Recent data have suggested that NT-proBNP might be a better, but by no means perfect marker of myocardial dysfunction and prognosis in patients with severe sepsis and septic shock compared to BNP (Table 4). After a pilot study had reported markedly elevated NT-proBNP levels in six patients with septic shock, the study by Roch et al evaluated NT-proBNP levels in 39 patients with septic shock who received mechanical ventilation and found higher median maximal NT-proBNP levels in nonsurvivors (34,028 pg/mL; interquartile range, 11,735 to 49,320 pg/mL) compared to survivors (7,856 pg/mL; interquartile range, 1,291 to 12,972 pg/mL; p = 0.002). An NT-proBNP level of > 13,600 pg/mL during the 24-h period after study inclusion has been shown to predict ICU mortality with a sensitivity of 73% and a specificity of 83% (area under the curve, 0.8). The NT-proBNP level was higher in nonsurvivors than in survivors at each time between study inclusion and day 7.
In addition, a weak inverse correlation between NT-proBNP level and LVSWI (r = -0.34) was found. The lowest LVSWI value during the first 24 h after study inclusion was the only independent predictor of an NT-proBNP level of > 13,600 pg/mL. These results are supported by those of a recent study indicating that NT-proBNP level was better correlated with LVSWI than BNP level in patients with respiratory failure of septic and nonseptic origin, and that in contrast to BNP levels, NT-proBNP levels were significantly higher in patients with an LVSWI of < 35 g/m/m2 than in those with an LVSWI of > 35 g/m//m2. However, NT-proBNP level could neither differentiate between high-PCWP vs low-PCWP respiratory failure nor predict the prognosis in this study setting, which was not restricted to patients with sepsis. An advantage of NT-proBNP over BNP might be its longer half-life (NT-proBNP half-life, 2 h; BNP half-life, 20 min). NT-proBNP may reflect hemodynamics and inflammatory stimuli over a longer period and thus might be more representative of the presence or absence of myocardial dysfunction and prognosis.
Table 4—Studies on the Impact of NT-proBNP Measurement in Critically III Patients
|Study||Study Population(Age, t y)||Severity of Disease||Serum Creatinine Level j;||Assessment of LV Performance||NT-pro-BNP Levels||Mortality||Relationships Among BNP, LV Performance, and Outcome|
|Roch et al||39 pts with septic shock and mechanical ventilation (63 ± 12 y)||SAPS II score on ICU admission, 52 ±21||Nonsurvivors, 225 ± 77 jiniol/l, survivors, 161 ± 81 jiniol/l,||РАС: LVSWI at 12 h: Nonsurvivors, 28 ± 11 g/m/m2: Survivors,42 ± 18 g/m/m2 LVSWI at 24 h: Nonsurvivors, 30 ± 12 g/m/m2: survivors,43 ± 23 g/m/m2||Highest level in the first 24-h period: nonsurvivors, 34,028 pg/mL (range, 11,735-49,320 pg/mL) survivors, 7,856 pg/mL (range, 1,291-12,972 pg/mL: Pts with lowest LVSWI < 35g g/m/m2, 16,122 pg/mL (range, 8,414-48^839 pg/mL): Pts with lowest LVSWI > 35 g/m/m2, 4,799 pg/mL (range, 2,090-9,966 pg/mL)||In hospital, 24/39 pts(56%)||Correlation between NT-pro-BNP, and LVSWI (r = —0.34): the highest NT-proBNP level in the 24-h period after study inclusion was an independent predictor of ICU mortality: NT-proBNP > 13,600 pg/mL predicted ICU mortality with an accuracy of 77%|
|Jefic et al||41 pts with respiratory failure defined as hypoxemia§ and infiltrates on chest x-ray (66.5 ± 16 y): sepsis/septic shock in 20 pts||APACHE II score, 18,5 ± 1: 39/41 pts (95%) mechanically ventilated||NA: creatinine clearance, 60.5 ± 7 mL/min||РАС: 34 pts with LVSWI < 35 g/m/m2: 18 pts with PCWP > 15 mm Hg||LVSWI < 35 g/m/m2, 13,528 ± 2,399 pg/mL: LVSWI > 35 g/m/nr, I,236 ± 83 pg/mL: nonsurvivors, II,777 ± 2,990 pg/mL: survivors, 11,630 ± 3,182 pg/mL||30 d, 17/40 pts (43%)||NT-proBNP correlated with LVSWI (r = —0.62) and CI (r = —0.44), no correlation with PCWP: NT-proBNP cannot differentiate low vs high PCWP respiratory failure: NT-proBNP correlated with creatinine clearance (r = —0.6): NT-proBNP without prognostic value|
|Brueckmann et al||57 pts with severe sepsis (55 ± 16.3 y)||APACHE II score, 26 (range, 19-30): 45/57 pts (79%) needed vasopressors||97 jiniol/l, (range, 81-163 |xmol/L)||РАС: LVSWI, 36.6 + 11.1 g/m/m2: CI, 4,5 I/min/nr- ТГЕ (n = 29): 6 pts with LVEF 35-50%, 4 pts with LVEF < 35%||Nonsurvivors, 1,431 pg/ niL (range, 712-1,920 pg/mL): survivors, 493 pg/mL (range, 314-1,126 pg/mL)||28 d, 16/57 pts (28%)||NT-proBNP correlated with creatinine (r = 0.58), cTnI levels (r = 0.68), APACHE II score (r = 0.42): higher NT-proBNP levels in nonsurvivors than in survivors: NT-proBNP level significant predictor for mortality|