Reconsidering critical illness as an uncharacterised acquired mitochondrial disorder

We welcome the inclusion in this edition of JICS of the article from Ivin and colleagues regarding the clinical consequences and evidence-based management of a mitochondrial disorder: carnitine palmitoyltransferase 2 (CPT2) deficiency. This article is a timely reminder of the intricate machinery upon which cell integrity and function is based and the consequences of perturbations to it. CPT2 deficiency may be an extremely rare condition, but by exploring the aftermath of insufficient function of a single enzyme in the mass of interweaving metabolic pathways involved in cellular ATP production, the authors draw our attention to the clinical relevance of processes that, for many intensive care clinicians, have long since faded into obscurity. The complex web of the Krebs cycle, interminable iterations of β-oxidation and peculiar apparatus responsible for oxidative phosphorylation are frequently decried by medical students as irrelevant to future practice, and rarely feature in postgraduate education in our speciality.

Support of cellular bioenergetic capacity is the desired end point of almost all supportive therapy on the intensive care unit (ICU), the raison d'être for intensive care itself, and yet familiarity with the complex processes underpinning bioenergetics is not prioritised. As clinicians, we aggressively target the supply end of the bioenergetic equation: supplementing inspired oxygen concentrations, manipulating ventilator pressures and rates, titrating intravenous fluid, inotropes and vasopressors and delivering enteral or parenteral nutrition, all with the avowed aim of bolstering mitochondrial oxidative phosphorylation. However, we are unable to determine if we have achieved our desired cellular end point, and the methods we use to judge the success of our interventions are limited to indirect indicators of global metabolic state, such as plasma lactate concentrations. Such measures are influenced by many factors (for example, circulating lactate concentrations depend upon the rate of hepatic lactate metabolism, as well as the rate of anaerobic activity) and also cannot differentiate between metabolic state in individual tissues.^1,2 Multiple randomised controlled trials in intensive care have demonstrated that the achievement of systemic targets, based on typical parameters associated with health, do not result in better patient outcomes ³ and may in fact be associated with harm.^4,5 Similarly, randomised controlled trials have demonstrated neither survival benefit from delivering nutrition to meet calculated energy requirements ⁶ nor harm from significant underfeeding. ⁷ It should, however, be noted that the composition of the nutrition administered was not considered, and cellular energy expenditure was not measured. In addition to failing to scrutinise the therapeutic effect on the cells we hope to support, we also neglect to monitor for potential side effects that our well-intentioned interventions may have at this level, or for the body as a whole.

In addition to producing ATP, the mitochondria play a key role in maintaining and adjusting cell redox balance, with the Krebs cycle and electron transport chain being significant generators of reactive oxygen species. ⁸ These highly reactive molecules are at the heart of a complex signalling network, spanning intracellular and extracellular compartments, and their interaction with other molecules is crucial for the co-ordination of stress signalling. ⁹ In excess, they also have the potential to cause oxidative damage to cell components, such as protein channels, lipid membranes and DNA. ¹⁰ Increased blood and tissue levels of oxidative stress markers have been demonstrated in patients with critical illness, and even associated with mortality, ¹¹ but the source and potential function of these oxidation products are not fully understood. Bioenergetic pathways thus appear to lie at the intersection between two processes fundamental to cell survival: energy metabolism and redox signalling.

It has been argued that critical illness itself, and the route to multiple organ failure, represents an acquired mitochondrial disorder. ¹² Unlike CPT2 deficiency, however, the precise nature of this disorder, and mechanism(s) underlying it, has not yet been characterised. A handful of studies of tissue samples from critically ill patients have attempted to define bioenergetic capacity using steady state concentrations of single metabolites in isolation, which cannot capture the complexity of the entire system, or identify what part (or parts) of the system require support.^13,14 Revealingly, plasma metabolomic profiling of a cohort of 90 patients demonstrated that one third of 187 metabolites tested within 24–72 h of ICU admission differed between survivors and non-survivors. ¹⁵ Moreover, these divergent metabolites were not restricted to a specific metabolic pathway but included lipid, amino acid, carbohydrate and nucleotide metabolites. More recent work has identified evidence for dysregulation in skeletal muscle mitochondrial β-oxidation in critical illness, but broader characterisation of the entire metabolic network is required to identify the level (or levels) at which the deficiency occurs. ¹⁶ Supplementation of individual metabolites, from pyruvate to glutamine, has demonstrated potential benefit in critical illness, either in terms of reducing inflammation or muscle wasting,^17,18 but a harmonised approach, targeting multiple affected metabolic pathways, may be even more effective.

When the fundamental processes underlying cellular bioenergetic capacity in critical illness have been mapped out, we may be able to apply a more sophisticated approach to preventing or treating energy crises, such as is seen with the specific substrate manipulation used to treat CPT2 deficiency. Ability to target supportive therapy to a precise cellular deficiency (or more likely, deficiencies), and to monitor the bioenergetic end point, may prove to be more effective than the current strategy of indiscriminate provision of O₂ and substrate, titrated to global parameters. It may also protect the patient from unintended consequences of our current blunderbuss approach. Intensive care physicians have never been more cognisant of the potential harm of over-intervention, with evidence solidifying for the harm associated with supraphysiological levels of arterial oxygenation. ¹⁹ The solution may be to dust off our biochemistry textbooks, and refamiliarise ourselves with the labyrinthine twists and turns of intermediary metabolism and electron transport, in respect to the delicate pathways that stand between our patients and death.