Early lactate measurement is associated with better outcomes in septic patients with an elevated serum lactate level

Sepsis is one of the leading causes of in-hospital death in the intensive care unit (ICU), with a short-term mortality rate ranging from 30 to 50% [1, 2]. Elevation as well as slow clearance of serum lactate levels has been reported to be associated with increased mortality in sepsis [3, 4]. As a product of anaerobic metabolism, lactate elevation can to some extent reflect poor tissue perfusion, though it can be affected by other factors [5]. Thus, monitoring lactate levels may help clinicians understand the tissue perfusion situation and identify unrecognized shock, enabling them to adjust treatment in a timely fashion.

Because randomized controlled trials have demonstrated a significant reduction in mortality with lactate-guided resuscitation [6, 7], lactate measurement has been introduced into the Surviving Sepsis Campaign (SSC) guidelines as an important component of the treatment bundles. The 2012 SSC bundle emphasized that an initial lactate level must be measured between 6 h before and 3 h after sepsis onset, followed by a repeated measurement within 6 h of sepsis presentation if the initial value is elevated [8]. Levy et al. updated the SSC bundle to indicate that the initial lactate measurement should be performed within 1 h of onset and remeasured within 2–4 h to guide resuscitation if the initial lactate is > 2.0 mmol/L [9]. However, there are limited studies focused on the timing of lactate measurement and remeasurement and its effect on outcomes. One study by Han et al. [10] demonstrated that patients with severe sepsis or septic shock and initial lactate levels > 2.0 mmol/L had a 2% increase in the odds of death with each hour delay in lactate measurement. According to the analysis of 96 studies, Vincent et al. concluded that lactate measurement every 1–2 h is probably sufficient in most acute conditions, including sepsis [11]. However, the optimal timing of the initial lactate measurement for septic patients remains controversial, and whether earlier initial and repeated lactate measurements could make a difference for septic patients with a higher lactate level remains unexplored.

Therefore, we designed this study to examine the relationship between early lactate measurement (within 1 h after ICU admission) and the outcomes of septic patients with an elevated serum lactate level (> 2.0 mmol/L), as well as to characterize the association of delays in initial lactate measurement and remeasurement with 28-day mortality to better understand the benefit of lactate monitoring in sepsis management.

Our study revealed that an initial lactate level measurement made within 1 h after ICU admission was associated with lower risk-adjusted 28-day mortality rate in septic patients with an initial lactate level > 2.0 mmol/L, because these patients received earlier vasopressor administration after an early lactate measurement. Regarding the time course of lactate measurement, delays in initial lactate measurement and remeasurement in the EL group were associated with increased risk-adjusted 28-day mortality rates. Our study, for the first time, shows the benefit of early lactate measurement and remeasurement in septic patients with an initial lactate level > 2.0 mmol/L.

Lactate measurement is an important component in the SSC bundle, and the most recent SSC bundle recommends that the initial lactate level should be measured within 1 h to guide resuscitation [9]; however, the suggested timing of the initial lactate measurement is still controversial. Limited studies regarding early lactate measurement in septic patients are available. Seymour et al. [18] reported that a delay in the time to complete the 3-h SSC bundle, including serum lactate measurement, was associated with increased mortality in septic patients after emergency department admission, regardless of the initial lactate level. From 5072 patients with severe sepsis or septic shock, Pruinelli et al. concluded that a delay of 20 min in initial lactate level measurement significantly increased the risk of mortality [19]. Consistent with these findings, our study showed that the 738 patients (27.9%) who had lactate measurements completed within 1 h (EL group) had a lower 28-day mortality rate than did patients who had lactate measurements completed more than 1 h after ICU admission (LL group) after adjustment for confounders, and the relationship was robust after PSM. The present study shows timing evidence that an early lactate measurement is beneficial in sepsis management for septic patients with an initial lactate level > 2.0 mmol/L; specifically, there is a benefit of an initial measurement within 1 h after ICU admission.

Measuring the lactate level itself does not affect clinical outcomes unless it is tied to therapeutic interventions. Several studies have suggested that a timely lactate measurement is associated with more rapid intervention delivery in septic patients. Han et al. [10] demonstrated that early lactate measurement was associated with early antibiotic therapy and IVF in septic patients with initial intermediate or elevated lactate levels, the latter of whom have increased hospital mortality when treatment is delayed [20]. Unfortunately, which therapeutic interventions contribute to the beneficial effect of early lactate measurement on mortality has not been clearly determined. We tested several interventions to interpret the benefit of early lactate measurement and found that the EL group had shorter times to initial vasopressor administration and antibiotic therapy, and although there were no significant differences in the onset of IVF treatment, patients in the EL group received more IVF within 6 h and 24 h. We then used CMA and found that the beneficial effect of early lactate measurement on 28-day mortality in septic patients was partly due to the early administration of vasopressors.

The mediating effect of the time to initial vasopressor administration has been shown by multiple studies. A retrospective cohort study concluded that septic shock patients could benefit from the early administration of norepinephrine [21]. Further randomized trials demonstrated that early norepinephrine administration was strongly associated with more shock control at 6 h, although there was no significant effect on 28-day mortality [22]. Even though we found no pronounced mediation effect of the time to initial IVF and the time to initial antibiotic therapy, these data should not be employed as evidence in sepsis management. There are several explanations for our result. First, it is not the time to antibiotic therapy but the time to appropriate antibiotic therapy that determines the outcome of sepsis [23], and it is difficult for us to test the appropriateness of the antibiotics administered using data from the MIMIC database. Second, the initiation of IVF therapy on the one hand increases the stroke volume, while on the other hand, it contributes to organ edema and dysfunction and increases the duration of organ support [24, 25]. More research on the mediation of therapeutic interventions is needed to assess the effect of early initial lactate measurement on mortality in septic patients.

Our results confirmed the benefit of early lactate measurement, concurrently suggesting that factors that could impact the outcomes of septic patients were complex, such as the type of vasopressor use, appropriate use of antibiotics, choice and volume of resuscitation fluids, corticosteroid use, sedation, and analgesia [26]. Hence, the lactate level should not be seen as the only index to guide resuscitation; but rather, clinicians should make an assessment of the general state of each septic patient and make a comprehensive plan of treatment.

Additionally, multivariate logistic regression analysis in our study further demonstrated that the timing of lactate measurement was an important determinant of 28-day mortality. This association was consistent with the results of the current limited studies. In a retrospective study, the time to the measurement of serum lactate was associated with in-hospital mortality in septic patients (OR 1.04; 95% CI 1.02–1.06; p < 0.001). Similar findings were found in Han et al.’s study [10], in which septic patients with initial lactate levels > 2.0 mmol/L had a 2% increased probability of mortality. The OR reported is dissimilar possibly because of the heterogeneity of participants, different primary endpoints, and different adjusted covariates.

To the best of our knowledge, there is no direct information for determining the optimal time for lactate remeasurement in septic patients. The latest SSC bundle recommends [9] that the remeasurement of lactate should be conducted within 2–4 h if the initial lactate level is elevated (> 2.0 mmol/L). However, this recommendation is based on one study by Jasen et al. [6], in which patients with lactate levels > 3.0 mmol/L in the lactate group were treated with the objective of decreasing their lactate levels by 20% or more per 2 h in the initial 8 h of their ICU stays; these patients had a reduced hospital mortality compared to the control group. In our study, we found that remeasuring the lactate level 3 h after the first measurement could affect the outcome in septic patients with lactate levels > 2.0 mmol/L. Our findings provide new evidence supporting the need for clinicians to take earlier action for both the initial lactate measurement and the remeasurement in the management of sepsis.

Our study has several limitations. First, the definition of sepsis was consistent with the third sepsis definition (documented or suspected infection coupled with an acute change in total SOFA score ≥ 2 points), but the diagnoses of infection were unclear [27]. Hence, we tried to include septic patients with positive blood culture results in the sensitivity analysis. Second, data from the MIMIC databases encompass more than 10 years, and the recommendations of the severe sepsis guidelines were changed during the study period, which is a confounding factor affecting the current clinical application of our results. Therefore, we included the admission period (before 2008 and 2008–2012) in our model, and the results were adjusted for the admission period. Third, there are multiple unmeasured confounders [28] that could impact our findings, such as different treatment affecting the lactate level after initial measurement, immunosuppression status, and the interventions before lactate measurement. Fourth, because of the retrospective nature of this cohort study, it is difficult for us to investigate the association between the timing of lactate remeasurement and mortality in all participants; we assessed only the EL group. Fifth, the causes of high lactate levels could be multifactorial in addition to hypoxemia, such as liver disease and drugs, including epinephrine, metformin, and linezolid. Whether the elevated lactate was caused by hypoxemia, drug use, or other diseases was hardly illuminated in the MIMIC-III database. Finally, the causal relationship between early lactate measurements and 28-day mortality was not explored clearly, and the therapeutic interventions after initial lactate measurement were complex, multiple, and often depended on the decision of the clinicians on duty. We examined the mediation effects of the time to initial vasopressor, antibiotic, and IVF administration separately. These effects need to be explored in future studies.

In conclusion, early lactate measurement (within 1 h after ICU admission) in septic patients with elevated serum lactate levels (> 2.0 mmol/L) is associated with a lower risk-adjusted 28-day mortality. The decrease in time to the administration of vasopressors may have mediated this effect. Repeating the measurement within 3 h after the initial measurement is appropriate for patients whose lactate was measured within 1 h of admission.