Fluid Management Strategies in ARDS

Abstract & Commentary

By Uday Nanavaty, MD, Pulmonary and Critical Care Medicine, Rockville, MD, Assistant Director AICU, St Agnes Hospital, Baltimore, is Associate Editor for Critical Care Alert.

Dr. Nanavaty reports no financial relationships related to this field of study.

Synopsis: In a large randomized trial of two fluid management strategies in patients with ALI or ARDS, investigators found no difference in mortality between liberal or conservative fluid strategies. In spite of arguments about early liberation from the ventilator and shorter ICU stays, the benefits of the rather complex "conservative" fluid management strategy remain unclear.

Source: The ARDS Net Investigators. Comparison of two fluid-management strategies in acute lung injury. N Engl J Med. 2006;354:2564-2575.

Proper fluid management of patients with acute lung injury (ALI) or the acute respiratory distress syndrome (ARDS) is important. Although the pulmonary edema in ARDS patients is not due to increased hydrostatic pressures, it is believed that there is more fluid leakage at higher pressures. Further, if fluid administration results in lower blood oncotic pressure, it may result in an increased tendency towards edema formation. Whether restricting intravenous fluids or early diuresis can improve mortality in patients with ALI or ARDS is not clear, so the investigators conducted a randomized, controlled, clinical trial of a liberal fluid strategy (which might be considered usual practice) compared to a conservative strategy (the experimental strategy for this trial). Participants of this trial were additionally randomized to receive either a central venous catheter (CVC) or a pulmonary artery catheter (PAC) to further direct the fluid management. Since the outcomes were the same whether the patients received a PAC or a CVC, in the analysis the fluid management strategy was used to group the patients irrespective of which catheter was used.

Over 4 years, more than 11,000 patients were screened, and 1001 patients were enrolled in the study. Patients were of similar age and severity of illness. Within 48 hours of development of ALI or ARDS, patients were randomized to a conservative (503 patients) or liberal (497 patients) fluid management strategy with fluid management being directed by a pre-determined protocol. For patients in shock, liberal fluid use was allowed in the conservative strategy, and for patients with high oxygen requirements, fluid boluses were somewhat restricted at adequate filling pressures in the liberal strategy. The details of this fluid management strategy are beyond the scope of this review. The ventilator management was per ARDS Network protocol of lung protective ventilation.

At 60 days, there was no difference in mortality rate between the conservative management (25.5 ± 1.9%) or liberal management (28.4 ± 2.0%) strategies. The conservative strategy patients had more ventilator-free days, days free of central nervous system failure, and ICU-free days, during the first 28 days. There were no significant differences in the number of failure-free days for other organs during the first 28 days, although there was a small increase in the number of cardiovascular-failure free days during the first 7 days with the liberal strategy. Black patients seem to have experienced a higher mortality rate, but that difference was not significant once adjustment for co-morbidities was made. The numbers of patients needing renal replacement therapies were similar in the two groups.


In my honest opinion, this massive study proves nothing. When 11,000 patients are screened and only 10% are eligible or enrolled in a trial, generalization of the study's findings is impossible. The protocol seems tedious and almost unrealistic in clinical application. Although the authors have gone to great lengths to suggest that ventilator-free days and days out of the ICU were statistically significantly better with the conservative (experimental) strategy, those conclusions seem just statistics and are unlikely to change practice very much. The authors have tried to dig hard into the database to come up with some positive conclusion of this massive clinical exercise but have come up really empty handed.

The original ARDS network protocol looked at mortality at 30 days and showed the difference in mortality then. The protocol here had to be stretched to 60 days and yet they could not show a mortality difference. The ventilator-free days and ICU-free days are calculated at 28 days, hence the differences reach statistical significance. Let's stretch them to 45 or 60 days or look at 7 days (when study protocol ended) and there would be no differences, I presume.

The preachers of the low tidal volume strategy did not practice their teaching. At the time of randomization, about 48 hours into the ALI/ARDS, tidal volumes were close to 7.4 mL/kg of predicted body weight. That would mean that they either do not start the ALI-ARDS protocol early enough or do not believe in their own results.

There is an alternative explanation in my opinion about the reason why the ventilator-free days were increased in the conservative-strategy patients. I presume that this group of patients was developing contraction alkalosis. With a significantly high number of patients with metabolic alkalosis, it is possible that they tolerated higher arterial CO2 levels and hence were breathing at a slightly lower minute ventilation, and hence were able to come off the ventilator a little bit early.

The authors did not describe the problems of hypokalemia as well. With frequent use of diuretics, I suspect that a large number of the conservative strategy patients needed frequent potassium checks and/or replacement. The hemodynamic parameters also are not usual standard practice, as far as other studies are concerned. Previous studies of "dry lungs equal happy lungs" have failed to show truth as far as mortality rates are concerned, and this massive expenditure of our tax dollars joins that list in my mind. Conservative or liberal, keep doing the right things, doctor.