Paediatric Respiratory Reviews 15 (2014) 24–27

Mini-Symposium: Controversies in the Evaluation and Treatment of Sickle Cell Disease

Systemic Corticosteroids in Acute Chest Syndrome: Friend or Foe?

Folasade Ogunlesi 1, Matthew M. Heeney 2, Anastassios C. Koumbourlis 1,*1 Division of Pulmonary & Sleep Medicine, Children’s National Medical Center/George Washington University, Washington DC2 Division of Hematology/Oncology, Boston Children’s Hospital /Harvard Medical School, Boston MA

EDUCATIONAL AIMS

THE READER WILL:

� Learn the the pathophysiology of the Acute Chest Syndrome (ACS) and its various triggers� Review the inflammatory responses elicited by the various triggers of the ACS� Be able to evaluate the evidence on the benefit of systemic steroids in the treatment of Acute Chest Syndrome (ACS) and of the

potential side effects

A R T I C L E I N F O

Keywords:

Acute Chest Syndrome

Sickle Cell

Corticosteroids

S U M M A R Y

Acute chest syndrome(ACS) is the most common pulmonary complication of sickle cell disease (SCD), the

second most common cause of hospitalization and the primary cause of death in patients with sickle cell

disease. Its highest prevalence is in early childhood. The pathogenesis of ACS is unknown but many

predisposing conditions and mechanisms have been implicated including infections, pulmonary fat

embolism, asthma and ischemic reperfusion injury. These conditions are associated with inflammation

and therefore, the use of corticosteroids has been advocated because of their anti-inflammatory

properties. Although, significant benefits from their use have been shown, there is great reluctance in

using them because of reports of serious adverse effects, such as readmission to the hospital due rebound

pain crisis, stroke, renal infarction, coma and even death. The current article reviews the evidence in favor

and against the use of corticosteroids in ACS. Emphasis is given on the potential benefits vs. risks among

the different types of corticosteroids, the importance of the dosing regimen and the role of underlying co-

morbidities.

� 2013 Elsevier Ltd. All rights reserved.

Contents lists available at ScienceDirect

Paediatric Respiratory Reviews

INTRODUCTION

Acute chest syndrome (ACS) is the most common form of acutepulmonary disease among patients with sickle cell disease (SCD),occurring in as many as 50% of patients [1]. It is the second mostcommon cause for admission to the hospital (after vaso-oclusivecrises) and the leading cause of death accounting for up to 25% ofSCD-related deaths [2]. ACS occurs in all ages but it is morecommon in children with the highest prevalence occurringbetween the ages of 2 and 5 years [3]. Several conditions andmechanisms are known to predispose, exacerbate or complicatethe pathogenesis of ACS but the exact cause remains unknown [4].

* Corresponding author. Children’s National Medical Center, 111 Michigan Ave

NW, Washington DC 20010. Tel.: +1 203 476 2642; fax: +1 202 476 5864.

E-mail address: akoumbou@childrensnational.org (A.C. Koumbourlis).

1526-0542/$ – see front matter � 2013 Elsevier Ltd. All rights reserved.

http://dx.doi.org/10.1016/j.prrv.2013.10.004

As a result there is no specific therapy for ACS. Instead itsmanagement has focused on the treatment of presumed bacterialinfection with broad spectrum antibiotics, judicious administra-tion of intravenous fluids, and pain medications in order toalleviate the patient’s discomfort and to prevent atelectasis causedby the patients’ hypoventilation secondary to splinting. Simple orexchange blood transfusions decrease the percentage of sickled redblood cells, improve oxygenation and often stop the progression ofACS.

Given the increasing evidence that inflammation is animportant factor in the presentation of ACS [5,6], treatment withimmunomodulatory agents such as systemic corticosteroids hasbeen advocated [7]. However, reports of serious side effects hasmade many clinicians reluctant to use corticosteroids. The currentarticle reviews the available evidence in favor and against the useof systemic corticosteroids in SCD.

F. Ogunlesi et al. / Paediatric Respiratory Reviews 15 (2014) 24–27 25

ACS & Inflammation

The pathogenesis of ACS has been attributed to many diversemechanisms in which the presence or development of inflamma-tion feature prominently [4–6]. These conditions are brieflysummarized below.

Infection

Patients with SCD develop functional asplenia due to infarctionof the organ ‘in situ’ that predisposes them to infections especiallyby encapsulated organisms [eg pneumococcus]. Other mechan-isms that may affect the immunity of patients with SCD have beenalso proposed including evidence from transgenic sickle mice thataltered baseline immunity may be secondary to morphologicabnormalities of the splenic tissue [8]. Finally, there is clinicalevidence of abnormal function of the complement systemalthough the results have been contradictory [9]. Infants andyoung children are at higher risk for infections because of theirnaturally immature immune system.

ACS is a syndrome defined on the basis of a combined clinicaland radiographic characteristics (development of a new pulmon-ary infiltrate involving at least one complete lung segment; fever;and any other constellation of pulmonary symptoms includinghypoxemia, dyspnea, tachypnea, wheezing, chest pain, or cough)that is often difficult to distinguish from a typical bacterialpneumonia in a patient without SCD [2].

Although inflammation is characteristic of both entities,patients with SCD and ACS seem to develop a much more rapidlyprogressive and severe presentation suggesting a propensity todevelop a more significant inflammatory response than non-SCDindividuals. Indeed, studies in transgenic sickle mice have shownthat those with SCD tend to mount a much more violentinflammatory response, and they seem to be much moresusceptible to injury from the inflammatory mediators that arebeing released than compared with the non-SCD mice [10–13]. Inother words, it is possible that patients with SCD may suffer notonly by the increased susceptibility to infections but also by theirown exaggerated inflammatory response that may then becomethe main trigger for the ACS.

Pulmonary fat embolism

Pulmonary fat embolism (PFE) is currently recognized as one ofthe major mechanisms associated with ACS [14]. Fatty bonemarrow may be released into the blood as a result of bone marrownecrosis caused by a vaso-occlusive bony crisis. Embolic fatactivates secretory phospholipase A2 (sPLA2), an enzyme thatcleaves phospholipids and liberates free fatty acids and generatesarachidonic acid that in turn produces inflammatory leukotrienesand prostaglandins. These fatty acids injure the pulmonaryendothelium, increase the expression of VCAM-1 and promotethe adhesion of erythrocytes to endothelium in vitro providingevidence for pathologic adhesive interactions in PFE. Theconcentration of sPLA2 in peripheral blood has been proposedas a laboratory marker of ACS, because it correlates with the courseand severity of ACS. Specifically, the sPLA2 increases before ACSbecomes clinically apparent, it peaks at the onset, and declinesduring resolution [15].

The basis for the PFE as a mechanism for the development ofACS has been based on the presence of lipid-laden macrophages(LLM) in bronchoalveolar lavage fluid in patients with ACS [16].However, increased numbers of LLMs are not pathognomonic offatty bone marrow embolization. Increased numbers of LLMs canalso be found in cases of aspiration of fat containing food products,and most importantly they can be released from injured cells (e.g.in severe pneumonia or ARDS) [17]. Thus, it is possible that thepresence of increased LLMs in the BAL may be due to cellular

damage secondary to the ACS and not the underlying pathogenicmechanism of ACS.

Asthma

Several epidemiologic studies have reported an unusually highprevalence of airway hyperresponsiveness and asthma amongchildren and adults with SCD compared with the generalpopulation (even after adjusting for race, socioeconomic condi-tions etc) [18,19]. Patients with SCD and asthma seem to have amuch higher (as high as 6-fold) risk of recurrent ACS comparedwith those without history of asthma [19–22]. Most importantlypatients with SCD who are admitted to the hospital for pain crisesare more likely to develop ACS if they have asthma, and they seemto improve after treatment with bronchodilators [19–22]. Asthmais a classic inflammatory disease and its association with ACSsuggests a possible synergistic effect by the presence ofinflammation.

Animal data in transgenic sickle mice have shown thatexperimentally induced asthma is associated with greatermortality due to increased allergic lung inflammation (elevationsin eosinophils, eosinophil peroxidase, and IgE levels) comparedwith control mice without asthma. Eosinophils (a major source ofleukotrienes in asthma) and elevations of LTB4 and LTC4 in bloodand LTE4 in urine also occur in patients with sickle cell disease [5].

Ischemia/reperfusion injury

SCD has traditionally been considered a disorder of micro-vascular vaso-occlusion secondary to mechanical obstruction by ofdeformed RBCs and subsequent tissue hypoxia [23] However, morerecently a modified paradigm has emerged suggesting that thewide spectrum of clinical manifestations of SCA results fromrecurrent episodes of ischemia-reperfusion injury [5]. Ischemia-reperfusion triggers a multifactorial cascade including inflamma-tory response characterized by increased leukocyte and sickleerythrocyte adhesion to vascular endothelium and activation ofcoagulation, platelets and neutrophils. The data suggest that acutelung vaso-occlusive injury causes an inflammatory response thattriggers chemotaxis of leukocytes and secondary injury [24–27].

SYSTEMIC CORTICOSTEROIDS AND ACS

Corticosteroids are powerful anti-inflammatory medicationswith pleiotropic beneficial effects in a variety of diverse clinicalconditions including cancer (pain control and mood-elevation), intrauma (decrease the risk of fat-embolism) and complications/mortality from ARDS. [28–30] ACS shares many of the manifesta-tions and pathology with these conditions (e.g. pain, fat embolism,hypoxemia and acute lung injury). Thus, there has been greatinterest in the role of systemic corticosteroids as the means ofinhibiting the inflammatory response that accompanies tissueischemia/infarction. Several clinical studies using different pre-parations and doses of corticosteroids have held promising results.Griffin et al. [31] used high dose methylprednisolone in vaso-occlusive crisis and reported significant reduction in the durationof analgesic therapy and hospitalization.

Bernini et al. [32] investigated the efficacy of a lower dose of alonger acting glucocorticoid, dexamethasone in 43 children withmild to moderately severe ACS in a randomized, double-blind,placebo-controlled trial. They showed reduction in the length ofhospitalization by about 40%, in the need for transfusion due toworsening anemia, in the duration of fever, and in need of oxygenrequirement and pain treatment.

Unfortunately, the successes associated with the use ofcorticosteroids in the setting of SCD have been accompanied byreports of complications ranging from recurrence of pain requiringreadmission to the hospital, to episodes of severe vaso-occlusive

F. Ogunlesi et al. / Paediatric Respiratory Reviews 15 (2014) 24–2726

crises, ACS, stroke, renal infarction and even coma and death [33–36]. In the study by Griffin et al. [31] 15% of the patients who weretreated with corticosteroids had ‘‘rebound’’ pain crisis thatrequired readmission within 6 days after discharge. Similar resultswere reported by Bernini et al. [32], although in their study thedifference between those treated with corticosteroids was notstatistically significant (27% vs 5%; P = .095).

In a retrospective study, Strouse et al. [33] reported that theodds ratio of readmission within 14 days following treatment withcorticosteroids was 20 fold higher among those treated withcorticosteroids. In contrast, Isakoff et al. [37], also in a retrospectivestudy, reported no rebound pain episodes or other seriouscomplications treated with dexamethasone although thesepatients also received transfusion. Preliminary data from aprospective, multicenter, randomized trial for dexamethasoneby the Comprehensive Sickle Cell Centers (CSCC) network [38]showed that Dexamethasone shortened the duration of hospita-lization by almost a day (20 hours) and resulted in the decrease ofthe leucocyte activation marker sL-selectin. Rebound painoccurred in both groups with more episodes occurring in thedexamethasone group. The sL-selectin also decreased withdexamethasone and increased in patients who had rebound pain.It should be noted that these results were derived from only 12patients (the study was discontinued due to slow recruitment) andtherefore generalization may not be appropriate.

To some extent the discrepancies between the various studiesregarding the frequency and severity of the side effects may berelated to the type and/or the dose of the steroid preparation used. AsStrouse et al. showed, patients with asthma and SCD treated withdexamethasone and those treated with high doses of prednisolone(>2mg/kg/day) were more likely to be readmitted compared withthose treated with low dose prednisolone [33]. It is possible that theresults of the prospective study might have been better had theinvestigators used oral prednisone instead of dexamethasone. Itshould also be noted that some of the more serious complicationswere reported in patients who had other co-morbidities such asautoimmune and/or systemic diseases receiving treatments otherthan corticosteroids that have the potential of causing seriouscomplications on their own [34]. Similarly, the one corticosteroid-associated fatality occurred in a patient who had also receivedgranulocyte colony stimulating factor [35].

Understandably, the possibility of serious side effects hascreated serious concerns and has led many physicians to avoid theuse of systemic corticosteroids in patients with SCD. Using datafrom the Pediatric Health Information System (PHIS) databasefrom 41freestanding children’s hospitals, Sobota et al. [39]analyzed the variation between hospitals regarding the use ofcorticosteroids for ACS. They found that systemic corticosteroidwere used in only 17% of admission for ACS and determined thatthe likelihood of using corticosteroids varied widely amonghospitals, ranging from as low as 10% to as high as 86%.Corticosteroids were used more often in patients with comorbidasthma and in ‘‘severe’’ patients (defined as Hb-SS genotype,comorbid asthma, requiring ventilator support and ICU care). Incontrast to earlier studies, length of stay was found to be longer inthe steroid group but the readmission rate within 72 hours washigher in the non-steroid group compared with the steroid group(4.4% vs. 1.9%). In a subsequent study the same investigators foundthat the 30-day readmission rate after sickle cell crisis was 17%[40]. The factors associated with readmission were older age,admission only for pain and inpatient use of steroids; simpletransfusion had a protective effect for readmission [40].

The etiology of the ‘‘rebound’’ pain phenomenon aftercorticosteroid treatment is not clear. It is possible that a briefcourse of corticosteroids may temporarily suppress but notcompletely treat the underlying inflammation that is expected

to continue until vaso-occlusion is relieved and reperfusion injuryresolves. In such case, it is likely that there will be a ‘‘flare-up’’ ofthe inflammation/pain shortly after the steroids are discontinued.By this mechanism the observed ‘‘rebound’’ effect may be due tothe early discontinuation of the corticosteroids and not a side effectof their use.

Special consideration should be given to the combination ofblood transfusion with corticosteroids. Blood transfusions havebeen shown to decrease many inflammatory markers that areincreased during episodes of ACS although this effect is notsustained [41]. It is not known whether the blood transfusion hasany direct effect on the action of corticosteroids and it is morelikely that they act synergistically. Systemic corticosteroids arerelatively slow acting. Thus, by decreasing the amount ofcirculating inflammatory mediators, blood transfusion providesimmediate anti-inflammatory action until the steroids startworking (usually 12-24 hours after their administration), and itmay also allow the use of a lower dose. Support to this theory isprovided by the study by Isakoff et al. [37] who reported no sideeffects in their patients with ACS who were treated with steroidsand blood transfusions. Sobota et al. [40] also reported that bloodtransfusion had a protective effect against readmission. Unfortu-nately, their analysis did not assess specifically the outcome inpatients who received steroids plus transfusion.

CONCLUSION

It is clear that the role of corticosteroid therapy in SCD and itscomplications is anything but settled. That there is wide variabilityin their use clearly shows the need for further study. We believethat the legitimate concerns about possible complications shouldnot obscure the evidence for potential positive therapeutic effect inthe right setting and that the concern for protecting the patientfrom iatrogenic injury should not lead to the under-treatment ofappropriate patients. This is particularly important for SCDpatients with comorbid asthma that together are major riskfactors for ACS. As Sobota et al. showed, less than 40% of the SCDpatients with asthma received corticosteroids during theirhospitalization for ACS, which may explain why those patientswith co-morbid asthma had a greater relative risk of readmission[39,40].

We strongly recommend the evaluation of all patients with SCDfor asthma (especially those who have had a prior episode of ACS)and their comprehensive treatment according to the guidelines fortreatment of asthma. Based on the currently available evidence webelieve that corticosteroids should be considered for patients withconditions such as asthma that are known to respond tocorticosteroids. However, when used, it would be prudent to treatwith relatively low dose (<2mg/kg/day; max: 60 mg/day) of oralprednisone and avoid the abrupt discontinuation of the medicationbut continue with a relatively prolonged tapering course [42].Whether there is an ‘‘optimal’’ corticosteroid agent, dose andduration of treatment requires further investigation.

FUTURE RESEARCH DIRECTIONS

� Low dose steroid treatment for asthma exacerbation and ACS� The role of inhaled corticosteroid in patients with SCD and

comorbid asthma

References

[1] Hassell KL. Population estimates of sickle cell disease in the U.S.. AmericanJournal of Preventive Medicine 2010;38:S512–21.

[2] Miller AC, Gladwin MT. Pulmonary complications of sickle cell disease. Am JRespir Crit Care Med 2012;185:1154–65.

F. Ogunlesi et al. / Paediatric Respiratory Reviews 15 (2014) 24–27 27

[3] Gill FM, Sleeper LA, Weiner SJ, et al. Clinical events in the first decade in acohort of infants with sickle cell disease. Cooperative Study of Sickle CellDisease. Blood 1995;86:776–83.

[4] Vishinsky EP, Neumayr LD, Earles AN, et al. Causes and outcomes of acute chestsyndrome in sickle cell disease. National Acute Chest Syndrome Study Group.N Engl J Med 2000;342:1855–65.

[5] Platt OS. Sickle cell anemia as an inflammatory disease. J Clin Invest2000;106:337–8.

[6] Wallace KL, Marshall MA, Ramos SI, et al. NKT cells mediate pulmonaryinflammation and dysfunction in murine sickle cell disease through produc-tion of IFN-g and CXCR3 chemokines. Blood 2009;114:667–76.

[7] Field JJ, Lin G, Okam MM, et al. Sickle cell vaso-occlusion causes activation ofiNKT cells that is decreased by the adenosine A2A receptor agonist regade-noson. Blood 2013;121:3329–34.

[8] Szczepanek SM, McNamara JT, Secor Jr ER, et al. Splenic morphological changesare accompanied by altered baseline immunity in a mouse model of sickle-celldisease. Am J Pathol 2012;181:1725–34.

[9] Test ST, Woolworth VS. Defective regulation of complement by the sickleerythrocyte: evidence for a defect in control of membrane attack complexformation. Blood 1994;83:842–52.

[10] Kaul DK, Hebbel RP. Hypoxia/reoxygenation causes inflammatory response intransgenic sickle mice but not in normal mice. J Clin Invest 2000;106:411–20.

[11] Holtzclaw JD, Jack D, Aguayo SM, et al. Enhanced pulmonary and systemicresponse to endotoxin in transgenic sickle mice. Am J Respir Crit Care Med2004;169:687–95.

[12] Belcher JD, Mahaseth H, Welch TE, et al. Critical role of endothelial cellactivation in hypoxia-induced vasoocclusion in transgenic sickle mice. Am JPhysiol Heart Circ Physiol 2005;288:H2715–2.

[13] Sabaa N, de Franceschi L, Bonnin P, et al. Endothelin receptor antagonismprevents hypoxia-induced mortality and morbidity in a mouse model ofsickle-cell disease. J Clin Invest 2008;118:1924–33.

[14] Vichinsky E, Williams R, Das M, et al. Pulmonary fat embolism: a distinct causeof severe acute chest syndrome in sickle cell anemia. Blood 1994;83:3107–12.

[15] Ballas SK, Files B, Luchtman-Jones L, et al. Secretory phospholipase A2 levels inpatients with sickle cell disease and acute chest syndrome. Hemoglobin2006;30:165–70.

[16] Godeau B, Schaeffer A, Bachir D, et al. Bronchoalveolar lavage in adult sicklecell patients with acute chest syndrome: value for diagnostic assessment of fatembolism. Am J Respir Crit Care Med 1996;153:1691–6.

[17] Koumbourlis AC, Santiago MT, Mustafa U, et al. Lipid Laden Macrophages inbronchoalveolar lavage fluid: Aspiration or Infection? Proc Am Thorac Soc2006;3:A166.

[18] Newaskar M, Hardy KA, Morris CR. Asthma in sickle cell disease. ScientificWorld Journal 2011;11:1138–52.

[19] Knight-Madden JM, Forrester TS, Lewis NA, Greenough A. Asthma in childrenwith sickle cell disease and its association with acute chest syndrome. Thorax2005;60:206–10.

[20] Boyd JH, Macklin EA, Strunk RC, et al. Asthma is associated with acute chestsyndrome and pain in children with sickle cell anemia. Blood 2006;108:2923–7.

[21] Nordness ME, Lynn J, Zacharisen MC, et al. Asthma is a risk factor for acutechest syndrome and cerebral vascular accidents in children with sickle celldisease. Clin Mol Allergy 2005;3:2.

[22] Sylvester KP, Patey RA, Broughton S, et al. Temporal relationship of asthma toacute chest syndrome in sickle cell disease. Pediatr Pulmonol 2007;42:103–6.

[23] Hebbel RP, Moldow CF, Steinberg MH. Modulation of erythrocyte-endothelialinteractions and the vasocclusive severity of sickling disorders. Blood1981;58:947–52.

[24] Turhan A, Weiss LA, Mohandas N, et al. Primary role for adherent leukocytes insickle cell vascular occlusion: a new paradigm. PNAS 2002;99:3047–51.

[25] Francis Jr RB. Platelets, coagulation, and fibrinolysis in sickle cell disease: theirpossible role in vascular occlusion. Blood Coagul Fibrinolysis 1991;2:341–53.

[26] Setty BN, Stuart MJ. Vascular cell adhesion molecule-1 is involved in mediatinghypoxia-induced sickle red blood cell adherence to endothelium: potentialrole in sickle cell disease. Blood 1996;88:2311–20.

[27] Pujols L, Mullol J, Picado C. Importance of glucocorticoid receptors in upperand lower airways. Front Biosci 2010;15:789–800.

[28] Bruera E, Roca E, Cedaro L, et al. Action of oral methylprednisolone in terminalcancer patients: a prospective randomized double-blind study. Cancer TreatRep 1985;69:751–4.

[29] Bederman SS, Bhandari M, McKee MD, et al. Do corticosteroids reduce the riskof fat embolism syndrome in patients with long-bone fractures?. A meta-analysis. Can J Surg 2009;52:386–93.

[30] Tang BM, Craig JC, Eslick GD, et al. Use of corticosteroids in acute lung injuryand acute respiratory distress syndrome: a systematic review and meta-analysis. Crit Care Med 2009;37:1594–603.

[31] Griffin TC, McIntire D, Buchanan GR. High-dose intravenous methylpredni-solone therapy for pain in children and adolescents with sickle cell disease. NEngl J Med 1994;330:733–7.

[32] Bernini JC, Rogers ZR, Sandler ES, et al. Beneficial effect of intravenousdexamethasone in children with mild to moderately severe acute chestsyndrome complicating sickle cell disease. Blood 1998;92:3082–9.

[33] Strouse JJ, Takemoto CM, Keefer JR, et al. Corticosteroids and increased risk ofreadmission after acute chest syndrome in children with sickle cell disease.Pediatr Blood Cancer 2007;50:1006–12.

[34] Couillard S, Benkerrou M, Girot R, et al. Steroid treatment in children withsickle-cell disease. Haematologica 2007;92:425–6.

[35] Darbari DS, Castro O, Taylor VI JG, et al. Severe vaso-occlusive episodesassociated with use of systemic corticosteroids in patients with sickle celldisease. J Natl Med Assoc 2008;100:948–51.

[36] Adler BK, Salzman DE, Carabasi MH, et al. Fatal sickle cell crisis after gran-ulocyte colony-stimulating factor administration. Blood 2001;97:3313–4.

[37] Isakoff MS, Lillo JA, Hagstrom JN. A single-institution experience with treat-ment of severe acute chest syndrome: lack of rebound pain with dexametha-sone plus transfusion therapy. J Pediatr Hematol Oncol 2008;30:322–5.

[38] Quinn CT, Stuart MJ, Kesler K, et al. Investigators of the Comprehensive SickleCell Centers. Tapered oral dexamethasone for the acute chest syndrome ofsickle cell disease. Br J Haematol 2011;155:263–7.

[39] Sobota A, Graham DA, Heeney MM, et al. Corticosteroids for acute chestsyndrome in children with sickle cell disease: Variation in use and associationwith length of stay and readmission. Am J Hematol 2010;85:24–8.

[40] Sobota A, Graham DA, Neufeld EJ, Heeney MM. Thirty-day readmission ratesfollowing hospitalization for pediatric sickle cell crisis at freestanding chil-dren’s hospitals: risk factors and hospital variation. Pediatr Blood Cancer2012;58:61–5.

[41] Liem RI, O’Gorman MR, Brown DL. Effect of red cell exchange transfusion onplasma levels of inflammatory mediators in sickle cell patients with acutechest syndrome. Am J Hematol 2004;76:19–25.

[42] Hsu LL, Gee BE. Sickle Cell Pain After Abrupt Withdrawal of Glucocorticoids forAsthma. 2010: A6207, 10.1164/ajrccm-conference.2010.181.1 MeetingAbstracts. A6207.