care Med 2005;33(4):911-2. [12] Annane D. Glucocorticoids in the treatment of severe sepsis and septic shock. Curr Opin Crit Care 2005;11(5):449-53. [13] Ware L, Matthay MA. The acute respiratory distress syndrome. N Engl J Med 2000;342:1334-49. [14] Annane D, Sebille V, Troche G, et al. A 3 level prognostic classification in septic shock based on cortisol levels and cortisol response to corticotropin. JAMA 2000;283:1038-45. [15] Hatherill M, Tibby S, Hillard T, et al. Adrenal insufficiency in septic shock. Arch Dis Child 1999;80:51-5. [16] Annane D, Sebille V, Charpentier C, et al. Effect of treatment with low doses of hydrocortisone and fludrocortisone on mortality in patients with septic shock. JAMA 2002;288:862-71. [17] Yildiz O, Doganay M, Aygen B, et al. Physiologic dose steroid therapy in sepsis. Crit Care 2002;6:251-8. [18] Prigent H, Maxime V, Annane D. Clinical review: corticotherapy in sepsis. Crit Care 2004;8:122-9. [19] Sprung C, Caralis P, Marcial E, et al. The effect of high dose corticosteroids in patients with septic shock. N Engl J Med 1984;311:1137-43. [20] Bernard GR, Luuce J, Sprung C. High dose corticosteroids in patients with the ARDS. N Engl J Med 1987;317:1565-70. [21] Meduri G, Chinn A, Leeper K, et al. Corticosteroids rescue treatment of progressive fibroproliferation in late ARDS: patterns of response and predictors of outcome. Chest 1994;105:1516-27.