The Challenge and the Promise of Studying Mitochondrial Dysfunction in Humans with Sepsis

What is the etiology of serum lactic acid (lactate) accumulation in patients with sepsis, and why does it matter? The prognostic value of lactate to predict patient outcomes in sepsis is well documented (1), and the greater the reduction in lactate over the course of the initial resuscitation (clinically referred to as lactate clearance) (2), the better the chance of patient survival. Based on these observations, in conjunction with the results of two randomized trials demonstrating the value of targeting lactate clearance (of 10% [3] and 20% [4], respectively) as a resuscitation goal, the Surviving Sepsis Campaign gave a 2C recommendation to the targeting lactate normalization (5) during early sepsis resuscitation, a goal formally untested in clinical trials.

How exactly one “targets” lactate normalization remains unclear and seems to rest on the administration of various components of goal-directed therapy, such as intravenous fluids, blood products, and inotropes. Such an approach rests on the assumption that the etiology of lactate accumulation in sepsis is solely or at least predominantly the result of anaerobic metabolism secondary to impaired tissue perfusion, a paradigm that has been challenged and is likely incomplete (6). Given emerging evidence regarding the potential harms associated with excessive fluid (or at least chloride [7]) administration and the lack of benefit from excessive packed red blood cell transfusion (8), one can easily imagine how an overly eager clinician attempting to reach a goal of lactate normalization might inadvertently defy the dictum to “do no harm.”

If lactate were simply the result of anaerobic metabolism, perhaps such an approach would not be unjustified. However, numerous studies in both humans and animals have demonstrated other important etiologies of lactate accumulation in addition to increased anaerobic production, including decreased lactate metabolism and clearance (9), increased aerobic production due to adrenergic stimulation of the Na⁺/K⁺ ATPase (10), and altered mitochondrial function or “cytopathic hypoxia” (11). The relative contribution of each of these various mechanisms and whether they vary patient to patient or over the time course of the sepsis syndrome remains unclear, due in large part to the challenges of studying these phenomena in human patients.

In this issue of AnnalsATS, Nuzzo and colleagues (pp. 1662–1666) present a prospective, unmatched, case control study of pyruvate dehydrogenase activity in serial samples of peripheral blood mononuclear cells (PBMCs) from patients with sepsis versus healthy control subjects (12). Pyruvate dehydrogenase (PDH) is a key enzyme regulating the metabolism of pyruvate into acetyl-coA for entry into the tricarboxylic acid cycle and has been noted in at least some animal models to be inhibited after severe infection (13). In their study, the authors found that PDH was down-regulated in patients with sepsis compared with healthy control subjects and in nonsurvivors compared with survivors, helping to translate and validate previous findings in animal models to human patients.

Given the fact that inhibition of PDH will generally “back-up” the system and cause the conversion of unmetabolized pyruvate into lactate, the study of this enzyme has relevance to our question regarding the etiology of lactate accumulation in sepsis. Indeed, lactate levels were inversely associated with PDH activity, raising the possibility that inhibition of mitochondrial function (specifically PDH) may be contributing to lactatemia in humans with sepsis. Unfortunately, this associative study did not specifically test that hypothesis nor by its design clarify the relative contribution of such an inhibition to lactate levels.

This exploratory, hypothesis-generating study is also not without significant limitations. From a technical standpoint, the variation between plates used for assays of PDH functional activity noted by the authors is a critical limitation to be mitigated in future work. From a clinical perspective, although PDH quantity and activity appear to be impaired in patients with sepsis compared with control patients, the cohorts were poorly matched. Although a regression analysis provides reassurance that demographic characteristics are not solely responsible for the differences in PDH, the possibility remains that an unmeasured confounder or generalized inflammation is at least partly responsible.

A similar criticism can be made of the observed differences in sepsis survivors versus nonsurvivors; whether leukocyte PDH activity is pathologic or just another severity-of-illness measurement requires additional investigation. There remains the additional possibility that bioenergetic down-regulation is actually an adaptive mechanism (14), due to a reduction in reactive oxygen species generation from irrevocably damaged mitochondria that must await mitochondrial biogenesis for resolution. Finally, regarding the etiology of lactate accumulation, white blood cells represent a very small biomass and are unlikely to be the source of any significant amount of systemic lactate themselves, although it is possible that they serve as a surrogate marker of a more substantive source.

Apart from the aforementioned limitations of the study design, greater concerns about the underlying premise of the question itself must be addressed to move this line of investigation forward. The first major challenge in the development of peripheral biomarkers for the diagnosis of mitochondrial dysfunction is whether they truly represent the processes we clinicians are interested in. One of the major limitations of this line of investigation in human patients is whether or not mitochondria from PBMCs, platelets, or even muscle biopsies are adequate surrogate representations of the mitochondria found in vital organs such as the heart, kidneys, and liver. Demonstration of concordance between these various measurements would be a major advance, although discordance between leukocyte and the vital organs of an animal model of hemorrhage (15) raises serious concerns regarding this use of PBMCs for this purpose.

A second challenge involves the fact that mitochondrial and bioenergetic pathways are highly redundant, and inhibition of a single enzyme may or may not have significant effects on the ultimate downstream component of interest, namely ATP. Development of a feasible comprehensive approach to the study of the entire energy production cascade in humans would represent an even greater accomplishment, as it may allow for the better understanding of the source of lactate in different patients and over time, allowing for more focused clinical trials of novel therapeutics for the treatment of bioenergetic impairment.

Finally, an additional area of investigation specifically related to cell line chosen for this study remains tantalizing. That is, assuming decreased PDH activity results in bioenergetic impairment, what is the effect of PDH inhibition specifically and mitochondrial inhibition generally on white blood cell function in patients with life-threatening infections? In contrast to the paradigm of sepsis as a hyperactive immune-mediated inflammatory response, there is significant evidence for the development of a relative immune suppression and apoptosis-induced loss of adaptive immune cell function (16), which would seem a natural extension of this work. Further delineation of the functional consequences of PDH inhibition on PBMC function, the systemic inflammatory response, and ultimately on clinical outcomes such as secondary infections and multisystem organ failure may prove a fruitful area of investigation that has the potential to not only help reshape our understanding of the pathophysiology of sepsis but also provide novel therapeutic targets with measurable surrogate outcomes that may be amenable to future interventional trials.

The study by Nuzzo and colleagues, therefore, is a small but measurable step toward the further translation of the study of mitochondrial function and bioenergetics into human patients. Despite its limitations and the subsequent work that remains, this study lends excitement to the promise of bioenergetic manipulation as a novel therapeutic strategy in patients with sepsis.