Vital Signs and Clinical Endpoints

Shock has been defined in a multitude of ways, but can best be described as a lack of adequate tissue perfusion, and thereby an impairment of oxygen delivery and removal of waste products. The six basic advanced trauma life support (ATLS) physiologic parameters that have been used to identify shock are heart rate, respiratory rate, blood pressure, urine output, level of consciousness, and pulse pressure. Urine output and level of consciousness are direct measurements of tissue perfusion, and are defined by the classes of shock. Renal blood flow correlates with arterial pressure, but can be subject to significant autoregulation during periods of hypoperfusion. Level of consciousness is less reliable when influenced by intoxication, central nervous system injury, and medication. Heart rate and respiratory rate can be notoriously misleading (Table 1). Anxiety, pain, and stress secondary to the emotional impact of trauma can falsely elevate these physiologic parameters. This can confuse the picture and mask the underlying severity of shock. The diagnosis of shock is best made by observing the body’s main compensatory mechanism: redistribution of blood flow.

Due to austere conditions and inability to measure blood pressure, combat medics in Operation Iraqi Freedom and Operation Enduring Freedom have been trained to resuscitate patients until they are conscious, or when consciousness cannot be assessed, until they have a palpable radial pulse. During World War I, Cannon stated that 75 mm Hg is the critical systolic blood pressure to maintain. For patients with traumatic brain injury, many have advocated avoiding hypotension to ensure adequate cerebral perfusion pressure. The presence of hypotension is predictive of a bad outcome after head injury. Additionally, patients with previous hypertension may display symptoms of organ hypoperfusion despite a “normal” mean arterial pressure (MAP). Due to complex interactions of the patients’ pre-existing disease states and the severity of the injury, there is unfortunately no clear “goal” MAP to determine adequate resuscitation.

Invasive Monitoring

Oxygen delivery is a function of hemoglobin concentration, oxygen saturation, and cardiac output. Hemoglobin concentration and high oxygen concentration are relatively easy to manipulate and monitor, but cardiac output can be more problematic. Adequate cardiac performance is largely a function of preload. Multiple methods have been devised to best estimate the patient’s volume status indirectly by evaluating the venous return to the heart. With the introduction of central venous and pulmonary artery catheters, central venous and pulmonary capillary wedge pressures have been used as better measurements of volume status. While these measurements serve as guides in determining the accuracy and trend of the resuscitation, absolute values should be interpreted with caution. Valvular or global cardiac dysfunction, as well as restrictive pulmonary disease can dramatically alter these measurements.

Newer pulmonary artery catheters (volumetric or oximetric catheters) have the capability for dynamic measurement of additional hemodynamic parameters that were previously unobtainable. Cardiac output can be continuously monitored, and cardiac performance can be evaluated using calculations of ventricular power and end diastolic volume index. A recent study has demonstrated that the right-ventricular, end diastolic volume index (RVEDVI) is a more sensitive measurement of preload than central venous pressure or wedge pressure, particularly in the mechanically ventilated patient. Cheatham et al. demonstrated that cardiac index correlated better with RVEDVI than with pulmonary artery wedge pressure at up to very high levels of positive, end expiratory pressure. In a comparison of RVEDVI with splanchnic perfusion in trauma patients, both Miller et al. and Chang et al. found that supranormal resuscitation to a RVEDVI of greater than 120 ml/m² during shock was associated with better outcomes.

Despite the advances in pulmonary artery catheters (PACs), their effectiveness has been in question since the mid-1990s. Connors et al. published an observational study suggesting that PACs were associated with increased mortality and increased utilization of resources. Despite the study’s limitations, critically ill patients who had a PAC placed had a higher 30-, 60-, and 180-day mortality, increased hospital cost, and longer ICU stays. A recent meta-analysis of 13 randomized controlled trials evaluating PACs showed that the use of PACs did not improve survival or decrease length of stay in hospital. It did not show an increased mortality in the patients with a PAC. The meta-analysis included 5051 patients and is a combination of surgical and medical ICU patients. The results suggest that PACs should not be used routinely in surgical ICU patients unless effective therapies can be found that improve outcomes when used in conjunction with this diagnostic tool.

Less invasive techniques for measuring cardiac performance have been developed. Thoracic electrical bioimpedance measures the resistance of the chest to low voltage currents. It is inversely related to thoracic fluid content, thereby allowing calculation of cardiac output. Several studies have proven that this method correlates well with thermodilution measurements of cardiac output. However, a meta-analysis demonstrated clinical utility in trend analysis but not accuracy for diagnostic interpretation. There can also be significant imprecision with tachycardia or with pathologic fluid collections such as pleural effusions. Transesophageal echocardiography can assess preload and peak velocity measurements as well as continuous cardiac output monitoring. It has been validated with thermodilution techniques. In animal models of hemorrhagic shock, it has accurately reflected the magnitude of change on cardiac output. Pulse contour–derived CO measurement (PCCO) can also continuously measure cardiac output and it correlates with thermodilution. PCCO requires placement of an arterial catheter and a central venous catheter on opposite sides of the diaphragm. It may be a more accurate indicator of cardiac preload than central venous or wedge pressure. Unfortunately, rapid changes in hemodynamic status, including changes in blood pressure and systemic vascular resistance, can alter the monitoring, and recalibration is necessary.

Many have attempted to define the oxygen consumption/delivery endpoint itself, but with no clear results. The Fick equation states that DO₂ and VO₂ are functions of cardiac performance (CI), hemoglobin (Hb), and arterial and venous oxygen saturations (SaO₂ and SvO₂, respectively):

Unfortunately, DO₂ and VO₂ are both derived from cardiac output, and the stable plateau described in Figure 1 is virtually impossible to obtain. Goal-directed therapy, aimed at ensuring CI greater than 4.5 l/min/m², DO₂ index greater than 600 ml/min/m², and VO₂ index greater than 170 ml/min/m², has been advocated by Shoemaker and others, showing reduced morbidity and mortality in critically ill patients. However, Gattinoni in a multicenter randomized controlled trial, and Heyland in a meta-analysis, have shown that no such benefit exists. Furthermore, a prospective, randomized controlled trial by Velmahos comparing conventional versus supranormal endpoints demonstrated that despite all efforts, only 70% of patients were able to reach these endpoints. They concluded that regardless of the resuscitation strategy, the ability of the patient to achieve “optimal hemodynamic values” significantly affected outcome. Looking at O₂ delivery alone, McKinley et al. found that there was no difference in outcome between groups resuscitated to an O₂ delivery goal of 600 ml/min/m² versus 500 ml/min/m².

Figure 1 Relationship between oxygen delivery (DO₂) and oxygen extraction (VO₂).

(From Bilkovski RN, et al: Targeted resuscitation strategies after injury. Curr Opin Crit Care 10:529–538, 2004, with permission.)

Resuscitation to supranormal endpoints has been associated with numerous complications. Shoemaker’s protocol included volume loading with crystalloids and blood, and enhancement of cardiac output with dobutamine. Improvements in blood pressure and cardiac performance by vasoactive drugs can be negated by reduced tissue perfusion, and can often result in tissue ischemia. Hayes et al. found in medical and surgical critically ill patients that the use of dobutamine to augment O₂ delivery may actually increase mortality. Over-resuscitation with crystalloid solutions can lead to the development of compartment syndromes, coagulopathy, and congestive heart failure in patients with cardiac disease.

Lactate

Measurable metabolic endpoints allow the clinician to effectively assess the microcirculation. Accumulation of lactate occurs under anaerobic conditions, and therefore, is a marker of inadequate microcirculatory oxygen delivery. Thus, lactate levels can provide a measure of the extent of shock. While elevated levels indicate worsening degrees of shock, there is no clear cutoff value for when resuscitation is “satisfactory” and adequate oxygen delivery to the tissues is present. Manikas et al. found that initial and peak lactate levels, along with duration of hyperlactatemia, correlated with the development of multiple organ dysfunction syndrome after trauma. Lactate clearance over time is more predictive of mortality than isolated values. Serial lactate levels in trauma patients have been associated with 0%–10% mortality if cleared (lactate level <2 mmol/l) within 24 hours, 25% mortality if cleared by 24–48 hours, and 80%–86% mortality if cleared beyond 48 hours.

Base Deficit

Base deficit has been advocated as a useful clinical marker for assessing reduced tissue perfusion and as a convenient marker of elevated lactate levels. It is defined as the amount of base (millimoles) required to raise 1 liter of whole blood to a normal pH. Davis first classified the base deficit according to severity: mild (2–5 mmol/l), moderate (6–14 mmol/l), or severe (>15 mmol/l). The severity of the deficit directly correlated with the volume of crystalloid and blood replaced within the first 24 hours. It has been shown that serum bicarbonate levels, which may be more readily available, correlate well with base deficits, but they are affected by the patient’s ventilatory status. Similar to lactate, the absolute base deficit can estimate the severity of shock, but no single value can be used as an endpoint. Persistently high or worsening base deficits may be an indicator of complications such as abdominal compartment syndrome or ongoing hemorrhage. Greater severity of base deficit does predict diminished oxygen consumption, increased risk of multiple organ dysfunction syndrome, and greater mortality. However, base deficit secondary to hyperchloremic acidosis is not associated with increased mortality or complications as demonstrated by Brill et al. (Table 2). The inaccurate interpretation of an elevated base deficit that is truly related to hyperchloremic acidosis may result in unnecessary interventions such as ongoing fluid resuscitation, blood transfusion, or even operative interventions. As with lactate, the trend of the base deficit over time is more useful in predicting outcomes.