Clinical review: Circulatory shock - an update: a tribute to Professor Max Harry Weil

Circulatory shock is common and associated with high morbidity and mortality. Appropriate shock treatment relies on a good understanding of the pathophysiological mechanisms underlying shock. In this article, we provide an update on the description, classification, and management of shock states built on foundations laid by Dr Max Harry Weil, a key early contributor to this field.

vasoconstriction caused by the activated sympathetic nervous system may mask the fall in blood pressure. Th e usual lower limit for systolic arterial pressure is considered as 90 mmHg, but this is an arbitrary value and may vary from one patient to another -for example, the pressure threshold may be lower in younger than in older individuals.
Weil and colleagues highlighted the importance of blood lactate concentrations in patients with shock many years ago [7,8], and lactate concentrations remain one of the most useful biological tests in this setting. Normal concentrations are around 1 mEq/L (or mmol/L), and a value above 2 mEq/L is considered to refl ect the presence of shock ( Figure 1). Importantly, in a recent study, mortality was increased even in those who had small increases in lactate concentration to between 1.5 and 2.0 mEq/L [9]. Although generally associated with anaerobic metabolism, raised lactate concentrations may also occur as a result of excessive aerobic glycolysis (for example, during shivering, seizures, hyperventilation) and/or decreased utilization (for example, liver failure, mitochondrial inhibition). Nevertheless, in the context of altered tissue perfusion, the severity of hyperlactatemia is directly related to outcome [10,11]. In addition to single measurements, changes in lactate concentrations over time may have additional predictive value for organ failure and mortality [12].
When assessing the damage an earthquake or fi re has caused inside a building, one looks through the windows. Using this analogy, it would be useful to be able to see inside the body to view the damage caused by the shock process. Clearly this is not possible, but the skin, the kidneys and the brain provide us with three types of 'window' through which we can see the eff ects of the altered tissue perfusion: through the skin 'window' , we can see decreased capillary fl ow, slow refi ll, cold and clammy skin [13]; through the kidney 'window' , we typically see oliguria <0.5 mL/kg/h; and through the brain 'window' , we see obtundation/disorientation/ confusion that was not present before the shock episode ( Figure 2). Unfortunately, we currently have no other 'windows' (for example, it would be nice to visualize the gut and liver, but this is not possible practically; gastric tonometry and dye-clearance were tried but their role was never clearly defi ned and they are not commercially available). Th e sublingual microcirculation (see below) may provide us with a new 'window' . Th is 'window' has been used in several studies [14][15][16][17][18] and has been shown to have high sensitivity for identifying the presence of shock and response to therapy [15]. Although current equipment is not yet suitable for routine clinical use, improvements in the available technology together with supportive clinical trials may make this a valuable window to identify and treat diff erent states of shock. Figure 3 shows the interaction between arterial pressure, altered tissue perfusion, increased lactate and microvascular alterations.

Classifi cation of shock states
Shubin and Weil [19] defi ned the pathophysiological states of circulatory shock many years ago, using a classifi cation based on four mechanisms (Table 1, Figure 4). In the fi rst three types, cardiac output is low. In fact, each of the three types is represented by one of the determinants of cardiac output: decreased preload (hypovolemic), altered contractility (cardiogenic), and increased afterload (obstructive). In the fourth type of shock, the distributive defect is the result of the release of many mediators, including cytokines. Th ese mediators can have vasodilating and vasopressor eff ects, although vasodilating eff ects predominate in the central circulation. Some of these mediators decrease myocardial contract ility, accounting for the myocardial depression associated with sepsis. Despite this myocardial depression, distributive shock in humans is generally associated with an increase in cardiac output. Th ere is also microvascular obstruction because of activated leukocytes and platelets impairing the distribution of blood fl ow in the periphery. Moreover, because of a defect in the microvasculature, autoregulatory mecha nisms are no longer eff ective in matching oxygen need to oxygen supply and there is an increased shunt of the microcirculation. Th e resultant increased heterogeneity of micro circulatory perfusion creates areas of no fl ow in close proximity to areas of fl ow. Importantly, although the focus of this review is circulatory shock, it must be appreciated that infl ammatory mediators and oxidative stress from circulatory shock and reperfusion or due to other factors (for example, sepsis) can also directly cause tissue injury.
Put in very basic terms, something is wrong with the pump (cardiogenic), with the volume (hypovolemic), with the major vessels (high afterload/obstruction) or with the small vessels (distributive/shunting).
Importantly, diff erent types of shock may co-exist. For example, in sepsis there may be a combination of distributive, hypovolemic (sweating, diarrhea, extravasation and so on) and even cardiogenic forms; in anaphylactic shock the same pattern may be present, that is, distributive and hypovolemic (due to severe permeability alterations) with altered myocardial contractility.

Microvascular alterations
Microvascular alterations are common in all shock states. In distributive types of shock we expect these changes, but they can also be observed in cardiogenic shock states [20].
Microcirculatory alterations caused by pathogenic factors and hemodynamic changes are critically involved in the eff ects of shock on organ function as oxygen transport to the cells becomes compromised due to limitation of convective (fl ow) and/or diff usive (increased distance between cells and red blood cell-carrying capillaries)  transport of oxygen to the tissues [21]. Cellular alterations of the microcirculation include endothelial dysfunction [22], changes in the hemorheological properties of red blood cells [23], leukocyte activation, coagulopathy and vascular smooth muscle cell alterations causing autoregulatory dysfunction. Endothelial glyco calyx shedding [24], which is highly sensitive to oxidative stress, contributes to the compromise of the endothelial vascular barrier, resulting in tissue edema [25]. From this pers pective, the microcirculation could indeed be regarded as a target of shock. Microcirculatory areas with obstructions are shunted, resulting in patchy, hetero geneous hypoxic areas [26]. In addition, cellular changes occur, involving mitochondrial depression [27]. Although this had been clearly identifi ed from animal studies, the true extent to which the above occurred in the clinical setting remained unclear until the late 1990s when the introduction of hand-held video microscopes allowed direct bedside observations of the microcirculation [28]. Heterogeneity of microvascular fl ow among organs and within the microcirculation, independent of systemic hemodynamic variables, is a characteristic of the micro circulatory altera tions seen in human sepsis [16] and capillary obstruction is observed in the presence of normal fl ow in larger vessels [14,29]. Th ese observations are a direct demon stration of the presence of shunting occurring at the microcirculatory level and give new credence to Dr Weil's appreciation of circulatory shunt ing as being a key feature of distributive shock [19].
Importantly, studies have demonstrated that persistent sublingual microcirculatory alterations are associated with adverse outcomes in patients with septic shock [29], and that resuscitation therapies, which are eff ective in the early recruitment of the microcirculation, can improve organ function and outcome in septic shock patients [15,18]. It may, therefore, be that, for early goaldirected therapy to be eff ective in patients with shock, it must be able to recruit the microcirculation. However, current technology for monitoring the microcirculation is not yet ready for the clinical arena and further clinical trials in diff erent patient groups are needed before the microcirculation can really present itself as a window to monitor and treat shock.

Principles of therapy
Dr Weil introduced the VIP rule (V for ventilate, I for infuse, and P for pump) many years ago [30] for the initial resuscitation of shock, but it is still relevant today.

Ventilation
Adequate oxygenation is of course essential but there is some debate about the use of excessive PaO 2 with suggestions that it may alter the microcirculation,  primarily by inducing vasoconstriction and generating oxygen radicals. After cardiac arrest, in particular, high PaO 2 may be deleterious [31]. Th e problem in shock is that the widely used, readily available indication of arterial saturation, pulse oximetry, may not be reliable because of the altered skin perfusion that occurs with major vasoconstriction; hence, to avoid the well-known risks associated with hypoxia, we tend to be relatively generous with oxygen administration. Importantly, if there is any question about whether or not a patient needs endotracheal intubation, then this procedure should be performed and not delayed. Non-invasive mechanical ventilation should be used with caution in hemodynamically unstable patients.

Infuse
Fluids should not be withheld based just on the presence of edema, because edema formation may be the result of extravasation of fl uids, which decreases blood volume, an eff ect demonstrated by Dr Weil decades ago [32]. It is also well known that static values of eff ective fi lling (pressures or volumes) are a poor predictor of the response to fl uids [33,34], so fl uids should not be withheld based on these measures. Fluid challenges with pre-set limits can help determine the need for ongoing fl uid infusion, as suggested by Dr Weil [35]. Optimal choice of fl uid remains debated, although recent studies in patients with severe sepsis suggest that 4% albumin solutions may be of benefi t compared to normal saline [36] and hydroxyethyl starch solutions may increase mortality compared to Ringer's acetate [37]. Th ere is some controversy about the use of saline solutions in the presence of severe metabolic acidosis, because of the chloride load.

Pump
Pump eff ectively refers to the use of vasoactive agents. Vasopressors should be given fi rst to maintain a minimal perfusion pressure, even if there is cardiogenic shock, because dobutamine administration may result in hypotension if there is any degree of hypovolemia. Vasopressors are generally started early, at the same time as fl uids, but patients are weaned from vasopressor support as soon as possible. Norepinephrine is preferred over dopamine, as it is associated with lower mortality rates in cardiogenic [38] and in septic [39] shock.

Conclusion
Dr Weil set the basis for much of today's current knowledge of circulatory shock. Th erapy for shock should be based on pathophysiological alterations rather than on protocols. Monitoring of shock relies on assessment of arterial pressure, cardiac output, tissue perfusion abnormalities, and blood lactate concentrations. Monitoring of the microcirculation may help, but further study is needed to confi rm this.