Key Point:

Most clinicians view sepsis as undesirable, but there are data to suggest that it may actually benefit critically ill patients. Does sepsis “rev up” the body’s defenses, thereby allowing the host to survive an infectious assault? Or is it an example of the body turning against itself?

ORLANDO, FLA—Most physicians would prefer that sepsis not develop in their critically ill patients, but not John C. Marshall, MD. “Sepsis is good for the patient,” asserted Dr. Marshall during a pro-con debate at the annual meeting of the Society of Critical Care Medicine.[1]

Sepsis, he explained, is the host response to invasive systemic infection; its clinical manifestations, such as tachycardia, tachypnea, and hyperthermia, “rev up” physiologic defenses that combat the infectious process and allow the host to survive.

At the molecular level, gram-positive and gram-negative infections are recognized by toll-like receptors (TLRs), primarily TLR-2 and TLR-4. At the cellular level, infection precipitates the release of oxygen intermediates, proteases, and other nonspecific killer cells that play a part in eliminating infectious microbes. In addition, the coagulation cascade may be activated to help contain the infection and prevent involvement of remote organs, although at a cost of reduced tissue perfusion.

Infection is also associated with the production of beneficial inflammatory mediators, such as tumor necrosis factor (TNF), which has been shown to induce cell death. By decreasing local vascular tone, nitric oxide facilitates the influx of TNF and other cells necessary for host defense into areas of infection.

These host responses can be detrimental when they occur in the absence of infection, a situation that can ultimately lead to tissue injury in the form of necrosis or apoptosis, acknowledged Dr. Marshall, who is a Professor of Surgery at the University of Toronto. In patients with infection, however, these responses are entirely appropriate, he maintained.

The hypothesis that benefit can be effected by interfering with the host response to infection is biologically implausible, he claimed. Indeed, a trend toward increased mortality was reported in patients with gram-positive infections who were treated with a monoclonal antibody to endotoxin. Similarly, neutralizing TNF with a monoclonal antibody appeared to have little, if any, meaningful effect on survival in patients with infections.

Interestingly, in animal models of intestinal ischemia and reperfusion injury—a condition not caused by infection—survival increased significantly in the animals that were treated intratracheally with killed Escherichia coli; Dr. Marshall speculated that this treatment provided the microbial stimulus necessary to induce apoptosis in the neutrophils that had accumulated in response to the injury. Thus, it can be argued that infection may actually assist in “turning off” the inflammatory response, he said.

“Much of the harm that we attribute to the innate immune response actually results from what we as physicians do to try to sustain physiology,” suggested Dr. Marshall. For example, the histologic findings associated with acute respiratory distress syndrome may have as much to do with mechanical ventilation as they do with the underlying disease, he posited.

“SEPSIS IS BAD FOR PATIENTS”

The American Heritage Dictionary defines sepsis as a poisoned condition resulting from the presence of pathogens or their toxins, related Edward Abraham, MD, Dr. Marshall’s opponent in the debate. “That clearly is not good at all,” stated Dr. Abraham, a Professor of Pulmonary and Critical Care Medicine at the University of Colorado Health Sciences Center in Denver.

Sepsis is associated with many undesirable clinical findings and adverse outcomes, Dr. Abraham pointed out; these include arterial hypotension, organ failure, and increased mortality. In animal models, sepsis has been shown to decrease resistance to secondary Pseudomonas aeruginosa colonization, leading to both lung infection and bacteremia.

In other animal models, sepsis-triggered elevations in pro-inflammatory cytokine concentrations and, surprisingly, the accompanying increase in levels of the anti-inflammatory molecule interleukin 10 did nothing to attenuate the inflammatory response. In fact, blocking interleukin10 increased survival in these models.

A similarly dysregulated inflammatory response has been observed in humans with sepsis, noted Dr. Abraham. In these patients, elevated levels of the pro-inflammatory cytokines nuclear factor kappa B (NF-kappa B) and TNF have been shown to significantly increase mortality; treatment with anti-TNF antibodies reduces sepsis-related mortality by about 10%. Due to its central role in signaling pathways that are important in sepsis, NF-kappa B may also be an appropriate therapeutic target.

In a study that is awaiting publication, the presence of endotoxemia was linked to greater ICU mortality in a group of primarily septic patients, Dr. Abraham added. The lesson to be learned from these findings, he stressed, is that the biochemical mediators released in sepsis contribute to an increased risk of death.

—Timothy Begany

Reference
1. Marshall J, Abraham E. Pro/Con: Sepsis is good for the patient. Presented at: 33rd Critical Care Congress; February 22, 2004; Orlando, Fla.

Return to table of contents