© 2005 American Thoracic Society
Hydrocortisone Infusion for Severe Community-acquired PneumoniaThe Role of Relative Adrenal InsufficiencyFrom the Authors:The letter from Dr. Restrepo and colleagues addresses an important limitation of our study (1)—the unplanned differences in distribution of patients receiving conventional ventilation and noninvasive positive pressure ventilation (NPPV) among treated and control groups. We are fully aware of the effect of conventional and noninvasive ventilation in patients with severe community-acquired pneumonia. In a previous randomized study of patients with severe community-acquired pneumonia and PaO2:FIO2 < 200, we reported that NPPV was associated, in comparison to conventional treatment, with an overall reduction in endotracheal intubation (21% versus 50%; p = 0.03) and duration of ICU stay (1.8 ± 0.7 days versus 6 ± 1.8 days; p = 0.04), without affecting mortality (2). The positive effects of NPPV, however, were limited to the subgroup of patients with COPD, a finding later supported by a larger, uncontrolled study (3). When pneumonia is associated with severe hypoxemia (PaO2:FIO2 < 120 sustained for > 6 hours) worse than that reported in our studies (1, 2), NPPV was shown to be beneficial in one recent trial (4). Patients with COPD were excluded from our study and contributed to a large number (n = 25) of ineligible patients. Other patients were excluded for asthma (n = 6), neoplasm (n = 11), or failure to meet all inclusion criteria (n = 27). Table 1 compares the response observed in the two groups of ventilated patients. On study day 8, the number of patients on conventional ventilation versus NPPV was 14 versus 1 and 3 versus 3, for the control and hydrocortisone-treated group, respectively. The incorrect conclusions reached by Dr. Restrepo and colleagues are the result of an incomplete explanation of the data in our article (1). Table 1 shows that the number of patients still on conventional ventilation on study day 8 was significantly different (14 [87%] versus 3 [43%]; p = 0.045) between the two groups.
The purpose of our study was to evaluate the effect of hydrocortisone infusion on organ dysfunction associated with pulmonary and systemic inflammation. Notably, improvement in PaO2:FIO2 and chest radiograph score by day 8 was observed with hydrocortisone treatment irrespective of the initial mode of ventilation. As a group, those receiving NPPV had, in comparison to patients receiving conventional ventilation, a shorter duration of mechanical ventilation (p = 0.001) and lower mortality (p = 0.07). The only NPPV failure was a control patient. These findings underscore the potential advantages of NPPV in this patient population and the need—in a future trial evaluating mortality as a primary end point—to use a dynamic allocation scheme that allows stratification by initial mode of mechanical ventilation (1). Dr. Restrepo and colleagues direct our attention to the subgroup of patients who initially received conventional ventilation and are at increased risk for ventilator-associated complications and mortality. A reasonable argument is made that the larger number of patients in the placebo group receiving conventional ventilation (16 versus 7) skewed the overall results in favor of hydrocortisone. A review of the variables shown in Table 1, however, shows within this group a response to hydrocortisone with a significant reduction in C-reactive protein level, multiple organ dysfunction syndrome (MODS), chest radiograph scores, and incidence of delayed septic shock. These findings suggest that modulation of systemic and pulmonary inflammation with prolonged low-dose hydrocortisone might benefit this group, at higher risk for sepsis-related complications, the most. We believe that our study supports the original hypothesis, and the data shown in Table 1 indicate that the findings also apply to patients initially placed on conventional ventilation. We recognize the limitations of this small study and made suggestions for a future larger trial with the primary aim to investigate the effect of hydrocortisone infusion on mortality. The letter by Dr. Jalloul addresses the issue of relative adrenal insufficiency and response to hydrocortisone treatment in patients with septic shock. Two recent meta-analyses (6, 7) have addressed this issue in extensive detail. In our study, only two hydrocortisone-treated patients had septic shock at study entry, and it is unlikely that relative adrenal insufficiency (which was not tested) might have occurred at study entry in patients without shock. Nevertheless, several patients randomized to placebo developed septic shock after study entry, and we cannot exclude that some might have had subclinical adrenal insufficiency at study entry or a blunted adrenal response to ACTH when they developed shock. The lack of an ACTH stimulation test prior to randomization or in conjunction with the development of delayed septic shock is a limitation of this study. Unfortunately, this issue was not fully appreciated in 1999 when the study was designed.
a University of Tennessee Health Science Center, Memphis, Tennessee FOOTNOTES Conflict of Interest Statement: G.U.M. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; R.U. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; M.C. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. REFERENCES
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
