In patients with unstable circulations one should assure that the airway and breathing are safe (simultaneously and continuously) as you resuscitate the circulation. To examine the circulation, determine if the hypotension is high output or low output (?warm extremities with brisk capillary refill and bounding pulses). If high output=septic shock until proven otherwise; refill the circulation, immediate antibiotics and drain loculated foci of infection. If low-output determine if heart is full (?S3, crackles, elevated JVP). If overfilled=cardiogenic shock; determine pathogenesis and treat the cause. If underfilled=hypovolemic shock; determine cause and refill the system.

The initial step in all approaches to the circulation is to refill the system. In our hospital, we do this empirically: we administer fluid (saline) until clinical signs of overfilling result. If at the outset, it is unclear whether the circulation is adequately filled and/or the patient has a history of heart problems, we place large vein catheters to guide resuscitation (filling to a pressure of 8-10 mmHg). Our preferred route has been femoral vein access, though some recent data suggest that this site may be prone to infection if it remains in beyond the initial resuscitation phase (we take out femoral lines within 24 hours in most cases). A single recent report suggests that outcomes may be improved if central vein catheters are used to sample oxygen saturations of the venous admixture (see reference below); we are awaiting a second confirmatory study before adopting this strategy. If the blood pressure remains low despite fluid administration (>2-3 L) and/or when vasoactive medications are used, we place arterial catheters to monitor blood pressure more accurately and continuously. The current literature suggests that pulmonary artery catheterization (PAC) should not be used routinely. There are no clinical situations in which PAC definitively improves outcomes; in general, we use it when we've formulated a hypothesis, tested the hypothesis with a trial of therapy and things are not improving or worsening. The most common reasonable indication is resuscitation of patients with shock who appear to be adequately filled (CVP>10 mmHg), who are oliguric and having increasing hypoxemia preventing further unmonitored volume resuscitation.

There are no data to suggest that any inotrope or pressor improves outcomes (for any given disorder). There are data to show that "low-dose dopamine" is ineffective (i.e. it does not improve outcomes). In the absence of data we have adopted the following strategy during resuscitations:

While assuring that the circulation is adequately filled, dopamine can be used up to 20 mcg/kg/min (this allows titration of effect from dopaminergic, 1-5 mcg/kg/min, ino- and chronotropic, 5-10 mcg/kg/min, and vasopressor >10 mcg/kg/min). If the circulation has been stabilized (SBP>90 mmHg), then customize management based on the lesion:

1. High output states: squeeze the vessels with an alpha agonist like neosynephrine, vasopressin or norepinephrine;
2. low output, heart full (cardiogenic)=revascularize the heart if appropriate, dobutamine is preferred to dopamine to increase cardiac output when tolerated because it is relatively neutral energetically (less positively chronotropic than dopa and afterload reducing). Amrinone is infrequently used in critically ill patients because of its relatively long duration of action compared to dopamine/dobutamine (hypotension caused by the medication does not reverse immediately upon cessation).