The identification of both occult and inadequately resuscitated shock in critically ill and injured patients continues to be a major clinical problem. Occult shock (that is, shock that is not immediately clinically apparent) is of particular concern in the care of elderly trauma patients (who may be in early sepsis and are frequently characterized by multiple comorbidities and/or medications that may mask the conventional signs and symptoms of shock), and wounded war fighters (where diagnostic and treatment resources are limited). Shock occurring in even the relatively young and healthy victim of blunt trauma (the classical trauma patient) may be difficult to recognize because of occult hemorrhage occurring in the thorax, abdomen, retroperitoneum, pelvis, or soft tissue.
Part of the difficulty inherent to shock assessment is that there is still considerable controversy as to what criteria determine the most appropriate end points for assessing shock depth and duration. Conventional vital signs are clearly inadequate (1), and attempts to gauge the extent of shock and resuscitation using more physiologically relevant measures of perfusion such as oxygen delivery (2) have led to equivocal results (3, 4). Most resuscitation strategies seem to be heavily weighted toward efforts to restore normal oxygen delivery to the tissues (5), whereas much of current resuscitation research has centered on methods of controlling inflammation or coagulopathy (6-10). Perhaps what has been lost in translation is the fact that inflammatory and coagulation systems (once thought separate) are instead a single system that is uniformly and rapidly activated and modulated by tissue injury and hypoperfusion (6, 11). The degree to which this system is activated and subsequent occurrence of complications are unlikely to be simply corrected either by using oxygen delivery as an end point in itself or by modulation of certain specific pathways. We suggest that all these efforts have lost sight of the major physiological underpinnings of the shock state. Instead, we propose a return to three fundamental physiological principles underlying shock and shock treatment: 1) prevention of further oxygen debt accumulation, 2) repayment of oxygen debt, and 3) minimization of the time to oxygen debt resolution.
There is a considerable literature on the subject of oxygen debt and shock-the reader is referred to an excellent review on the topic (12)-so we provide only an outline of the relevant information required to understand the next two principles. Principle 2 is an obvious extension of principle 1, but there have been very few experimental or clinical investigations concerning this concept, and we expand on this topic below. Principle 3 is more novel and will be the emphasis of this review. The time to oxygen debt resolution will be referred to hereafter as the oxygen debt repayment schedule.
THE PHYSIOLOGY OF SHOCK AND REPERFUSION: THE BASICS
Shock is a state of hypoperfusion at the cellular level that occurs when the delivery of oxygen (DO2) to the tissues falls below the tissue oxygen consumption (VO2) requirements, and thus represents an imbalance or mismatch between tissue DO2 and VO2. Oxygen delivery is dependent on blood flow (traditionally assessed globally by cardiac output) and arterial oxygen content. Clinically, multiple organ dysfunction is associated with a persistent inadequate balance of DO2 and VO2 of specific tissue or organ beds. Conventionally, perfusion status is assessed by whole-body end points such as mental status and the standard cardiovascular parameters of heart rate, palpable pulses, and systemic blood pressure (5). However, data from both animal models and clinical studies indicate that these measures are very poorly correlated with perfusion of specific tissue beds (1). Thus, organ beds may have inadequate DO2 even if gross systemic hypotension has been corrected. As a result, even if the subject is normotensive, unequal distribution of DO2 to various tissue beds may result in isolated organ ischemia before the occurrence of whole-body ischemia. The gut in particular seems to be especially susceptible to ischemic injury; there is increasing evidence to suggest that ischemic changes in the gut drive the systemic activation of inflammatory cascades (13). Continuing systemic hypoperfusion has been implicated in ischemic cellular injury and cell death, which, unless corrected, leads to systemic inflammatory response syndrome and irreversible multiple organ dysfunction syndrome (MODS). Although the total incidence of MODS has decreased over the last several decades, MODS remains a leading cause of late morbidity and mortality in trauma, and the mortality rate still remains high (50%-80%) (14).
Both animal and clinical data support the findings that first, late outcome is strongly related to both the severity and duration of shock, and second, oxygen debt and its metabolic surrogates are the best predictors of outcome (12). To understand the concept of oxygen debt, it is useful to describe the relationship between oxygen delivery and oxygen consumption during normal perfusion and in shock. In the normal healthy subject, whole-body oxygen consumption (VO2) is independent of cardiac output (and, hence, DO2) because of the ability of the tissues to modulate oxygen extraction from the blood at the level of the microcirculation. However, if DO2 is decreased below a certain threshold (critical oxygen delivery, DO2crit), extraction is no longer adequate, and VO2 declines in proportion to the reduction in DO2; ischemic metabolic insufficiency then follows. A marker of this insufficiency is the increase in the concentration of metabolites such as lactate in the peripheral blood (Fig. 1).
Approximately 20 years ago, it was suggested that organ failure resulted from the subject crossing this critical DO2 threshold to a prolonged and often unrecognized state of flow-dependent VO2 (15). Accordingly, it was proposed that resuscitation should involve "pushing" DO2 back over this threshold so that VO2 reaches the flow-independent "plateau" (15, 16) (Fig. 1). In the clinical setting, DO2crit cannot be measured easily and will vary widely among patients; as a result, the concept of "supranormal" resuscitation as a therapeutic goal soon followed. Supranormal resuscitation involved provision of fluid and inotropic support to ensure that DO2crit could be exceeded, and therefore the patient could be maintained at an oxygenation level commensurate with the flow-independent plateau. Supranormal levels were defined a priori as the median values of hemodynamic and oxygen transport variables values observed for survivors in a previous clinical evaluation of critically ill, high-risk surgical patients (17). The use of supranormal oxygen delivery as a metric was at one time particularly widespread in treatment of septic patients; this approach was subsequently applied to both sepsis and trauma patients who had in all probability already accumulated substantial oxygen debt but for whom debt could not be readily measured or quantified (3, 18). Although initial data indicated patient outcomes improved with early optimization to supranormal DO2 (15), subsequent studies have not shown any advantage over more conventional resuscitation protocols (3, 4). A further concern was that, in practice, the volume of crystalloid administered during shock resuscitation did not seem to be closely coupled to the levels of DO2 actually measured in patients. Supranormal delivery protocols were characterized by substantially larger amounts of crystalloid required to achieve the higher DO2 end point, with subsequent risks of fluid overloading, volume shifts (3), and inflammatory responses (19). Possible reasons for these failures will be discussed in the following sections. It seems that driving resuscitation DO2 to supranormal values is now less popular as a widespread resuscitation goal. In cases of traumatic hemorrhagic shock, current Advanced Trauma Life Support guidelines (5) state that the goal of resuscitation is a "balance" between restoration of organ perfusion via fluid infusion and minimization of the risk of rebleeding; suggested end points are blood pressure and recommendations for the serial monitoring of base deficit and/or lactate. Preload-directed resuscitation with sequential fluid boluses remains the standard of care for trauma resuscitation in the intensive care unit (20). These strategies, although nominally based on principles of oxygen transport, often are inadequate in practice, as will be demonstrated below.
OXYGEN DEFICIT AND OXYGEN DEBT
Clearly, oxygen delivery alone is still not an ideal indicator of either perfusion status or the adequacy of resuscitation. To resolve this problem, we must consider what is actually being measured during the trajectory of shock and resuscitation. When DO2 is reduced below DO2crit, an oxygen deficit is incurred because the amount of oxygen demanded by the tissues is inadequately matched by supply; this is the standard definition of shock. Therefore, oxygen deficit can be calculated as the difference between baseline "normal" oxygen consumption VO2, and the VO2 measured at a given time during the shock period. However, because there is a significant associated time dimension, shock cannot be evaluated merely by the oxygen deficit "snapshot" of perfusion status at any one time; the shock state must account for the amount of deficit accumulated over time from the point of injury. Deficit accumulated over time is debt. In other words, oxygen debt is the accumulation of multiple oxygen deficits over time and thus represents the sum of all deficits incurred. As an example, suppose that baseline VO2 (an estimate of tissue oxygen demand) is 200 mL min−1 and is followed by a reduction in VO2 by slightly more than one third to 134 mL min−1. Because oxygen deficit is the change in VO2 from baseline, oxygen deficit is therefore equal to the difference between baseline VO2 (VO2,0) and the VO2 at this new time point t, or
In this example, the reduction in VO2 results in an oxygen deficit of (200 − 134) = 66 mL min−1. If this deficit is sustained for a period of 1 h, the resulting oxygen debt will be equal to the product of the oxygen deficit integrated over time (66 mL min−1 × 60 min) or 3.96 L. Figure 2 further illustrates the distinction between oxygen deficit and oxygen debt and how there may be little correlation between deficit at any one time and the amount of debt already accumulated. At time point 1, the deficit is 30 mL min−1, and debt incurred to that point is 359 mL. At time point 2, deficit is at a maximum (48 mL min−1), and debt is increasing. At time point 3, deficit is beginning to resolve with resuscitation and is equivalent to that measured at t1; however, debt is still increasing, although the rate at which it is accumulating is declining. At time point 4, deficit is 0, and no more debt is being incurred; however, the overall oxygen debt is still substantial (1,871 mL).
The concept of oxygen debt has been known since the early 1960s but has not been applied uniformly in the clinical setting. Oxygen debt has been shown to be the only physiological variable that can quantitatively predict survival and the development of multiple organ failure after hemorrhage (17, 21, 22). Implicit in the concept of oxygen debt is that the probability that multiple organ dysfunction and death are influenced primarily by the accumulated debt. Early animal experiments indicated that there was a minimum threshold of oxygen debt below which all animals survived and above which mortality increased until a universally lethal threshold of debt was attained (23). Subsequent animal and clinical studies showed that increasing probability of mortality was directly associated with total oxygen debt, and this debt could be estimated from key metabolic markers, namely, base deficit and lactate (12, 22, 24). It follows that if resuscitation is initiated before a clinically significant oxygen debt is incurred and the debt is then repaid, cellular damage will be slight or nonexistent. Conversely, the likelihood of cellular damage and subsequent organ failure is substantially increased if the period of increased oxygen debt is prolonged and/or resuscitation is inadequate (i.e., failure to repay oxygen debt) (12, 21). Although there are limited data as to the precise sequence of cellular events that occur during debt repayment in the clinical setting, evidence of shock resolution should consist, at a minimum, of the complete repayment of oxygen debt.
Unfortunately, none of the original oxygen debt studies made any assumptions as to the timeframe within which accumulated debt is to be "forgiven" or repaid. In theory, morbidity and/or mortality should not be affected by the repayment schedule as long as no more debt is allowed to accumulate. However, in practice, it is likely that debt repayment will be slower when lower volumes of resuscitation fluid are administered or if there is a delay in the onset of definitive resuscitation. It has been observed that prolonged hemorrhagic shock coupled with inadequate resuscitation causes a relatively small proportion of immediate deaths but nevertheless accounts for more than one fourth of hospital deaths, primarily from organ failure (25). There is now evidence that outcome from traumatic brain injuries is greatly worsened if resuscitation is inadequate (26). This has profound implications for war fighters because traumatic brain injury is the signature injury of the current military conflicts in Iraq and Afghanistan. The recent push toward both low-volume (and hypotensive) and delayed resuscitation in the prehospital environment means that it is even more important that we reevaluate these resuscitation strategies in terms of debt repayment schedule.
CASE STUDY: A CANINE MODEL OF OXYGEN DEBT REPAYMENT
There are few data that pertain to the potential effects of oxygen debt repayment schedule on mortality and morbidity resulting from hemorrhagic shock. The only systematic experimental study of which we are aware is a previously reported canine model of hemorrhagic shock described by Siegel et al. (27). This study represents a unique effort to characterize oxygen debt repayment during early resuscitation at the level of the whole organism. In brief, this model was essentially an "assay" of different volumes of colloid fluid (5% albumin) administered during the immediate posthemorrhage period and their relative effectiveness for maximizing survival and minimizing organ damage. This model was relatively severe: dogs were hemorrhaged to a predicted mortality of 50%. The resulting oxygen debt of approximately 104 mL O2 kg−1 resulted in a realized mortality of 40% within the first 3 days postshock. Fluids were administered postshock according to one of five randomly assigned resuscitation regimes: "full" resuscitation (120% of shed blood volume [SBV] replaced by fluid during a 20-min period), no fluid (0% SBV) for 2 h postshock, followed by full resuscitation, or by three "partial" resuscitation regimes (8.4%, 15%, and 30% SBV during a period of 10 min, followed by the remaining SBV as colloid at 2 h posthemorrhage). Outcome was scored according to the amount of physiological compromise incurred at 7 days posthemorrhage, as determined by organ function and histological examination; a good outcome indicated recovery at 7 days with intact motor and neurological function, stabilization of vital signs, and minimal to no cellular damage. Although survivors did not differ appreciably in visible clinical signs of impairment, cellular damage in liver and kidney was most severe in surviving animals receiving the lowest partial immediate resuscitation volumes (8.4% and 15% SBV) and minimal to absent in animals receiving at least 30% SBV immediately postshock. None of the animals in the 0% immediate resuscitation group survived.
IMPLICATIONS FOR RESUSCITATION TIMING AND FLUID VOLUMES
If oxygen debt repayment alone determines mortality and morbidity, there should have been no differences in outcome between groups of animals receiving 15%, 30%, or 120% SBV immediately after the shock period. This is because these volumes of fluid were sufficient to prevent additional debt accumulation, and any remaining volume was returned at 2 h so that debt was eventually repaid in all groups. However, histological and organ function data showed that there was instead a graded response in outcomes; cell and organ damage was successively minimized with each volume increment returned immediately posthemorrhage. Accordingly, we determined the cell damage threshold by estimating the percent of oxygen debt repaid for each hemorrhage group from the reported values and comparing to the chart of histological outcomes (Table 1). The percentage of oxygen debt repaid over 2 h for each group was estimated from the ratio of oxygen debt repaid to that incurred. For example, animals in the 30% SBV group incurred a debt of 108.8 mL kg−1 by the end of hemorrhage. They received approximately 30% SBV immediately posthemorrhage, which repaid approximately 80 mL kg−1, which is approximately 80/108.8 or 74% of the accumulated debt over 2 h.
Two deductions can be inferred from this study. First, our calculations suggest that it was in fact the amount of debt repaid within a restricted time window, rather than the absolute repayment of debt over a prolonged period of time, which had a major impact on organ damage (Table 1). Animals receiving 120% SBV repaid all of the debt within 2 hours and incurred no cellular damage (Fig. 3A). Animals receiving 30% SBV were still able to repay approximately three fourths of the sustained debt in a 2-h period, although they incurred minimal to moderate organ damage (Fig. 3B). Animals receiving an immediate resuscitation volume equal to 15% of SBV repaid less than a third of the accumulated debt in 2 hours, with more severe organ damage (Fig. 3C). In all these groups, all of the incurred oxygen debt was eventually repaid, and no additional debt was incurred. In contrast, animals receiving an immediate resuscitation volume equal to only 8.4% of SBV continued to accrue debt for a short period and then repaid less than a third of the accumulated debt in 2 hours, resulting in subsequent organ failure and death. In summary, assessment of our calculated debt repayment proportions in the context of the reported histological data suggested that at least 30% of the shed blood volume, and approximately 75% of oxygen debt, had to be repaid within 2 h of the hemorrhagic injury to prevent or minimize organ damage. These data suggest that there is in fact a relatively short time window (probably less than 2 h) within which a minimum proportion of oxygen debt must be repaid to avoid mortality and major morbidity after moderate to severe shock.
The second consideration relates to the minimum volume of resuscitation fluid required for oxygen debt repayment. For the animals in the study of Siegel et al., we estimated the minimum immediate fluid volume requirements that permitted survival with at least an "acceptable" amount of organ dysfunction. As previously discussed, this cutoff was equivalent to a 75% oxygen debt repayment or approximately 30% of the SBV (Table 1). Approximate SBV for each group was estimated by assuming that total blood volume (TBV) was approximately equivalent to 8.7% body mass (27). Mass-specific fluid volumes were then estimated as the fraction of SBV returned divided by body mass. Therefore, estimated TBV for animals in the 30% SBV group was approximately 0.087 mL g−1 × 21,300 g = 1,853 mL. The SBV removed to incur an oxygen debt of 108.8 mL kg−1 in this group was 69.8% of TBV or 0.698 × 1,853 mL = 1,293 mL. The volume of fluid returned immediately posthemorrhage was 30% of SBV or 0.3 × 1,293 mL = 388 mL. Thus, the estimated minimum mass-specific volume associated with minimal organ damage was (388 mL/21.2 kg) or 18.2 mL kg−1. In summary, these data suggest a fluid bolus of approximately 20 mL kg−1 may be necessary to avoid mortality and major morbidity after moderate to severe shock.
The canine model previously described is obviously not an ideal representation of the clinical situation-albumin was used for the resuscitation fluid, hemorrhage was controlled, there was no additional trauma imposed, animals were anesthetized and heparinized, and only a single level of oxygen debt was characterized. Despite these limitations, our deductions from these data have several major implications for current resuscitation protocols and for the design of subsequent clinical trials. First, to minimize the probability of both death and organ damage after moderate to severe shock, the immediate repayment of at least two-thirds to three-fourths of the incurred oxygen debt within a relatively short period of time (approximately 2 h) may be warranted. In some circumstances, this will not be an achievable goal. For example, in the combat arena, delayed evacuation to definitive resuscitation is almost inevitable given the reality of hostile urban, austere, and/or remote environments. However, so-called "delayed" resuscitation strategies for civilian victims of penetrating trauma have been advocated by the apparent necessity to prevent clot dislodgment and the subsequent increase in blood loss caused by immediate and/or excessive fluid administration before hemorrhage control (28, 29). To our knowledge, none of the published rationales for delayed administration have considered implications for the oxygen debt repayment schedule. Our calculations suggest that resuscitation protocols must balance minimization of blood loss and clotting failure with the need to resolve oxygen debt as soon as possible after injury.
A second consideration is that "adequacy" of fluid resuscitation must be gauged by the extent to which oxygen debt has been repaid within a given time window, rather than in terms of a set fluid volume. Current Advanced Trauma Life Support and PHTLS protocols deemphasize the amount of fluid to be administered until hemorrhage can be controlled. Both PHTLS and Tactical Combat Casualty Care guidelines (30) recommend partial (hypotensive) resuscitation with titration of fluid to maintain SBP at approximately 80 to 90 mmHg or MAP of 60 to 65 mmHg until definitive care is reached. The stated goal of so-called low-volume resuscitation strategies is the provision of enough fluid after severe blood loss to raise blood pressure sufficiently to maintain adequate oxygen delivery to tissue but not high enough to dislodge clots and increase blood loss (31, 32). To meet current resuscitation goals, relatively small volumes of hypertonic crystalloids or synthetic colloid solutions have been promoted for use in-theater (33). Low-volume resuscitation with hypertonic fluids and/or colloids has the obvious physiological advantages of volume expansion and hyperosmotic vasodilation (which promotes reperfusion of ischemic tissues) (34) while minimizing the absolute volume of fluids to be administered (35). Small volume requirements also result in considerable logistic advantages, such as the minimization of weight carried by medics and conservation of scarce fluid resources in austere environments (30, 33, 36, 37). Tactical Combat Casualty Care volume guidelines suggest administration of colloids up to two boluses of 500 mL each (30). This is equivalent to approximately 7 to 14 mL kg−1 for a 70-kg subject. However, weight-based recommendations predicated on the assumption of the hypothetical 70-kg male subject are themselves misleading because many recent potential military recruits have greatly exceeded this weight (http://www.armytimes.com/news/2009/01/ap_fat_camp_recruits_011209/), with the result that even lower volumes will be provided. More extreme fluid goals have been proposed by the DARPA Surviving Blood Loss Program (http://www.darpa.mil/DSO/thrusts/bio/mainhuman/sbl/index.html) with the objective of reducing fluid loads to 4 mL kg−1 (the equivalent of a 250-mL bag of fluid for a 70-kg subject) before definitive resuscitation. However, our calculations indicate the minimum colloid requirements to repay even moderate debt may be substantially higher than these suggested volumes. Based on our analysis of oxygen debt repayment kinetics, the minimum fluid requirements of 18 mL kg−1 estimated here with a colloid (albumin) as the primary fluid were somewhat higher than the Tactical Combat Casualty Care-recommended maximum, and more than four times greater than the DARPA/Surviving Blood Loss target; volume requirements with Hextend would likely be no less. In contrast, use of crystalloid would dramatically increase the fluid needed to repay even a minimal amount of oxygen debt.
Our calculations regarding debt repayment schedule also indicate why many prior historical values of critical oxygen delivery as an end point (e.g., 600 mL min−1 m−2) gave such equivocal results. Critical oxygen delivery marks the transition at which oxygen debt may accumulate, but without an estimate of how much oxygen debt was actually incurred beforehand and how much repaid on a per patient basis, historical target values are essentially an arbitrary end point. In many studies, the actual duration of the shock state was either unknown or unreported. Therefore, the time at which resuscitation was initiated relative to the insult, and consequently the rate of oxygen debt repayment, will also be unknown. In the original studies using hospitalized high-risk surgical patients, it is possible that the surviving patients with targeted oxygen delivery may never have passed the critical DO2 threshold or, alternatively, the level of oxygen debt was considerably less than expected so that resuscitation was adequate to repay debt and thus favor a positive outcome. Alternatively, critical illness may induce an increase of VO2 from baseline (38), thus resulting in a threshold DO2 occurring at higher levels of DO2 compared with those produced by nonsurgical insults or injuries. In this scenario, attempts to restore oxygen delivery to threshold by a fixed-volume load would fail in that critical threshold would not be achieved, and the individual subject would be underresuscitated. For a third subset of patients, no resuscitation will be adequate either because the initial incurred oxygen debt was lethal or the debt repayment was so delayed that death or MODS was inevitable. In trials designed to evaluate patient end points, it is the patient-specific end point that is relevant. Because use of a fixed oxygen delivery target will result in some patients being underresuscitated, and some grossly overresuscitated, we argue that "one-size-fits-all" guidelines are not useful and do not allow for individualization of care. Basing resuscitation protocols on specified end points without regard to the physiological status of the individual patient may do more harm than good. Clearly, time relative to injury must be accounted for when assessing efficacy of resuscitation protocols. In practice, however, determining this time point will be a problem especially for sepsis patients where the point of insult is unknown.
SOME MODEST PROPOSALS AND A CAVEAT
A reevaluation of current experimental approaches to the problem of adequate fluid resuscitation is clearly warranted. In 1999, the Committee on Fluid Resuscitation for Combat Casualties concluded that hemorrhagic shock research concentrating on correction of either hemodynamics or single biochemical abnormalities associated with hemorrhage was "unlikely to be successful because multiple pathways lead to the cell death result from severe hemorrhagic shock. Rather, novel therapies should be aimed at the multiple metabolic and cellular derangements that accompany traumatic shock" (39). We propose that resuscitation to correct immediate volume deficits and hypoperfusion, and to control inflammation, should involve not only a consideration of anti-inflammatory adjuvants but also place increased emphasis on clinical strategies to rapidly estimate and repay oxygen debt.
Methods of directly measuring oxygen debt in the prehospital setting are not currently feasible. However, various biochemical correlates such as lactate and base deficit have been proposed as indices of posttraumatic hypovolemic shock, postinjury resolution or deterioration, and the probability of mortality (12, 40). Serum lactate levels have been used for many years to identify and manage patients in shock secondary to trauma and sepsis. Elevated lactated levels (>4 mM) are associated with increased patient mortality (41). However, single time-point measurements are limited in their use unless measured very soon after injury; this is because a single lactate measurement either prior to, or some time after, institution of resuscitation cannot provide the information necessary to quantitatively estimate the extent of shock. Therefore, it has been suggested that sequential values will be of much greater clinical use as an index to patient condition and response to fluid resuscitation (5, 42-44). Because the estimated half-life of serum lactate is approximately 20 min, persistently high or elevated lactate levels have been considered to be a good indicator of persistent impairment of oxygen delivery, early activation of inflammation, and subsequent development of MODS (45). Several studies have shown an association between rates of lactate reduction and outcome (41-43, 46, 47). Other studies have indicated that lactate kinetics, or time to lactate clearance, discriminated between survivors and nonsurvivors of trauma, sepsis, burns, and cardiac arrest (41-44, 47, 48). In an animal trial of acute severe hemorrhage shock followed by low-volume resuscitation, survivors could be discriminated from nonsurvivors by lower absolute lactate concentrations at the end of hemorrhage and during resuscitation, and sequential lactate measurements obtained even over very short time frames (minutes to hours) provided a useful rapid indication of subject response to resuscitation (49). Therefore, given the importance of oxygen debt repayment during the early postshock period, rapid sequential assessment of oxygen debt repayment markers could increase the effectiveness of early resuscitation especially in the prehospital or battlefield setting. The recent evolution of point-of-care lactate/base deficit monitors indicates that rapid sequential assessment of metabolic markers in the field will be feasible in the very near future.
The studies previously cited implicitly assume a close association between high levels of marker concentration on the one hand and the extent of the shock state on the other. In one sense, use of a surrogate marker is essentially a problem of calibration; that is, the magnitude of the surrogate measurement is assumed to vary in a predictable fashion with that of the "standard." However, unlike a true calibration problem, the requirements for determining a calibration standard for shock are not met; that is, the "standard" condition (shock) is still poorly understood, and there are few objective measures of the shock state presented in the literature that can be used to predict mortality and morbidity in the laboratory or clinical setting. The mechanisms underlying oxygen debt at the cellular level have not been well characterized. Most investigators would agree that "shock" likely represents the integration of the degree and duration of tissue ischemia from multiple tissue beds (50). Nevertheless, cellular mechanisms underlying the progression from low to high death probability (loss of high-energy phosphate pools, failure of membrane pumps, etc) have received far less attention than simpler models of ischemia/reperfusion. Consequently, it is unknown how changes in surrogate markers may be expected to vary with changes in patient condition during shock and resuscitation. If the change in marker concentration with the degree of ischemia is in fact intrinsically nonlinear (as might be expected because the debt mortality is logarithmic), and without indications of either limits to detection or areas of insensitivity, there will be difficulty in interpreting the absolute magnitude of the marker and its relevance to clinical condition. For instance, lactate levels increase not only in response to hypoxia but are also modulated by skeletal muscle ATPase activity (51, 52). Therefore, it is important to assess how well a given surrogate marker tracks changes in oxygen deficit and/or debt in real time.
As part of a larger trial performed at our institution, we examined the extent to which changes in lactate levels correspond to changes in oxygen debt in a swine model of hemorrhagic shock and resuscitation (approved in advance by the Virginia Commonwealth University Institutional Animal Care and Use Committee). In these trials, swine were subjected to controlled hemorrhage under anesthesia to a fixed oxygen debt of approximately 80 mL kg−1. At the end of hemorrhage, animals were resuscitated with a single bolus of a given resuscitation fluid (such as whole blood, lactated Ringer, hypertonic saline, and colloids, e.g., Hespan). Sequential arterial lactates were measured simultaneously with oxygen debt. As demonstrated with other models of oxygen debt (24), the level of debt was not closely associated with hemorrhage volumes and hemoglobin levels. Figure 4 demonstrates several oxygen debt-lactate response curves observed for three animals selected to illustrate (A) a poor response to resuscitation, (B) near-complete resolution of shock, and (C) an intermediate or partial resolution of shock with failure to completely repay oxygen debt. It is apparent that during the hemorrhage phase, each animal showed increases in lactate corresponding to an increase in oxygen debt. However, responses to resuscitation were variable. In Figure 4A, both oxygen debt and lactate continued to accumulate; fluid resuscitation (with lactated Ringer) was clearly inadequate, and the animal died shortly before the end of the prescribed 3-h postresuscitation period. In Figure 4B, fluid resuscitation (with whole blood) seemed to be adequate to completely repay oxygen debt, and lactate was reduced to near-baseline levels. However, in the intermediate case (Fig. 4C), oxygen debt was not fully repaid; lactate had begun to resolve but still remained above baseline.
It is apparent from these examples that any two individuals may start resuscitation from the same level of initial oxygen debt and yet may end up on either extreme of the resuscitation response continuum. Here, resolution of oxygen debt and lactate clearly depend on the relative response to a given resuscitation fluid. Second, although not perfect indicators of oxygen debt status, sequential lactate determinations clearly give promise of the first real calibration of a surrogate marker with the shock condition. In this study, because the period of shock was relatively short, two of these fluid regimens were sufficient to partially or completely repay oxygen debt in the short term. By "pushing" oxygen delivery back to the flow-independent portion of the VO2 curve, it is assumed that anaerobiosis, and therefore the accumulation of further oxygen debt, would be halted. Similarly, lactate levels are assumed to track oxygen debt because the cessation of anaerobiosis would result in the return of lactate production to normal, with subsequent clearing of accumulated lactate, most likely by the liver and renal cortex (53, 54). However, if the shock state was sufficiently severe or prolonged beyond a certain point, the ensuing ischemic damage to liver and kidney would result in an inability to clear lactate. Further loss of high-energy phosphate pools, mitochondrial disruption, and disruption of ionic gradients would lead to cellular damage generated in multiple organ beds (55), driving the progression of shock that would in turn influence potential responses to subsequent resuscitation efforts. As illustrated in Figure 4C, this implies that in certain cases significant oxygen debt may remain despite lactate returning to clinically normal levels. This creates a practical difficulty in determining if patients whose lactate levels are apparently normalized with resuscitation have in fact repaid their oxygen debt. There are some clinical data for sepsis patients suggesting that lactate levels may normalize without oxygen debt resolution (56).
Crucial from the standpoint of managing the individual patient is the question as to how much accumulated oxygen debt can be tolerated while avoiding the transition to an irreversible state of cell damage. The intermediate conditions previously described represent the greatest problem for personnel charged with the initial resuscitation of the critically ill and injured because at present, there are no useful clinical methods for fully tracking debt accumulation and clearance. Even if VO2 could be measured (as is the case for the ventilated sedated patient in the intensive care unit setting), both baseline VO2 and the actual duration of the shock state before treatment-essential components of the oxygen debt calculation-will still be unknown for a given individual. Furthermore, hypothermia, sedation, anesthesia, and other factors that serve to reduce VO2 will make it difficult to assess baseline VO2 relative to levels of VO2 at the time of injury. Therefore, both the time at which resuscitation was actually initiated relative to initiation of debt accumulation and the absolute rate of oxygen debt repayment will also be unknown. Further investigations of potential surrogate markers and serial correlates for oxygen debt repayment may be a good start. However, because organ blood flow is determined by perfusion pressure and vascular resistance, we need to investigate other tissue oxygenation variables (such as oxygen extraction ratios of accessible tissue beds) for monitoring the relative persistence of oxygen debt during the postresuscitation period. Tissue oxygenation, although thought to be an important determinant of organ dysfunction (57), could be an unreliable end point if VO2 is decreased, not because of a reduction in DO2, but because mitochondria are unable to use available oxygen, a state known as cytopathic hypoxia (58). Potential areas for future study could include determinations of patterns of organ system-specific accumulation of oxygen debt during hemorrhage, debt repayment upon resuscitation, and the consequences for long-term organ function during the postresuscitation period.
Use of low-volume expanders in the treatment of trauma and combat casualties may require the addition of pharmacologic adjuncts to enhance repayment of debt, especially those that promote microcirculatory blood flow, reduce VO2 requirements of tissues, or provide alternative substrate use until such time that adequate DO2 is assured. In the military context, repayment of minimal oxygen debt requirements within the modest volume constraints currently proposed will likely require a combination of colloid, hypertonic saline, and possibly hemoglobin or a non-hemoglobin-based oxygen carrier. Single administrations of hypertonic saline combined with dextran (HSD) not exceeding 4 mL kg−1 seem to be safe (35, 59), although in vitro coagulopathic changes have been noted at higher loadings (60). HSD has been used clinically for several years in Europe as RescueFlow (www.rescueflow.com). Higher concentrations of sodium and colloids may serve as effective resuscitation agents even at levels less than 4 mL kg−1 (35, 59). Dilutional anemia resulting from repeated dosing could be overcome, in theory at least, by combining HSD (or other hypertonic saline/colloid cocktails) with a hemoglobin-based oxygen carrier. Data from animal studies indicate that incorporation of hemoglobin-based oxygen carrier into colloid solutions result in improved short-term survival (61, 62), compared with colloid alone.
In this article, we first review the concept of the oxygen supply-delivery relationship with that of oxygen debt and show how this is relevant to the current understanding of shock and resuscitation. We then build on this overview to introduce the concept of oxygen debt repayment schedule. There is obviously a great need for appropriate clinical data to assess this concept for hemorrhagic shock patients. However, we used some of the only available data from a canine model of shock and resuscitation (27) to argue that this metric for shock resuscitation provides a more rigorously defined and clinically relevant end point for assessing the therapeutic benefit of various resuscitation strategies. The concept of the oxygen debt repayment schedule also provides an unambiguous and potentially objective standard for quantifying the behavior of various postulated shock "markers" in response to resuscitation. We demonstrate its heuristic value in providing a predictive framework for both the optimum therapeutic time window and optimum fluid loadings before critical transitions to an irreversible shock state can occur. A quantitative metric such as oxygen debt should allow more rigorous assessment of potential associations between the shock state and the various physiological and biochemical alterations induced by hemorrhage, such as the degree of inflammation, immune suppression, and coagulopathy. Future work should be directed toward the correlation of downstream injuries with oxygen debt repayment kinetics. For the clinician who normally cannot know absolute oxygen debt, the practical application of the oxygen debt model will be manifested primarily as the prevention of additional debt accumulation. Normally, this would be accomplished first by control of bleeding, and second, by initial resuscitation to increase oxygen delivery to a point sufficient to support baseline oxygen consumption. Once definitive resuscitation can occur, we argue that repayment of debt should occur as quickly as possible. As the canine data suggest, the "golden hour" may in reality consist of "two silver hours"; that is, in moderate to severe shock, approximately three fourths of the accumulated oxygen debt must be repaid within 2 h. Finally, although our discussion has concentrated on hemorrhagic shock and its resuscitation, oxygen debt repayment kinetics may find wider applicability in other types of shock and resuscitation protocols.
The authors thank John Siegel, M.D., F.A.C.S., Emeritus Professor, New Jersey Medical School (retired), for valuable comments on the manuscript.
1. Ward KR, Ivatury RR, Barbee RW: Endpoints of resuscitation for the victim of trauma
. J Intensive Care Med
2. Shoemaker WC, Patil R, Appel PL: Hemodynamic and oxygen transport patterns for outcome prediction, therapeutic goals, and clinical algorithms to improve outcome: feasibility of artificial intelligence to customize algorithms. Chest
3. McKinley BA, Kozar RA, Cocanour CS, Valdivia A, Sailors RM, Ware DN, Moore FA: Normal versus supranormal oxygen delivery goals in shock resuscitation: the response is the same. J Trauma
4. Gattinoni L, Brazzi L, Pelosi P, Latini R, Tognoni G, Fumagalli R: A trial of goal-oriented hemodynamic therapy in critically ill patients. N Engl J Med
5. American College of Surgeons: Advanced Trauma Life Support® (ATLS®) for Doctors
. Chicago IL: American College of Surgeons, 2008.
6. Brohi K, Singh J, Heron M, Coats T: Acute traumatic coagulopathy. J Trauma
7. DeLoughery TG: Coagulation defects in trauma
patients: etiology, recognition, and therapy. Crit Care Clin
8. Marshall JC: Inflammation, coagulopathy, and the pathogenesis of multiple organ dysfunction syndrome. Crit Care Med
9. Chow CC, Clermont G, Kumar R, Lagoa C, Tawadrous Z, Gallo D, Betten B, Bartels J, Constantine G, Fink MP, et al.: The acute inflammatory response in diverse shock states. Shock
10. Lagoa CE, Bartels J, Baratt A, Tseng G, Clermont G, Fink MP, Billiar TR, Vodovotz Y: The role of initial trauma
in the host's response to injury and hemorrhage: insights from a comparison of mathematical simulations and hepatic transcriptomic analysis. Shock
11. Hess JR, Brohi K, Dutton RP, Hauser CJ, Holcomb JB, Kluger Y, Mackway-Jones K, Parr MJ, Rizoli SB, Yukioka T, et al.: The coagulopathy of trauma
: a review of mechanisms. J Trauma
12. Rixen D, Siegel JH: Bench-to-bedside review: oxygen debt and its metabolic correlates as quantifiers of the severity of hemorrhagic and post-traumatic shock. Crit Care
13. Schmid-Schönbein GW: A journey with Tony Hugli up the inflammatory cascade towards the auto-digestion hypothesis. Int Immunopharmacol
14. Kauvar DS, Lefering R, Wade CE: Impact of hemorrhage on trauma
outcome: an overview of epidemiology, clinical presentations, and therapeutic considerations. J Trauma
15. Fiddian-Green RG, Haglund U, Gutierrez G, Shoemaker WC: Goals for the resuscitation of shock. Crit Care Med
16. Shoemaker WC, Appel PL, Kram HB, Bishop M, Abraham E: Hemodynamic and oxygen transport monitoring to titrate therapy in septic shock. New Horizons
17. Shoemaker WC, Appel PL, Kram HB, Waxman K, Lee T-S: Prospective trial of supranormal values of survivors as therapeutic goals in high-risk surgical patients. Chest
18. McKinley BA, Sucher JF, Todd SR, Gonzalez EA, Akozar R, Sailors RM, Moore FA: Central venous pressure vs pulmonary artery catheter directed shock resuscitation. Shock
19. Rhee P, Koustova E, Alam HB: Searching for the optimal resuscitation method: recommendations for the initial fluid resuscitation
of combat casualties. J Trauma
20. Marr AB, Moore FA, Sailors RM, Valdivia A, Selby JH, Kozar RA, Cocanour CS, McKinley BA: Preload optimization using "Starling curve" generation during shock resuscitation: can it be done? Shock
21. Shoemaker WC, Appel PL, Kram HB: Tissue oxygen debt as a determinant of lethal and nonlethal postoperative organ failure. Crit Care Med
22. Dunham CM, Siegel JH, Weireter L, Fabian M, Goodarzi S, Guadalupi P, Gettings L, Linberg SE, Vary TC: Oxygen debt and metabolic acidemia as quantitative predictors of mortality and the severity of the ischemic insult in hemorrhagic shock
. Crit Care Med
23. Crowell JW, Smith EE: Oxygen deficit
and irreversible hemorrhagic shock
. Am J Physiol
24. Rixen D, Raum M, Holzgraefen B, Sauerland S, Nagelschmidt M, Neugerbauer EAM: A pig hemorrhagic shock
model: oxygen debt and metabolic acidemia as indicators of severity. Shock
25. Shoemaker WC, Peitzman AB, Bellamy R, Bellomo R, Bruttig SP, Dubick MA, Kramer GC, McKenzie JE, Pepe PE, Safar P, et al.: Resuscitation from severe hemorrhage. Crit Care Med
26. Cooper DJ, Myles PS, McDermott FT, Murray LJ, Laidlaw J, Cooper G, Tremayne AB, Bernard SS, Ponsford J: Prehospital hypertonic saline resuscitation of patients with hypotension and severe traumatic brain injury: a randomized controlled trial. J Am Med Assoc
27. Siegel JH, Fabian M, Smith JA, Kingston EP, Steele KA, Wells MR: Oxygen debt criteria quantify the effectiveness of early partial resuscitation after hypovolemic hemorrhagic shock
. J Trauma
28. Bickell WH, Wall MJ, Pepe PE, Martin RR, Ginger VF, Allen MK, Mattox KL: Immediate versus delayed fluid resuscitation
for hypotensive patients with penetrating torso injuries. N Engl J Med
29. Sondeen JL, Coppes VG, Holcomb JL: Blood pressure at which rebleeding occurs after resuscitation in swine with aortic injury. J Trauma
30. Salomone JP, Pons PT: Tactical field care. In: Salomone JP, Pons PT, eds.: PHTLS Prehospital Trauma Life Support. Military Version
. St. Louis, MO: Mosby-Elsevier, 516-537, 2007.
31. Stern SA: Low-volume fluid resuscitation
for presumed hemorrhagic shock
: helpful or harmful? Curr Opin Crit Care
32. Stern SA, Wang X, Mertz M, Chowanski ZP, Remick DG, Kim HM, Dronen SC: Under-resuscitation of near-lethal uncontrolled hemorrhage: effects on mortality and end-organ function at 72 hours. Shock
33. Holcomb JB: Fluid resuscitation
in modern combat casualty care: lessons learned from Somalia. J Trauma
34. Cabrales P, Tsai AG, Intaglietta M: Hyperosmotic-hyperoncotic versus hyperosmotic-hyperviscous small volume resuscitation in hemorrhagic shock
35. Kramer GC: Hypertonic resuscitation: physiologic mechanisms and recommendations for trauma
care. J Trauma
36. Champion HL: Combat fluid resuscitation
: introduction and overview of conferences. J Trauma
37. Pearce FJ, Lyons WS: Logistics of parenteral fluids in battlefield resuscitation. Mil Med
38. Tuchschmidt J, Oblitas D, Fried JC: Oxygen consumption in sepsis and septic shock. Crit Care Med
39. Pope A, French G, Longnecker DE: Novel approaches to treatment of shock. In: Pope A, French G, Longnecker DE, eds.: Fluid Resuscitation: State of the Science for Treating Combat Casualties and Civilian Injuries
. Washington DC: National Academy Press, 208, 1999.
40. Kirschenbaum LA, Astiz ME, Rackow EC: Interpretation of blood lactate
concentrations in patients with sepsis. Lancet
41. Meregalli A, Oliveira RP, Friedman G: Occult hypoperfusion is associated with increased mortality in hemodynamically stable, high-risk, surgical patients. Crit Care
42. Bakker J, Gris P, Coffernils M, Kahn R, Vincent JL: Serial blood lactate
levels can predict the development of multiple organ failure
following septic shock. Am J Surg
43. Manikis P, Jankowski S, Zhang H, Kahn RJ, Vincent J-L: Correlation of serial blood lactate
levels to organ failure and mortality after trauma
. Am J Emerg Med
44. McNelis J, Marini CP, Jurkiewicz A, Szomstein S, Simms H, Ritter G, Nathan IM: Prolonged lactate
clearance is associated with increased mortality in the surgical intensive care unit. Am J Surg
45. Nast-Kolb D, Waydhas C, Gippner-Steppert C, Schneider I, Trupka A, Ruchholtz S, Zettl R, Schweiberer L, Jochum M: Indicators of the posttraumatic inflammatory response correlate with organ failure in patients with multiple injuries. J Trauma
46. Abramson D, Scalea TM, Hitchcock R, Trooskin SZ, Henry SM, Greenspan J: Lactate
clearance and survival following injury. J Trauma
47. Nguyen HB, Rivers EP, Knoblich BP, Jacobsen G, Muzzin A, Ressler JA, Tomlanovich MC: Early lactate
clearance is associated with improved outcome in severe sepsis and septic shock. Crit Care Med
48. Donnino MW, Miller J, Goyal N, Loomba M, Sankey SS, Dolcourt B, Sherwin R, Otero R, Wira C: Effective lactate
clearance is associated with improved outcome in post-cardiac arrest patients. Resuscitation
49. Reynolds PS, Barbee RW, Ward KR: Lactate
profiles as a resuscitation assessment tool in a rat model of battlefield hemorrhage-resuscitation. Shock
50. Walley KR: Heterogeneity of oxygen delivery impairs oxygen extraction by peripheral tissues: theory. J Appl Physiol
51. Levy B: Lactate
and shock state: the metabolic view. Curr Opin Crit Care
52. Luchette FA, Friend LA, Brown CC, Upputuri RK, James JH: Increased skeletal muscle Na+
-ATPase activity as a cause of increased lactate
production after hemorrhagic shock
. J Trauma
53. Bellomo R: Bench-to-bedside review: lactate
and the kidney. Crit Care
54. Leverve XM, Mustafa I: Lactate
: a key metabolite in the intercellular metabolic interplay. Crit Care
55. Levy RJ: Mitochondrial dysfunction, bioenergetic impairment, and metabolic down-regulation in sepsis. Shock
56. Sakr Y, Dubois MJ, De Backer D, Creteur J, Vincent JL: Persistent microcirculatory alterations are associated with organ failure and death in patients with septic shock. Crit Care Med
57. Cohn SM, Nathens AB, Moore FA, Rhee P, Puyana JC, Moore EE, Beilman GJ: Tissue oxygen saturation predicts the development of organ dysfunction during traumatic shock resuscitation. J Trauma
58. Fink MP: Bench-to-bedside review: cytopathic hypoxia. Crit Care
59. Dubick MA, Bruttig SP, Wade CE: Issues of concern regarding the use of hypertonic/hyperoncotic fluid resuscitation
of hemorrhagic hypotension. Shock
60. Coats TJ, Heron M: The effect of hypertonic saline dextran on whole blood coagulation. Resuscitation
61. Fitzpatrick CM, Biggs KL, Atkins BZ, Quance-Fitch FJ, Dixon PS, Savage SA, Jenkins DH, Kerby JD: Prolonged low-volume resuscitation with HBOC-201 in a large-animal survival model of controlled hemorrhage. J Trauma
62. Reynolds PS, Barbee RW, Skaflen MD, Ward KR: Low-volume resuscitation cocktail extends survival after severe hemorrhagic shock