Affiliations: Denver Health Medical Center, Denver, CO,
Carolinas Medical Center, Charlotte, NC
Correspondence to: Stephen J. Wolf, MD, Department of Emergency Medicine, Denver Health Medical Center, 777 Bannock St, Mail Code 0108, Denver, CO 80204; e-mail: Stephen.Wolf@dhha.org
We read the article, “Prospective External Validation of the Clinical Effectiveness of an Emergency Department-Based Early Goal-Directed Therapy Protocol for Severe Sepsis and Septic Shock,” by Jones and colleagues1in a recent issue of CHEST (August 2007) with interest. Demonstrating clinical effectiveness is a crucial step in establishing the “real-world” benefit of therapies previously shown to be efficacious.3 Although the randomized controlled trial is considered the criterion standard, quasi-experimental study designs have been successfully used to demonstrate effectiveness.
Unlike randomized controlled trials, in which the groups are comparable across all measured and unmeasured variables (assuming true randomization), in before-and-after studies they often are not, and require additional statistical procedures to “control” for such variations. In this study, several variables (eg, end-stage renal disease and intubation) appeared to be significantly different between cohorts and were not controlled for. It is possible that these comorbidities contributed to the observed survival difference.
Additionally, by performing only a before-and-after study it is impossible to account for the effects of changes in practice or patient demographics over time (ie, secular trends). The authors’ conclusions would have been strengthened had a second control phase been conducted with an associated increase in mortality following the intervention phase, thereby bolstering the argument that the observed benefit was from the intervention and not from secular trends.
Finally, by stating, “we demonstrate the clinical effectiveness … of implementing [early goal-directed therapy] EGDT … ” the authors appear to overstate the definitiveness of their results. Even in the absence of the potential systematic error described above, the wide 95% confidence interval surrounding the mortality benefit and the inclusion of no difference within the interval (absolute mortality difference, 9%; 95% confidence interval, – 5 to 21%) implies that random error alone may account for the observed mortality benefit and that there may have been no true effect.4
The authors have reported to the ACCP that no significant conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article.
The authors have no conflicts of interest to disclose.
We would like to thank Dr. Wolf and colleagues for their interest in our work.1 Their first concern was that our choice of study design resulted in unequal measured and unmeasured variables between the groups. While we agree that other study designs may produce a lower possibility of confounding, there is no guarantee of equality between groups with any study design, including a randomized controlled trial. Furthermore, as our study outlines, we made a change in the standard-of-care for all of our emergency department severe sepsis and septic shock patients, thus having a concurrent control group would simply have been unethical. We would also point out that our post-early goal directed therapy (EGDT) group had higher severity of illness and organ dysfunction scores, lower systolic BP, lower oxygen saturation, and both a higher heart and respiratory rate as compared to our pre-EGDT group. Thus, our observed survival advantage was realized in a group with a higher severity of illness and organ dysfunction.
Their second concern was regarding our conclusion of demonstrating the clinical effectiveness of EGDT. Dr. Wolf is of the opinion that we overstated our findings given that the confidence interval surrounding the mortality difference “implies random error alone may account for the observed mortality benefit and that there may have been no true effect.” When examining this assertion we would offer three considerations. First, it is important to recognize that we defined clinical effectiveness a priori as a 33% relative mortality reduction, which is the same mortality reduction in the EGDT efficacy trial.2Second, our findings are similar to every previous full-length EGDT effectiveness publication,3–5 and we are aware of no publication demonstrating a lack of effectiveness of EGDT. Finally, although we are aware of no universally accepted definition of the term random error, we would point out that random error is just what it implies, random, and therefore no study can completely account or control for randomness.
In conclusion, Dr. Wolf and colleagues have every right to not agree with our findings. However, given the publications of the efficacy and effectiveness of EGDT to date, it is difficult to argue with both early resuscitation and a goal-oriented approach (whatever the goals may be) in patients with severe sepsis and septic shock.
Become a CHEST member and receive a FREE subscription as a benefit of membership.
Individuals can purchase this article on ScienceDirect.
Individuals can purchase a subscription to the journal.
Individuals can purchase a subscription to the journal or buy individual articles.
Learn more about membership or Purchase a Full Subscription.
Institutional access is now available through ScienceDirect and can be purchased at myelsevier.com.
Some tools below are only available to our subscribers or users with an online account.
Download citation file:
Web of Science® Times Cited: 2
Customize your page view by dragging & repositioning the boxes below.
Enter your username and email address. We'll send you a reminder to the email address on record.
Athens and Shibboleth are access management services that provide single sign-on to protected resources. They replace the multiple user names and passwords necessary to access subscription-based content with a single user name and password that can be entered once per session. It operates independently of a user's location or IP address. If your institution uses Athens or Shibboleth authentication, please contact your site administrator to receive your user name and password.