Introduction: The Sequential Organ Failure Assessment (SOFA) Score is a score validated to assess prognosis in septic patients. The initial SOFA score predicts in-hospital mortality: values above 11 are correlated with a mortal- ity more than 80%. Troponin I (TnI) is a marker of myocardial injury, but its role in sepsis is still controversial: it has been supposed that TnI could iden- tify patients with a worse prognosis. We aimed to evaluate if adding TnI to SOFA score resulted in a better prognostic performance in the septic patient. Materials and Methods: 81 patients affected by severe sepsis or septic shock and admitted to our Internal Medicine Department were enrolled. For each patient, SOFA score and serum TnI, procalcitonin (PCT), C-reactive protein (CRP) were evaluated at the admission. Outcome was defined as in-hospital death, and collected as a binary variable. TnI was recorded both as a continuous and as a dichotomous variable (≤0.05 or >0.05 ng/ml). Age and days of in-hospital admission were collected as continuous variables. Sex was treated as dichotomous. SOFA score was synthesized as an ordi- nal variable, considering partial pressure of oxygen, fraction of inhaled O2, platelet count, Glasgow coma scale, serum bilirubin and creatinine and the degree of hypotension (1) as single items of the scale. SOFA-T was calcu- lated adding 1 point to SOFA score if TnI levels were higher than 0.05 ng/ ml. The difference in levels of TnI between survivors and non survivors was analyzed with a Univariate/GLM including SOFA, CRP, PCT, age, sex and days of admission as covariates. ROC curve analysis was performed considering SOFA and SOFA-T to predict the main outcome. Analysis was performed with SPSS 13.0 for Windows systems. Results: Mean age was 77,7±11,24 years, males representing 53,8% of the sample. 33,8% of the patients died during the hospitalization. Mean SO- FA score was 6,42±3,24. The mean days of admission were 16,07±10,25. TnI had a mean concentration of 2,31±7,14 ng/ml, PCT 11,51±21,33 pg/ ml, CRP 14,96±11,1 mg/dl. At univariate/GLM model estimated mean TnI level was 1,028 ng/ml (SE: 0,722) among survivors, while it was 4,779 ng/ ml (SE:1,243) among non survivors, and this difference was statistically Figure 1. ROC curve analysis in predicting in-hospital death for patients with sepsis with SOFA or SOFA-T score significant (p<0.05). SOFA showed an AUC of 0,677 (SE 0,062; 95%CI 0,554-0,799) while SOFA-T had an AUC of 0,723 (SE 0,723; 95%CI 0,609- 0,837) in predicting the main outcome (Figure 1). The difference between the two curves was statistically significant (p<0.05). Discussion: TnI presence in a septic patient is deemed to be a marker of both a multi-organ dysfunction and of a septic cardiomyopathy. In both cases, the presence of these complications in an older, frail patient is associated to a worse prognosis. TnI levels were significantly higher in people who died of sepsis, and this was independent to other markers of prognosis, such as SOFA, CRP, procalcitonin and age. When included in SOFA score, TnI increased the AUC and the predictive value of this index. However, larger samples are required to further validate SOFA-T score index.
Tarquinio N, F.L. (2014). Myocardial injury markers in the septic patient: is it time for a SOFA-T score?. Springer.
Myocardial injury markers in the septic patient: is it time for a SOFA-T score?
Falsetti L
Writing – Original Draft Preparation
;
2014
Abstract
Introduction: The Sequential Organ Failure Assessment (SOFA) Score is a score validated to assess prognosis in septic patients. The initial SOFA score predicts in-hospital mortality: values above 11 are correlated with a mortal- ity more than 80%. Troponin I (TnI) is a marker of myocardial injury, but its role in sepsis is still controversial: it has been supposed that TnI could iden- tify patients with a worse prognosis. We aimed to evaluate if adding TnI to SOFA score resulted in a better prognostic performance in the septic patient. Materials and Methods: 81 patients affected by severe sepsis or septic shock and admitted to our Internal Medicine Department were enrolled. For each patient, SOFA score and serum TnI, procalcitonin (PCT), C-reactive protein (CRP) were evaluated at the admission. Outcome was defined as in-hospital death, and collected as a binary variable. TnI was recorded both as a continuous and as a dichotomous variable (≤0.05 or >0.05 ng/ml). Age and days of in-hospital admission were collected as continuous variables. Sex was treated as dichotomous. SOFA score was synthesized as an ordi- nal variable, considering partial pressure of oxygen, fraction of inhaled O2, platelet count, Glasgow coma scale, serum bilirubin and creatinine and the degree of hypotension (1) as single items of the scale. SOFA-T was calcu- lated adding 1 point to SOFA score if TnI levels were higher than 0.05 ng/ ml. The difference in levels of TnI between survivors and non survivors was analyzed with a Univariate/GLM including SOFA, CRP, PCT, age, sex and days of admission as covariates. ROC curve analysis was performed considering SOFA and SOFA-T to predict the main outcome. Analysis was performed with SPSS 13.0 for Windows systems. Results: Mean age was 77,7±11,24 years, males representing 53,8% of the sample. 33,8% of the patients died during the hospitalization. Mean SO- FA score was 6,42±3,24. The mean days of admission were 16,07±10,25. TnI had a mean concentration of 2,31±7,14 ng/ml, PCT 11,51±21,33 pg/ ml, CRP 14,96±11,1 mg/dl. At univariate/GLM model estimated mean TnI level was 1,028 ng/ml (SE: 0,722) among survivors, while it was 4,779 ng/ ml (SE:1,243) among non survivors, and this difference was statistically Figure 1. ROC curve analysis in predicting in-hospital death for patients with sepsis with SOFA or SOFA-T score significant (p<0.05). SOFA showed an AUC of 0,677 (SE 0,062; 95%CI 0,554-0,799) while SOFA-T had an AUC of 0,723 (SE 0,723; 95%CI 0,609- 0,837) in predicting the main outcome (Figure 1). The difference between the two curves was statistically significant (p<0.05). Discussion: TnI presence in a septic patient is deemed to be a marker of both a multi-organ dysfunction and of a septic cardiomyopathy. In both cases, the presence of these complications in an older, frail patient is associated to a worse prognosis. TnI levels were significantly higher in people who died of sepsis, and this was independent to other markers of prognosis, such as SOFA, CRP, procalcitonin and age. When included in SOFA score, TnI increased the AUC and the predictive value of this index. However, larger samples are required to further validate SOFA-T score index.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.


