Introduction:

Sepsis is one of the main causes of morbidity and mortality in the ICU, and when associated with SCM, the prognosis worsens; it can occur in up to 68% of patients. ECHO parameters have been proposed useful for the assessment of the LV systolic function, such as STRAIN-GLS and STRAIN-CS. In this study, the association of these with mortality in patients with sepsis in the ICU was assessed.

Material and methods:

A prospective, cross-sectional, cohort study was conducted in patients with sepsis admitted to the ICU of a hospital center in Mexico City from January 1st to July 31st, 2018. Transthoracic ECHO was taken within the first 24 hours of the diagnosis of sepsis.

Results:

Thirty patients were included, 17 were male (56.7%); the median age was 68.5 years (RIQ 56-84); the median hospital stay was 12 days (RIQ 7-17). The death rate was of 23.3%. The average length of stay in the ICU was 19.1 days (3-185 days). Independently assessed by CS, it was found that 36% of the population presented LV systolic dysfunction, while assessed by GLS, it was 40%. When comparing non-survivors and survivors, there were no significant demographic differences; a difference of proportions was observed in the number of patients with diabetes. Differences in medians of systolic pulmonary artery pressure (SPAP), VAI, CS and GLS were observed. When testing the ability to discriminate survivors against non-survivors, it was found that CS was marginally superior to GLS, APACHE (Acute Physiology Age and Chronic Health Evaluation) II, SAPS (Simplified Acute Physiology Score) II, and SOFA (Sequential Organ Failure Assessment). When testing the ability of CS and GLS to discriminate hospital stay greater than seven days, it was observed that the echocardiographic measures were superior. Cut-off points were selected for discrimination of survivors against non-survivors by analyzing sensitivity and specificity for the following echocardiographic measurements: CS ≥ -15.1 (S: 71.43%, E: 83.33%, LR(+) 4.3, LR(-) 0.34) and GLS ≥ -15.4 (S: 85.71%, E: 73.91%, LR(+) 3.3, LR(-) 0.19). In the multivariate analysis, it was found that a CS ≥ -15.1 was predictive of mortality during hospitalization in the study period and population, adjusted for other echocardiographic variables such as LVEF < 55%, GLS ≥ -15.4 and confounders such as sex [RM = 10.23 (95% CI, 1.01-103.2), p < 0.049]. In linear regression models, no predictive echocardiographic variables were found for days of hospital stay in the study period and population.

Conclusions:

The development of new ECHO techniques such as speckle tracking echocardiography (STE) has facilitated the ability to evaluate LV function through the quantitative evaluation of myocardial deformation; although they are complex techniques, if performed and interpreted appropriately, they are very useful in different clinical conditions. MD can be an early marker of mortality in patients with sepsis; obtained by measuring the GLS and CS, it could be a reliable predictor of the outcomes of patients in the ICU, and it can also potentiate scales such as APACHE II, SOFA and SAPS II to allow early identification of septic patients at high risk. There are still some obstacles to the regular clinical application of GLS and CS in septic patients in the ICU. The optimal GLS limit for the prediction of mortality in these patients remains uncertain, and the intrinsic differences between the populations could contribute to the observed differences. However, it is known to be an effective parameter for the quantification of left ventricular function, even more sensitive than the LVEF by bidimensional ECHO, depending relatively less on the operator and loading conditions. Current results should be confirmed in additional large-scale and multi-center studies. Therefore, it still remains in the field of clinical research for patients in critical condition.

        · resumen en Español | Portugués     · texto en Español     · Español ( pdf )