ARDS of Early or Late Onset: Does It Make a Difference? FREE TO VIEW

Acute respiratory failure due to acute lung injury (ALI) or ARDS is a common problem in ICU patients. Even though mortality rates may have decreased over time,¹ they remain high, at around 30% to 40%.^2-4 Unfortunately, despite considerable research, no specific treatment has been shown to improve outcomes from ARDS⁵; the only really clear therapeutic indication is that tidal volumes should be kept low.⁶

ARDS is a complex syndrome that can affect many groups of patients and is associated with a number of pulmonary and/or extrapulmonary factors, which can impact a patient’s prognosis.^7,8 Clinical trials evaluating the effects of therapeutic interventions in ARDS are complicated by the heterogeneity of the patient population studied. In addition to the underlying diseases, the lead time may be an important issue for prognosis, and hence for clinical trials assessing the effects of potential therapies. In renal failure, a later onset is associated with a worse outcome.^9,10 Similarly, ICU patients with late-onset septic shock (>24 h after ICU admission) had worse prognoses than patients with early-onset septic shock (<24 h after admission),¹¹ and in a multivariate analysis of patients with shock from the Sepsis Occurrence in Acutely Ill Patients (SOAP) database, late-onset shock was an independent predictor of worse outcome (odds ratio, 2.6; 95% CI, 1.6-4.3, P<.001).¹² However, this pattern may not apply to patients with ARDS. In a single-center study of 178 patients with posttraumatic ARDS, Croce et al¹³ reported no differences in outcomes between patients with early-onset (within 48 h of admission) and those with late-onset (>48 h after admission) ARDS. Similarly, in a single-center study in a mixed medico-surgical ICU, Lobo et al¹⁴ reported no differences in mortality rates in patients with early-onset (within 48 h of admission) compared with those with late-onset (>48 h after admission) acute respiratory failure. Flaatten et al¹⁵ also reported no differences in mortality in patients diagnosed with acute respiratory failure on admission and those developing acute respiratory failure later during their ICU stay. We investigated the differences in outcomes from ALI/ARDS according to the time of onset in a larger, international cohort of ICU patients.

This report is the result of a substudy from the SOAP database, a prospective, multicenter, observational study that was designed to evaluate the epidemiology of sepsis in European countries and was initiated by a working group of the European Society of Intensive Care Medicine. Since this epidemiologic, observational study did not require any deviation from routine medical practice, institutional review board approval was either waived or expedited in participating institutions and informed consent was not required. Full details of recruitment, data collection and management, and definitions have been published previously.¹⁶ Briefly, during a 14-day period from May 1 to May 15, 2002, all patients older than 15 years and newly admitted to one of 198 units participating in the SOAP network in 24 European countries (see Appendix for a list of participating centers) were included in the study. Patients were followed until death, hospital discharge, or for 60 days. Patients admitted for routine postoperative observation who stayed in the ICU for less than 24 h were excluded.

Data were collected prospectively using preprinted case report forms. Data collection on admission included demographic data and comorbidities. Clinical and laboratory data for the Simplified Acute Physiology Score II (SAPS II)¹⁷ were reported as the worst value within 24 h after admission. Microbiologic and clinical infections were reported daily, as well as the antibiotics administered. A daily evaluation of organ function according to the Sequential Organ Failure Assessment (SOFA) score¹⁸ was performed, with the most abnormal value for each of the six organ systems (respiratory, renal, cardiovascular, hepatic, coagulation, and neurologic) collected on admission and every 24 h thereafter. In mechanically ventilated patients, values of tidal volume, positive end-expiratory pressure, plateau pressure, and Fio₂ were recorded at the same time each day for every 24-h period; the mode of mechanical ventilation was not recorded. The predicted body weight of male patients was calculated as equal to 50+0.91 (height in centimeters, 152.4), and that of female patients as equal to 45.5+0.91 (height in centimeters, 152.4).^6,19,20

Patients were defined as having ALI or ARDS if the arterial oxygen pressure to inspiratory oxygen fraction ratio (Pao₂/Fio₂) was less than 300 for ALI and less than 200 for ARDS²¹ and all of the following were present: bilateral infiltrates on the chest radiograph, no clinical evidence of heart failure, no chronic pulmonary disorders, and mechanical ventilation. Early-onset ALI/ARDS was defined as ALI/ARDS occurring within 48 h of ICU admission; late-onset ALI/ARDS was defined as ALI/ARDS occurring more than 48 h after ICU admission. Sepsis was defined according to standard definitions.²²

Data were analyzed using SPSS 13.0 for Windows (SPSS Inc.; Chicago, IL). Descriptive statistics were computed for all study variables. A Kolmogorov-Smirnov test was used, and histograms and normal-quantile plots were examined to verify the normality of distribution of continuous variables. Discrete variables were expressed as counts (percentage) and continuous variables as means ± SD or median (25th-75th percentiles). Difference testing between groups was performed using the two-tailed t test, Mann-Whitney U test, Χ² test, and Fisher’s exact test, as appropriate. Survival curves were constructed and adjusted for SAPS II score and for initial SOFA score and age. We performed a multivariate Cox proportional hazard model, with time to in-hospital death as the dependent factor. Variables included in the Cox regression analysis were age, gender, comorbid diseases, SAPS II and SOFA scores on admission, type of admission (medical or surgical), source of admission, mean Fio₂, presence of infection, and presence of sepsis. Variables were introduced in the model if they were significantly associated with a higher risk of in-hospital death on a univariate basis at a P<.2. Colinearity between variables was excluded prior to modeling. A stepwise approach was used, and presence in the early- or late-onset ALI/ARDS group was forced as the last step in the model. All statistics were two-tailed, and a P<.05 was considered to be significant.

Of the 3,147 patients included in the SOAP study, 393 (12.5%) had a diagnosis of ALI/ARDS; 334 of these (85%) met the criteria for ARDS, and 59 had only ALI (15.0%). Of the 393 patients with ALI/ARDS, 254 (64.6%) had early-onset ALI/ARDS (31, early-onset ALI; 223, early-onset ARDS) and 139 (35.45%) late-onset (28, late-onset ALI; 111, late-onset ARDS) (Table 1). As expected, patients with early-onset ALI/ARDS had a higher SAPS II score on admission and a higher initial SOFA score (see Table 1). Patients with late-onset ALI/ARDS had longer ICU and hospital lengths of stay than patients with early-onset ALI/ARDS.

More patients with early ALI/ARDS were infected on admission than patients with late-onset ALI/ARDS (52% vs 38.8%, P<.001), and more patients with late-onset ALI/ARDS developed infection during their ICU stay (34.5% vs 8.7%, P<.001) (Table 2). However, there were no differences between groups in the incidence of septic shock during the ICU stay.

ICU and hospital mortality rates were, if anything, lower in late-onset ALI/ARDS than in early-onset ALI/ARDS, but these differences were not statistically significant. Figure 1 shows the survival rates over time in all patients and adjusted for SAPS II and initial SOFA scores. In a multivariate analysis, SOFA scores (without respiratory component) on admission and age, but not time of onset of ALI/ARDS, were associated with an increased hazard of in-hospital mortality (Table 3).

Our results show that late-onset ALI/ARDS does not have a worse prognosis than early-onset ALI/ARDS. This is in contrast with the patterns seen in renal failure^9,10 and shock,^11,12 possibly because ARDS is intimately related to sepsis. In the SOAP database, 73% of patients with ALI/ARDS had sepsis, compared with 46% of patients with renal failure and 60% of patients with shock (of any cause). In the SOAP database as a whole, there were no differences in mortality rates in patients with sepsis present on admission, those who developed sepsis within 48 h after admission, and those who developed sepsis more than 48 h after admission (27% vs 20% vs 28%, P = .562).¹⁶

This leads us to a basic question: What is the primary cause of death in patients with ALI/ARDS? In 1985, Montgomery et al²³ reported that less than 20% of patients with ARDS died of refractory hypoxemia. More recent studies made similar observations,^24-26 supporting the concept that ARDS is a syndrome that is rarely, by itself, a cause of death. Hence, it is less likely that the time of onset of ARDS would influence outcome than the time of onset of shock or renal failure, which are important causes of death per se.

The fact that ARDS is itself not a primary cause of death but rather accompanies other disease processes also explains, at least in part, the difficulty that has been encountered in developing specific therapies for ARDS. Therapies aimed at improving oxygenation, such as the use of nitric oxide,²⁷ surfactant,²⁸ and partial liquid ventilation,²⁹ have not been shown to improve outcomes. The only approach that has been shown to have any effect is avoidance of the additional insult caused by excessive tidal volumes.⁶

The reasons for the longer ICU and hospital lengths of stay in patients with late-onset ALI/ARDS cannot be determined from the present data, but are likely to be multifactorial and possibly in part linked to the increased rate of hospital-acquired infections in these patients.

The advantage of our study is that it involves a large database from multiple centers with systematic collection of data. One limitation is that ALI/ARDS was diagnosed from specific criteria, rather than simply by asking doctors whether the patient had ALI/ARDS. However, doctors may underdiagnose ALI/ARDS when asked to identify patients clinically,³⁰ and definitions were the same for the late- and early-onset groups, so this should not have influenced our results. The development of ARDS may also be insidious, so that the precise time of onset is not clearly defined and could vary. A second limitation is that a low-tidal-volume strategy was not applied uniformly in our patients, so that ventilator-induced lung injury may have contributed to a poor outcome in some patients. However, this would have involved patients with early- and late-onset ARDS, so it should not have influenced our conclusions. Finally, another limitation is that we are unable to identify the precise cause of death in these patients.

In conclusion, we were unable to demonstrate any difference in patient survival between early- and late-onset ARDS. These findings are reassuring for clinical trial development in that the lead-time bias does not seem to be important for ARDS, and we can continue to enroll patients at any time during the course of their ICU stay.

Author contributions:Dr Vincent: conceived the initial SOAP study, participated in the design and coordination of the SOAP study, drafted the manuscript, and read and approved the final manuscript.

Dr Sakr: participated in the design and coordination of the SOAP study, performed the statistical analyses, and read and approved the final manuscript.

Dr Groeneveld: collected data for the study, revised the manuscript, and read and approved the final manuscript.

Dr Zandstra: collected data for the study, revised the manuscript, and read and approved the final manuscript.

Dr Hoste: collected data for the study, revised the manuscript, and read and approved the final manuscript.

Dr Malledant: collected data for the study, revised the manuscript, and read and approved the final manuscript.

Dr Lei: collected data for the study, revised the manuscript, and read and approved the final manuscript.

Dr Sprung: collected data for the study, revised the manuscript, and read and approved the final manuscript.

Financial/nonfinancial disclosures: The authors have reported to CHEST that no potential conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article.

Role of sponsors: The companies that funded the SOAP study (Abbott, Baxter, Eli Lilly, GlaxoSmithKline, and NovoNordisk) had no involvement at any stage of the study design, in the collection and analysis of data, in writing the manuscript, or in the decision to submit for publication.

Austria: University Hospital of Vienna (G. Delle Karth); LKH Steyr (V. Draxler); LKH-Deutschlandsberg (G. Filzwieser); Otto Wagner Spital of Vienna (W. Heindl); Krems of Donau (G. Kellner, T. Bauer); Barmherzige Bruede of Linz (K. Lenz); KH Floridsdorf of Vienna (E. Rossmann); University Hospital of Innsbruck (C. Wiedermann); Belgium: CHU of Charleroi (P. Biston); Hôpitaux Iris Sud of Brussels (D. Chochrad); Clinique Europe Site St Michel of Brussels (V. Collin); CHU of Liège (P. Damas); University Hospital Ghent (J. Decruyenaere, E. Hoste); CHU Brugmann of Brussels (J. Devriendt); Centre Hospitalier Jolimont-Lobbes of Haine St Paul (B. Espeel); CHR Citadelle of Liege (V. Fraipont); UCL Mont-Godinne of Yvoir (E. Installe); ACZA Campus Stuivenberg (M. Malbrain); OLV Ziekenhuis Aalst (G. Nollet); RHMS Ath-Baudour-Tournai (J. C. Preiser); AZ St Augustinus of Wilrijk (J. Raemaekers); CHU Saint-Pierre of Brussels (A. Roman); Cliniques du Sud-Luxembourg of Arlon (M. Simon); Academic Hospital Vrije Universiteit Brussels (H. Spapen); AZ Sint-Blasius of Dendermonde (W. Swinnen); Clinique Notre-Dame of Tournai (F. Vallot); Erasme University Hospital of Brussels (J. L. Vincent); Czech Republic: University Hospital of Plzen (I. Chytra); U SV Anny of Brno (L. Dadak); Klaudians of Mlada Boleslav (I. Herold); General Faculty Hospital of Prague (F. Polak); City Hospital of Ostrava (M. Sterba); Denmark: Gentofte Hospital, University of Copenhagen (M. Bestle); Rigshospitalet of Copenhagen (K. Espersen); Amager Hospital of Copenhagen (H. Guldager); Rigshospitalet, University of Copenhagen (K.-L. Welling); Finland: Aland Central Hospital of Mariehamn (D. Nyman); Kuopio University Hospital (E. Ruokonen); Seinajoki Central Hospital (K. Saarinen); France: Raymond Poincare of Garches (D. Annane); Institut Gustave Roussy of Villejuif (P. Catogni); Jacques Monod of Le Havre (G. Colas); CH Victor Jousselin of Dreux (F. Coulomb); Hôpital St Joseph & St Luc of Lyon (R. Dorne); Saint Joseph of Paris (M. Garrouste); Hôpital Pasteur of Nice (C. Isetta); CHU Brabois of Vandoeuvre Les Nancy (J. Larché); Saint Louis of Paris (J.-R. LeGall); CHU de Grenoble (H. Lessire); CHU Pontchaillou of Rennes (Y. Malledant); Hôpital des Hauts Clos of Troyes (P. Mateu); CHU of Amiens (M. Ossart); Hôpital Lariboisière of Paris (D. Payen); CHD Félix Gyuon of Saint Denis La Reunion (P. Schlossmacher); Hôpital Bichat of Paris (J.-F. Timsit); Hôpital Saint Andre of Bordeaux (S. Winnock); Hôpital Victor Dupouy of Argentueil (J.-P. Sollet); CH Auch (L. Mallet); CHU Nancy-Brabois of Vandoeuvre (P. Maurer); CH William Morey of Chalon (J.-M. Sab); Victor Dupouy of Argenteuil (J.-P. Sollet); Germany: University Hospital Heidelberg (G. Aykut); Friedrich Schiller University Jena (F. Brunkhorst); University Clinic Hamburg-Eppendorf (A. Nierhaus); University Hospital Mainz (M. Lauterbach); University Hospital Carl Gustav Carus of Dresden (M. Ragaller); Hans Sushemihl Krankenhaus of Emden (R. Gatz); Vivantes-Klinikum Neukoelln of Berlin (H. Gerlach); University Hospital RWTH Aachen (D. Henzler); Kreisklinik Langen-Seligenstadt (H.-B. Hopf); GKH Bonn (H. Hueneburg); Zentralklinik Bad Berka (W. Karzai); Neuwerk of Moenchengladbach (A. Keller); Philipps University of Marburg (U. Kuhlmann); University Hospital Regensburg (J. Langgartner); ZKH Links der Weser of Bremen (C. Manhold); University Hospital of Dresden (M. Ragaller); University of Wuerzburg (B. Reith); Hannover Medical School (T. Schuerholz); Universitätsklinikum Charité Campus Mitte of Berlin (C. Spies); Bethanien Hospital of Moers (R. Stögbauer); KhgmbH Schongau (J. Unterburger); Greece: Thriassio Hospital of Athens (P.-M. Clouva-Molyvdas); Sismanoglion General Hospital of Athens (G. Giokas); KAT General Hospital of Athens (E. Ioannidou); G. Papanikolaou General Hospital of Thessaloniki (A. Lahana); Agios Demetrios of Thessaloniki (A. Liolios); Onassis Cardiac Surgery Center of Athens (K. Marathias); University Hospital of Ioannina (G. Nakos); Tzanio Hospital of Athens (A. Tasiou); Athens Gen. Hosp. Gennimatas (H. Tsangaris); Hungary: Peterfy Hospital of Budapest (P. Tamasi); Ireland: Mater Hospital of Dublin (B. Marsh); Beaumont Hospital of Dublin (M. Power); Israel: Hadassah Hebrew University Medical Center (C. Sprung); Italy: Azienda Ospedaliera Senese o Siena (B. Biagioli); S. Martino of Genova (F. Bobbio Pallavicini); Osp Regionale of Saronno (C. Capra); Ospedale Maggiore-University A. Avogadro of Novara (F. Della Corte); Osp. Molinette of Torino (P. P. Donadio); A. O. Umberto I Ancona, Rianimazione Clinica (A. Donati); Azienda Ospedaliera Universitaria Policlinico of Palermo (A. Giarratano); San Giovanni Di Dio of Florence (T. Giorgio); H San Raffaele IRCCS of Milano (D. Giudici); Ospedale Di Busto Arsizio (S. Greco); Civile Di Massa (A. Guadagnucci); San Paolo of Milano (G. Lapichino); S. Giovanni Bosco Torino (S. Livigni); Osp. San Giovanni of Sesto (G. Moise); S Camillo of Roma (G. Nardi); Vittorio Emanuele of Catania (E. Panascia); Azienda Ospedaliera S. Gerardo dei Tintori of Monza (A. Pesenti); Hospital of Piacenza (M. Pizzamiglio); Universita di Torino-Ospedale S. Giovanni Battista (V. M. Ranieri); Policlinico Le Scotte of Siena (R. Rosi); Ospedale Maggiore Policlinico IRCCS of Milano (A. Sicignano); A. Uboldo of Cernusco Sul Naviglio (M. Solca); P. O. Civile Carrara of Massa (G. Vignali); San Giovanni of Roma (I. Volpe Rinonapoli); The Netherlands: Boven IJ Ziekenhuis of Amsterdam (M. Barnas); UMC St Radboud of Nijmegen (E. E. De Bel); Academic Medical Center of Amsterdam (A.-C. De Pont); VUMC of Amsterdam (J. Groeneveld); Groningen University Hospital (M. Nijsten); Waterlandziekenhuis of Purmerend (L. Sie); OLVG of Amsterdam (D. F. Zandstra); Norway: Sentralsjukehuset i Rogaland of Stavanger (S. Harboe); Sykehuset Østfold of Fredrikstad (S. Lindén); Aker University Hospital of Oslo (R. Z. Lovstad); Ulleval University Hospital of Oslo (H. Moen); Akershus University Hospital of Nordbyhagen (N. Smith-Erichsen); Poland: Paediatric University Hospital of Lodz (A. Piotrowski); Central Clinic Hospital SLAM of Katowice (E. Karpel); Portugal: Garcia de Orta of Almada (E. Almeida); Hospital de St António dos Capuchos of Lisboa (R. Moreno); Hospital de Santa Maria of Lisboa (A. Pais-De-Lacerda); Hospital S. Joao of Porto (J. A. Paiva); São Teotonio Viseu (A. Pimentel); Fernado Fonseca of Masama (I. Serra); Romania: Institute of Cardiovascular Diseases of Bucharest (D. Filipescu); Serbia and Montenegro: Military Medical Academy of Belgrade (K. Jovanovic); Slovakia: SUSCH of Bratislava (P. Malik); Slovenia: General Hospital of Novo Mesto (K. Lucka); General Hospital of Celje (G. Voga); Spain: Hospital Universitario Rio Hortega of Valladolid (C. Aldecoa Alvarez-Santullano); Sabadell Hospital (A. Artigas); Hospital Clinic of Barcelona (E. Zavala, A. Escorsell, J. Nicolas); Virgen del Camino of Pamplona (J. J. Izura Cea); Virgen de la Salud of Toledo (L. Marina); 12 de Octubre of Madrid (J. Montejo); Gregorio Maranon of Madrid (E. Palencia); General Universitario de Elche (F. Santos); Puerta del Mar of Cadiz (R. Sierra-Camerino); Fundación Jiménez Díaz of Madrid (F. Sipmann); Hospital Clinic of Barcelona (E. Zavala); Sweden: Central Hospital of Kristianstad (K. Brodersen); Stockholm Soder Hospital (J. Haggqvist); Sunderby Hospital of Luleå (D. Hermansson); Huddinge University Hospital of Stockholm (H. Hjelmqvist); Switzerland: Kantonsspital Luzern (K. Heer); Hirslanden Klinik Beau-Site of Bern (G. Loderer); University Hospital of Zurich (M. Maggiorini); Hôpital de la Ville of La Chaux-de-Fonds (H. Zender); United Kingdom: Edinburgh Western General Hospital (P. Andrews); Peterborough Hospitals NHS Trust of Peterborough (B. Appadu); University Hospital Lewisham, London (C. Barrera Groba); Bristol Royal Infirmary (J. Bewley); Queen Elizabeth Hospital Kings Lynn (K. Burchett); Milton Keynes General (P. Chambers); Homerton University Hospital of London (J. Coakley); Charing Cross Hospital of London (D. Doberenz); North Staffordshire Hospital of Stoke On Trent (N. Eastwood); Antrim Area Hospital (A. Ferguson); Royal Berkshire Hospital of Reading (J. Fielden); The James Cook University Hospital of Middlesbrough (J. Gedney); Addenbrookes of Cambridge (K. Gunning); Rotherham DGH (D. Harling); St Helier of Carshalton (S. Jankowski); Southport & Formby (D. Jayson); Freeman of Newcastle Upon Tyne (A. Kilner); University Hospital of North Tees at Stockton on Tees (V. Krishna-Kumar); St Thomas Hospital of London (K. Lei); Royal Infirmary of Edinburgh (S. Mackenzie); Derriford of Plymouth (P. Macnaughton); Royal Liverpool University Hospital (G. Marx); Stirling Royal Infirmary (C. McCulloch); University Hospital of Wales, Cardiff (P. Morgan); St George’s Hospital of London (A. Rhodes); Gloucestershire Royal Hospital (C. Roberts); St Peters of Chertsey (M. Russell); James Paget Hospital of Great Yarmouth (D. Tupper-Carey, M. Wright); Kettering General Hospital (L. Twohey); Burnley DGH (J. Watts); Northampton General Hospital (R. Webster); Dumfries Royal Infirmary (D. Williams)