corticosteroids in acute respiratory distress syndrome · corticosteroids in acute respiratory...

13
Braz J Med Biol Res 38(2) 2005 Corticosteroids in acute respiratory distress syndrome Laboratórios de 1 Investigação Pulmonar, and 2 Fisiologia da Respiração, Instituto de Biofísica Carlos Chagas Filho, Universidade Federal do Rio de Janeiro, Rio de Janeiro, RJ, Brasil A.B.S. Fernandes 1 , W.A. Zin 2 and P.R.M. Rocco 1 Abstract Improving the course and outcome of patients with acute respira- tory distress syndrome presents a challenge. By understanding the immune status of a patient, physicians can consider manipulat- ing proinflammatory systems more rationally. In this context, corti- costeroids could be a therapeutic tool in the armamentarium against acute respiratory distress syndrome. Corticosteroid therapy has been studied in three situations: prevention in high-risk patients, early treatment with high-dose, short-course therapy, and pro- longed therapy in unresolving cases. There are differences be- tween the corticosteroid trials of the past and recent trials: today, treatment starts 2-10 days after disease onset in patients that failed to improve; in the past, the corticosteroid doses employed were 5- 140 times higher than those used now. Additionally, in the past treatment consisted of administering one to four doses every 6 h (methylprednisolone, 30 mg/kg) versus prolonging treatment as long as necessary in the new trials (2 mg kg -1 day -1 every 6 h). The variable response to corticosteroid treatment could be attributed to the heterogeneous biochemical and molecular mechanisms acti- vated in response to different initial insults. Numerous factors need to be taken into account when corticosteroids are used to treat acute respiratory distress syndrome: the specificity of inhibi- tion, the duration and degree of inhibition, and the timing of inhibition. The major continuing problem is when to administer corticosteroids and how to monitor their use. The inflammatory mechanisms are continuous and cyclic, sometimes causing dete- rioration or improvement of lung function. This article reviews the mechanisms of action of corticosteroids and the results of experi- mental and clinical studies regarding the use of corticosteroids in acute respiratory distress syndrome. Correspondence P.R.M. Rocco Laboratório de Investigação Pulmonar Instituto de Biofísica Carlos Chagas Filho, CCS, UFRJ 21949-900 Rio de Janeiro, RJ Brasil Fax: +55-21-2280-8193 E-mail: [email protected] Research supported by PRONEX- MCT, CNPq and FAPERJ. Received April 12, 2004 Accepted December 8, 2004 Key words Acute respiratory distress syndrome Collagen Cytokine Methylprednisolone Experimental models Clinical trials Introduction Effective management of acute respira- tory distress syndrome (ARDS) has been problematic since this syndrome was first described in 1967. Ashbaugh and colleagues (1) described 12 patients with acute respira- tory distress, cyanosis refractory to oxygen therapy, decreased lung compliance, and diffuse infiltrates evident on the chest radio- Brazilian Journal of Medical and Biological Research (2005) 38: 147-159 ISSN 0100-879X Review

Upload: hanhi

Post on 02-Apr-2019

225 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Corticosteroids in acute respiratory distress syndrome · Corticosteroids in acute respiratory distress syndrome Laboratórios de 1Investigação Pulmonar, and 2Fisiologia da Respiração,

147

Braz J Med Biol Res 38(2) 2005

Corticosteroids in ARDS

Corticosteroids in acute respiratorydistress syndrome

Laboratórios de 1Investigação Pulmonar, and 2Fisiologia da Respiração,Instituto de Biofísica Carlos Chagas Filho, Universidade Federal do Rio de Janeiro,Rio de Janeiro, RJ, Brasil

A.B.S. Fernandes1,W.A. Zin2

and P.R.M. Rocco1

Abstract

Improving the course and outcome of patients with acute respira-tory distress syndrome presents a challenge. By understandingthe immune status of a patient, physicians can consider manipulat-ing proinflammatory systems more rationally. In this context, corti-costeroids could be a therapeutic tool in the armamentarium againstacute respiratory distress syndrome. Corticosteroid therapy hasbeen studied in three situations: prevention in high-risk patients,early treatment with high-dose, short-course therapy, and pro-longed therapy in unresolving cases. There are differences be-tween the corticosteroid trials of the past and recent trials: today,treatment starts 2-10 days after disease onset in patients that failedto improve; in the past, the corticosteroid doses employed were 5-140 times higher than those used now. Additionally, in the pasttreatment consisted of administering one to four doses every 6 h(methylprednisolone, 30 mg/kg) versus prolonging treatment aslong as necessary in the new trials (2 mg kg-1 day-1 every 6 h). Thevariable response to corticosteroid treatment could be attributedto the heterogeneous biochemical and molecular mechanisms acti-vated in response to different initial insults. Numerous factorsneed to be taken into account when corticosteroids are used totreat acute respiratory distress syndrome: the specificity of inhibi-tion, the duration and degree of inhibition, and the timing ofinhibition. The major continuing problem is when to administercorticosteroids and how to monitor their use. The inflammatorymechanisms are continuous and cyclic, sometimes causing dete-rioration or improvement of lung function. This article reviews themechanisms of action of corticosteroids and the results of experi-mental and clinical studies regarding the use of corticosteroids inacute respiratory distress syndrome.

CorrespondenceP.R.M. Rocco

Laboratório de Investigação Pulmonar

Instituto de Biofísica Carlos

Chagas Filho, CCS, UFRJ

21949-900 Rio de Janeiro, RJ

Brasil

Fax: +55-21-2280-8193

E-mail: [email protected]

Research supported by PRONEX-MCT, CNPq and FAPERJ.

Received April 12, 2004

Accepted December 8, 2004

Key words• Acute respiratory distress

syndrome• Collagen• Cytokine• Methylprednisolone• Experimental models• Clinical trials

Introduction

Effective management of acute respira-tory distress syndrome (ARDS) has beenproblematic since this syndrome was first

described in 1967. Ashbaugh and colleagues(1) described 12 patients with acute respira-tory distress, cyanosis refractory to oxygentherapy, decreased lung compliance, anddiffuse infiltrates evident on the chest radio-

Brazilian Journal of Medical and Biological Research (2005) 38: 147-159ISSN 0100-879X Review

Page 2: Corticosteroids in acute respiratory distress syndrome · Corticosteroids in acute respiratory distress syndrome Laboratórios de 1Investigação Pulmonar, and 2Fisiologia da Respiração,

148

Braz J Med Biol Res 38(2) 2005

A.B.S. Fernandes et al.

graph. Although the term ARDS is often usedinterchangeably with acute lung injury, bystrict criteria ARDS should be reserved forthe most severe end of the spectrum. Acutelung injury is characterized by acute develop-ment of bilateral infiltrates detected on chestradiographs, pulmonary capillary wedge pres-sure of 18 mmHg or less or the absence ofclinically evident left atrial hypertension, anda ratio of partial pressure of arterial oxygen tothe fraction of inspired oxygen (PaO2/FiO2)of 300 or less. The definition of ARDSdiffers only by the presence of worsenedoxygenation, represented by a PaO2/FiO2

ratio of 200 or less (2). Despite recentadvances in the understanding of the patho-physiology of ARDS and improved life sup-port of patients with ARDS, the mortalityrates persist at 30 to 60% (3,4).

ARDS is thought to be a uniform expres-sion of a diffuse and overwhelming inflam-matory reaction of the pulmonary paren-chyma to a variety of serious underlyingdiseases. In 1994, the American-EuropeanConsensus Conference (2) defined twopathogenetic pathways leading to ARDS as adirect (“primary” or “pulmonary”) insult,that directly affects lung parenchyma, and anindirect (“secondary” or “extrapulmonary”)insult, that results from an acute systemicinflammatory response (5,6). The patho-logic features of the lung in ARDS derivefrom severe injury to the alveolocapillaryunit. The morphologic picture of the lung inARDS has been labeled diffuse alveolar dam-age, and extravasations of intravascular fluiddominate the onset of the disease. Micro-scopic findings are dependent on the stage ofthe illness. Traditionally, ARDS has beendivided into three stages in which an initialinflammatory phase (exudative) is followedby fibro-proliferation, which can lead toestablished interstitial and intra-alveolar fi-brosis, the final phase. The histological fea-tures of the exudative phase are: a) hyalinemembranes, b) alveolar collapse, and c)swollen type I pneumocytes with cytoplas-

mic vacuoles. The endothelial cells swell, theintercellular junctions widen, and pinocyticvesicles increase, causing disruption in thecapillary membrane and resulting in capillaryleakage and edema formation (7). The prolif-erative phase has been described to begin asearly as the third day but was most prominentin the second and third weeks after symptomonset. However, recently, some investiga-tors have reported that fibro-proliferation isan early response to lung injury (8-10). Thus,inflammatory and repair mechanisms occurin parallel rather than in series. In the fibroticphase, extensive remodeling of the lung bysparsely cellular collagenous tissue occurs,air spaces are irregularly enlarged and thereis alveolar duct fibrosis. Type III collagen isreplaced by type I collagen, leading to a stifflung over time (7).

The recognition that neutrophils, macro-phages, and other components of the inflam-matory cascade participate in the progres-sion of ARDS has resulted in the use of anti-inflammatory agents, particularly glucocor-ticoids (11,12). Although glucocorticoidtreatment might increase the risk of infec-tion, careful limited administration of thecorticosteroid could promote the resolutionof injured tissue (13,14).

Corticosteroid therapy in ARDS has beenstudied in three main different situations: 1)prevention in high risk patients (15,16), 2)early treatment with high-dose, short-coursetherapy (17), and 3) prolonged therapy inunresolving cases (18-22). It is clear thatthere is no evidence to support the use ofcorticosteroids for the prevention of ARDSin all patients at high risk. In addition, al-though there is intense inflammatory activityin the early stages of ARDS, early cortico-steroid therapy similarly does not appear tobe justified. The effects of corticosteroidshave been investigated in unresolving ARDS.The failure of the beneficial effect of cortico-steroids in the early phase of ARDS may bedue to the population studied, in particular tothe fact that those studies were multicentric

Page 3: Corticosteroids in acute respiratory distress syndrome · Corticosteroids in acute respiratory distress syndrome Laboratórios de 1Investigação Pulmonar, and 2Fisiologia da Respiração,

149

Braz J Med Biol Res 38(2) 2005

Corticosteroids in ARDS

ones, highly heterogeneous in term of casemix and of patient management. In addition,negative effects due to the profound immu-nodepression or other side effects inducedby high doses of steroids could counterbal-ance positive effects, so that the overalleffect could be neutral or even deleterious.Corticosteroid therapy would be ineffectiveif many of the patients who were consideredto have ARDS based on clinical definitionsdid not develop activation of inflammatorycascades in their lungs.

This article reviews the mechanisms ofaction of corticosteroids and the results ofexperimental and clinical studies for the useof corticosteroids in ARDS.

Mechanisms of action

Cortisol is the predominant corticoster-oid secreted by the adrenal cortex in humans.In a healthy, unstressed person, cortisol issecreted according to a diurnal pattern underthe influence of corticotropin released fromthe pituitary gland. Corticotropin secretion,in turn, is under the influence of hypothalam-ic corticotropin-releasing hormone, and bothhormones are subject to negative feedbackcontrol by cortisol itself. Circulating cortisolis bound to corticosteroid-binding globulin,with less than 10% in the free, bioavailableform. With severe infection, trauma, burns,illness, or surgery, there is an increase incortisol production that is proportional to theseverity of the illness. Additionally, diurnalvariation in cortisol secretion is lost. Theseeffects are due to increased production ofcorticotropin-releasing hormone and corti-cotropin and a reduction in negative feed-back from cortisol. Stimulation of the hypo-thalamic-pituitary-adrenal axis is caused byelevated levels of circulating cytokines, amongother factors (23).

Corticosteroids modulate the host de-fense response at virtually all levels, protect-ing the host from immune system overreac-tion. Corticosteroids are mainly transported

in the blood complexed to transcortin (corti-costeroid-binding globulin) and albumin, al-though a small portion is in a free, metaboli-cally active state. The free corticosteroidmolecules cross the plasma membrane intothe cytoplasm, where they bind to a specificreceptor, the glucocorticoid receptor (GR).The GR is located in the cytoplasm of nearlyall human cells (14). After hormone binding,the GR complex migrates to the cell nucleusand inhibits inflammatory gene transcription,including nuclear factor (NF)-κB and activa-tor protein-1, which are activated by extra-cellular inflammatory signals received by cellsurface receptors.

When not bound to its ligand, GR issequestered in the cytoplasm as an inactivecomplex with two molecules of heat shockprotein (HSP-90), and other cytosolic pro-teins. Upon binding glucocorticoids, GRundergoes conformational changes whichallow it to dissociate from HSP-90 mol-ecules. The hormone-bound GR translo-cates to the nucleus, where it transientlyassociates with another HSP, HSP-56, andlater dissociates from it and binds as a dimerto conserved palindromic DNA sequencesnamed glucocorticoid response elements(GRE; Figure 1) (24).

NF-κB is the central transcription factorthat drives the inflammatory response toinsults. It is found in essentially all cell typesand is involved in activation of an exception-ally large number of target genes. NF-κBactivation is an essential step in the experi-mental development of neutrophilic lung in-flammation (25).

NF-κB consists of two subunits arrangedas homodimers or heterodimers. The mostcommon form of activated NF-κB consistsof a p65 and p50 heterodimer. Under basalconditions NF-κB is retained in the cyto-plasm in an inactive state by a related inhib-itory protein known as IκBα. Currently themost commonly accepted mechanism lead-ing to the activation of NF-κB involves theactivation of the recently described IκB kinase,

Page 4: Corticosteroids in acute respiratory distress syndrome · Corticosteroids in acute respiratory distress syndrome Laboratórios de 1Investigação Pulmonar, and 2Fisiologia da Respiração,

150

Braz J Med Biol Res 38(2) 2005

A.B.S. Fernandes et al.

which rapidly phosphorylates IκBα in re-sponse to various pro-inflammatory signalssuch as endotoxin, tumor necrosis factor(TNF)-α, IL-1ß, oxidants, bacteria, viruses,and phorbol esters. The liberated NF-κB thentranslocates into the nucleus and binds topromoter regions of target genes to initiatethe transcription of multiple cytokines in-cluding TNF-α, and the interleukins (IL)-1ß,IL-2, IL-6, chemokines such as IL-8, celladhesion molecules, growth factors, inter-feron, receptors involved in immune recog-nition, proteins involved in antigen presenta-tion, receptors required for neutrophil adhe-sion and migration, and inflammation-asso-ciated enzymes (25). Products of the genesthat are stimulated by NF-κB activate thistranscription factor. Thus, TNF-α and IL-1ß both activate and are activated by NF-κBby forming a positive regulatory loop thatamplifies and perpetuates the inflammation.

Activated GRs mediate transcriptionalinterference via the following mechanisms:a) by physically interacting with the p65subunit and forming an inactive GRα-NF-κBcomplex, b) by inducing the transcription ofthe inhibitory protein IκBα gene, which traps

NF-κB in inactive cytoplasmic complexescatabolized by the ubiquitin-proteosoma path-way, c) by blocking degradation of IκBα viaenhanced synthesis of IL-10, d) by impairingTNF-α-induced degradation of IκBα, and e)by competing for limited amounts of GR co-activators (26,27). GRs may also interactdirectly with protein transcription factors inthe cytoplasm and nucleus and thereby influ-ence the synthesis of certain proteins inde-pendently of an interaction with DNA in thecell nucleus.

The inhibitory effects of corticosteroidson cytokine synthesis are particularly impor-tant. Corticosteroids inhibit NF-κB and con-sequently the expression of the NF-κB-de-pendent proinflammatory gene (27). Thus,they inhibit the transcription of several cyto-kines that are relevant to ARDS, includingIL-1, IL-3, IL-4, IL-5, IL-6, IL-8, TNF-α,and granulocyte-macrophage colony-stimu-lating factor. Corticosteroids also have aninhibitory effect on fibrogenesis (18), andact on the antagonist of IL-1 receptor and onthe anti-inflammatory cytokines IL-4, IL-10,and IL-13 (28) with synergy to control thehost defense response. Glucocorticoidsstimulate apoptosis of T-cells, eosinophils,and monocytes, inhibit neutrophil activation,and are important in maintaining endothelialintegrity and vascular permeability (29).

The response of a single cell exposed tocorticosteroids is the result of the interplay ofthe following factors: 1) the concentration ofthe free hormone; 2) the relative potency ofthe hormone (influenced by biological activ-ity, affinity for the GR in the nucleus), and 3)the ability of the cell to receive and transducethe hormonal signal (30). The endogenousglucocorticoids are not always effective insuppressing life-threatening systemic inflam-mation, even though the degree of cortisole-mia frequently correlates with severity ofillness and mortality rate (25). Failure tosuppress inflammation could be due to tissueresistance to corticosteroids, and inadequacyof the level and duration of endogenous

Pro-inflammatory cytokinesAdhesion moleculesChemokinesIκBα

Chemokines,cytokines, endotoxin,oxidants,ischemia/reperfusion

Cytoplasm

Kinases

NF-κBIκBα

NF-κB

GC

GC GR

Cytoplasm

Pro-inflammatory cytokines

Adhesion molecules

Chemokines

GC

GRE

Gene

Gene

Nucleus

Nucleus

P

κB binding site

Figure 1. Mechanisms of action of corticosteroids. Glucocorticoids activate the cytoplas-mic glucocorticoid receptor, which translocates to the nucleus and serves as an importantregulator of the proinflammatory process that is initiated through NF-κB signaling. NF-κB =nuclear factor-κB; IκBα = inhibitory protein of NF-κB; GC = glucocorticoid; GR = glucocor-ticoid receptor; GRE = glucocorticoid response elements.

Page 5: Corticosteroids in acute respiratory distress syndrome · Corticosteroids in acute respiratory distress syndrome Laboratórios de 1Investigação Pulmonar, and 2Fisiologia da Respiração,

151

Braz J Med Biol Res 38(2) 2005

Corticosteroids in ARDS

glucocorticoid elevation (31). The increasedblood cortisol levels in patients with sepsismay reflect a block of steroidal activity ortransport as a consequence of the infection.In this situation, a small increase in bloodlevels with a low dose of exogenous cortico-steroids was believed to be sufficient tofacilitate the passage of steroids into hostcells. In vitro studies have shown that cyto-kines may induce resistance to glucocorti-coids by reducing GR binding affinity tocortisol and/or GREs (32). Since it is thereceptor that ultimately controls the expres-sion of pertinent genes, anything that affectsits concentration or binding affinity, translo-cation to the nucleus, and/or final transacti-vating or transrepressing conformation canalso influence the response of the cells tocorticosteroids. If one of these factors isdefective, the system does not respond prop-erly.

The various corticosteroid preparationsavailable for systemic use differ in theirrelative anti-inflammatory potency, potentialfor sodium retention, and plasma and bio-logic half-lives. In general, short-acting prepa-rations such as prednisone and prednisoloneare preferable to longer-acting agents suchas dexamethasone because tapering to analternate-day schedule cannot be accom-plished with drugs with prolonged (i.e., >24h) biologic half-lives. In addition, hydrocor-tisone and cortisone are rarely used to treatinflammatory and immunologically mediateddiseases because of the considerable miner-alocorticoid activity that accompanies theiruse. Methylprednisolone may be better con-centrated in the lungs than prednisolone be-cause it has a larger volume of distribution,longer mean residence time, and greaterretention in the epithelial lining fluid of thealveoli (14).

Animal research

The results of animal studies have greatlyincreased our knowledge about lung injury,

but considerable caution is necessary whenextrapolating from experimental models tohumans. In addition, results obtained fromcorticosteroid therapy in animal models withARDS have been controversial.

A wide variety of animal models havebeen developed to explore the effects ofcorticosteroids on ARDS. Both direct andindirect insults to the alveolar-capillary mem-brane have been used in various animal spe-cies to model ARDS (33). Several studiesusing animal models of sepsis and lung injuryhave shown that corticosteroids decreasemorbidity and mortality if given simulta-neously with or before the experimental in-sult (21,34). Pretreatment with corticoste-roids was also shown to increase the survivalrate in acute lung injury induced by aspirationpneumonitis and oleic acid-induced pulmo-nary edema (35,36). In a rat model of butyl-ated-hydroxytoluene-induced ARDS, the tim-ing of corticosteroid administration had strik-ingly different effects. Early administrationof steroids resulted in increased collagendeposition, increased lung damage, and inhi-bition of type II pneumocyte proliferation.Late administration, on the other hand, pre-vented excessive collagen deposition (33).

Systemic methylprednisolone or dexa-methasone at doses of 20 or 4 mg kg-1

day-1, respectively, improved pulmonary in-flammation and mechanics in animals withacute lung injury (36,37). On this basis,dexamethasone (a single dose of 2.5 mg/kg)significantly reduced inflammatory cell pro-tein content in bronchoalveolar lavage fluid(BALF), and improved lung compliance 24 hafter injury induced by oleic acid (38).

Kuwabara and co-workers (39), how-ever, showed that prophylactic treatmentwith methylprednisolone (60 mg kg-1 h-1 30min before oleic acid infusion, followed bycontinuous infusion until the end of theexperiments) did not attenuate oleic acid-induced acute lung injury or the increasedlevel of phospholipase A2 activity, leukotri-ene B4 and thromboxane B2 in the BALF, but

Page 6: Corticosteroids in acute respiratory distress syndrome · Corticosteroids in acute respiratory distress syndrome Laboratórios de 1Investigação Pulmonar, and 2Fisiologia da Respiração,

152

Braz J Med Biol Res 38(2) 2005

A.B.S. Fernandes et al.

greatly reduced IL-8 (39).Corticosteroid at a dose of 10 mg/kg,

starting on day 5 after the infection withreovirus 1/L, and given daily until the end ofthe time course of the disease, also did notattenuate the infiltration of inflammatory leu-kocytes, did not suppress key cytokine/chemokine expression, and did not inhibit thedevelopment of fibrotic changes in the lungs(40).

Rocco and colleagues analyzed the ef-fects of corticosteroids in the early phase ofparaquat-induced acute lung injury (41). Theyobserved that corticosteroids acted differ-ently depending on the degree of acute lunginjury, leading to a complete maintenance ofnormal tissue mechanics and collagen con-tent in a mild lesion, whereas they minimizedthe changes in tissue impedance and extra-cellular matrix components in a severe le-sion. In addition, the early beneficial effectsof corticosteroids on extracellular matrixremained unchanged 30 days after paraquat-induced acute lung injury.

Clinical trials

Early acute respiratory distress syndrome

Based on results of animal research, me-thylprednisolone was subjected to clinicaltrials as adjunctive therapy for patients withARDS (Table 1). Weigelt and colleagues (15)performed a prospective, double-blind, ran-domized study of early corticosteroid thera-py on acutely ill, mechanically ventilatedpatients felt to be at high risk for ARDS. Ofthe 81 patients, 39 received intravenousmethylprednisolone (30 mg/kg every 6 h for48 h) and 42 received placebo. ARDS devel-oped in significantly more methylpredniso-lone-treated patients (25/39 (51%) and pla-cebo, 14/42 (33%)), but no significant dif-ferences in mortality were observed betweenthese two groups.

At least three subsequent randomizedcontrolled trials failed to show any protectiveeffects of high dose corticosteroids in pa-tients with sepsis at risk for ARDS (16,17,42).

Table 1. Clinical trials of glucocorticoids for the prevention of acute respiratory distress syndrome or for thetreatment of the early phase of the disease.

Reference Population Dose Results

Schonfeld et al. Fat embolism MP 7.5 mg/kg every MP decreases the risk of(44) syndrome 6 h x 12 doses respiratory failure

Weigelt et al. (15) Septic shock MP 30 mg/kg every Greater progression to ARDS with6 h x 2 days no change in mortality

Luce et al. (42) Septic shock MP 30 mg/kg every Similar progression to ARDS and6 h x 4 days overall mortality rate for MP and

placebo groups

Bernard et al. (17) Established ARDS MP 30 mg/kg every Similar mortality rate for MP and6 h x 4 days placebo groups

Bone et al. (16) Septic syndrome MP 30 mg/kg every MP impedes the reversal of ARDS6 h x 4 days and increases the mortality rate

Bozzette et al. (45) Pneumocystis carinii Prednisone 40 mg every Prednisone reduces the risks ofpneumonia 12 h x 2 doses respiratory failure and death

Gagnon et al. (46) Pneumocystis carinii MP 40 mg every MP improves survival and decreasespneumonia 6 h x 7 days the occurrence of respiratory failure

Some trials have reported that short-course therapy with high-dose glucocorticoids was ineffective, andsometimes harmful. Other trials have shown that corticosteroids given prophylactically decrease the risk ofrespiratory failure in patients with ARDS or at risk for developing the syndrome. ARDS = acute respiratorydistress syndrome; MP = methylprednisolone.

Page 7: Corticosteroids in acute respiratory distress syndrome · Corticosteroids in acute respiratory distress syndrome Laboratórios de 1Investigação Pulmonar, and 2Fisiologia da Respiração,

153

Braz J Med Biol Res 38(2) 2005

Corticosteroids in ARDS

Bone and colleagues (16) conducted a mul-ticenter, double-blind, randomized controlledtrial to determine whether corticosteroid thera-py could prevent the development of ARDSin high-risk patients. The patients with sepsisreceived methylprednisolone (30 mg/kg) orplacebo infusions every 6 h for a total of fourdoses. Treatment was initiated within 2 h ofthe onset of sepsis. There was an increasedincidence of ARDS in the steroid group. Inthose patients developing ARDS, 14-daymortality was significantly higher in the ste-roid group than in the placebo group. As inthe studies of Bone and co-workers (16), theresults of a clinical trial conducted by Luceand colleagues (42) failed to demonstrate anybenefit of corticosteroid on ARDS develop-ment or subsequent death. These investiga-tors conducted a prospective, double-blind,randomized trial evaluating the efficacy ofsteroid in preventing ARDS in patients withseptic shock. Patients received four doses ofmethylprednisolone (30 mg/kg every 6 h) orplacebo. In the steroid group, 13 of 38patients (34%) developed ARDS comparedwith 14 of 37 (38%) in the placebo group.Mortality was 58% in the steroid groupversus 54% in the placebo group. Thus, it isclear that there is no evidence to support theuse of corticosteroids for the prevention ofARDS in all patients at high risk.

In 1987, Bernard and colleagues (17)performed a double-blind, randomized con-trolled trial to determine the effect of high-dose, short-course corticosteroid on mortal-ity and several physiological variables in earlyARDS. They studied adult patients who metARDS criteria (2) and excluded those withrecent corticosteroid use, evidence of activeinfection, another indication for corticoste-roids, extensive burns, or pregnancy. Theyrandomized 99 adult patients within 30 h ofthe onset of ARDS to receive high-dosemethylprednisolone or placebo. They ob-served that corticosteroid had no effect onoxygenation, hemodynamics, ARDS rever-sal, or mortality. In addition, there was a

trend toward increased infections in the cor-ticosteroid group, as well as significantlytreatment-related hyperglycemia. There isno evidence of benefit from a short-courseof high-dose therapy early in the course ofARDS.

It is not surprising that corticosteroid willnot be effective as ARDS therapy. Thisfailure could be due to the population understudy, in particular to the fact that thosestudies were multicentric ones, highly het-erogeneous in terms of case mix and patientmanagement. The effect of the corticoste-roids could indeed be very different accord-ing to the dosage used. In addition, negativeeffects due to profound immunodepressionor other side effects induced by high doses ofsteroids could counterbalance positive ef-fects, so that the overall effect could beneutral or even deleterious. While such dis-appointing results may be due to the “wrong”mediator being targeted, another possibilityis the heterogeneity of pulmonary processesin patients included in clinical trials. In addi-tion, corticosteroid therapy may be ineffec-tive if many of the patients that were consid-ered to have ARDS based on clinical defini-tions do not have activation of inflammatorycascades in their lungs. Donnelly and Bucala(43) presented other evidence that may ex-plain why corticosteroids fail in early ARDS(43). Macrophage inhibitor factor has beenshown to override corticosteroid-mediatedinhibition of cytokine secretion. Interest-ingly, macrophage inhibitor factor is presentin the BALF of ARDS patients and enhancesboth TNF-α and IL-8 secretion by alveolarmacrophages from ARDS patients. Thus,macrophage inhibitor factor may act as amediator that promotes and sustains thepulmonary inflammatory response in ARDS,and thus may explain, in part, why glucocor-ticoids have been ineffective in the acutephase of ARDS.

Corticosteroids have been shown to re-duce mortality in two groups of patients withARDS or at risk for developing the syn-

Page 8: Corticosteroids in acute respiratory distress syndrome · Corticosteroids in acute respiratory distress syndrome Laboratórios de 1Investigação Pulmonar, and 2Fisiologia da Respiração,

154

Braz J Med Biol Res 38(2) 2005

A.B.S. Fernandes et al.

drome. The first includes patients at risk forfat embolism syndrome, in whom cortico-steroids given prophylactically have beenshown to decrease the risk of respiratoryfailure. Schonfeld and co-workers (44) ran-domly assigned 64 consecutive patients whohad one or more lower-extremity, long-bonefractures to receive intravenous placebo ormethylprednisolone (7.5 mg/kg every 6 h for12 doses, total dose of 90 mg/kg). One of thepatients who received placebo died, whereasnone of the patients who received cortico-steroid died. Nine of the patients who re-ceived placebo developed fat embolism syn-drome, which was associated with alveolarinfiltrates in three of them, whereas none ofthe patients who received methylpredniso-lone developed fat embolism syndrome withor without infiltration. These data supportthe value of high-dose corticosteroids inpatients at risk for the fat embolism syn-drome.

The second group of patients with ARDSor at risk for developing the syndrome, whobenefit from corticosteroids, are those withAIDS and Pneumocystis carinii pneumonia.Two randomized trials have assessed theefficacy of adjunctive therapy with cortico-steroids in patients with AIDS and pneumo-nia (45,46). Patients in these studies received

placebo or a combination of intravenous andoral corticosteroids when pneumonia initiallywas treated with antibiotics. The active drugswere associated with diminished respiratorydeterioration, improved survival, or both, inall three studies.

Late acute respiratory distress syndrome

In contrast to early ARDS, there is evi-dence that corticosteroids may be beneficialin the fibroproliferative phase, or late phaseof ARDS (Table 2). Ashbaugh and Maier(47) described 10 patients who did not re-spond to conventional treatment. These pa-tients underwent open-lung biopsies, andhistological examination showed that all hadcellular proliferation, obliteration of alveoli,and fibrosis without infection. Patients weretreated with intravenous methylprednisolone(125 mg every 6 h) beginning 6 to 22 daysafter onset of ARDS, followed by oral pred-nisolone tapered over 3 to 6 weeks. Eightpatients recovered and 2 died of sepsis.Hooper and Kearl (48) studied the effects ofcorticosteroids in 10 patients who survivedthe early phase of ARDS. ARDS was present3 to 40 days before steroid therapy. Theinitial dose of methylprednisolone (125 to250 mg every 6 h) was based on the severity

Table 2. Clinical trials of glucocorticoids in the late phase of acute respiratory distress syndrome.

Reference Population Dose Results

Ashbaugh and Maier (47) Idiopathic pulmonary MP 125 mg every 6 h MP increases survivalfibrosis

Hooper and Kearl (48) Established ARDS MP 125 to 250 mg MP improves theevery 6 h respiratory parameters

Meduri et al. (50) Established ARDS MP 200 mg followed MP increases the PaO2/FiO2

by 2 to 3 mg kg-1 day-1 ratio and decreases LISevery 6 h until extubation

Meduri et al. (52) Established ARDS MP 2 mg/kg followed by Changes in PaO2/FiO2 ratio,2 mg kg-1 day-1 from day mean pulmonary artery1 to 14 every 6 h pressure, and MODS score

These clinical trials suggest a beneficial effect of prolonged use of glucocorticoids in late ARDS. ARDS = acuterespiratory distress syndrome; MP = methylprednisolone; LIS = lung injury score; MODS = multiple-organdysfunction syndrome.

Page 9: Corticosteroids in acute respiratory distress syndrome · Corticosteroids in acute respiratory distress syndrome Laboratórios de 1Investigação Pulmonar, and 2Fisiologia da Respiração,

155

Braz J Med Biol Res 38(2) 2005

Corticosteroids in ARDS

of respiratory involvement, and was main-tained for 72 to 96 h. All patients showedimprovement in respiratory parameters. Over-all survival was 81%. Biffl and co-workers(49) used methylprednisolone (1 to 2 mg/kgevery 6 h) in 6 patients with refractory lateARDS. Steroids were instituted after 16 daysof mechanical ventilatory support. By day 7of steroid therapy, there was a clinicallysignificant improvement in the ratio of arte-rial oxygen pressure to fraction of inspiredoxygen (PaO2/FiO2) and a decrease in lunginjury score. Overall survival was 83% andthe mean duration of corticosteroid therapywas 21.3 days.

A large uncontrolled series of patientswith late ARDS treated with corticosteroidswas studied by Meduri and colleagues (18,50).They initially described 8 patients followedby an additional 17 patients. All patients hadprogressively worsening respiratory failure7 days or more after the onset of ARDS.Patients were treated with methylpredniso-lone (200 mg in bolus followed by 2 to 3 mgkg-1 day-1 in divided doses every 6 h untilextubation, after which the steroid was ta-pered slowly). By day 7 of treatment, thePaO2/FiO2 ratio increased and lung injuryscore decreased (50). Three patterns ofresponse were noted: a) rapid respondersshowed improvement by day 7, b) delayedresponders showed improvement by day 14,and c) nonresponders exhibited no improve-ment by day 14. Overall survival rate was72%. Intensive care unit survival was 87%among rapid responders, 83% among de-layed responders, and 25% among nonre-sponders. The average duration of cortico-steroid treatment was 36 days. Pneumoniadeveloped in 38% of responders and 75% ofnonresponders.

According to Meduri et al. (51), a com-parison of physiological data with cytokinelevels over time may explain the differencesbetween responders and nonresponders. Therate of improvement in lung function duringtreatment may be related to the magnitude of

the inflammatory response at initiation ofcorticosteroid therapy. After initiation of treat-ment, nonresponders had significantly worselung function. Although nonresponders hada cytokine profile at initiation of treatmentsimilar to the one of delayed responders, thebehavior of IL-6 at the onset of ARDS wassignificantly different. Third, during cortico-steroid therapy, improvements occurred ingas exchange, lung mechanics, and indicesof increased endothelial permeability parallelto reductions in both plasma and bronchoal-veolar lavage TNF-α, IL-1ß, IL-6, and IL-8levels.

In 1998, Meduri and co-workers (52)speculated that the early removal of highdose corticosteroid treatment in the previousrandomized trials, which used short-termcorticosteroid treatment, might have reversedany early beneficial effect of treatment oroverturned the ability to detect a beneficialeffect. In order to determine the effects ofprolonged steroid therapy on lung functionand mortality in non-resolving ARDS, theyconducted a randomized, double-blind, pla-cebo-controlled trial in four medical inten-sive care units. Twenty-four patients whofulfilled criteria for the diagnosis of ARDS(2) had been mechanically ventilated for 7days and had no evidence of untreated infec-tion. Sixteen patients were randomized totreatment with methylprednisolone while eightpatients received placebo. The methylpred-nisolone dose was initially 2 mg/kg adminis-tered intravenously followed by 2 mg kg-1

day-1 from day 1 to day 14 and then byprogressively lower doses until day 32 of thestudy. One fourth of the total daily methyl-prednisolone dose was given every 6 h.Significant changes were observed in PaO2/FiO2 ratio, lung injury score, mean pulmo-nary artery pressure, and multiple-organdysfunction syndrome score in the cortico-steroid-treated group versus the placebogroup. Intensive care unit survival was 100%in the steroid group versus 37% in the pla-cebo group, and overall survival was 87

Page 10: Corticosteroids in acute respiratory distress syndrome · Corticosteroids in acute respiratory distress syndrome Laboratórios de 1Investigação Pulmonar, and 2Fisiologia da Respiração,

156

Braz J Med Biol Res 38(2) 2005

A.B.S. Fernandes et al.

versus 37%, respectively.Meduri (53) hypothesized that if endoge-

nous glucocorticoid inadequacy and/or pe-ripheral tissue resistance were important patho-physiologic factors in a deregulated, protractedsystemic inflammatory response in ARDS,then prolonged glucocorticoid therapy mightbe useful, not as an anti-inflammatory treat-ment per se, but as hormonal supplementationnecessary to compensate for the host’s inabil-ity to produce appropriately elevated amountsof cortisol in relation to the degree of peripheralglucocorticoid resistance.

The host defense response to insults issimilar regardless of the tissue involved andresults from an interactive network of spe-cialized and interconnected pathways thatact in synergy to increase the host’s chanceof survival. Meduri (54) reviewed the hostdefense response in the progression of ARDSand how this response may be affected bycorticosteroid therapy. Continued elevatedproduction of host defense response media-tors, such as TNF-α, IL-1ß, and IL-6 pre-vents effective restoration of lung anatomyand function by sustaining inflammation,coagulation and fibro-proliferation resultingin fibrosis (55,56). The degree of initial hostdefense response may determine the progressof ARDS. In another study by Meduri andco-workers (51), the effects of corticoster-oid treatment on plasma and BALF cytokinelevels in 9 patients with late ARDS werecompared with 12 previous non-survivorsfrom ARDS, who had undergone cytokineconcentration measurements. Baseline plasmaand BALF cytokine levels in the corticoster-oid-treated patients were similar to those ofthe other group. The surviving patients treatedwith corticosteroids were found to havesignificant reductions in plasma and BALFTNF-α, IL-1ß, IL-6, and IL-8 concentra-tions. The decreases in various cytokinelevels were seen only after 5 to 14 days ofsteroid administration. On the other hand,ARDS non-survivors have been reported tohave higher initial and persistent elevation of

plasma and BALF cytokines compared withsurvivors. These findings mandate a reap-praisal of the role of anti-inflammatory treat-ment of ARDS (51).

Ineffective lung repair in patients withunresolving ARDS is accompanied by pro-gressive fibro-proliferation, inability to im-prove lung injury score, progressive multipleorgan dysfunction syndrome, and an unfa-vorable outcome. Meduri and co-workers(57) also examined the effects of corticoste-roids on plasma and BALF levels of procol-lagen aminoterminal propeptide type I (PINP)and type III (PIIINP) in patients with non-resolving ARDS. PINP and PIIINP are se-creted by fibroblasts and reflect collagensynthesis at the site of disease. Previousstudies reported that non-survivors of ARDShad persistent elevations of plasma and BALFPIIINP levels (58,59). Meduri and co-work-ers (57) found elevated plasma levels ofPINP and PIIINP in their patients at the timeof study enrollment and observed that theconcentrations of PINP and PIIINP increasedover time in non-survivors, as opposed tosurvivors whose levels did not change sig-nificantly. Bronchoalveolar lavage concen-trations of PINP and PIIINP were also notedto be higher in non-survivors compared withsurvivors, although the differences were notstatistically significant. In that study 11 pa-tients who had not shown an improvement inlung injury score greater than 1 point wererandomized to receive methylprednisoloneusing the same protocol as in their previousrandomized trial (52), and six patients re-ceived placebo. Patients treated with methyl-prednisolone showed significant decreasesin plasma and BALF PINP and PIIINP levels,whereas no changes in these concentrationswere seen in patients receiving placebo. De-creases in plasma and BALF PINP and PIIINPlevels correlated with improvements in lunginjury score and PaO2/FiO2 ratio.

Recently, Meduri and co-workers (60)investigated whether unresolving ARDS isassociated with systemic inflammation-in-

Page 11: Corticosteroids in acute respiratory distress syndrome · Corticosteroids in acute respiratory distress syndrome Laboratórios de 1Investigação Pulmonar, and 2Fisiologia da Respiração,

157

Braz J Med Biol Res 38(2) 2005

Corticosteroids in ARDS

duced glucocorticoid resistance and whetherprolonged methylprednisolone administrationaccelerates the suppression of systemic in-flammatory indices and normalizes the sen-sitivity of the immune system to glucocorti-coids. Patients enrolled into a randomizedtrial evaluating prolonged methylpredniso-lone administration in unresolving ARDS hadserial plasma samples collected before andafter randomization. Patients treated withcorticosteroid had progressive and sustainedreductions of TNF-α, IL-1ß, IL-6, ACTH,and cortisol concentrations over time. Nor-mal peripheral blood leukocytes exposed toplasma samples collected during methyl-prednisolone treatment also exhibited rapidand progressive significant increases in GR-α-mediated activities and significant reduc-tions in NF-κB binding and transcription ofTNF-α and IL-1ß. These findings providesupport for the presence of endogenousglucocorticoid inadequacy in the control ofinflammation and systemic inflammation-induced peripheral glucocorticoid resistancein ARDS. Prolonged methylprednisolone ad-ministration accelerated the resolution ofboth systemic inflammation and peripheralacquired glucocorticoid resistance in ARDS.

The variable response of ARDS to corti-costeroid treatment could be attributed to theheterogeneous biochemical and molecularmechanisms of the different initial insults.Furthermore, numerous important factorsneed to be taken into account when cortico-steroid is used as a strategy to treat ARDS:the specificity of inhibition, the duration anddegree of inhibition, and the timing of inhibi-tion. Predictors of poor outcome in ARDSare manifestations of a persistent and exag-gerated host defense response. Suppressionof this inflammatory response may lead toresolution of ARDS. Continued elevated pro-duction of host defense response mediators,such as TNF-α, IL-1ß, and IL-6, preventseffective restoration of lung anatomy andfunction by sustaining inflammation, coagu-lation and fibro-proliferation. The degree of

initial host defense response may determinethe progress of ARDS. Spontaneous or treat-ment-induced suppression of the exagger-ated and autonomous pulmonary host de-fense response is associated with improve-ment in lung function and outcome. Thereaf-ter, excessive inflammatory activity in pa-tients with sepsis, septic shock, or ARDSmay induce noncompensated glucocorticoidresistance in target organs, thereby negatingthe beneficial suppressive influence of aninadequately secreted endogenous cortisolon the deregulated host defense response.Treatment with exogenous glucocorticoidsmay be necessary to compensate adequatelyfor the inability of target organs to respond toendogenous cortisol and for the inability ofthe host to produce appropriately elevatedlevels of corticosteroids.

In conclusion, improving the course andoutcome of patients with ARDS presents aconsiderable challenge. An important com-ponent of meeting this challenge is a morecomprehensive understanding of the hetero-geneous pathophysiology of ARDS and thebiological response of the individual patient.By understanding the immune status of agiven patient at a given point in the diseaseprocess, physicians can consider manipulat-ing proinflammatory systems more ratio-nally. Thus, ARDS need to be strongly re-considered as a consistent response to itsdiverse etiology. The great problem remainsthe time when to start the use of a cortico-steroid and, mainly, how to monitor its use.The inflammatory mechanisms are continu-ous and cyclic, sometimes causing deterio-ration or improvement of lung function.Experimental models have shown that treat-ment with a single low-dose of methylpred-nisolone in the early phase of acute lunginjury might help prevent fibroelastogenesis,avoiding the side effects related to prolongedand high doses of steroid. Thus, these exper-imental studies may help the design of aclinical trial focused on diminishing lunginjury with steroids.

Page 12: Corticosteroids in acute respiratory distress syndrome · Corticosteroids in acute respiratory distress syndrome Laboratórios de 1Investigação Pulmonar, and 2Fisiologia da Respiração,

158

Braz J Med Biol Res 38(2) 2005

A.B.S. Fernandes et al.

References

1. Ashbaugh DG, Bigelow DB, Petty TL & Levine BE (1967). Acuterespiratory distress syndrome. Lancet, 2: 319-323.

2. Bernard GR, Artigas A, Bringham KL, Carlet J, Falke K, Hudson L,Lamy M, Legall JR, Morris A & Spragg R (1994). The American-European consensus conference on ARDS: definitions, mechan-isms, relevant outcomes, and clinical trial coordination. AmericanJournal of Respiratory and Critical Care Medicine, 149: 818-824.

3. Kallet RH (2004). Evidence-based management of acute lunginjury and acute respiratory distress syndrome. Respiratory Care,49: 793-809.

4. Crimi E & Slutsky AS (2004). Inflammation and the acute respiratorydistress syndrome. Best Practice & Research. Clinical Anaesthesi-ology, 18: 477-492.

5. Hoelz C, Negri EM, Lichtenfels AJ, Concecao GM, Barbas CS,Saldiva PH & Capelozzi VL (2001). Morphometric differences inpulmonary lesions in primary and secondary ARDS. A preliminarystudy in autopsies. Pathology, Research and Practice, 197: 521-530.

6. Pelosi P, Gamberoni C, Aspesi M, Severgnini P, Chiumello D,Caironi P, Chiaranda L & Gattinoni L (2003). Pulmonary and extra-pulmonary forms of acute respiratory distress syndrome are differ-ent? European Respiratory Journal, 22 (Suppl II): II-48-II-56.

7. Tomashefski Jr JF (2000). Pulmonary pathology of acute respira-tory distress syndrome. Clinics in Chest Medicine, 21: 435-466.

8. Chesnutt AN, Matthay MA, Tibayan FA & Clark JG (1997). Earlydetection of type III procollagen peptide in acute lung injury.Pathogenic and prognostic significance. American Journal of Res-piratory and Critical Care Medicine, 156: 840-845.

9. Pugin J, Verghese G, Widmer MC & Matthay MA (1999). Thealveolar space is the site of intense inflammatory and profibroticreactions in the early phase of acute respiratory distress syndrome.Critical Care Medicine, 27: 304-312.

10. Marshall RP, Bellingan G, Webb S, Puddicombe A, Goldsack N,McAnulty RJ & Laurent GJ (2000). Fibro-proliferation occurs early inthe acute respiratory distress syndrome and impacts on outcome.American Journal of Respiratory and Critical Care Medicine, 162:1783-1788.

11. Luce JM (2002). Corticosteroids in ARDS. An evidence-basedreview. Critical Care Clinics, 18: 79-89.

12. Brower RG, Ware LB, Berthiaume Y & Matthay MA (2001). Treat-ment of ARDS. Chest, 120: 1347-1367.

13. Tasaka S, Hasegawa N & Ishizaka A (2002). Pharmacology of acutelung injury. Pulmonary Pharmacology and Therapeutics, 15: 83-95.

14. Jantz MA & Sahn AS (1999). Corticosteroids in acute respiratoryfailure. American Journal of Respiratory and Critical Care Medicine,160: 1079-1100.

15. Weigelt JA, Norcross JR, Borman KR & Snyder WH (1985). Earlysteroid therapy for respiratory failure. Archives of Surgery, 120:536-540.

16. Bone RC, Fisher Jr CJ, Clemmer TP, Slotman GJ & Metz CA(1987). Early methylprednisolone treatment for septic syndromeand the adult respiratory distress syndrome. Chest, 92: 1032-1036.

17. Bernard GR, Luce JM, Sprung CL, Rinaldo JE, Tate RM, SibbaldWJ, Kariman K, Higgins S, Bradley R & Metz CA (1987). High dosecorticosteroids in patients with the adult respiratory distress syn-drome. New England Journal of Medicine, 317: 1565-1570.

18. Meduri GU, Belenchia JM, Estes RJ, Wunderink RG, el Torky M &Leeper Jr KV (1991). Fibroproliferative phase of ARDS: clinical

findings and effects of corticosteroids. Chest, 100: 943-952.19. Demling RH, Proctor R, Grossman J, Duy N & Starling J (1980).

Comparison of the systemic and pulmonary vascular response toendotoxin with plasma and lung lymph lysosomal enzyme release:effect of steroid pretreatment. Circulatory Shock, 7: 317-331.

20. Brigham KL, Bowers RE & McKeen CR (1981). Methylpredniso-lone prevention of increased lung vascular permeability followingendotoxemia in sheep. Journal of Clinical Investigation, 67: 1103-1110.

21. Thomas CS & Brockman SK (1968). The role of adrenal corticoster-oid therapy in Escherichia coli endotoxin shock. Surgery, Gynecol-ogy and Obstetrics, 126: 61-69.

22. Wertzberger JJ & Peltier LF (1968). Fat embolism: the effect ofcorticosteroids on experimental fat embolism in the rat. Surgery,64: 143-147.

23. Cooper MS & Stewart PM (2003). Corticosteroid insufficiency inacutely ill patients. New England Journal of Medicine, 348: 727-734.

24. Almawi WY & Melemedjian OK (2002). Negative regulation ofnuclear factor-kappaB activation and function by glucocorticoids.Journal of Molecular Endocrinology, 28: 69-78.

25. Meduri GU, Carratu P & Freire AX (2003). Evidence of biologicalefficacy for prolonged glucocorticoid treatment in patients withunresolving ARDS. European Respiratory Journal, 42 (Suppl I): I-57-I-64.

26. Payne DNR & Adcock IM (2001). Molecular mechanisms of cortico-steroid actions. Paediatric Respiratory Review, 2: 145-150.

27. Wissink S, van Heerde EC, vand der Burg B & van der Saag PT(1998). A dual mechanism mediates repression of NF-kappa Bactivity by glucocorticoids. Molecular Endocrinology, 12: 355-363.

28. Hart PH, Whitty GA, Burgess DR, Croatto M & Hamilton JA (1990).Augmentation of glucocorticoid action on human monocytes byinterleukin-4. Lymphokine Research, 9: 147-153.

29. Thompson BT (2003). Glucocorticoids and acute lung injury. Criti-cal Care Medicine, 31: 5253-5257.

30. Bamberger CM, Shulte HM & Chrousos GP (1996). Moleculardeterminants of glucocorticoid function and tissue sensitivity toglucocorticoids. Endocrine Reviews, 17: 245-261.

31. Briegel J, Forst H, Hellinger H & Haller M (1991). Contribution ofcortisol deficiency to septic shock. Lancet, 338: 507-508.

32. Almawi WY, Lipman ML, Stevens AC, Zanker B, Hadro ET & StromTB (1991). Abrogation of glucocorticoid-mediated inhibition of Tcell proliferation by the synergistic action of IL-1, IL-6, and IFN-gamma. Journal of Immunology, 146: 3523-3527.

33. Jones RJ & King RG (1975). The effects of methylprednisolone onoxygenation in experimental hypoxemic respiratory failure. Journalof Trauma, 15: 297-303.

34. White GI, Archer LJ, Beller BK & Hinshaw LB (1978). Increasedsurvival with methylprednisolone treatment in canine endotoxinshock. Journal of Surgical Research, 25: 357-364.

35. Lawson DW, Defalco AJ, Phelps JA, Bradley BE & McClenathan JE(1966). Corticosteroids as treatment for aspiration of gastric con-tents: an experimental study. Surgery, 59: 845-852.

36. Shiue ST & Thrall RS (1991). Effect of corticosteroid therapy on theacute injury and recovery stage of oleic acid induced lung injury inthe rat. Experimental Lung Research, 17: 629-638.

37. Cheney FW, Hung TH & Gronka R (1979). Effects of methylpred-nisolone on experimental pulmonary injury. Annals of Surgery,190: 236-242.

Page 13: Corticosteroids in acute respiratory distress syndrome · Corticosteroids in acute respiratory distress syndrome Laboratórios de 1Investigação Pulmonar, and 2Fisiologia da Respiração,

159

Braz J Med Biol Res 38(2) 2005

Corticosteroids in ARDS

38. Volpe BT, Lin W & Thrall RS (1994). Effect of intratracheal dexa-methasone on oleic acid-induced lung injury in the rat. Chest, 106:583-587.

39. Kuwabara K, Furue S, Tomita Y et al. (2001). Effect of methylpred-nisolone on phospholipase A2 activity and lung surfactant degrada-tion in acute lung injury in rats. European Journal of Pharmacology,433: 209-216.

40. London L, Majeski E, Altman-Hamamdzic S, Enockson C, PaintliaMK, Harley RA & London SD (2002). Respiratory Reovirus 1/Linduction of diffuse alveolar damage: pulmonary fibrosis is notmodulated by corticosteroids in acute respiratory distress syn-drome in rats. Clinical Immunology, 103: 284-295.

41. Rocco PRM, Souza AB, Faffe DS, Passaro CP, Santos FB, NegriEM, Lima JG, Contador RS, Capelozzi VL & Zin WA (2003). Effect ofcorticosteroid on lung parenchyma remodeling at an early phase ofacute lung injury. American Journal of Respiratory and Critical CareMedicine, 168: 677-684.

42. Luce JM, Montgomery AB, Marks JD, Turner J, Metz CA & MurrayJF (1988). Ineffectiveness of high-dose methylprednisolone inpreventing parenchymal lung injury and improving mortality inpatients with septic shock. American Review of Respiratory Dis-ease, 138: 62-68.

43. Donnelly SC & Bucala R (1997). Macrophage migration inhibitoryfactor: a regulator of glucocorticoid activity with a critical role ininflammatory disease. Molecular Medicine Today, 3: 502-507.

44. Schonfeld SA, Ploysongsang Y, DiLisio R, Crissman JD, Miller E,Hammerschmidt DE & Jacob HS (1983). Fat embolism prophylaxiswith corticosteroids: a prospective study in high risk patients.Annals of Internal Medicine, 99: 438-443.

45. Bozzette SA, Sattler FR, Chiu J, Wu AW, Gluckstein D, Kemper C,Bartok A, Niosi J, Abramson I & Coffman J (1990). A controlled trialof early adjunctive treatment with corticosteroids for Pneumocystiscarinii pneumonia in the acquired immunodeficiency syndrome.New England Journal of Medicine, 323: 1451-1457.

46. Gagnon S, Boota AM, Fischl MA, Baier H, Kirksey OW & La Voie L(1990). Corticosteroids as adjunctive therapy for severe Pneumo-cystis carinii pneumonia in the acquired immunodeficiency syn-drome: A double-blind, placebo-controlled trial. New England Jour-nal of Medicine, 323: 1444-1450.

47. Ashbaugh DG & Maier RV (1985). Idiopathic pulmonary fibrosis inadult respiratory distress syndrome. Diagnosis and treatment. Ar-chives of Surgery, 120: 530-535.

48. Hooper RG & Kearl RA (1990). Established ARDS treated with asustained course of adrenocortical steroids. Chest, 97: 138-143.

49. Biffl WL, Moore FA, Moore EE, Haenel JB, McIntyre Jr RC & BurchJM (1995). Are corticosteroids salvage therapy for refractory acuterespiratory distress syndrome? American Journal of Surgery, 170:591-596.

50. Meduri GU, Chinn AJ, Leeper KV, Wunderink RG, Tolley E, Winer-Muram HT, Khare V & Eltorky M (1994). Corticosteroid rescuetreatment of progressive fibro-proliferation in late ARDS: patterns ofresponse and predictors of outcome. Chest, 105: 1516-1527.

51. Meduri GU, Headley S, Tolley E, Shelby M, Stentz F & PostlethwaiteA (1995). Plasma and BAL cytokine response to corticosteroidrescue treatment in late ARDS. Chest, 108: 1315-1325.

52. Meduri GU, Headley S, Golden E, Wunderink RG, Tolley E, Winer-Muram HT, Khare V & Eltorky M (1998). Effect of prolongedmethylprednisolone therapy in unresolving acute respiratory dis-tress syndrome: a randomised controlled trial. Journal of theAmerican Medical Association, 280: 159-165.

53. Meduri GU (1999). A historical review of glucocorticoid treatment insepsis: disease pathophysiology and the design of treatmentinvestigation. Sepsis, 3: 21-28.

54. Meduri GU (1996). The role of the host defense response in theprogression and outcome of ARDS: pathophysiologic correlationsand response to glucocorticoid treatment. European RespiratoryJournal, 9: 2650-2670.

55. Meduri GU, Headly S, Kohler G, Stentz F, Tolley E, Umberger R &Leeper K (1995). Persistent elevation of inflammatory cytokinespredicts a poor outcome in ARDS: plasma IL-1ß and IL-6 areconsistent and efficient predictors of outcome over time. Chest,107: 1062-1073.

56. Meduri GU, Kohler G, Headley S, Tolley E, Stentz F & PostlethwaiteA (1995). Inflammatory cytokines in the BAL of patients withARDS: persistent elevation over time predicts poor outcome.Chest, 108: 1303-1314.

57. Meduri GU, Tolley EA, Chinn A, Stentz F & Postlethwaite A (1998).Procollagen types I and III aminoterminal propeptide levels duringacute respiratory distress syndrome and in response to methyl-prednisolone treatment. American Journal of Respiratory and Criti-cal Care Medicine, 158: 1432-1441.

58. Entzian P, Huckstadt A, Kreipe H & Barth J (1990). Determination ofserum concentrations of type III procollagen peptide in mechani-cally ventilated patients: pronounced augmented concentrations inthe adult respiratory distress syndrome. American Review of Respi-ratory Disease, 142: 1079-1082.

59. Clark JG, Milberg JA, Steinberg KP & Hudson LD (1995). Type IIIprocollagen peptide in the adult respiratory distress syndrome.Annals of Internal Medicine, 122: 17-23.

60. Meduri GU, Tolley EA, Chrousos GP & Stentz F (2002). Prolongedmethylprednisolone treatment suppresses systemic inflammationin patients with unresolving acute respiratory distress syndrome.Evidence for inadequate endogenous glucocorticoid secretion andinflammation-induced immune cell resistance to glucocorticoids.American Journal of Respiratory and Critical Care Medicine, 165:983-991.