Interpreting Elevated Lactate Levels in Shock States
- In ICU
- Tue, 20 May 2025
ICU Management & Practice, Volume 25 - Issue 2, 2025
Elevated lactate levels indicate higher morbidity and mortality but require careful interpretation. In the presence of tissue hypoperfusion, they should prompt urgent haemodynamic optimisation. Without such signs, they warrant reassessing diagnosis and treatment. Serial lactate levels guide overall management but have limitations, and clinicians must understand the complexities of lactate metabolism in patient care.
The presence of lactate in human blood was first documented by Joseph Scherer in 1848, who described lactic acidosis in the blood of a young woman who had succumbed to fulminant septic shock secondary to puerperal fever. A decade later, in 1858, Carl Folwarczny reported elevated lactate levels in a living patient (Kompanje et al. 2007). Today, blood lactate monitoring is a standard clinical practice in critically ill patients, primarily to detect tissue hypoxia and guide resuscitation strategies (Evans et al. 2021). However, the relationship between hyperlactataemia and inadequate tissue perfusion is complex and multifactorial. In this article, we review the current understanding of how to appropriately interpret lactate levels in critically ill patients.
Metabolism of Lactate
As a product of glycolysis, all cells in the body can produce lactate. Glycolysis is a process that occurs in the cytosol and converts one molecule of glucose into two molecules of pyruvate through a series of catalytic reactions. This process generates two molecules of adenosine triphosphate (ATP) (Figure 1) (Levy 2006). Glycolysis is the main energy process for all cells operating under hypoxic conditions, such as during tissue hypoperfusion. Pyruvate can be directly reduced to lactate during glycolysis in a process dependent on lactate dehydrogenase, or it can move into a different series of enzymatic reactions (Krebs cycle) depending on oxygen availability. In the presence of oxygen, pyruvate enters the mitochondria after being decarboxylated to acetyl-coenzyme-A, which subsequently feeds the tricarboxylic acid (TCA) cycle in a series of successive decarboxylation producing CO2, H2O, and 18 ATP molecules (when NADH+ and FADH+ residues, product of such decarboxylations, attain the oxidative phosphorylation) (Levy 2006). In normal steady-state conditions, metabolic processes produce lactate and pyruvate in a 10:1 ratio. Nevertheless, during critically low oxygen availability, pyruvate does not enter the mitochondria (as the accumulation of NADH/FADH residues and other TCA intermediate metabolites block the pyruvate dehydrogenase), thus limiting the aerobic energy generation, and then, glycolysis becomes the only available pathway to maintain ATP production. This causes the lactate to pyruvate ratio to increase.
Nevertheless, lactate can be converted back into glucose, which can then be reconverted to lactate, being the basis of the Cori cycle. This regeneration of glucose from lactate plays a crucial role in restoring glucose levels and lactate reuptake from systemic circulation, particularly after periods of prolonged tissue hypoxia, such as during cardiac arrest. Lactate utilisation involves both intraorgan and interorgan lactate shuttles, processes facilitated by monocarboxylate transporters (MCTs) that manage the influx and efflux of lactate along with their co-transported protons. Under normal physiological conditions, lactate production and consumption are balanced, maintaining a stable lactate concentration in the blood (Kraut and Madias 2014).
The following reaction can summarise the bioenergetic process of lactate production:
glucose + 2(ADP + inorganic phosphate) → 2 lactate + 2 H+ + 2 ATP
During glycolysis, generation of lactate ions is closely associated with the release of protons of an equivalent number, resulting from ATP hydrolysis. Indeed, such ATP hydrolysis represents a main source of free H+ in the cells. Conversely, lactate consumption removes an equal number of protons, effectively aiding in maintaining acid-base equilibrium in the body.
Mechanisms of Increased Blood Lactate Levels
Elevated blood lactate levels occur when lactate production exceeds lactate consumption. Hyperlactataemia (>2 mmol/L) can solely be induced by increased anaerobic or aerobic lactate production, possibly accompanied by decreased lactate reuptake by the cells (Table 1).
Anaerobic production
Numerous experimental studies have established a clear link between tissue hypoxia and lactate production, particularly when oxygen supply falls below the level necessary to meet cellular oxygen demands (Cain 1965; Zhang and Vincent 1993). At this critical threshold, oxygen consumption (VO₂) becomes dependent on oxygen delivery (DO₂), signalling a shift to anaerobic metabolism, which results in a significant rise in lactate levels. This relationship between VO₂/DO₂ dependency and hyperlactataemia has also been observed clinically in patients undergoing end-of-life care, where DO₂ reduction proceeded until circulatory arrest (Ronco et al. 1993).
Additional clinical evidence suggests the anaerobic origin of hyperlactataemia in critically ill patients. For instance, during the early stages of septic shock, hyperlactataemia apparently coincided with oxygen supply dependency (Friedman et al. 1998). Ronco et al. (1993) further observed that lowering DO₂ below a critical threshold coincided with decreased oxygen consumption and was accompanied by increased blood lactate levels. Similarly, Vincent et al. (1990) observed that increasing DO₂ through dobutamine infusion elevated VO₂ exclusively in critically ill patients with high blood lactate concentrations, likely favouring the anaerobic origin of hyperlactataemia in these patients (VO₂/DO₂ dependency).
Aerobic production
Lactate can be elevated in situations without tissue hypoxia. Indeed, as lactate is a normal product of glucose and pyruvate metabolism, any rise in glucose metabolism or reduction in pyruvate metabolism will increase lactate production even under adequate tissue oxygenation conditions (Levy 2006). Sepsis-induced inflammatory cytokines can accelerate aerobic glycolysis, resulting in quantities of pyruvate exceeding the capability of pyruvate dehydrogenase to catalyse the conversion of pyruvate into acetyl coenzyme A. Therefore, pyruvate is inevitably converted into lactate by the lactate dehydrogenase, causing hyperlactataemia. Epinephrine can overaccelerate ATP hydrolysis by increasing the sodium-potassium-adenosine triphosphatase (Na-K-ATPase) activity through β2 stimulation, which ultimately leads to increased adenosine diphosphate (ADP) levels, free H+, and phosphofructokinase activity, with the subsequent acceleration of aerobic glycolysis and thus, increased lactate concentration (Vincent et al. 1990). Indeed, increased endogenous epinephrine concentrations observed in both animal models and patients with septic shock contribute to hyperlactataemia by enhancing Na-K-ATPase activity, which in turn accelerates aerobic glycolysis (Bundgaard et al. 2003; Levy et al. 2008; McCarter et al. 2002). These findings have been supported by observations showing that lactate production can be reduced by β2 antagonists, such as esmolol, and by Na-K-ATPase inhibitors, such as ouabain, during septic shock or when glycolysis is stimulated by epinephrine (James et al. 1996; James et al. 1999; Levy et al. 2005). Additionally, thiamine deficiency (beriberi disease) impedes pyruvate dehydrogenase activity or mitochondrial dysfunction (Svistunenko et al. 2006), altering pyruvate metabolism and causing hyperlactataemia. Furthermore, dysfunction of the pyruvate dehydrogenase enzyme has been documented in experimental and clinical sepsis, and it could also be related to increased lactate levels in septic patients (Nuzzo et al. 2015; Vary 1991; Vary 1996).
Diminished lactate clearance
Lactate is primarily metabolised by the liver, with additional kidney contribution. Cardiac myocytes also utilise lactate as an energy source under specific conditions, including exercise, β-adrenergic stimulation, and shock (Kline et al. 2000; Stanley 1991). Additionally, the brain can metabolise lactate during periods of increased metabolic demand (van Hall et al. 2009). Therefore, impaired reuptake of lactate by the cells represents another mechanism contributing to increased lactate levels, independent of tissue hypoxia. Conditions such as liver dysfunction or failure, cardiac surgery, and sepsis have all been associated with a reduced capacity to clear lactate (Almenoff et al. 1989; Mustafa et al. 2003; Tapia et al. 2015).
Indeed, hyperlactataemia may result primarily from impaired lactate reuptake rather than excessive production in haemodynamically stable patients with sepsis (Levraut et al. 1998). Although hepatic impairment of lactate reuptake is not universally observed in septic patients, it becomes clinically significant in those with existing or newly developed hepatic dysfunction (Revelly et al. 2005).
Lactate, Metabolic Acidosis, and Acidaemia
According to Stewart's approach to acid-base balance, three independent variables determine pH (H+ concentrations in the blood): (1) carbon dioxide, (2) strong ion difference (SID), which is basically the difference between strong cations and anions (i.e., those anions fully dissociated at physiological pH, including lactate and chloride), and (3) total nonvolatile weak acids (not fully dissociated at physiological pH, including albumin, globulins, and inorganic phosphate) (Stewart 1983). Thus, an elevation in blood lactate concentration invariably leads to metabolic acidosis due to a reduction in SID, reflecting over accumulation of strong negative ions. However, the relationship between lactate concentration and acidaemia (defined as an abnormally elevated H⁺ concentration [low pH]) is complex. Acidaemia may not always occur if concurrent processes reduce the concentration of other negative strong ions, thereby increasing the SID and driving the pH back toward normal. Indeed, Gattinoni et al. (2019) observed that among 1,741 septic patients, hyperlactataemia accompanied by acidaemia occurred exclusively in those with renal dysfunction, highlighting the critical role of the kidney to compensate for the acidosis mediated by lactate ions.
Lactate Levels and Prognosis
Irrespective of their origin, elevated lactate levels have been associated with increased short-term mortality in both sepsis and unselected critically ill patients (Zhang et al. 2014). In emergency patients presenting with infection, grouping lactate concentrations into low, intermediate, and high ranges revealed corresponding in-hospital mortality rates of 15%, 25%, and 38%. This dose-response relationship highlights the association between elevated serum lactate levels and increased mortality (Trzeciak et al. 2007). Likewise, Mikkelsen et al. (2009) demonstrated that intermediate (2.0–3.9 mmol/L) and high (≥4 mmol/L) serum lactate levels were independently associated with an increased mortality risk, even in the absence of organ failure or shock. Moreover, lactate levels > 4.0 mmol/L have been independently linked to increased mortality in septic patients without hypotension, thus emphasising the utility of lactate levels for prompt identification of patients at high risk of death (Casserly et al. 2015).
A post-hoc analysis from a randomised controlled trial found that septic patients admitted to the emergency department with hypotension and lactate levels above 2 mmol/L experienced significantly greater in-hospital mortality compared to those with hypotension but with lactate ≤ 2.0 mmol/L (26% vs. 9%; P < .0001) (Sterling et al. 2013). This reinforces the role of elevated lactate concentrations as a prognostic indicator independent of blood pressure. After adjusting for risk factors, septic patients with both fluid-resistant hypotension requiring vasopressors and elevated lactate levels experienced significantly higher in-hospital mortality compared to those presenting with either isolated hyperlactataemia or fluid-resistant hypotension without elevated lactate (lactate < 2.0 mmol/L) (Singer et al. 2016). Even lactate concentrations at the upper limit of the normal range (1.4–2.3 mmol/L) were associated with a worse prognosis than lower normal lactate levels in patients with sepsis (Wacharasint et al. 2012). Thus, increased lactate levels should represent an alert situation in patients with sepsis.
Lactate-Guided Resuscitation
The 2021 Surviving Sepsis Campaign guidelines recommend targeting reductions in serum lactate concentrations during initial resuscitation. However, despite lactate-guided fluid resuscitation being widely practiced, the supporting evidence remains limited, leading the guideline panel to issue only a weak recommendation (Evans et al. 2021).
In a multi-centre randomised controlled trial including 300 septic patients in an early phase, a resuscitation strategy aimed to normalise central venous oxygenation saturation (ScvO2) > 70% was compared with a strategy aimed to decrease lactate levels by at least 10% in the 6-hour duration study protocol (Jones et al. 2010). The hospital mortality rates were not significantly different between the lactate-guided clearance and ScvO2-guided groups (17% vs. 23%, respectively). No differences in fluid administration were observed between the two groups. In another randomised trial that enrolled 348 critically ill patients with a lactate level > 3.0 mmol/L, a lactate-guided resuscitation approach aiming to decrease lactate by ≥ 20% per 2 hours within the first 8 hours of ICU admission, was associated with lower in-hospital mortality when corrected for baseline imbalances (hazard ratio 0.61 [CI: 0.43–0.87]) (Jansen et al. 2010). However, the volume of fluid administered within the initial 8 hours differed minimally between groups (2194 ± 1669 mL vs. 2697 ± 1965 mL, p = 0.01), and even lactate levels decreased identically between groups, which makes it improbable that fluid differences alone accounted for the observed outcomes. Notably, patients allocated to lactate-guided resuscitation were more likely to receive vasodilators than controls (43% vs. 20%, p < 0.001).
However, the recent ANDROMEDA-SHOCK trial compared a lactate-guided vs a peripheral perfusion-guided resuscitation strategies in 424 patients with septic shock across 28 intensive care units. The lactate-guided group aimed to achieve a lactate reduction ≥ 20% every 2 hours within the 8-hour intervention period, while the perfusion-guided resuscitation group aimed to attain a peripheral capillary refill time (CRT) ≤ 3.0 seconds (Hernández et al. 2019) also within the 8-hour intervention period. At 28 days, mortality was 34.9% among the perfusion-guided resuscitation group and 43.4% among those allocated to lactate-guided resuscitation (hazard ratio 0.75 [95% CI, 0.55–1.02]; p = 0.06). The peripheral perfusion-guided resuscitation group received significantly less resuscitation fluids during the initial 8 hours and had a faster recovery of organ failure when compared with the lactate-guided resuscitation group. An ulterior post hoc Bayesian analysis of ANDROMEDA-SHOCK indicated a potential survival advantage in patients allocated to the peripheral perfusion-guided resuscitation group (Zampieri et al. 2020).
Collectively, these findings cast doubt on the effectiveness of using repeated lactate measurements to guide resuscitation strategies, particularly fluid administration (Legrand et al. 2024).
How to Use Lactate in Clinical Practice
A recent post-hoc analysis of serial lactate levels from the international multicentre CLASSIC trial compared restrictive vs liberal intravenous fluid strategies in patients with septic shock (Ahlstedt et al. 2024). In the original CLASSIC study (Meyhoff et al. 2022), the restrictive approach allowed fluid boluses exclusively to cases of significant hypotension or hypoperfusion. Median fluid volumes administered from days 1 to 3 were 5334 mL (3476–7578) in the restrictive group vs. 6919 mL (4721–9744) in the liberal group. Remarkably, the restrictive fluid strategy did not significantly alter the rate of hyperlactataemia resolution compared to standard fluid therapy (hazard ratio: 1.21 [95% CI: 0.89–1.65] for hyperlactataemia resolution on days 2–3). This finding remained consistent even in patients with elevated baseline lactate levels. Overall, the authors concluded that a liberal fluid strategy offers no advantages over a restrictive approach regarding lactate clearance or clinical outcomes. Accordingly, the authors claimed that lactate may not reliably reflect tissue hypoperfusion in patients with septic shock.
Therefore, persistent hyperlactataemia poses significant interpretive challenges due to multiple potential underlying mechanisms implied in lactate elevation. These include anaerobic glycolysis secondary to tissue hypoperfusion, particularly when severe microcirculatory dysfunction is present (Hernandez et al. 2013); increased aerobic glycolysis driven by adrenergic stress (Levy et al. 2005); reduced lactate reuptake by the liver; and impaired mitochondrial function or restricting pyruvate metabolism (Alegría et al. 2017; Hernandez et al. 2014). Identifying hyperlactataemia caused specifically by hypoperfusion is crucial, as optimising systemic perfusion under these conditions may reverse tissue ischaemia, thus potentially improving clinical outcomes. Conversely, inappropriate resuscitation efforts in cases where hyperlactataemia is unrelated to hypoperfusion could lead to harmful fluid overload and increased risk of complications (Hernandez et al. 2019). This underscores a key challenge in shock resuscitation: determining when persistent hyperlactataemia indicates ongoing inadequate tissue perfusion.
It has been reported that the time course of lactate normalisation during a successful resuscitation follows a biphasic curve: an early rapid decrease (between 0 and 6 hours) associated with normalisation of other flow-responsive variables such as ScvO2, CRT, and central venous-to-arterial PCO2 gradient (DPCO2), and followed by a later slower recovery trend potentially explained by non-flow-dependent mechanisms (Hernandez et al. 2014). Therefore, flow-sensitive variables such as ScvO₂, ΔPCO₂, CRT, and the peripheral perfusion index should be employed to aid in interpreting hyperlactataemia by identifying ongoing tissue hypoperfusion in this clinical context.
Several algorithms have been suggested to tackle this issue, utilising an approach that integrates multiple perfusion monitoring parameters. Flow-sensitive variables should be assessed whenever elevated lactate levels (> 2.0 mmol/L) are detected. Alterations in these variables indicate tissue hypoperfusion, prompting clinicians to enhance blood flow either through fluid administration in fluid-responsive patients or by initiating inotropic support in the presence of acute cardiac dysfunction (Figure 2).
Persistent hyperlactataemia accompanied by normal flow-sensitive parameters (such as ScvO₂, ΔPCO₂, and peripheral perfusion indices) may suggest a lower likelihood of ongoing tissue hypoperfusion. However, additional research is required to confirm this interpretation. Current evidence supports repeating measurements at intervals of every 1 to 2 hours (Vincent et al. 2016). Persistently elevated lactate levels should prompt clinicians to reassess the underlying diagnosis and evaluate whether the current therapeutic approach effectively addresses the cause of lactate elevation.
