Aggressive Sepsis Treatment Lowers Morbidity, Mortality
Back in 2003, after activated protein C received approval for treating severe sepsis and septic shock, I reported on the pathophysiology and treatment of sepsis/SIRS. Since then, there has been much discussion surrounding that important study by Rivers, et al. (N Engl J Med 2001;345:1368.) Simply put, the study demonstrated that septic patients do better if you treat aggressively. This concept has been coined early goal-directed therapy (EGDT).
This study, along with earlier recommendations from the Society of Critical Care Medicine (Crit Care Med 1999;27:723), led to an aggressive approach to septic patients. It has been shown that the first six hours of the patient's treatment has large implications on mortality. (Crit Care Med 2004;32;858.)
Sepsis is defined as a systemic response to infection where there is fever or hypothermia, tachycardia, tachypnea, and evidence of decreased blood flow to internal organs. (Taber's Medical Dictionary. F.A. Davis Company.) Severe sepsis and septic shock are not uncommon, with one study citing an annual average of 282,800 ED visits a year. (Ann Emerg Med 2006;48:326.) Some statistics put it at approximately three percent of all hospital admissions and 10 percent of ICU admissions. (Can Med Assoc J 2005;173:1054.)
Depending on which study one reviews, mortality ranges from 30 percent to 50 percent. The basic pathophysiology concerning sepsis is that SIRS (systemic inflammatory response syndrome) is a host response to either an infectious or non-infectious cause. (Sepsis mimics thyroid storm, toxins/drugs, or neuroleptic malignant syndrome, to name a few.) SIRS includes two or more of the following: temperature greater than 38°C or less than 36°C, heart rate more than 90 bpm (unless on a beta blocker, calcium channel blocker, or a paced rhythm), respiratory rate greater than 20, or leukocyte count greater than 12,000, less than 4,000, or more than 10% bands. Sepsis is defined as two or more SIRS criteria in the presence of a documented or presumed infection. Severe sepsis is further defined as sepsis plus organ dysfunction or hypoperfusion. (Can Med Assoc J 2005;173:1054.)
On a microscopic level, the initial response to the infectious agent is both humoral and cellular with the coagulation and compliment system becoming activated. The endothelium then becomes involved, leading to local ischemia, hypoxia, and increased permeability. This endothelial response is what is thought to lead to the multi-system organ failure often seen in severe sepsis.
Another important concept directly relating to EGDT is oxygen delivery once the above cascade has begun (known as supply-demand mismatch). This is measured by either the Sv02 (using a Swan-Ganz catheter) or Scv02 (using a central venous catheter). Low readings here suggest the cells have to extract a higher percentage of the oxygen delivered, resulting in a lower reading on the venous side. When the supply cannot meet demand, the cells resort to anaerobic metabolism, which results in increased lactate production. Measuring both lactate and Scv02, therefore, are two important components of detecting early sepsis and are crucial in following EGDT.
The most obvious presentation in the septic patient usually relates to blood pressure. Hypotension would be considered by most to be one of the classic criteria of a septic patient. EGDT obviously addresses this by requiring earlier and more aggressive use of vasoactive agents. The protocol suggests 500 ml boluses of either normal saline or colloid up to approximately 2 liters (20-40 ml/kg) with a goal of a central venous pressure of 8-12 mm Hg. A large study compared albumin with normal saline, and found no mortality benefit overall but a trend toward benefit in the sickest patients receiving albumin. (N Engl J Med 2004:350:2247.)
Once the CVP goal is met (in other words, volume resuscitated) and there is continued hypotension, vasopressors are the next step. Mean arterial pressure measurement is what is used in the EGDT protocol, with a target of 65 mm Hg. Although dopamine has been our pressor of choice for years, norepinephrine (Levophed) has had a resurgence recently, partly as a result of a 1998 article showing no end-organ damage in healthy volunteers. (Crit Care Med 1998;26:260.) Therefore, norepinephrine and dopamine are considered first-line agents.
If the patient has persistent hypotension after one pressor, a second is added. Here, one can add either the other first-line agent or epinephrine or vasopressin. Vasopressin has become an agent of interest recently (see current ACLS guidelines), and some advocate its use in sepsis in doses considered physiologic replacement (0.01-0.04 units/min). More about vasopressin will be borne out with the Canadian multicentre Vasopressin and Septic Shock Trial (VASST).
Other recommendations coming out of the Rivers article include early intubation, even with seemingly adequate oxygenation. The thought process behind this is to decrease the work of breathing, to maximize oxygen delivery, and to decrease consumption. Another part of the EGDT protocol is attention to the patient's hematocrit. The recommendation is keeping the hematocrit above 30%, and if below, using PRBC transfusion and inotropic drugs (i.e., dobutamine or milrinone) if the patient is persistently hypotensive after fluids, blood (if necessary), and vasopressors.
Sepsis can cause myocardial depression in approximately 10 percent to 15 percent of patients, and Rivers et al found that adding an inotrope helped raise the MAP. Several other adjunctive therapies which are part of the protocol are tight glycemic control, physiological steroid administration, lung protection while on the ventilator, and the use of activated protein C. In one study, tighter glucose control (80-110 mg/dl vs. 180-200 mg/dl) led to improved morbidity at one year. (N Engl J Med 2001;345:1359.) Low-dose or physiologic-dosed steroids, which have been studied for decades related to sepsis, have been shown to decrease vasopressor requirements (Crit Care Med 1999;27:723) and mortality. (JAMA 2002;288:862.) Lastly, lower tidal volumes (6 ml/kg) also have been shown to decrease mortality. (Circulation 1997;95:1122.) I will discuss activated protein C in next month's column.
Putting all these things together, there are many ways to optimize the patient's outcome.
Isaacs, Lawrence MDEmergency Medicine News:Volume 29(4)April 2007p 3