Negative Studies Deserve More Attention

To the Editor:

With regard to their recent article, Kacmarek and colleagues (1) are to be congratulated not only for their trial on partial liquid ventilation (PLV) but also for their persistence in reporting the results of an important "negative" clinical study with liquid perfluorocarbons for the therapy of acute lung injury. The fact that this study failed in demonstrating beneficial effects of PLV has important implications for both study design and the peer-review process itself.

Most of the previous experimental studies that led to acceptance of PLV as a possible alternative in the therapy of acute lung injury were conducted with mechanical ventilation strategies suggested to be hazardous, namely, relatively high tidal volumes and low positive end-expiratory pressure levels (2). When evidence accumulated that such strategies were associated with further lung injury (ventilator-associated lung injury), phase II trials with PLV were conducted, and control groups received the best possible mechanical ventilation (3), which aimed at minimizing the mechanical stress on lung tissues through over-distension and/or cycling collapse/reopening of lung units. Two important questions arise: Should the experimental evidence on the benefits of PLV not have been put in question at that time, when the approach to mechanical ventilation changed considerably? Were there any "negative" experimental studies on PLV that used alternate ventilatory modes in their control groups and were rejected, using the argument: "You cannot affirm PLV does not work if so many other studies show it does"? Clearly, the latter is a highly speculative and provocative question.

As pointed out by Kacmarek and colleagues (1), it is difficult to convince reviewers and editors to publish "negative" results. Moreover, authors often are not motivated to publish their work when a given therapy is not successful. I consider this an "antiscientific" view that should be banned from publication politics and suggest that much more attention should be paid to so-called negative experimental studies. In addition, editors and reviewers should be more careful when considering the publication of "positive" studies that do not use state-of-the-art approaches in their control groups. We cannot change the past, and also we probably cannot save PLV from death, but we can learn from this experience and avoid similar mistakes in the future.

Conflict of Interest Statement: M.G.d.A. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript.

MARCELO GAMA DE ABREU

University Clinic Carl Gustav Carus

Technical University Dresden

Dresden, Germany

References

1. Kacmarek RM, Wiedemann HP, Lavin PT, Wedel MK, T

2. Mead J, Takishima T, Leith D. Stress distribution in lungs: a model of pulmonary elasticity. J Appl Physiol 1970;28:596-608.

3. Hirschl RB, Croce M, Gore D, Wiedemann H, Davis K, Bartlett RH. Prospective, randomized, controlled pilot study of partial liquid ventilation in adult acute respiratory distress syndrome. Am J Respir Crit Care Med 2002;165:781-787.

From the Authors:

We thank Dr. Gama de Abreu for his letter regarding our recent article (1). He raises important questions regarding the methodology used to ventilate patients during our trial and the potential utility of perfluorocarbon therapy in general for the management of adult acute respiratory distress syndrome (ARDS). When a clinical trial is designed, investigators struggle to ensure that the control group receives an approach to therapy that is considered at the time of study design to be standard or the best possible based on the best available data. At the time our study began (1998), we based our ventilatory strategies on the best available data that addressed lung protection (1). The mortality in all groups was 15% (control arm) to 26%, equivalent to or better than any other study evaluating ventilatory strategy in ARDS. Although we only enrolled patients 65 yr of age or younger, we obtained these low mortality rates even though our oxygenation entry criteria were more rigorous than those in published ARDS trials, except that of Villar and coworkers (2). Specifically, patients had to first demonstrate a Pa^sub O2^/FI^sub O2^ less than 200 mm Hg on an FI^sub O2^ of = 0.5 with = 5 cm H2O positive end-expiratory pressure (PEEP). They were then placed on = 13 cm H2O PEEP with an FI^sub O2^ = 0.5. On these settings they had to maintain a Pa^sub O2^/FI^sub O2^ < 300 mm Hg to be randomized. The ARDSnet only required a Pa^sub O2^/FI^sub O2^ < 300 mm Hg regardless of PEEP or FI^sub O2^ (3).

In response to Dr. Gama de Abreu's second question, at the time we began our trial there were laboratory data available on partial liquid ventilation (PLV) with high-frequency ventilation (4-6). These studies compared high-frequency ventilation and conventional ventilation with and without PLV. In two of the studies, PLV was better able to improve gas exchange (both high-frequency and conventional) (5, 6), and in the third, no differences were observed (4). Thus, to our knowledge there were no important negative experimental data on PLV that used alternate ventilatory modes in the control group.

As Dr. Gama de Abreu indicates, we felt strongly that the negative results from our trial needed to be published so that others would not utilize the approach to perfluorocarbon therapy in ARDS that we used. We agree that it is critical for negative trials to be published, since we can learn just as much from a negative trial as we do from a positive trial, even though the results may not be as exciting.

Conflict of Interest Statement: R.M.K. received a $65,000 grant from Alliance Pharmaceuticals in 1999 to support studies of partial liquid ventilation in animal models. A.S.S. has been a paid consultant to Maquet in the field of mechanical ventilation ($10,000/yr). He chaired a DSMB for Leo Pharma in relation to a surfactant trial ($10,000/yr), and was on the Alliance advisory board for the trial presented in this publication, but received no financial compensation for this.

ROBERT M. KACMAREK

Massachusetts General Hospital

Boston, Massachusetts

ARTHUR S. SLUTSKY

University of Toronto

Toronto, Ontario, Canada

References

1. Kacmarek RM, Wiedemann HP, Lavin PT, Wedel MK, Tütüncü AS, Slutsky AS. Partial liquid ventilation in adult patients with acute respiratory distress syndrome. Am J Respir Crit Care Med 2006;173:882-889.

2. Villar J, Kacmarek RM, Pérez-Méndez L, Aguirre-Jaime A. A high positive end-expiratory pressure, low tidal volume ventilatory strategy improves outcome in persistent acute respiratory distress syndrome: a randomized, controlled trial. Crit Care Med 2006;34:1311-1319.

3. The Acute Respiratory Distress Syndrome Network. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med 2000;342:1301-1308.

4. Baden HP, Mellema JD, Bratton SL, O'Rourke PP, Jackson JC. High-frequency oscillatory ventilation with partial liquid ventilation in a model of acute respiratory failure. Crit Care Med 1997;25:299-302.

5. Smith KM, Bing DR, Meyers PA, Connett JE, Boros SJ, Mammel MC. Partial liquid ventilation: a comparison using conventional and high-frequency techniques in an animal model of acute respiratory failure. Crit Care Med 1997;25:1179-1186.

6. Smith KM, Mrozek JD, Simonton SC, Bing DR, Meyers PA, Connett JE, Mammel MC. Prolonged partial liquid ventilation using conventional and high-frequency ventilatory techniques: gas exchange and lung pathology in an animal model of respiratory distress syndrome. Crit Care Med 1997;25:1888-1897.

© 2006 American Thoracic Society Provided by ProQuest LLC. All Rights Reserved.