| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
Dr. Manthous is Assistant Clinical Professor of Medicine, Divisions of Pulmonary and Critical Care, Bridgeport Hospital and Yale University School of Medicine.
Correspondence to: Constantine A. Manthous, MD, FCCP, Bridgeport Hospital, 267 Grant St, PO Box 5000, Bridgeport, CT 06610; email: pcmant{at}bpthosp.org
In this issue of CHEST (see page 855), Dr. Paul Marik presents a clinical approach to fever in critically ill patients. This concise review reiterates that all fevers are not necessarily caused by infections and presents a reasonable, albeit not evidence-based, algorithm for approaching this common clinical problem. At some hospitals, intensivists and infectious disease physicians may differ in their general approaches to fever in the ICU. There appears to be a pervasive notion that intensivists administer antibiotics, and particularly powerful broad-spectrum antibiotics, more liberally than infectious disease (ID) experts. I am unaware of any data to support this contention. Nonetheless, if this is true, and it is reasonable for each of us who practices critical care medicine to contemplate, we may be contributing to the problem of multidrug resistance, a modern plague in the ICU. It is also worthwhile for us to think about why it may be so (if it is).
Sepsis may be defined as the systemic effects of an infectious organism and/or its toxins on a host. So the major determinants of the development of sepsis are factors related to the organism (eg, virulence and innoculum size) and to the host (eg, immunocompetence, mechanical clearance). Whether a patient develops sepsis, sepsis syndrome and, finally, septic shock, and the rate of that progression depend on the balance of these factors and when, in the temporal progression of sepsis, antibiotics are administered to tip the balance in favor of the host. Few young healthy women develop septic shock from cystitis, a relatively common infectious disease. However, most experienced clinicians have seen patients with diabetes, alcoholism, and what starts as a simple urinary tract infection develop fulminant untreatable septic shock unto death. Even worse, most of us have witnessed young healthy people with no coexisting diseases die rapidly of fulminant meningococcal septic shock, despite administration of appropriate antibiotics.
There are two general approaches to selection of antibiotics in infectious diseases. One approach is to carefully obtain the history and physical examination, to gather appropriate specimens for staining and culture, and to await culture results that guide therapies. The second approach posits that in some situations, one cannot wait for culture results; that waiting could allow the infection to become excessively advanced. The second approach is to treat broadly, with antibiotics, for the most likely pathogens (based on the available data), and when cultures and sensitivities return, to narrow the coverage. Obviously, the two approaches are not mutually exclusive. The choice of antibiotics should be based on efficacy, toxicity, and cost in that order of priority. Recent data suggest that it may also be important to consider the effects of chosen antibiotics on development of resistant pathogens.1 It appears that the schism between ID specialists and intensivists is that the former gravitate toward awaiting culture/sensitivity results and the latter toward empiric therapy. Moreover, and perhaps as importantly, we must consider whether intensivists are more likely to approach isolated fever as an infectious disease.
Our Chief of Infectious Diseases mentioned to me that "everyone knows how to start antibiotics"; when he assesses a case, he also tries to consider reasons not to start them. I tend to use broad-spectrum antibiotics earlier and more liberally than he for empiric therapy of fever and a questionable site of infection. Why are we so different? I suspect that we (intensivists and ID specialists) sometimes differ in our approaches because of inherent differences in our patient populations. Infectious disease physicians care for the full spectrum of severity of infectious diseases, and critically ill patients comprise a minority of most practices. Don’t get me wrong; that doesn’t imply that intensivists know better. But, intensivists care for patients, at the extreme of the spectrum, with very severe perturbations of host cardiopulmonary status, in whom there is less physiologic margin for delay or error. Our patients frequently have comorbid conditions that render them less than fully immunocompetent. Endotracheal tubes serve as unnatural conduits for microbes to the lower airways, Foley catheters provide routes of entry to the urinary tract, and IV lines into the vessels and heart. Natural mechanisms of protection such as cough, gag, swallow, mucociliary clearance, and gut permeability to microbes may be altered. Sepsis is the leading cause of mortality among our patients; we do share this with our infectious disease colleagues. The inherent relative fragility of critically ill patients may warrant an earlier, more empiric "attack" than, say, would be warranted in a well person with an uncomplicated infection. In many cases, this may warrant initial selection of an antibiotic with the broadest spectrum available. And, ignoring the potential for selecting resistant pathogens, some have suggested that this is a cost-effective strategy.2 However, are we so hypervigilant that we see the monster behind every fever? And do we inappropriately use a bazooka to kill a mosquito (which sometimes isn’t even there)? And do we continue the bazooka even when the sensitivity results subsequently suggest a less-destructive weapon would be as effective? Obviously, the answers to these questions are different for every practitioner. We devise treatment pathways and antibiotic "formularies," of unproven efficacy, in the hope of reducing individual caprice. But ultimately, each case will be different, and thresholds for when to work up and treat fever and how hard to hit will depend on many of the variables (of host and microbe) listed above. To a certain degree, each physician’s recent experience may also impact the level of aggression (eg, if one of our patients recently died of a resistant Escherichia coli, maybe we would be more inclined to cover our next case with a "bigger gun" until the sensitivity report returns).
Finally, this "lower threshold" phenomenon may also affect our propensity to "work up" fevers. I must confess also that my trainees order seemingly innumerable cultures nearly every time a patient "spikes a fever," without a complete thoughtful approach guided by data derived from the bedside. Perhaps, I have not been vigilant in correcting them and may be contributing to a next generation of physicians who will abuse laboratory resources. I’ve been meaning to create (or borrow) an algorithm or policy or some such pathway, but haven’t gotten around to it yet.
I suggest that intensivists should consider these issues carefully. Begin by reading Paul Marik’s article. Then read "Practice Guidelines for Evaluating New Fever in Critically Ill Adult Patients,"3 which was formulated by a joint task force of ID and critical care specialists. Then, examine your practice patterns, and in future cases, balance the good (to the individual patient) of attenuating sepsis by a thoughtful workup and early administration of antibiotics against the evil (to the individual and the tribe) of treating every fever, thereby contributing to multidrug resistant organisms. There may be no categorically right answer. However, such self-examination may reduce the likelihood of "antibiotic abuse" while maximizing what is, after all, our shared goal: helping patients to return to their families in good health.
References
This article has been cited by other articles:
|
|
 |
|
  P. Eggimann and D. Pittet Infection Control in the ICU Chest, December 1, 2001; 120(6): 2059 - 2093. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
