0
Communications to the Editor |

FDA Evaluation of Antimicrobials : Subgroup Analysis FREE TO VIEW

John H. Powers, MD; Daphne Lin, PhD; David Ross, MD, PhD
Author and Funding Information

Affiliations: Center for Drug Evaluation and Research, US Food and Drug Administration, Rockville, MD,  Northwestern University Feinberg School of Medicine, Chicago, IL,  Washington University, St. Louis, MO,  Joan XXIII University Hospital, Tarragona, Spain

Correspondence to: John H. Powers, MD, HFD 104, 9201 Corporate Blvd, Rockville, MD 20850; e-mail: powersjoh@cder.fda.gov



Chest. 2005;127(6):2298-2301. doi:10.1378/chest.127.6.2298
Text Size: A A A
Published online

To the Editor:

In response to Wunderink and colleagues, we would like to clarify some apparent misunderstandings of how the US Food and Drug Administration (FDA) utilizes data from subgroup analyses in evaluating the safety and efficacy of antimicrobials. In their article on the treatment of nosocomial pneumonia (NP) comparing linezolid and vancomycin, Wunderink et al1use a subgroup analysis to make a fundamentally different conclusion than shown in the trial as a whole. That is, they use a subgroup of patients with NP who are infected with one pathogen to claim superior efficacy in trials that overall show noninferiority of linezolid to vancomycin.23 In contrast, the FDA uses subgroup analyses to investigate the robustness and confirm the overall conclusions of a trial. The FDA also uses subgroup analyses in a risk-benefit assessment of which populations may be most likely benefit from a drug. This is not the same as drawing different conclusions about the study drug relative to the control drug. For instance, when the overall results of a trial show superiority of the study drug to the control drug, it may be appropriate to look at subgroups of patients treated with the study drug to determine whether there is a consistently favorable risk/benefit across subgroups. This is the case in the example the authors cite with drotrecogin alfa.4 In this situation, the study drug was overall superior in efficacy to placebo in treating sepsis, but the safety risk of intracranial hemorrhage justified its approval in a subgroup of more severely ill patients where the benefit justified the risk. This did not change the fundamental scientific conclusion that drotrecogin alfa was superior in efficacy to placebo in the trial. However, it did affect the public health conclusion as to the patient population for which the drug could be approved based on safety concerns.

It is fundamentally more difficult to draw a conclusion of superiority in a subgroup of patients when the trial as a whole demonstrates noninferiority or inferiority of the study drug to the control, for the reasons enumerated in our previous correspondence.5 The results of the trials23 in NP showed linezolid was noninferior to vancomycin, yet Wunderink and colleagues claim that a subgroup shows superiority for the subset of patients infected with methicillin-resistant Staphylococcus aureus (MRSA).,1 This is a very different way of using subgroup analyses than done by the FDA.

The FDA uses an evaluation of microbiologically evaluable (ME) and clinically evaluable (CE) subgroups of patients as one of the analyses in an overall assessment of drug efficacy in a trial. The authors are incorrect in suggesting that these analyses are “favored” by the FDA, and the previous guidance on design and analysis of clinical trials for NP is currently under revision based on discussions held at a public workshop in November 2002. While drug sponsors may specify the ME and CE subgroups as the “primary” analyses and present this as such in publications, the FDA also evaluates outcomes in a modified intention-to-treat (mITT) population of all randomized patients with the disease under study who receive at least one dose of study medication. The conclusions in the ME and CE subgroups should support the conclusions in the mITT population, and vice versa. The FDA does not analyze the results in either the mITT population or the subgroups in isolation. In noninferiority trials, neither the mITT nor the evaluable populations are optimal; therefore, one should consider the results in both the mITT population and the evaluable subgroups.6 An analysis of the mITT population, while conservative in a superiority trial, may make the efficacy of drugs appear more similar in the setting of a noninferiority trial. This has the potential to lead to a false-positive conclusion of noninferiority. However, the ME and CE subgroups have all the issues discussed previously in relation to subgroup analyses.5 In the linezolid NP trials, almost one half and three quarters of patients were not evaluable in the CE and ME subgroups, respectively. This is especially important when the reasons for nonevaluability may be related to prognosis.7 Conclusions based solely on subgroup analyses of ME and CE populations ignore the results in a large proportion of patients entered in the trial. Therefore, one should exercise caution in drawing conclusions of superiority for a drug unless analyses of both mITT and the subgroups support similar conclusions of superiority.

Wunderink and colleagues are incorrect in assuming that FDA has granted approval of antimicrobials for community-acquired pneumonia (CAP) due to multidrug-resistant Streptococcus pneumoniae (MDRSP) based on subgroup analyses alone. Firstly, the approvals for MDRSP do not make relative claims that one drug is superior to another, only that the labeled drug is effective in the treatment of CAP caused by MDRSP. Secondly, these approvals were based on the overall efficacy of the drug in treating CAP, the efficacy in CAP due to susceptible S pneumoniae, and finally the efficacy in CAP caused by MDRSP. Only if the data from all three groups are consistent and support the efficacy of the drug will it be approved for disease due to a resistant pathogen in this setting.

Wunderink and colleagues are incorrect that our recommendations about subgroups would lead one to use quinupristin-dalfopristin for the treatment of MRSA pneumonia. The FDA determines that a clinical trial is safe to proceed but does not “approve” individual trials. Rather, the FDA-approved labeling for quinupristin-dalfopristin specifically states that the drug is approved only for those cases of complicated skin and skin structure infections caused by methicillin-susceptible Staphylococcus aureus (MSSA) strains.8 In addition, quinupristin-dalfopristin is not FDA approved for the treatment of pneumonia of any kind, so we do not imply that the drug should be used off-label for this purpose.

Finally, we disagree that clinicians do not need to understand the overall efficacy of a drug in the treatment of NP, including cases due to Gram-positive pathogens. At the present time, treatment of NP is largely empirical and clinicians do not know the causative pathogen at the time of diagnosis. The overall efficacy of a drug in treating NP is just as important as the efficacy in specific subsets of patients, as clinicians initially will not know the causative pathogen. As the authors note in their article,1 initial prescribing of an inappropriate drug in NP has been associated with increased mortality. The question of whether linezolid has superior efficacy in the treatment of NP due to MRSA can be answered by a prospective, randomized, controlled trial, and we applaud the sponsor of the drug for pursuing such a trial. However, until we have diagnostic testing that can quickly and easily determine the causative pathogen prior to initiation of treatment, it is unclear how this knowledge will aid the clinician when empirically prescribing drugs for NP. The authors assume that a larger trial would not show linezolid to be less effective for pathogens other than MRSA. For example, in trials in sepsis, follow-up trials of patients treated with an experimental monoclonal antibody did not confirm the efficacy seen in initial trials in subgroups of patients with Gram-negative bacteremia.9 In the setting of empirical treatment, with NP due to various pathogens, linezolid was found to be noninferior to vancomycin. The “hard” data show no difference in the proportions of clinical successes as defined in the study in the overall population of patients enrolled. This is the most relevant clinical question and the one that clinicians must grapple with at the bedside.

We would also like to point out that the authors’ reply did not address our primary concerns that we enumerated regarding subset analyses in our previous letter.5 This includes the issues of the biological plausibility of a difference between drugs for disease caused by MRSA but not for MSSA despite similar in vitro activity. Also, the issues of not correcting for multiple comparisons still remain.,10It appears that these authors have published further subgroup analyses in another journal11 from these same NP trials in a subgroup of patients with ventilator-associated pneumonia, so it is clear that other comparisons were made beyond those presented in the article in CHEST. Our point remains that the results in these subgroups may still represent chance findings, and a p value of 0.03 as presented in the article in CHEST in the setting of numerous comparisons does not represent a “significant” result. The authors suggest that clinicians should accept the results because they appear favorable, regardless of the issues with the methods of how those results were obtained. We would suggest that well-informed clinicians know better.

The authors are federal government employees. This letter does not necessarily represent the official policy of the US Food and Drug Administration.

Wunderink, RG, Rello, J, Cammarata, SK, et al (2003) Linezolid vs vancomycin: analysis of two double-blind studies of patients with methicillin-resistantStaphylococcus aureusnosocomial pneumonia.Chest124,1789-1797. [CrossRef] [PubMed]
 
Rubinstein, E, Cammarata, S, Oliphant, T, et al Linezolid (PNU-100766) versus vancomycin in the treatment of hospitalized patients with nosocomial pneumonia: a randomized, double-blind, multicenter study.Clin Infect Dis2001;32,402-412. [CrossRef] [PubMed]
 
Wunderink, RG, Cammarata, SK, Oliphant, TH, et al Continuation of a randomized, double-blind, multicenter study of linezolid versus vancomycin in the treatment of patients with nosocomial pneumonia.Clin Ther2003;25,980-992. [CrossRef] [PubMed]
 
Bernard, GR, Vincent, JL, Laterre, PF, et al Efficacy and safety of recombinant human activated protein C for severe sepsis.N Engl J Med2001;344,699-709. [CrossRef] [PubMed]
 
Powers, JH, Ross, DB, Lin, D, et al Linezolid and vancomycin for methicillin-resistantStaphylococcus aureusnosocomial pneumonia: the subtleties of subgroup analyses.Chest2004;126,314-315. [CrossRef] [PubMed]
 
Lachin, JM Statistical considerations in the intent-to-treat principle.Control Clin Trials2000;21,167-189. [CrossRef] [PubMed]
 
Montori, VM, Guyatt, GH Intention-to-treat principle.Can Med Assoc J2001;165,1339-1341
 
 Product labeling for quinupristin-dalfopristin (Synercid). Physicians Desk Reference. 2004; ,.:2186 -2189 Thomson PDR. Montvale, NJ:.
 
Siegel, JP Biotechnology and clinical trials.J Infect Dis2002;185(Suppl),S52-S57
 
Bland, JM, Altman, DG Multiple significance tests: the Bonferroni method. BMJ. 1995;;310 ,.:170. [CrossRef] [PubMed]
 
Kollef, MH, Rello, J, Cammarata, SK, et al Clinical cure and survival in Gram-positive ventilator-associated pneumonia: retrospective analysis of two double-blind studies comparing linezolid with vancomycin.Intensive Care Med2004;30,388-394. [CrossRef] [PubMed]
 
To the Editor:

Once again, we are surprised by the criticisms of Drs. Powers, Ross, and Lin, US Food and Drug Administration (FDA) employees, who suggest that they have a better sense of what clinicians should be concerned about than physicians who spend approximately 50% or more of their time actually caring for patients. The writers suggest that they (the FDA) use subgroup analysis in a fundamentally different way than we did in our analysis of methicillin-resistant Staphylococcus aureus (MRSA) pneumonia treatment1because their intent is different. We know of no reason why statistical rules for subgroup analysis operate differently depending on the intent of their use. Our point was that the FDA does not make a statistical adjustment for multiple comparisons in any of the analyses they perform on data presented to them for approval of a drug for a specific indication. Yet, no warning from the FDA about the dangers of subgroup analysis accompanied their approval of drotrecogin alfa activated only for the subgroup of patients at high risk of death, such as Powers and associates2 suggest for the use of linezolid rather than vancomycin.

The writers also use the example of their “type” of subgroup analysis with a fundamentally different study than the linezolid/vancomycin studies.34 They illustrate with a placebo-controlled study5 that from the outset was designed as a superiority study. When that study showed overall superiority, the FDA believed it was legitimate to examine subgroups to determine if there was a consistent favorable risk/benefit across subgroups. We actually agree with the concept and the appropriateness of that analysis, although we may disagree with the specific recommendation, as did the European regulatory agencies.

The situation with the linezolid/vancomycin studies is very different. These studies were a priori powered and designed to demonstrate equivalence, and each independently confirmed that assumption. Examination of subgroups to determine if the risk/benefit ratio is equivalent across all subgroups is every bit as important in that setting as it is in superiority trials. As noted in their letter, Dr. Powers and associates acknowledge that they do perform subgroup analysis for these studies also as part of their regulatory responsibilities.

Where we part ways from Dr. Powers and his associates at the FDA, as well as with the pharmaceutical industry, is on the types of data “cuts” on which to do the subgroup analysis. The subgroups analyzed by the FDA, like clinically evaluable, microbiologically evaluable, and modified intention-to-treat, are meaningless to clinicians. Clinicians are interested in two populations: the first is the full intention-to-treat population,6 since this group corresponds to the patients in whom clinicians initiate therapy before culture results return. We therefore included this population from both studies in our analysis and found that linezolid was equivalent overall to vancomycin.

The other subgroups clinicians are interested in are those based on culture results. Here once again we differ from Dr. Powers and his FDA associates. They combine all patients with a positive culture finding into a single “microbiologically-evaluable” subgroup, as if pneumonia due to α-hemolytic Streptococci, penicillin-susceptible pneumococci, and S aureus, both methicillin sensitive and methicillin resistant, are equivalent. Clinicians would not continue either of these drugs for anything but MRSA pneumonia.7Both should be stopped even if no microbiologic etiology for pneumonia is documented on culture, at least for ventilator-associated pneumonia (VAP), since a negative Gram stain and culture makes MRSA VAP very unlikely.8 Because MRSA pneumonia is the only pneumonia population in which these drugs should be continued, this subgroup is therefore clearly the most important subgroup to analyze, at least for clinicians. We would have thought the FDA would agree.

It is instructive to reverse the situation. Assume linezolid is the already FDA-approved drug and vancomycin is the experimental therapy but that the results were the same as in the trials we analyzed: an absolute difference of 23.5% in clinical cure rates and an absolute mortality difference of 16.5% favoring linezolid for MRSA pneumonia.1 In their careful “subgroup analysis of risk/benefit to determine which population may be most likely to benefit,” Dr. Powers’ associates at the FDA focus on this MRSA pneumonia subgroup. Further subgroup analysis to “investigate the robustness of the data and confirm the overall conclusion” demonstrates that this inferior outcome of vancomycin was found in patients who had a more accurate diagnosis, either by quantitative cultures of respiratory secretions or positive blood cultures findings; that it was true even in the more vulnerable VAP population, as well as the nonintubated population; and that the inferiority even stood up to multivariate analysis with other factors previously demonstrated to be of significance in the treatment of MRSA pneumonia. Since MRSA pneumonia is the patient population most likely to be treated with vancomycin and that for every six patients treated with vancomycin rather than linezolid, one extra patient would die (not just fail treatment), we would expect that the FDA panel would not recommend approval for pneumonia, or at least for MRSA VAP.

Others would agree, since this scenario is exactly what was found with a nearly identical FDA-approved study comparing vancomycin to quinupristin/dalfopristin. Even though the drugs were equivalent in each of the subgroup analyses that Powers and colleagues would have us emphasize,9 a higher rate of clinical failure in the quinupristin/dalfopristin for the subgroup of MRSA pneumonia (19.4% clinical success in the intention-to-treat population vs 40% success for vancomycin) led the sponsoring company to not even submit the study to the FDA for a pneumonia indication, despite the significant financial investment in this study. Whether is was because the company believed that the FDA would not accept this “evidence” of equivalency or whether they realized that they would have no market for the drug since clinicians would be reluctant to prescribe this drug for pneumonia, even if the FDA gave it an approved pneumonia indication, isn’t clear. We would hope it was the former but are afraid it was the latter. We cite this example not to imply that Dr. Powers and associates are suggesting that quinupristin/dalfopristin should be used for pneumonia but to demonstrate that the same type of subgroup analysis of a trial demonstrating equivalency, that they disparage when performed by us,2was used to appropriately deny approval of a drug for the same indication. We therefore disagree with Powers and co-workers that our analysis is fundamentally different than that performed routinely by the FDA. The failure rates of vancomycin for MRSA pneumonia in studies FDA approved for safety,34,9as well as other observational studies,10 are so consistent that we wonder if the Dr. Powers and his associates at the FDA should in fact consider a moratorium on vancomycin as the comparator drug for studies that may include MRSA pneumonia.

Dr. Powers, Ross, and Lin also question our analysis because we offer no biological plausibility for an advantage for linezolid in MRSA pneumonia but not in methicillin-sensitive S aureus (MSSA) pneumonia. They assume that the only difference in outcome between these groups is related to antibiotic efficacy. Clinicians would however readily recognize that patients with MSSA pneumonia are fundamentally different than patients with MRSA pneumonia.7 The host response in the less-ill MSSA pneumonia patients is better able to handle an infectious challenge, no matter what the antibiotic. Even in MRSA pneumonia patients, the difference in outcome between linezolid and vancomycin in those with lower APACHE (acute physiology and chronic health evaluation) II scores was negligible.3

We will again state, as the authors of the article, that we also have concerns about subgroup analysis.11 Although multivariate analysis adds support to the possibility that the differences are truly related to the drugs, baseline differences between the two groups may still explain all or some of the differences in outcome. We, along with other clinicians who were presented this data, were the ones who urged the company to do a follow-up head-to-head comparison of linezolid with vancomycin, not the FDA. FDA approval of a drug for a specific indication only allows the drug on the market. In contrast to what Powers and his FDA coworkers believe, we think that the question that we and other clinicians have to “grapple with at the bedside” is how best to use the various FDA-approved drugs, both for the specific patient in front of us and for the patients we will be treating in a few years who will also need effective antibiotics for the same indications. Dr. Powers and his colleagues at the FDA do not give any guidance regarding these issues when they approve drugs.

Until the direct comparison is completed, clinicians will want information from subgroup analysis in order to help with the difficult clinical decisions regarding which drug to use for MRSA pneumonia. In response to concerns expressed to us regarding differences between VAP and nonintubated patients, we have published the results in that subgroup of MRSA pneumonia patients.12Once again, clinically important information (that the efficacy of linezolid persisted in the sicker VAP patients) was demonstrated. Concern has been expressed regarding vancomycin in the subgroup of patients with renal insufficiency or receiving other nephrotoxic drugs. Conversely, others have suggested that the subgroup of patients infected with MRSA isolates with low minimal inhibitory concentrations can be safely treated with vancomycin.13 Other clinicians have suggested that the dose of vancomycin was inappropriately low10 and that the differences could be explained by underdosing (although Dr. Powers and his colleagues at the FDA demand that only FDA-approved doses be used in clinical trials for registry purposes). All of these reflect legitimate attempts by clinicians to define subgroups of patients in which one drug may be preferable or in which they may be interchangeable; other issues, such as cost and avoiding overuse, become the most significant concern.

The original article1 concluded with the facts (patients with MRSA pneumonia randomized to receive linezolid were associated with lower mortality and higher clinical success rates than those randomized to vancomycin) and a very conservative impression (“linezolid should be considered in patients with suspected nosocomial pneumonia … at risk for … MRSA”). Well-informed clinicians not only recognize the statistical problems of subgroup analysis, but can also tease out what is relevant for patient care from the clinically irrelevant analysis of registry trials required by the FDA.

References
Wunderink, RG, Rello, J, Cammarata, SK, et al Linezolid vs vancomycin: analysis of two double-blind studies of patients with methicillin-resistantStaphylococcus aureusnosocomial pneumonia.Chest2003;124,1789-1797. [CrossRef] [PubMed]
 
Powers, JH, Ross, DB, Lin, D, et al Linezolid and vancomycin for methicillin-resistantStaphylococcus aureusnosocomial pneumonia: the subtleties of subgroup analyses.Chest2004;126,314-315. [CrossRef] [PubMed]
 
Wunderink, RG, Cammarata, SK, Oliphant, TH, et al Continuation of a randomized, double-blind, multicenter study of linezolid versus vancomycin in the treatment of patients with nosocomial pneumonia.Clin Ther2003;25,980-992. [CrossRef] [PubMed]
 
Rubinstein, E, Cammarata, S, Oliphant, T, et al Linezolid (PNU-100766) versus vancomycin in the treatment of hospitalized patients with nosocomial pneumonia: a randomized, double- blind, multicenter study.Clin Infect Dis2001;32,402-412. [CrossRef] [PubMed]
 
Bernard, GR, Vincent, JL, Laterre, PF, et al Efficacy and safety of recombinant human activated protein C for severe sepsis.N Engl J Med2001;344,699-709. [CrossRef] [PubMed]
 
Montori, VM, Guyatt, GH Intention-to-treat principle.Can Med Assoc J2001;165,1339-1341
 
Gonzalez, C, Rubio, M, Romero-Vivas, J, et al Bacteremic pneumonia due toStaphylococcus aureus: a comparison of disease caused by methicillin-resistant and methicillin-susceptible organisms.Clin Infect Dis1999;29,1171-1177. [CrossRef] [PubMed]
 
Fagon, JY, Chastre, J, Wolff, M, et al Invasive and noninvasive strategies for management of suspected ventilator-associated pneumonia: a randomized trial.Ann Intern Med2000;132,621-630. [PubMed]
 
Fagon, J, Patrick, H, Haas, DW, et al Treatment of gram-positive nosocomial pneumonia: prospective randomized comparison of quinupristin/dalfopristin versus vancomycin. Nosocomial Pneumonia Group.Am J Respir Crit Care Med2000;161,753-762. [PubMed]
 
Moise, PA, Forrest, A, Bhavnani, SM, et al Area under the inhibitory curve and a pneumonia scoring system for predicting outcomes of vancomycin therapy for respiratory infections byStaphylococcus aureus.Am J Health Syst Pharm2000;57(Suppl 2),S4-S9. [PubMed]
 
Wunderink, RG, Kollef, MH, Rello, J Linezolid and vancomycin for methicillin-resistantStaphylococcus aureusnosocomial pneumonia [letter].Chest2004;125,315-316. [CrossRef] [PubMed]
 
Kollef, MH, Rello, J, Cammarata, SK, et al Clinical cure and survival in Gram-positive ventilator-associated pneumonia: retrospective analysis of two double-blind studies comparing linezolid with vancomycin.Intensive Care Med2004;30,388-394. [CrossRef] [PubMed]
 
Sakoulas, G, Moise-Broder, PA, Schentag, J, et al Relationship of MIC and bactericidal activity to efficacy of vancomycin for treatment of methicillin-resistantStaphylococcus aureusbacteremia.J Clin Microbiol2004;42,2398-2402. [CrossRef] [PubMed]
 

Figures

Tables

References

Wunderink, RG, Rello, J, Cammarata, SK, et al (2003) Linezolid vs vancomycin: analysis of two double-blind studies of patients with methicillin-resistantStaphylococcus aureusnosocomial pneumonia.Chest124,1789-1797. [CrossRef] [PubMed]
 
Rubinstein, E, Cammarata, S, Oliphant, T, et al Linezolid (PNU-100766) versus vancomycin in the treatment of hospitalized patients with nosocomial pneumonia: a randomized, double-blind, multicenter study.Clin Infect Dis2001;32,402-412. [CrossRef] [PubMed]
 
Wunderink, RG, Cammarata, SK, Oliphant, TH, et al Continuation of a randomized, double-blind, multicenter study of linezolid versus vancomycin in the treatment of patients with nosocomial pneumonia.Clin Ther2003;25,980-992. [CrossRef] [PubMed]
 
Bernard, GR, Vincent, JL, Laterre, PF, et al Efficacy and safety of recombinant human activated protein C for severe sepsis.N Engl J Med2001;344,699-709. [CrossRef] [PubMed]
 
Powers, JH, Ross, DB, Lin, D, et al Linezolid and vancomycin for methicillin-resistantStaphylococcus aureusnosocomial pneumonia: the subtleties of subgroup analyses.Chest2004;126,314-315. [CrossRef] [PubMed]
 
Lachin, JM Statistical considerations in the intent-to-treat principle.Control Clin Trials2000;21,167-189. [CrossRef] [PubMed]
 
Montori, VM, Guyatt, GH Intention-to-treat principle.Can Med Assoc J2001;165,1339-1341
 
 Product labeling for quinupristin-dalfopristin (Synercid). Physicians Desk Reference. 2004; ,.:2186 -2189 Thomson PDR. Montvale, NJ:.
 
Siegel, JP Biotechnology and clinical trials.J Infect Dis2002;185(Suppl),S52-S57
 
Bland, JM, Altman, DG Multiple significance tests: the Bonferroni method. BMJ. 1995;;310 ,.:170. [CrossRef] [PubMed]
 
Kollef, MH, Rello, J, Cammarata, SK, et al Clinical cure and survival in Gram-positive ventilator-associated pneumonia: retrospective analysis of two double-blind studies comparing linezolid with vancomycin.Intensive Care Med2004;30,388-394. [CrossRef] [PubMed]
 
Wunderink, RG, Rello, J, Cammarata, SK, et al Linezolid vs vancomycin: analysis of two double-blind studies of patients with methicillin-resistantStaphylococcus aureusnosocomial pneumonia.Chest2003;124,1789-1797. [CrossRef] [PubMed]
 
Powers, JH, Ross, DB, Lin, D, et al Linezolid and vancomycin for methicillin-resistantStaphylococcus aureusnosocomial pneumonia: the subtleties of subgroup analyses.Chest2004;126,314-315. [CrossRef] [PubMed]
 
Wunderink, RG, Cammarata, SK, Oliphant, TH, et al Continuation of a randomized, double-blind, multicenter study of linezolid versus vancomycin in the treatment of patients with nosocomial pneumonia.Clin Ther2003;25,980-992. [CrossRef] [PubMed]
 
Rubinstein, E, Cammarata, S, Oliphant, T, et al Linezolid (PNU-100766) versus vancomycin in the treatment of hospitalized patients with nosocomial pneumonia: a randomized, double- blind, multicenter study.Clin Infect Dis2001;32,402-412. [CrossRef] [PubMed]
 
Bernard, GR, Vincent, JL, Laterre, PF, et al Efficacy and safety of recombinant human activated protein C for severe sepsis.N Engl J Med2001;344,699-709. [CrossRef] [PubMed]
 
Montori, VM, Guyatt, GH Intention-to-treat principle.Can Med Assoc J2001;165,1339-1341
 
Gonzalez, C, Rubio, M, Romero-Vivas, J, et al Bacteremic pneumonia due toStaphylococcus aureus: a comparison of disease caused by methicillin-resistant and methicillin-susceptible organisms.Clin Infect Dis1999;29,1171-1177. [CrossRef] [PubMed]
 
Fagon, JY, Chastre, J, Wolff, M, et al Invasive and noninvasive strategies for management of suspected ventilator-associated pneumonia: a randomized trial.Ann Intern Med2000;132,621-630. [PubMed]
 
Fagon, J, Patrick, H, Haas, DW, et al Treatment of gram-positive nosocomial pneumonia: prospective randomized comparison of quinupristin/dalfopristin versus vancomycin. Nosocomial Pneumonia Group.Am J Respir Crit Care Med2000;161,753-762. [PubMed]
 
Moise, PA, Forrest, A, Bhavnani, SM, et al Area under the inhibitory curve and a pneumonia scoring system for predicting outcomes of vancomycin therapy for respiratory infections byStaphylococcus aureus.Am J Health Syst Pharm2000;57(Suppl 2),S4-S9. [PubMed]
 
Wunderink, RG, Kollef, MH, Rello, J Linezolid and vancomycin for methicillin-resistantStaphylococcus aureusnosocomial pneumonia [letter].Chest2004;125,315-316. [CrossRef] [PubMed]
 
Kollef, MH, Rello, J, Cammarata, SK, et al Clinical cure and survival in Gram-positive ventilator-associated pneumonia: retrospective analysis of two double-blind studies comparing linezolid with vancomycin.Intensive Care Med2004;30,388-394. [CrossRef] [PubMed]
 
Sakoulas, G, Moise-Broder, PA, Schentag, J, et al Relationship of MIC and bactericidal activity to efficacy of vancomycin for treatment of methicillin-resistantStaphylococcus aureusbacteremia.J Clin Microbiol2004;42,2398-2402. [CrossRef] [PubMed]
 
NOTE:
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

  • CHEST Journal
    Print ISSN: 0012-3692
    Online ISSN: 1931-3543