Prognostic Value of Serum Lactate Levels in Critically Ill Patients in an Intensive Care Unit

The seriously ill patient is currently a challenge for both the medical staff and the health system, given its complexity and variability. New strategies and therapies to reduce mortality, hospitalization time, sequelae and costs are required [1].

Severity indexes or prognostic scales are necessary methods to estimate the severity of anatomical lesions, physiological alterations and the probability of survival of the patients. Validation of these indices has allowed their widespread use and the adoption of a universal language on the subject [2].

There are different scales to assess the clinical status of a patient, which allow a diagnosis of the existing severity and which help anticipate possible imminent and late complications, including death [3].

For several years the use of a predictor of organic dysfunction in critically ill patients (CIP) has been proposed, recommending that the ideal should meet the criteria in the context of critical care, be simple, routinely reproducible and promptly evaluable [3].

Serum lactate levels can be a useful tool in confirming tissue hypoperfusion and organic dysfunction. Knowledge of this marker allows the establishment of values ​that allow an early indication of the severity of the patient’s clinical condition and the ability to monitor the progression of the critical situation, thus impacting on the patient’s prognosis and possibly reducing mortality [4,5].

According to the published literature, multiple studies relate the serum lactate levels in patients with septic [6] and polytraumatized states [7], but data concerning the general context of intensive care remain scarce.

The objective of this study is to determine the prognostic value of serum lactate in patients admitted to the ICU.

During the study period, 170 patients were admitted to the ICU, of which 15 patients were excluded because they did not present serum lactate level determinations in the first six hours after admission and a further twenty due to incomplete clinical records in the database. A total of 145 subjects were enrolled in the study (85%)of which twenty-eight died in the ICU (19%).

The general characteristics of the population studied, stratified according to serum lactate levels in the first six hours after admission to the ICU are shown in (Table 1). Traumatic (p = 0.05) and infectious (p = 0.041) causes and the distribution of the APACHE II prognostic score (p = 0.002) were significantly related to high serum lactate levels and the highest probability of death in the ICU. A good association was also observed in patients who required hemodynamic support (p = 0.008) and mechanical ventilation (p<0,0001).

The median serum lactate was 2.1 mmol / l (RIQ 1.4-4.2 mmol / l). Table 2 shows the analytical values and physiological variables in the first 6 hours after admission to the ICU. It was found that the distribution of serum lactate was higher in the group of patients who presented high creatinine levels while the mean blood pressure was lower in that group, with a statistically significant relationship in relation to mortality in the ICU.

When the analysis of the Kaplan-Meier curve was performed (figure 1), patients who presented serum lactate ≤ 2mmol / l had a higher probability of survival with a median stay of 14 days in the ICU, which revealed a statistically significant difference in relationship to the group of patients who presented elevated serum lactate with a Log Rank (Mantel-Cox) (p <0.0001).

Fig. 1

In the Cox regression analysis (Table 3), of all the variables studied, only serum lactate was an independent predictor of mortality with a statistically significant relationship (AHR 1.28 [1.07-1.53], p = 0.008).

In every patient, but especially in the critically ill patient, a doctor has to establish a diagnosis, determine the severity of the disease, decide on a treatment, predict and monitor the degree of response in order to make all necessary adjustments before the critical condition take the patient to the point of no return. To achieve this, it is essential to have a biomarker as an aid to treatment planning in those critical moments. Serum lactate levels have been proposed as a useful tool in such situations [4].

In critically ill patients, elevated serum lactate levels at the time of admission to the hospital as a static value over time are related to higher mortality. In 2010, Soliman et al studied the lactate levels of 433 patients of a medical-surgical ICU defining hyperlactatemia as a serum concentration greater than or equal to 2 mEq / L 45% of the patients had hyperlactatemia and found a direct relationship between lactate levels and risk of death, reaching a mortality of 17% with lactate concentrations between 2-4 mEq / l and 64% in those with concentrations greater than 8 mEq / l. Also, it was correlated with a longer hospital stay in the ICU and higher APACHE II and SOFA scores [4,8]. The same results were found in the present study, in which the median Apache II score was 14 points in patients with serum lactate greater than 2 mmol / l, obtaining a powerful association between this group of patients and the probability of death in ICU.

At present, it is proposed that in the patient with polytrauma, serum lactate levels rise not only due to the hemodynamic instability that arises as a result of circulatory shock, but also due to the trauma caused by an increase in stimulation of β2 receptors, big producers of adrenaline. All this leads to an increase in pyruvate that saturates the capacity of the enzyme pyruvate dehydrogenase, which causes the metabolism to deviate and produce an increase in lactate levels.[5].

A previous study on traumatized patients showed that an initial lactate level of more than 4 mmol / L is associated with an increase in the probability of death in the acute phase [9]. Zachary et al. retrospectively evaluated lactate levels in patients with closed or open trauma, at least twice in the first twenty-four hours of admission [10]. The recorded mortality was 7.3%. The average initial lactate levels of the survivors were lower than the deceased, with a statistically significant difference [10]. In a recent study at the San Vicente Foundation University Hospital (Medellin, Colombia), 251 polytrauma patients were evaluated, who were given lactate measures at admission, at 6 and 24 hours, 15.5% died, serum lactate upon admission was 4.6 mmol / L (IQR = 2.9-6.9) and an important association with mortality was found in the adjusted logistic regression model [11]. In our study, traumatic causes and serum lactate greater than 2 mmol / l were significantly related to death in the ICU.

According to world literature, there are multiple reports that show the relationship between the elevation of serum lactate and the occurrence of septic states [6,12, 13, 14, 15], this is due to the presence of an imbalance between the genesis and clearance of serum lactate. It is proposed that there is a state of critical tissue hypoperfusion that affects the synthesis of ATP by blocking the pyruvate dehydrogenase enzyme due to hypoxia this, in turn, alters with the conversion of pyruvate to lactate [16]. In one study, 830 patients admitted to an emergency ward, with a diagnosis of severe sepsis, were classified into three groups according to the level of lactate in low risk. The three groups were defined as; a level of lactate less than 2; intermediate from 2 to 3.9 and high greater than 4. At twenty-eight post-admission days, mortality was of 15.4%, 37% and 46.9%, respectively for the three defined lactate levels. In contrast, in patients who did not present with shock, mortality was 8.7%, 16.4% and 31.8%, respectively [4].

Our results concur with the literature, since the cause of admission of infectious origin, and the presence of elevated serum lactate levels showed a significant relationship with mortality in the ICU.

Concerning the need for mechanical ventilation, hemodynamic support and presence of arterial hypotension, in the present study, an association with hyperlactatemia and death in ICU was also found. Despite the effect that mechanical ventilation has on cardiac output and tissue perfusion, this alone did not influence the elevated levels of serum lactate. It is concluded that the cause may be related to the fact that many of these patients presented in a state of shock, with hemodynamic instability and need for support with vasoactive amines, which in turn results in a decreased blood flow to the tissues, favouring anaerobic metabolism, with lactic acidosis and hyperlactatemia.

Concerning the ICU survival time, patients who presented high serum lactate levels died more quickly than those who presented normal parameters, with significant statistical differences between both groups of studies (figure 1). On the other hand, in the Cox regression analysis, serum lactate levels were only associated with an increased risk of dying in the ICU (AHR 1.28 [1.07-1.53], p = 0.008).

It is accepted that observational studies such as the present one, do not allow stratified randomization. The study was restricted to one ICU, but never the less addressed an ICU population with diverse medical, surgical, traumatic and infectious conditions, which we feel moderates its limitations.

The assessment of a critically ill patient by determining serum lactate levels provides useful and valuable information. Its measurement is not expensive, it is fast, and it is available.