| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
Electronic Letters to:
|
|
Electronic letters published:
![[Read eLetter]](/icons/misc/sectionDOWN.gif){kind=icon}
Diurese to live or live to diurese?
|
|
|||
|
Andre CKB Amaral, Physician Heart Institute (InCor/DF) - Zerbini Foundation - Brazil
Send letter to journal:
andre.amaral{at}incordf.zerbini.org.br Andre CKB Amaral
|
Great interest is brought by the recent article by Sakr and colleagues(1). The author’s are to be congratulated for their huge effort in collecting data from over 3000 ICU patients, and describing the epidemiology of ALI/ARDS in European Critical Care Units. The study provides very important data that might bring ideas for future research both in management and resource allocation. However the author’s provide a multivariate analysis that might mislead some readers. The inclusion of mean fluid balance as an independent predictor of mortality should not be used to disclose aggravating factor’s, as stated by the authors, but only severity of illness during the whole ICU stay (such as the mean SOFA). An important question then arises, did the patients undergo a negative fluid balance because they lived or did they live because there was a negative fluid balance? Critically ill patients often display a two-phase response regarding fluid -balance, with an initial positive fluid balance (ressuscitation phase) followed by a negative fluid balance in those who survive. However, non- survivors may not undergo this second phase, leading this study to a time- dependent bias(2). Adding to the complexity of the analysis, a positive fluid balance may be observed due to renal failure, which, by itself, is an independent predictor of mortality(3). Also, the use of stepwise logistic regression analysis might lead to unexpected and clinically unreasonable results, and should be used mainly as an exploratory tool(4). Furthermore, for an explanatory model (such as this one) their use to infer causal association should be avoided. A new evaluation would be welcome, trying to focus on a clinical model, forcing the following variables into a Cox porpotional hazards model: 1. High tidal volume; 2. Daily fluid balance (as a time-dependent covariate); 3. Initial SAPS score; 4.Shock during ICU stay; 5. Worst SOFA renal score; 6. PaO2 /FiO2 value on the day of ALI/ARDS diagnosis; 7. Age; 8. Admission source; 9. Type of admission; 10. Comorbid diseases. Also, interaction terms between fluid balance and both shock and renal failure could be included. This paper is very important due to reinforcement of the importance of ventilatory strategies in ARDS management, however it is bound to a great question, with potential clinical implications for the management of patients in septic shock and ARDS: what should we do? Agressive fluid ressuscitation(5) or negative fluid balance?. References (1) Sakr Y, Vincent JL, Reinhart K et al. High Tidal Volume and Positive Fluid Balance Are Associated With Worse Outcome in Acute Lung Injury. Chest 2005; 128(5):3098-3108. (2) van Walraven C, Davis D, Forster AJ et al. Time-dependent bias was common in survival analyses published in leading clinical journals. Journal of Clinical Epidemiology 2004; 57(7):672-682. (3) Levy EM, Viscoli CM, Horwitz RI. The effect of acute renal failure on mortality. A cohort analysis. JAMA 1996; 275(19):1489-1494. (4) Katz MH. Multivariable Analysis: A Primer for Readers of Medical Research. Ann Intern Med 2003; 138(8):644-650. (5) Rivers E, Nguyen B, Havstad S et al. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med 2001; 345(19):1368-1377. |
|||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH |
eLetters: