Acute heart failure (AHF) leads to a significant impaired organ oxygen delivery caused by a reduced cardiac output (CO). Severe AHF and cardiogenic shock (CS) are associated with a high mortality rate and require intensive care unit (ICU) admission for treatment with catecholamines and/or mechanical ventilation (1).
Interestingly, patients may develop an inflammatory activation and multi-organ failure despite restoration of cardiac output (2–4). Besides the kidneys and the brain, one of the very first organs, which is impaired by reduced CO during AHF or CS, is the intestinal tract, an organ system characterized by a strong sensitivity for hypoxic epithelial injury (5). It is not only one of our biggest organs; its perfusion also takes almost 30% of the CO of a healthy person. The increased systemic vascular resistance (SVR) during AHF and CS in particular leads to an impaired perfusion of the intestinal tract, which may be additionally worsened by catecholamine treatment. An inflammatory response, transmigration of bacteria, or the release of cytoplasmic proteins into the circulation may be a result (6).
Among these cytoplasmic proteins are the fatty-acid-binding proteins (FABP), which belong to a group of cytoplasmic proteins with high organ specificity. iFABP, a protein specifically located in small bowel enterocytes, constitutes up to 2% of the cytoplasmic protein content of the mature enterocyte and is liberated into the circulation upon intestinal epithelial injury (7, 8). Therefore, plasma iFABP levels may reflect the extent of intestinal epithelial cell damage in patients with hypoxic cell injury during hypoperfusion. It has been shown that urinary concentrations of iFABP are increased in patients with septic shock (9) and that plasma levels of iFABP are associated with presence of shock in a mixed cohort of critically ill patients (10).
The aim of this study was to investigate whether circulating levels of iFABP are associated with mortality in patients with severe AHF or CS, requiring ICU admission.
PATIENTS AND METHODS
Subjects and study design
This was a single-center, prospective observational study (11, 12). All consecutive patients with AHF or CS who were admitted to the medical ICU of the Department of Internal Medicine II, Medical University of Vienna, between August 2012 and August 2013 were included. The study was approved by the ethical committee of the Medical University of Vienna and complies with the Declaration of Helsinki. Informed consent was obtained from conscious patients, for unconscious patients, the need for informed consent was waived by the ethical committee. Baseline demographics and clinical history were recorded in all patients as well as all major interventions preceding ICU admission or taking place within the first 72 h of ICU stay (including mechanical ventilation, surgery, intra-aortic balloon counterpulsation, extracorporeal membrane oxygenation, extracorporeal renal replacement therapy, and use of catecholamines). 30-day survival was recorded in all patients.
A patient was considered to be in cardiogenic shock if he or she had a systolic blood pressure of less than 90 mmHg for more than 30 min or was on catecholamines to maintain a systolic pressure above 90 mmHg. Further if he or she had clinical signs of pulmonary congestion, reduced cardiac output (reduced left ventricular ejection fraction by echocardiography or reduced cardiac index) and had impaired end-organ perfusion (altered mental status; cold, clammy skin and extremities; oliguria with urine output of less than 30 mL per hour; or serum lactate level higher than 2.0 mmol per liter) (13).
Blood was drawn within 24 h after admission and at day 3 from the arterial or central venous line into a serum separator tube and a 3.8% sodium citrate vacuette tube (both Greiner Bio-One) and immediately centrifuged (4°C, 3,000 RPM, 15 min) and stored at −80°C for later analysis.
For the detection of standard laboratory markers including procalcitonin, C-reactive protein, NT-proBNP, and liver and kidney parameters, blood was analyzed in the Central Laboratory of the General Hospital of Vienna. Lactate was measured using a benchtop blood gas analyzer (ABL800 FLEX, Radiometer Medical, Brønshøj, Denmark).
Measurement of iFABP
Serum levels of iFABP were measured by specific ELISA (RND Systems, Minneapolis, Minn) according to the manufacturer's instruction. Assay range was 15.6 pg/mL to 1,000 pg/mL. Higher values were determined after dilution with assay diluent. Intra-assay coefficient of variation was 4.1% and inter-assay coefficient of variation was 11.1%.
Sample size calculation analysis revealed that in a cohort with a mortality rate of 30%, given a power of 0.8 and a significance level of 0.05, we would need 90 patients to detect a difference of 50% in iFABP serum levels between survivors and non-survivors. Categorical variables are summarized as counts or percentages and are compared by the χ2 or by Fisher exact test as appropriate. Continuous variables are expressed as median (interquartile range). Parametric data was compared by the unpaired Student t test. Non-parametric data (assessed by the Kolmogorov–Smirnov test) was compared by Mann–Whitney test. Cox proportional hazard regression analysis was performed to assess the effect of iFABP and other intestinal markers on survival. We adjusted the multivariate model for confounding effects of baseline variables that differ between survivors and non-survivors by a P value < 0.05. Kaplan–Meier analysis (log-rank test) was applied to verify the time-dependent discriminative power of quartiles of iFABP. Two-sided P values of <0.05 indicated statistical significance. SPSS 22.0 (IBM Corporation, Armonk, NY) and R (The R Foundation for Statistical Computing, Vienna, Austria) were used for all analyses.
We included 90 consecutive patients with cardiogenic shock (74.4%) or severe acute heart failure (25.6%) admitted to a cardiac ICU. 30-day survival was 64.4%. Baseline characteristics of patients are presented in Table 1. 76.7% of patients were male and the median age was 64.7 (49.4–74.3) years. 75.3% of patients were treated with vasopressors and 61.8% of patients needed mechanical ventilation. Primary diagnosis of patients was non-ST-elevation myocardial infarction (NSTEMI) in 28.9%, ST-elevation myocardial infarction (STEMI) in 13.3%, decompensated congestive heart failure in 43.3%, and other conditions leading to severe acute heart failure or cardiogenic shock in 14.5%.
Serum levels of iFABP, primary diagnosis, and markers of organ perfusion
Serum levels of iFABP at admission were 272.8 IQR 129.6–565.9 pg/mL and at day 3 iFABP levels were 444.6 IQR 191.8–780.5 pg/mL (P = 0.21 vs. day 0). Serum levels of iFABP were similar in patients with AHF and in patients with CS at admission (P = 0.21) and at day 3 (P = 0.73). In addition, iFABP levels were not associated with primary diagnosis. At admission iFABP levels were not different in patients after cardiopulmonary resuscitation (CPR) compared with patients without need for CPR (P = 1.0). Interestingly, patients with CPR showed significantly lower levels of iFABP at day 3 compared with patients without CPR (281.8 IQR 138.5–538.0 pg/mL vs. 532.0 IQR 325.4–1294.1 pg/mL; P < 0.001). Serum levels of iFABP at admission showed associations with lactate levels (R = 0.29; P = 0.006), serum creatinine (R = 0.27; P = 0.004), and creatinine clearance (R = −0,34; P = 0.002). In contrast, iFABP at admission did not correlate with NT-proBNP and was not associated with liver parameters at day 0 but showed some correlations with bilirubin (R = 0.32; P = 0.003) and aspartate aminotransferase (R = 0.24; P = 0.03) and an inverse correlation with prothrombin time (R = −0.24; P = 0.03) at day 3. At admission, serum levels of iFABP correlated with norepinephrine (R = 0.23; P = 0.03) but not dobutamine (−0.10; P = 0.44) dosing. In contrast, serum levels of iFABP at day 3 were not associated with kidney or liver parameters and were not associated with vasopressor treatment (data not shown).
iFABP levels and 30-day mortality
Patients who died within 30 days after admission showed significantly higher levels of iFABP at day 0 as compared with survivors (286.0 IQR 206.2–3413.9 pg/mL vs. 239.7 IQR 105.8–479.6 pg/mL; P < 0.05). In contrast, iFABP levels at day 3 were not associated with mortality (422.9 IQR 102.3–567.2 pg/mL vs. 452.7 IQR 216.9–918.3 pg/mL; P = 0.19; Fig. 1). Patients with serum levels of iFABP at admission in the highest quartile (iFABP ≥ 588.4 pg/mL) had a 2.6-fold risk (P = 0.01) of dying as compared with patients in the lower quartiles (Fig. 2). Cox regression analysis revealed that this was independent of demographics, NT-proBNP levels, serum lactate, vasopressor use, and creatinine clearance (Table 2). Interestingly, extensively elevated admission levels of iFABP above the 90th percentile (iFABP ≥ 10208.4 pg/mL) showed an excessive mortality rate of 88.9%. The c-statistic for iFABP levels was 0.72 (0.62–0.82). Addition of iFABP levels to a clinical model including NT-proBNP, age, and gender improved the c-statistic from 0.71 (0.60–0.81) to 0.78 (0.68–0.89); P < 0.05. Determination of iFABP in addition to NT-proBNP at day 0 showed enhancement in individual risk prediction indicated by a net reclassification improvement of 28.2% (P < 0.05). The integrated discrimination increment for iFABP at day 0 was 0.131 (P < 0.05).
Comparison of the predictive value of iFABP with markers of organ perfusion
After adjustment for age and gender, the hazard ratio for 1-SD increase of iFABP was 1.58 (95% CI 1.26–2.36, P < 0.0001). Cox-regression analysis revealed that at day 0 iFABP was the second strongest predictor of mortality after serum lactate (Fig. 3A) with a hazard ratio of 1.71 (95% CI 1.30–2.24; P < 0.0001). In addition, kidney parameters and pancreas enzymes at admission were associated with survival whereas liver parameters at day 0 did not predict outcome. In contrast to day 0, at day 3 iFABP serum levels, serum creatinine and blood urea nitrogen (BUN) were not associated with outcome whereas an increase of bilirubin and aspartate transaminase (AST) were predictors of mortality at the latter timer point (Fig. 3B).
Association of iFABP levels with central venous pressure and central venous oxygen saturation
To examine whether venous congestion or hypoperfusion is associated with increased iFABP levels, we correlated central venous pressure (CVP) and central venous oxygen saturation (ScvO2) with iFABP levels at 0 and day 3. CVP measurement at day 0 was available in 71 and at day 3 in 60 patients whereas ScvO2 was available in 52 patients at day 0 and in 37 patients at day 3. ScvO2 at day 0 was not correlated with iFABP (R = 0.1; P = 0.51), however ScvO2 at day 3 showed an inverse correlation with iFABP levels (R = −0.35; P < 0.05). In contrast, CVP did not correlate with iFABP levels at day 0 (R = −0.035; P = 0.77) and at day 3 (R = 0.17; P = 0.25).
Association of iFABP levels at admission with markers of inflammation and infection
Patients with iFABP levels at admission in the fourth quartile did not show increased levels of the acute phase marker C-reactive protein (data not shown). However, these patients showed a significant increase of procalcitonin, a more specific marker for bacterial infection (14), at day 0 [0.34 (0.14–1.2) vs. 2.6 (0.29–9.9) ng/mL; P = 0.023) and day 3 [0.35 (0.16–0.97) vs. 1.9 (0.44–7.2) ng/mL; P = 0.007), respectively.
Association of iFABP serum levels with intestinal complications
When clinical suspicion of intestinal complications was present, an abdominal multidetector computer tomography was performed to rule out gut perforation, acute mesenteric ischemia, nonocclusive mesenteric ischemia, or ileus. Patients who underwent abdominal multidetector computer tomography (n = 8, 8.9% of the total cohort) did not have higher levels of iFABP as compared with patients without clinical suspicion of intestinal complications (data not shown). Of these 8 patients who underwent abdominal multidetector computer tomography only one patient showed an intestinal complication, namely signs of nonocclusive mesenteric ischemia (NOMI) and a perforation of the colon. Interestingly, this patient showed iFABP levels at admission >45,000 pg/mL.
In our prospective study including 90 consecutive patients with severe acute heart failure or cardiogenic shock, patients who died within 30 days after admission to the cardiac ICU, showed significantly higher levels of iFABP at day 0 as compared to survivors. Patients in the highest quartile had a 2.5-fold risk of dying, independent of demographics, NT-proBNP levels and vasopressor use. In addition, extensively elevated admission levels of iFABP above the 90th percentile showed an excessive mortality rate of 88.9%. At day 0, iFABP levels showed the highest hazard ratio for 1-SD increase compared with markers for kidney and liver function. In contrast, at day 3 iFABP plasma levels were not associated with outcome but increased liver and pancreas parameters were predictors of mortality. In addition, iFABP significantly improved the c-statistic of a clinical model including NT-proBNP, age, and gender, suggesting that measurement of iFABP adds complimentary prognostic information to this powerful biomarker for acute heart failure (15).
iFABP is specifically located in small bowel enterocytes and it displays a high affinity binding for long chain fatty acids and other hydrophobic ligands, thus it is believed to be involved with uptake and trafficking of lipids in the enterocyte, however any extracellular function of the protein is not known so far (16). iFABP is released into the circulation upon destruction of those cells (7). It has been shown that urinary concentrations of iFABP are increased in patients with septic shock (9) and that plasma levels of iFABP are associated with presence of septic, hemorrhagic, or cardiogenic shock and with 28-day mortality in a mixed cohort of critical ill patients (10). Interestingly, although iFABP correlated with serum lactate levels in our cohort, serum levels of iFABP were similar in patients with acute heart failure and in patients with cardiogenic shock.
Acute heart failure and cardiogenic shock are associated with decreased CO and increased SVR that may be additionally worsened by treatment with catecholamines. This is reflected in our study by the fact that iFABP levels correlated with norepinephrine dosing. Decreased CO and increased SVR may lead to an imbalance between the demand and the delivery of oxygen to the splanchnic area leading to acute mesenteric ischemia causing enterocyte damage and increased intestinal permeability (17). For most patients with acute decompensation of congestive heart failure, venous congestion and not CO is the driver for organ dysfunction (18). Therefore, we tested whether iFABP is associated with venous congestion as measured by CVP or whether iFABP is more associated with tissue hypoperfusion as reflected by ScvO2. In our study, iFABP was not correlated with CVP but iFABP at day 3 correlated inversely with ScvO2 suggesting that not venous congestion but a decreased CO leading to tissue hypoperfusion may cause cellular leakage of iFABP.
Recently, it has been shown that iFABP levels correlate with whole blood endotoxin activity in patients after cardio-pulmonary resuscitation (8). Interestingly, the unspecific acute phase marker C-reactive protein did not correlate with iFABP levels in our study, but procalcitonin, that is a more specific and sensitive marker for bacterial infection than for the acute phase response (14), was increased in patients in the fourth quartile of iFABP.
It has been shown that patients may develop an inflammatory activation and multi-organ failure despite restoration of cardiac output (2–4). Whether an increase in iFABP solely reflects decreased CO and increased SVR and is thereby a marker for the severity of acute heart failure and cardiogenic shock or whether enterocyte damage and increased intestinal permeability leading to endotoxemia is causative involved in a complicated course leading to multi-organ failure and increased mortality cannot be clearly answered by our study. Profound mesenteric hypoperfusion may lead to NOMI, a condition with necrosis of the intestine that occurs despite preserved patency of the large mesenteric vessels and that is associated with a very high mortality (6). In our study cohort, 8.9% of patients underwent abdominal multidetector computer tomography because a suspicion of abdominal complications. However, only one patient was diagnosed with NOMI that was associated with a very high level of iFABP.
Interestingly, an early increase of iFABP was associated with mortality, whereas high levels of iFABP at day 3 did not predict outcome. This could be caused by several factors: iFABP is eliminated by the kidneys and a worsening of kidney function could lead to an accumulation of iFABP. Therefore, iFABP on day 3 may not only be determined by gut hypoperfusion but also by kidney function. One could also speculate that an improved splanchnic perfusion after treatment may lead to a “wash out” effect similar to the observed increase of cardiac enzymes after successful coronary reperfusion (19). If this is the case, increased iFABP levels on day 3 could either reflect a splanchnic hypoperfusion due to prolonged shock or could reflect a successful resuscitation of splanchnic perfusion leading to a “wash out” of the protein. In addition, we want to point out that similar to iFBAP, surrogates of renal function that are well known for their importance regarding the prognosis of the patients (i.e., creatinine and BUN) were predictive only at day 0 but not at day 3.
It has to be pointed out that iFABP showed the highest hazard ratio for 1-SD increase at day 0, suggesting that it is a very sensitive and early marker for tissue hypoperfusion. In contrast, a shock liver or hypoxic hepatitis with elevation of liver parameters occurred at the later time point suggesting that these markers together with pancreas enzymes may reflect a prolonged splanchnic hypoperfusion (20) whereas iFABP may be a very early and sensitive marker for reduced intestinal perfusion.
One limitation of this study is that iFABP is eliminated by the kidneys. Therefore, acute renal failure could cause increased levels of iFABP. However, we were able to demonstrate that, although iFABP and creatinine showed a univariate correlation, the predictive value of iFABP was independent of creatinine clearance. To understand the mechanisms that lead to increase of iFABP, we correlated CVP and ScvO2 with iFABP levels; however, these parameters were not measured in all patients and were not available when the central line was placed in the groin. As this was an observational study, we can only analyze the predictive abilities of circulating levels of iFABP but cannot draw any functional conclusions of this association. However, we want to emphasize the role of splanchnic hypoperfusion in the course of severe acute heart failure and cardiogenic shock.
In conclusion, we hereby provide evidence for the predictive value of iFABP serum levels for 30-day survival of patients with acute heart failure or cardiogenic shock admitted to a cardiac ICU. This data suggests that early splanchnic hypoperfusion may play an important role for the outcome of these patients.
1. Ponikowski P, Voors AA, Anker SD, Bueno H, Cleland JGF, Coats AJS, Falk V, Gonzalez-Juanatey JR, Harjola VP, Jankowska EA, et al. ESCSD Group. 2016 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure: The Task Force for the diagnosis and treatment of acute and chronic heart failure of the European Society of Cardiology (ESC)Developed with the special contribution of the Heart Failure Association (HFA) of the ESC. Eur Heart J
37 27:2129–2200, 2016.
2. Hochman JS. Cardiogenic shock
complicating acute myocardial infarction: expanding the paradigm. Circulation
107 24:2998–3002, 2003.
3. Geppert A, Dorninger A, Delle-Karth G, Zorn G, Heinz G, Huber K. Plasma concentrations of interleukin-6, organ failure, vasopressor support, and successful coronary revascularization in predicting 30-day mortality of patients with cardiogenic shock
complicating acute myocardial infarction. Crit Care Med
34 8:2035–2042, 2006.
4. Geppert A, Steiner A, Zorn G, Delle-Karth G, Koreny M, Haumer M, Siostrzonek P, Huber K, Heinz G. Multiple organ failure in patients with cardiogenic shock
is associated with high plasma levels of interleukin-6. Crit Care Med
30 9:1987–1994, 2002.
5. Brandt LJ, Feuerstadt P, Longstreth GF, Boley SJG. American College of: ACG clinical guideline: epidemiology, risk factors, patterns of presentation, diagnosis, and management of colon ischemia (CI). Am J Gastroenterol
110 1:18–44, 2015.
6. Bourcier S, Oudjit A, Goudard G, Charpentier J, Leblanc S, Coriat R, Gouya H, Dousset B, Mira JP, Pene F. Diagnosis of non-occlusive acute mesenteric ischemia in the intensive care unit
. Ann Intensive Care
6 1:112, 2016.
7. Thuijls G, van Wijck K, Grootjans J, Derikx JP, van Bijnen AA, Heineman E, Dejong CH, Buurman WA, Poeze M. Early diagnosis of intestinal ischemia using urinary and plasma fatty acid binding proteins. Ann Surg
253 2:303–308, 2011.
8. Grimaldi D, Guivarch E, Neveux N, Fichet J, Pene F, Marx JS, Chiche JD, Cynober L, Mira JP, Cariou A. Markers of intestinal injury are associated with endotoxemia in successfully resuscitated patients. Resuscitation
84 1:60–65, 2013.
9. van Haren FM, Pickkers P, Foudraine N, Heemskerk S, Sleigh J, van der Hoeven JG. The effects of methylene blue infusion on gastric tonometry and intestinal fatty acid binding protein levels in septic shock patients. J Crit Care
25 2:358, 2010.
10. Piton G, Belon F, Cypriani B, Regnard J, Puyraveau M, Manzon C, Navellou JC, Capellier G. Enterocyte damage in critically ill patients is associated with shock condition and 28-day mortality. Crit Care Med
41 9:2169–2176, 2013.
11. Krychtiuk KA, Lenz M, Koller L, Honeder MC, Wutzlhofer L, Zhang C, Chi L, Maurer G, Niessner A, Huber K, et al. Monocyte subset distribution is associated with mortality in critically ill patients. Thromb Haemost
116 5:949–957, 2016.
12. Krychtiuk KA, Ruhittel S, Hohensinner PJ, Koller L, Kaun C, Lenz M, Bauer B, Wutzlhofer L, Draxler DF, Maurer G, et al. Mitochondrial DNA and toll-like receptor-9 are associated with mortality in critically ill patients. Crit Care Med
43 12:2633–2641, 2015.
13. Thiele H, Zeymer U, Neumann FJ, Ferenc M, Olbrich HG, Hausleiter J, Richardt G, Hennersdorf M, Empen K, Fuernau G, et al. I.-S. I. T. Investigators. Intraaortic balloon support for myocardial infarction with cardiogenic shock
. N Engl J Med
367 14:1287–1296, 2012.
14. Simon L, Gauvin F, Amre DK, Saint-Louis P, Lacroix J. Serum procalcitonin and C-reactive protein levels as markers of bacterial infection: a systematic review and meta-analysis. Clin Infect Dis
39 2:206–217, 2004.
15. Santaguida PL, Don-Wauchope AC, Oremus M, McKelvie R, Ali U, Hill SA, Balion C, Booth RA, Brown JA, Bustamam A, et al. BNP and NT-proBNP as prognostic markers in persons with acute decompensated heart failure: a systematic review. Heart Fail Rev
19 4:453–470, 2014.
16. Gajda AM, Storch J. Enterocyte fatty acid-binding proteins (FABPs): different functions of liver and intestinal FABPs in the intestine. Prostaglandins Leukot Essent Fatty Acids
17. Oldenburg WA, Lau LL, Rodenberg TJ, Edmonds HJ, Burger CD. Acute mesenteric ischemia: a clinical review. Arch Intern Med
164 10:1054–1062, 2004.
18. Mullens W, Abrahams Z, Francis GS, Sokos G, Taylor DO, Starling RC, Young JB, Tang WH. Importance of venous congestion for worsening of renal function in advanced decompensated heart failure. J Am Coll Cardiol
53 7:589–596, 2009.
19. Abe S, Arima S, Yamashita T, Miyata M, Okino H, Toda H, Nomoto K, Ueno M, Tahara M, Kiyonaga K, et al. Early assessment of reperfusion therapy using cardiac troponin T. J Am Coll Cardiol
23 6:1382–1389, 1994.
20. Fuhrmann V, Kneidinger N, Herkner H, Heinz G, Nikfardjam M, Bojic A, Schellongowski P, Angermayr B, Kitzberger R, Warszawska J, et al. Hypoxic hepatitis: underlying conditions and risk factors for mortality in critically ill patients. Intensive Care Med
35 8:1397–1405, 2009.