Adrenomedullin: a marker of impaired hemodynamics, organ dysfunction, and poor prognosis in cardiogenic shock

Abstract

Background

The clinical CardShock risk score, including baseline lactate levels, was recently shown to facilitate risk stratification in patients with cardiogenic shock (CS). As based on baseline parameters, however, it may not reflect the change in mortality risk in response to initial therapies. Adrenomedullin is a prognostic biomarker in several cardiovascular diseases and was recently shown to associate with hemodynamic instability in patients with septic shock. The aim of our study was to evaluate the prognostic value and association with hemodynamic parameters of bioactive adrenomedullin (bio-ADM) in patients with CS.

Methods

CardShock was a prospective, observational, European multinational cohort study of CS. In this sub-analysis, serial plasma bio-ADM and arterial blood lactate measurements were collected from 178 patients during the first 10 days after detection of CS.

Results

Both bio-ADM and lactate were higher in 90-day non-survivors compared to survivors at all time points (P < 0.05 for all). Lactate showed good prognostic value during the initial 24 h (AUC 0.78 at admission and 0.76 at 24 h). Subsequently, lactate returned normal ([less than or equal to]2 mmol/L) in most patients regardless of later outcome with lower prognostic value. By contrast, bio-ADM showed increasing prognostic value from 48 h and beyond (AUC 0.71 at 48 h and 0.80 at 5-10 days). Serial measurements of either bio-ADM or lactate were independent of and provided added value to CardShock risk score (P < 0.001 for both). Ninety-day mortality was more than double higher in patients with high levels of bio-ADM (>55.7 pg/mL) at 48 h compared to those with low bio-ADM levels (49.1 vs. 22.6%, P = 0.001). High levels of bio-ADM were associated with impaired cardiac index, mean arterial pressure, central venous pressure, and systolic pulmonary artery pressure during the study period. Furthermore, high levels of bio-ADM at 48 to 96 h were related to persistently impaired cardiac and end-organ function.

Conclusions

Bio-ADM is a valuable prognosticator and marker of impaired hemodynamics in CS patients. High levels of bio-ADM may show shock refractoriness and developing end-organ dysfunction and thus help to guide therapeutic approach in patients with CS.

Study identifier of CardShock study NCT01374867 at clinicaltrials.gov

Tolppanen et al. Ann. Intensive Care (2017) 7:6 DOI 10.1186/s13613-016-0229-2

Adrenomedullin: a marker ofimpaired hemodynamics, organ dysfunction, andpoor prognosis incardiogenic shock

Heli Tolppanen1,2,3*http://orcid.org/0000-0002-3364-8554

Web End = , Mercedes RivasLasarte1,4, Johan Lassus3, Jordi SansRosell4, Oliver Hartmann5, Matias Lindholm6, Mattia Arrigo1,19,20, Tuukka Tarvasmki7, Lars Kber6, Holger Thiele8, Kari Pulkki9,10, Jindrich Spinar11,12, John Parissis13, Marek Banaszewski14, Jose SilvaCardoso15, Valentina Carubelli16, Alessandro Sionis4, VeliPekka Harjola7 and Alexandre Mebazaa1,17,18

http://orcid.org/0000-0002-3364-8554

Web End = *Correspondence: heli.tolppanen@helsinki.

2 Heart Center, PijtHme Central Hospital, Lahti, FinlandFull list of author information is available at the end of the article

The Author(s) 2017. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/

Web End =http://creativecommons.org/licenses/by/4.0/ ), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

Tolppanen et al. Ann. Intensive Care (2017) 7:6

Page 2 of 10

Background

Cardiogenic shock (CS) is a state of global tissue hypoperfusion caused by severe cardiac dysfunction. In spite of advances in therapeutic options, the short-term mortality associated with CS remains unacceptably high [1, 2]. While very early mortality is largely related to sudden and severe circulatory failure, subsequent death is strongly inuenced by activation of neurohumoral and inammatory responses leading to multiorgan failure [3]. Risk stratication is crucial in order to accurately identify patients that could potentially benet from more aggressive strategies, and moreover, to identify advanced stages of shock when restoring cardiac function may not reverse end-organ failure. The CardShock risk score was recently introduced to help risk stratication in the early phase of CS. The score includes lactate levels and six other clinical variables available at the time of detection of shock [2]. However, as based on baseline parameters, the score may not reect the change in mortality risk in response to initial therapies.

At present, few biomarkers have been proven benecial in risk stratication of patients with CS. Lactate is an established marker of hemodynamic instability and prognosis in critically ill patients [46]. Adrenomedullin (ADM) has been shown as prognosticator in CS after an acute coronary syndrome [7]. The study was small, monocentric, and based only on one sample measured at 24h, however. In patients with septic shock, higher ADM levels were associated with hemodynamic instability, requirement of vasopressor therapy, and increased mortality [8]. The aim of the present study was to evaluate the prognostic value and association with hemodynamic parameters of serial measurements of mature bioactive ADM (bio-ADM) in patients with CS, in order to help risk assessment and support clinical decision in CS.

Methods

Study population andendpoints

CardShock study (NCT01374867) is a prospective European multicenter and multinational cohort study that enrolled consecutive CS patients in 9 centers in 8 countries between October 2010 and December 2012. The inclusion criteria were systolic blood pressure <90mmHg for 30min despite uid administration or need for vaso-active therapy, and one or more signs of organ hypoperfusion (cool extremities, confusion or altered mental status, oliguria <0.5ml/kg/h for the previous 6h, or blood lactate >2mmol/l), cardiac origin of the state of hypoper-fusion, and age over 18years. Study inclusion was within the rst 6h of the detection of shock. Exclusion criteria were shock caused by ongoing hemodynamically signi-cant arrhythmias and shock after cardiac or non-cardiac

surgery. For this sub-study, 178 patients with biomarker data (lactate and bio-ADM) available were included.

The primary endpoint of the study was to determine the prognostic value of serial measurements of bio-ADM on mortality prediction at 90days. Secondary endpoint was to describe the relationship of bio-ADM and lactate with hemodynamic parameters.

Study protocol

Detailed medical history and patient characteristics were collected. Clinical signs with routine laboratory measurements, including lactate which was measured locally, were registered at presentation to the hospital. A total of 69 (39%) patients had pulmonary artery catheter, and additional 42 (24%) patients had central venous pressure monitoring. As per study protocol, all patients had echo-cardiography performed at baseline and at 72h. Patients were treated according to local practice in each hospital. Vital status during follow-up was determined through direct contact with the patient or next of kin, or through population and hospital registries. Three patients were lost to follow-up; in the mortality analyses their cases were censored at the time of hospital discharge. Serial plasma samples were taken at various time points after presentation and immediately frozen and stored at 80C. Both bio-ADM and arterial lactate were measured at 0, 12, 24, 48, 72, and 96h, and bio-ADM again at 510days.

The CardShock study was approved by local ethics committees at the participating centers and conducted in accordance with the Declaration of Helsinki. All patients or their next of kin gave informed consent.

BioADM measurement

All bio-ADM measurements were taken blinded for clinical data in the laboratories of Sphingotec GmbH, Hennigsdorf, Germany, with a previously described immunoassay [8]. Mid-regional pro-Adrenomedullin (MR-proADM), a non-bioactive precursor of ADM, has been used in recent years to overcome the obstacles of mature ADM measurement relating to analyte stability and interference with complement factor H in the measurement [911]. In our study, the novel immunoassay allowed reliable ultrasensitive measurement of bioactive ADM peptide from small sample volume (50 uL of plasma), contrary to the earlier measurement of mature ADM levels [12]. Briey, a one-step sandwich-coated tube chemiluminescence immunoassay was used based on acridinium NHS-ester labeling for the detection of human ADM in plasma. More detailed description of the bio-ADM measurement is provided in Additional le1. The upper limit of normal values of bio-ADM with the assay used is 43pg/mL [8].

Tolppanen et al. Ann. Intensive Care (2017) 7:6

Page 3 of 10

Statistical analysis

Results are presented as numbers (n) and percentages (%), and means with standard deviations (SD) or medians with interquartile ranges (IQRs) as appropriate. Between groups, comparisons were made using Chi-square test, t test, or Wilcoxon rank-sum test as appropriate.

Cox proportional hazards regression was used to analyze the time-dependent eect of serial measurements of bio-ADM and lactate on 90-day survival in uni- and multivariable analyses [13, 14]. Hazard ratios (HRs) are given with 95% condential intervals (CIs). Both biomarkers were tested for independency from the previously developed CardShock risk score [2], which summarizes seven clinical parameters, which were associated with in-hospital mortality. The model included baseline lactate as its strongest component, as well as age over 75years, acute coronary syndrome as the etiology of CS, previous history of myocardial infarction or coronary artery bypass surgery, altered mental status at presentation, renal function, and left ventricular ejection fraction below 40% at baseline. The assumptions of proportional hazard were tested for all variables. For all analyses, biomarkers (bio-ADM and lactate) were log-transformed and HR was standardized to describe the HR for a biomarker change in one IQR. Wald statistics were used to investigate the prognostic value of each biomarker and their combination when measured at each time point. To give an eect measure for the prognostic value of bio-ADM and lactate in 90-day mortality, the receiver-operating characteristic (ROC) curve analysis was performed and areas under ROC curves (AUCs) were calculated. KaplanMeier curves were also used in survival analyses. Dichotomization of patients was based on bio-ADM level 55.7mg/ml, which was the optimal cuto with highest sensitivity and specicity for 90-day mortality when measured at 48h, and similar to the median values of bio-ADM during the rst 96h (range of medians at 096h 54.559.9pg/ml).

For comparison of biomarker levels with hemodynamic parameters, median of all biomarker measurements taken during the initial 96h of each patient was used. Dichotomization was based on bio-ADM level of 55.7pg/mL and lactate level of 1.63mmol/L, which was the median value of each patients median lactate level during the rst 96h. For comparison of hemodynamic measures and end-organ dysfunction at 4896 h, the median value of the measures between 48 and 96h of each patient was used, and dichotomization was based on median value of bio-ADM at 4896h with the cuto level of 55.7pg/mL. A two-sided P value <0.05 was regarded as statistically signicant. The statistical analyses were performed using R version 2.5.1 (http://www.r-project.org

Web End =http://www.r-project.org , library Design, Hmisc, ROCR), SPSS 21.0 statistical software (IBM Corp, Armonk, NY, USA) and STATA (version 13, Statacorp, Texas, USA).

Results

The mean age of the 178 patients included in this study was 6612years, and 137 (74%) were men. Most common etiology of CS was acute coronary syndrome (78%).

The overall 90-day mortality was 43% (n=75). Table1 describes the patient characteristics of the 90-day survivors and non-survivors. Twenty-nine (16%) patients died before 48h from the detection of shock, and the remaining 46 (26%) patients died between 48h and 90days. The earlier deaths tended to occur more often due to myocar-dial infarction (71 vs. 51%, P=0.086) and less often due to worsening heart failure (17 vs. 42%, P = 0.017). On the contrary, the later occurring deaths were numerically more often related to infection, renal failure, and stroke, although these dierences did not reach statistical signicance (Additional le2: Table S1).

BioADM andlactate levels insurvivors andnonsurvivors

Plasma bio-ADM levels and arterial blood lactate were higher in non-survivors compared to survivors at all time points. The highest lactate levels were observed at baseline both in survivors and non-survivors (2.2 and 5.0mmol/L, respectively, P<0.0001). The median levels of lactate returned to normal values within 12h in survivors and within 24h in non-survivors (Fig.1). Hence, at 24h 76% of all patients had normal lactate levels. The time course of plasma bio-ADM levels was divergent between survivors and non-survivors; bio-ADM levels stayed close to the upper normal limit (43pg/mL) in survivors while remained elevated in non-survivors (Fig.1).

Prognostic value ofbioADM andlactate levels

Serial measurement of the biomarkers showed that for both bio-ADM and lactate, a normalization of concentration was associated with a decrease in mortality risk, while a continuing high concentration or increasing concentrations were associated with a high mortality risk. In time-dependent Cox model, serial bio-ADM and lactate measures were associated with increased 90-day risk of death in univariate time-dependent Cox analysis (HR 2.22, CI 1.762.80, P<0.001 and HR 3.83, CI 2.735.37, P<0.001, respectively) and after adjustment for the Card-Shock risk score (HR 1.62, 95% CI 1.262.09, P<0.001, and HR 2.78, 95% CI 1.943.97, P<0.001, respectively). Time-dependent Cox model for serial bio-ADM and lac-tate was associated with increased risk of 90-day mortality also when selecting only patients with CS caused by ACS (HR 1.49, CI 1.102.02, P=0.01 for bio-ADM and

HR 2.76, CI 1.943.92, P<0.001 for lactate).

In the early phase of CS, lactate had good prognostic value (AUC at baseline 0.76, 95% CI 0.690.82) that rapidly decreased, whereas bio-ADM had incremental prognostic value with an AUC of 0.71 at 48 h (95% CI

Tolppanen et al. Ann. Intensive Care (2017) 7:6

Page 4 of 10

Table 1 Characteristics of90-day survivors andnon-survivors

Table 1 continued

Survivors (n=103)

Nonsurvivors (n=75)

P value

P value

Survivors (n=103)

Nonsurvivors (n=75)

ECMO 2 (2%) 1 (1%) 0.8 LVAD 1 (1%) 4 (6%) 0.08

Age 63 (13) 71 (11) <0.001 Male gender 82 (80%) 51 (68%) 0.08 CardShock risk score 3.4 (1.7) 5.5 (1.5) <0.001 Medical history

Hypertension 59 (57%) 51 (68%) 0.15 Hyperlipidemia 41 (40%) 44 (59%) 0.013 Diabetes 23 (22%) 30 (40%) 0.011 Smoker 48 (47%) 23 (31%) 0.032 Ischemic heartdisease

23 (22%) 36 (48%) <0.001

Previous infarction 16 (16%) 29 (39%) <0.001 Previous CABG 1 (1%) 10 (13%) 0.001 Chronic heart failure 13 (13%) 16 (21%) 0.12 Stroke or TIA 8 (8%) 8 (11%) 0.5 Peripheral arterydisease

Continuous variables expressedasmean (standard deviation) or median (interquartile range), as appropriate; categorical variables expressed as number (percentage). BMI body mass index, CABG coronary artery bypass graft surgery, COPD chronic obstructive pulmonary disease, TIA transient ischemic attack,BP blood pressure, LVEF left ventricular ejection fraction, hs-TnT high sensitive troponin T, NT-proBNP N-terminal pro-brain natriuretic peptide, eGFR estimated glomerular ltration rate, PCI percutaneous coronary intervention, IABP intraaortic balloon pump, LVAD left ventricular assist device, ECMO extracorporeal membrane oxygenation

6 (6%) 13 (17%) 0.014

Asthma or COPD 11 (11%) 9 (12%) 0.8 Status at inclusion

Altered mental status 56 (55%) 61 (81%) <0.001 Systolic BP, mmHg 80 (7085) 75 (6680) 0.016 Mean BP, mmHg 58 (5364) 53 (4760) 0.011 Heart rate 88 (27) 89 (31) 0.94 LVEF, % 36 (15) 29 (12) <0.001 hsTnT, ng/L 1366 (1834191) 2862 (11247842) 0.008 NTproBNP, ng/L 2026 (4437101) 5174 (144716,547) 0.001 eGFR, ml/min/1.72m2 71 (29) 51 (27) <0.001 Postresuscitation 20 (19%) 27 (36%) 0.013 Etiology of cardiogenicshock

Acute coronary syndrome

0.620.79) up to 0.80 (95% CI 0.780.91) at 510 days (Fig. 2). The 90-day mortality was more than double higher in patients with high levels of bio-ADM at 48 h compared to those with low levels of bio-ADM (mortality 49.1 vs. 22.6%, P=0.001), as shown in Fig.3.

A more in-depth analysis of individual time points

revealed that for lactate, its measurement at early time points provided added value to risk prediction, and later time points showed poor prognostic ability. For bio-ADM, the later time points provided the most added value and best discriminatory power (Table2).

BioADM andhemodynamic alterations

Overall, both high bio-ADM levels and high lactate levels during the study period were associated with low cardiac index and low mean arterial pressure. In addition, high bio-ADM levels, but not high lactate levels, were associated with high central venous pressure and high systolic pulmonary artery pressure (Fig. 4). Furthermore, high bio-ADM levels at 4896h were associated with impaired cardiac and end-organ dysfunction, as shown in Fig.5. Of note, at that time period, of the hemodynamic parameters only cardiac index was a good prognosticator of later outcome (Additional le3: Table S2).

Discussion

The present study shows that bio-ADM has strong prognostic value in CS when measured after the initial phase of management (at 48h or later), and is associated with impaired hemodynamics and persistently impaired cardiac and end-organ dysfunction.

Lactate is a well-known marker of hemodynamic instability and disease severity in patients with shock. It is a marker of poor outcome if measured during the

77 (75%) 65 (87%) 0.051

Left main stenosis 11 (14%) 15 (25%) 0.09 Threevesseldisease

17 (22%) 25 (42%) 0.008

PCI 66 (88%) 52 (80%) 0.2 Thrombolysis 12 (16%) 5 (8%) 0.13 Myocarditis or Takot

subo

8 (8%) 0 (0%)

Valvular cause 6 (6%) 6 (8%) 0.6 Chronic cardiomyo

pathy/heart failure

11 (11%) 4 (5%) 0.2

Inhospital manage ment

Any inotrope 74 (76%) 54 (81%) 0.5 Any vasopressor 75 (76%) 64 (94%) 0.002 Invasive ventilation 54 (52%) 56 (75%) 0.003 IABP treatment 51 (50%) 45 (60%) 0.17

Tolppanen et al. Ann. Intensive Care (2017) 7:6

Page 5 of 10

initial 24h of intensive care unit admission in CS [5] or in overall critically ill patient population [6]. Our study conrmed the prognostic value of lactate during the rst 24h in CS. Later, in patients surviving the early phase of shock, arterial blood lactate levels returned normal in the majority of patients regardless of outcome, and the association with mortality was less signicant.

In recent years, mature ADM, a hormone with potent vasodilatory and inotropic properties, or its precursor protein MR-proADM as its surrogate, have evolved as powerful prognostic markers in patients presenting with

acute chest pain [15], dyspnea [1618] and in those with acute heart failure [19]. In patients with acute myocardial infarction, high ADM levels have been associated with impaired left ventricular function and death [7, 20, 21]. Moreover, in patients with refractory CS requiring extra-corporeal membrane oxygenation (ECMO) support, the levels of MR-proADM were found steadily elevated during rst seven days and did not dier regardless of weaning success [22]. In our study, using a novel ultrasensitive method for bio-ADM measurement [8, 12], we showed that bio-ADM had good prognostic value in patients with CS. Indeed, bio-ADM was elevated during the whole study period in non-survivors, and high levels of bio-ADM were associated with increased short-term death, especially after 48h, time when lactate had lower prognostic value than at baseline.

The exact source and the role of bio-ADM in CS are unknown. Plasma ADM is mainly derived from vascular endothelial cells, smooth muscle cells, and adventitial broblasts. Catecholamines, angiotensin II, and aldosterone, all of which are highly elevated in CS, are potent stimulators of ADM production [23]. Inamma-tory cytokines, such as interleukins and TNF, appearing in CS complicated by systemic inammatory response syndrome [3], have also been advocated to stimulate ADM secretion [2426]. Furthermore, in septic shock, high ADM levels are associated with decreased vascular

Tolppanen et al. Ann. Intensive Care (2017) 7:6

Page 6 of 10

Table 2 Predictive value for 90-day mortality with Wald statistics of lactate and bio-ADM at each time point afterthe detection ofshock

x2 P value

0 h

Lactate 38.44 <0.0001 BioADM 0.09 0.8 Total 42.76 <0.0001 12 h

Lactate 23.99 <0.0001 BioADM 4.28 0.039 Total 46.69 <0.0001 24 h

Lactate 38.03 <0.0001 BioADM 2.41 0.12 Total 44.82 <0.0001 48 h

Lactate 3.14 0.077 BioADM 9.79 0.002 Total 20.58 <0.0001 72 h

Lactate 2.52 0.11 BioADM 8.01 0.0047 Total 15.08 0.0005 96 h

Lactate 0.18 0.7 BioADM 17.01 <0.0001 Total 19.95 <0.0001

tone and requirement of vasopressor therapy [8, 27]. In this study, we found that in contrast to lactate, bio-ADM levels are not only related to hypoperfusion (low cardiac index and low mean arterial pressure) but also with high cardiac lling pressures (central venous pressure and pulmonary artery pressure). We hypothesize that myocardial stunning is responsible for the activation of neurohu-moral response (catecholamines, angiotensin II, inter-leukins), leading to bio-ADM production. Bio-ADM with its potent vasodilatory properties may act perpetuating shock and contributing to end-organ damage associated with poor prognosis. The exact pathways implicated in these processes need further investigation.

Interestingly, causes of death diered between patients who died before 48h and later. The early deaths tended to occur more often due to myocardial infarction and less often due to worsening heart failure. Initial management in CS, as it was recently published in international recommendations [2831], includes stabilization with volume expansion, inotropes, and vasopressors. This aggressive resuscitation in patients surviving the initial phase may be enough to reestablish a correct perfusion allowing lactate levels to decrease and even to normalize, as we found in our work. Nevertheless, in patients with activations of systemic inammatory response, mortality has been reported to remain high related to other causes of death [32]. After initial medical stabilization, in many centers, mechanical assist devices are an increasingly used alternative to support circulation and allow

recovery of stunned or hibernating myocardium if clinical signs of recovery are absent [3335]. It seems that to increase survival, these advanced therapies should be started before irreversible end-organ dysfunction has occurred to carefully selected patients, considering the costs and possible complications of these therapies [36]. High levels of bio-ADM at 48h or later may reect a state of refractory shock with end-organ damage, despite normalization of lactate levels, and may help the clinician in a more accurate patient selection for advanced therapies, or guide in the difficult process of limiting the therapeutic eort.

Our study carries several limitations. Plasma samples were not available in all patients and at all time points. The high early mortality further decreased the number of subsequent samples. Nevertheless, considering the difculties in prospectively studying patients with CS with timely plasma sampling, this is one of the largest cohorts of biomarker studies in patients with CS. As we used a novel technique for the identication of plasma bio-ADM,

Lactate and bio-ADM levels were log10-transformed for the analysis. During the rst 24h, only lactate contributes to mortality prediction, later only bio-ADM contributes to prediction

Tolppanen et al. Ann. Intensive Care (2017) 7:6

Page 7 of 10

Tolppanen et al. Ann. Intensive Care (2017) 7:6

Page 8 of 10

the levels of bio-ADM were directly comparable with only a few studies. However, the novel technique used is accurate and allows measures from small amounts of plasma and thus has the potential to become the technique of choice to assess ADM pathway in the future. Being still in experimental use, the measurement of bio-ADM is currently less available and a considerably more expensive laboratory test compared to lactate. Nevertheless, it is expected to become available on widespread fully automated platforms in near future. As this was an observational multinational study, the management of patients was not guided per protocol. Management in this study, however, reects the real-world practice in European tertiary care university hospitals with high rate of accordance to the international recommendations. There were

only a few patients treated with circulatory assist devices, thus preventing sub-analyses of these patients.

Conclusions

As a conclusion, our study has a potentially important clinical implication, suggesting that bio-ADM measurement could be added to CS evaluation because of its prognostic value after the initial phase of management in patients with CS. Elevated levels of bio-ADM seem to be related to persistent cardiac and end-organ dysfunction and may support clinical decision when choosing therapeutic approach in patients with refractory CS. Whether risk estimation based on bio-ADM levels may help to optimize therapies and improve outcome needs further investigation.

Tolppanen et al. Ann. Intensive Care (2017) 7:6

Page 9 of 10

Additional les Hospital Center, University of Porto, Porto, Portugal. 16 Division of Cardiol ogy, Department of Medical and Surgical Specialties Radiological Sciences and Public Health, University and Civil Hospital of Brescia, Brescia, Italy.

17 Department of Anesthesia and Critical Care, University Hospital Saint Louis Lariboisire, APHP, Paris, France. 18 University Paris Diderot, Sorbonne Paris Cit, Paris, France. 19 Department of Cardiology, University Heart Center, 8091 Zrich, Switzerland. 20 Department of Cardiology, University Hospital

Zrich, 8091 Zrich, Switzerland.

Acknowledgements

Tom Bcklund, Raija Jurkko, Kristiina Jrvinen, Tuomo Nieminen, Leena Soininen, Reijo Sund, Ilkka Tierala, Jukka Tolonen, Marjut Varpula, Tuomas Korva, Mervi Pietil, and Anne Pitkl were local CardShock investigatorsin Helsinki University Hospital; Alexandra Sousa, Carla Sousa, Mariana Paiva, Ins Rangel, Rui Almeida, Teresa Pinho, Jlia Maciel in So Joo Hospital Center; Valentina Lazzarini, Carlo Lombardi, and Marco Metra in University and Civil hospital of Brescia.

Competing interests

Sphingotec GmbH performed freely bioADM measurements, with no access to clinical data. O. Hartmann reports to be an employee of Sphingotec GmbH. V.P.H. has served on advisory boards for Bayer, BMS/Pzer, BoehringerIngelheim, Roche Diagnostics, Novartis, and Servier and received lecture fees from Bayer, Orion Pharma, Resmed, and Roche Diagnostics. J.L. has received consulting and/or lecture fees from BoehringerIngelheim, Roche Diagnostics, Novartis, Orion Pharma, Pzer, Servier, and Vifor Pharma. J. Parissis received honoraria for lectures from Novartis and Orion Pharma. J. SilvaCardoso has consulted and received speaker fees, or investigational grants for Abbott, AstraZeneca Pharmaceuticals, Menarini, Merck, Merck Sharp & Dohme, Novartis, Orion, Pzer, and Sano. A. Mebazaa received speakers honoraria from Abbott, Novartis, Orion, Roche, and Servier, and fee as member of advisory board and/or Steering Committee from Cardiorentis, Adrenomed, MyCartis, ZS Pharma and Critical Diagnostics. All other authors reported no competing interest.

Availability of data and materials

All data are presented within the manuscript or in the additional les.

Ethics approval and consent to participate

The CardShock study was approved by all local ethics committees at the participating centers and conducted in accordance with the Declaration of Helsinki. All patients or their next of kin gave informed consent.

Funding

CardShock study was funded by unrestricted research grant from Finnish Foundation for Cardiovascular Research. H. Tolppanen also received personal research grants from Aarne Koskelo Foundation, Finska Lkaresllskapet and from the Finnish Medical Foundation.

Received: 5 July 2016 Accepted: 20 December 2016

Abbreviations

CS: cardiogenic shock; ADM: adrenomedullin; bioADM: bioactive adre nomedullin; MRproADM: midregional proadrenomedullin; SD: standard deviation; IQR: interquartile range; HR: hazard ratio; CI: condential interval; LR: likelihood ratio; ROC: receiveroperating characteristics; AUC: area under curve; ECMO: extracorporeal membrane oxygenation.

Authors contributions

HT was involved in acquisition of data at Helsinki University Hospital, analysis and interpretation of data, and drafting of manuscript. MRL was involved in acquisition of data at Hospital de la Santa Creu i Sant Pau in Barcelone, analysis and interpretation of data, and drafting of manuscript. JL and VPH were involved in designing study protocol, acquisition of data at Helsinki University Hospital, and critical revision of manuscript. JSR and AS were involved in acquisition of data at Hospital de la Santa Creu i Sant Pau in Barcelone and crit ical revision of manuscript. OH was involved in analysis and interpretation of data (laboratory analyses of bioactive adrenomedullin) and statistical analyses. ML and LK were involved in acquisition of data at Rigshospitalet, University of Copenhagen, and critical revision of manuscript. MA was involved in analysis and interpretation of data and critical revision of manuscript. TT was involved in acquisition of data at Helsinki University Hospital, analysis and interpretation of data, and critical revision of manuscript. HT was involved in critical revision of manuscript. KP was involved in analysis and interpretation of data (labora tory analyses of study laboratory samples, excluding bioactive adrenomedul lin, at Islab, Finland) and critical revision of manuscript. JS was involved in acquisition of data at University Hospital Brno and critical revision of manu script. JP was involved in acquisition of data at Attikon University Hospital in Athens and critical revision of manuscript. MB was involved in acquisition of data at Institute of Cardiology in Warsaw and critical revision of manuscript. JSC was involved in acquisition of data at So Joo Hospital Center in Porto and critical revision of manuscript. VC was involved in acquisition of data at University and Civil Hospital of Brescia and critical revision of manuscript. AM was involved in designing study protocol, analysis and interpretation of data, drafting of manuscript, and critical revision of manuscript. All authors read and approved the nal manuscript.

Author details

1 INSERM UMRS942, Paris, France. 2 Heart Center, PijtHme Central Hospital, Lahti, Finland. 3 Heart and Lung Center, Helsinki University and Helsinki University Hospital, Helsinki, Finland. 4 Intensive Cardiac Care Unit, Cardiology Department, Hospital de la Santa Creu i Sant Pau, Biomedical Research Institute IIBSantPau, Universidad Autnoma de Barcelona, Barcelona, Spain.

5 Sphingotec GmbH, Hennigsdorf, Germany. 6 Department of Cardiology, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark. 7 Depart ment of Emergency Care, Helsinki University and Helsinki University Hospital, Helsinki, Finland. 8 Medical Clinic II, University Hospital SchleswigHolstein, University Heart Center Lbeck, Lbeck, Germany. 9 Department of Clinical Chemistry, University of Eastern Finland, Kuopio, Finland. 10 Eastern Finland Laboratory Centre, Kuopio, Finland. 11 Department of Internal Medicineand Cardiology, University Hospital Brno, Brno, Czech Republic. 12 International Clinical Research Centre (ICRC), Brno, Czech Republic. 13 Heart Failure Clinic and Secondary Cardiology Department, Attikon University Hospital, Athens, Greece. 14 Intensive Cardiac Therapy Clinic, Institute of Cardiology, Warsaw, Poland. 15 Department of Cardiology, CINTESIS, Porto Medical School, So Joo

References

1. Thiele H, Zeymer U, Neumann FJ, Ferenc M, Olbrich HG, Hausleiter J, et al. Intraaortic balloon support for myocardial infarction with cardiogenic shock. N Engl J Med. 2012;367(14):128796.

2. Harjola VP, Lassus J, Sionis A, Kober L, Tarvasmaki T, Spinar J, et al. Clinical picture and risk prediction of shortterm mortality in cardiogenic shock. Eur J Heart Fail. 2015;17(5):5019 Epub 2015/03/31.

3. Reynolds HR, Hochman JS. Cardiogenic shock: current concepts and improving outcomes. Circulation. 2008;117(5):68697.

4. Bakker J, Gris P, Coernils M, Kahn RJ, Vincent JL. Serial blood lactate levels can predict the development of multiple organ failure following septic shock. Am J Surg. 1996;171(2):2216.

5. Attana P, Lazzeri C, Chiostri M, Picariello C, Gensini GF, Valente S. Lactate clearance in cardiogenic shock following ST elevation myocardial infarc tion: a pilot study. Acute Card Care. 2012;14(1):206.

Tolppanen et al. Ann. Intensive Care (2017) 7:6

Page 10 of 10

6. Zhang Z, Xu X. Lactate clearance is a useful biomarker for the prediction of allcause mortality in critically ill patients: a systematic review and metaanalysis*. Crit Care Med. 2014;42(9):211825.

7. Katayama T, Nakashima H, Takagi C, Honda Y, Suzuki S, Yano K. Predictors of mortality in patients with acute myocardial infarction and cardiogenic shock. Circ J. 2005;69(1):838.

8. Marino R, Struck J, Maisel AS, Magrini L, Bergmann A, Di Somma S. Plasma adrenomedullin is associated with shortterm mortality and vasopressor requirement in patients admitted with sepsis. Crit Care. 2014;18(1):R34.

9. Pio R, Martinez A, Unsworth EJ, Kowalak JA, Bengoechea JA, Zipfel PF,et al. Complement factor H is a serumbinding protein for adrenomedul lin, and the resulting complex modulates the bioactivities of both partners. J Biol Chem. 2001;276(15):12292300.

10. Lewis LK, Smith MW, Yandle TG, Richards AM, Nicholls MG. Adrenomedul lin(152) measured in human plasma by radioimmunoassay: plasma concentration, adsorption, and storage. Clin Chem. 1998;44(3):5717.

11. Struck J, Tao C, Morgenthaler NG, Bergmann A. Identication of an Adre nomedullin precursor fragment in plasma of sepsis patients. Peptides. 2004;25(8):136972.

12. Di Somma S, Magrini L, Travaglino F, Lalle I, Fiotti N, Cervellin G, et al. Opinion paper on innovative approach of biomarkers for infectious dis eases and sepsis management in the emergency department. Clin Chem Lab Med. 2013;51(6):116775.

13. Harrell FE. Regression modeling strategies with applications to linear models, logistic regression, and survival analysis. New York: Springer; 2001.

14. Hartmann O, Schuetz P, Albrich WC, Anker SD, Mueller B, Schmidt T. Time dependent Cox regression: serial measurement of the cardiovascular biomarker proadrenomedullin improves survival prediction in patients with lower respiratory tract infection. Int J Cardiol. 2012;161(3):16673.

15. Tzikas S, Keller T, Ojeda FM, Zeller T, Wild PS, Lubos E, et al. MRproANP and MRproADM for risk stratication of patients with acute chest pain. Heart. 2013;99(6):38895.

16. Potocki M, Breidthardt T, Reichlin T, Morgenthaler NG, Bergmann A, Noveanu M, et al. Midregional proadrenomedullin in addition to btype natriuretic peptides in the risk stratication of patients with acute dysp nea: an observational study. Crit Care. 2009;13(4):R122.

17. Shah RV, Truong QA, Gaggin HK, Pfannkuche J, Hartmann O, Januzzi JL Jr. Midregional proatrial natriuretic peptide and proadrenomedullin test ing for the diagnostic and prognostic evaluation of patients with acute dyspnoea. Eur Heart J. 2012;33(17):2197205.

18. Maisel A, Mueller C, Nowak RM, Peacock WF, Ponikowski P, Mockel M,et al. Midregion prohormone adrenomedullin and prognosis in patients presenting with acute dyspnea: results from the BACH (Biomarkers in Acute Heart Failure) trial. J Am Coll Cardiol. 2011;58(10):105767.

19. Lassus J, Gayat E, Mueller C, Peacock WF, Spinar J, Harjola VP, et al. Incremental value of biomarkers to clinical variables for mortality prediction in acutely decompensated heart failure: the Multinational Observational Cohort on Acute Heart Failure (MOCA) study. Int J Cardiol. 2013;168(3):218694 Epub 2013/03/30.

20. Katayama T, Nakashima H, Honda Y, Suzuki S, Yano K. Relationship between adrenomedullin and leftventricular systolic function and mor tality in acute myocardial infarction. Angiology. 2005;56(1):3542.

21. Khan SQ, OBrien RJ, Struck J, Quinn P, Morgenthaler N, Squire I, et al. Prognostic value of midregional proadrenomedullin in patients with acute myocardial infarction: the LAMP (Leicester Acute Myocardial Infarc tion Peptide) study. J Am Coll Cardiol. 2007;49(14):152532.

22. Luyt CE, Landivier A, Leprince P, Bernard M, Pavie A, Chastre J, et al. Usefulness of cardiac biomarkers to predict cardiac recovery in patients on extracorporeal membrane oxygenation support for refractory cardio genic shock. J Crit Care. 2012;27(5):524e714.

23. Onitsuka H, Imamura T, Yamaga J, Kuwasako K, Kitamura K, Eto T. Angio tensin II stimulates cardiac adrenomedullin production and causes accu mulation of mature adrenomedullin independently of hemodynamic stress in vivo. Horm Metab Res. 2005;37(5):2815.

24. Sugo S, Minamino N, Shoji H, Kangawa K, Kitamura K, Eto T, et al. Interleu kin1, tumor necrosis factor and lipopolysaccharide additively stimulate production of adrenomedullin in vascular smooth muscle cells. Biochem Biophys Res Commun. 1995;207(1):2532.

25. Kato J, Tsuruda T, Kita T, Kitamura K, Eto T. Adrenomedullin: a pro tective factor for blood vessels. Arterioscler Thromb Vasc Biol. 2005;25(12):24807.

26. von Haehling S, Filippatos GS, Papassotiriou J, Cicoira M, Jankowska EA, Doehner W, et al. Midregional proadrenomedullin as a novel predic tor of mortality in patients with chronic heart failure. Eur J Heart Fail. 2010;12(5):48491.

27. Nishio K, Akai Y, Murao Y, Doi N, Ueda S, Tabuse H, et al. Increased plasma concentrations of adrenomedullin correlate with relaxation of vascular tone in patients with septic shock. Crit Care Med. 1997;25(6):9537.

28. Mebazaa A, Yilmaz MB, Levy P, Ponikowski P, Peacock WF, Laribi S, et al. Recommendations on prehospital & early hospital management of acute heart failure: a consensus paper from the Heart Failure Association of the European Society of Cardiology, the European Society of Emer gency Medicine and the Society of Academic Emergency Medicine. Eur J Heart Fail. 2015;17(6):54458 Epub 2015/05/23.

29. Thiele H, Ohman EM, Desch S, Eitel I, de Waha S. Management of cardio genic shock. Eur Heart J. 2015;36(20):122330 Epub 2015/03/04.

30. Mebazaa A, Tolppanen H, Mueller C, Lassus J, DiSomma S, Baksyte G, et al. Acute heart failure and cardiogenic shock: a multidisciplinary practical guidance. Intensive Care Med. 2016;42(2):14763.

31. Ponikowski P, Voors AA, Anker SD, Bueno H, Cleland JG, Coats AJ, et al. 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 Asso ciation (HFA) of the ESC. Eur Heart J. 2016;18(8):891975.

32. Kohsaka S, Menon V, Lowe AM, Lange M, Dzavik V, Sleeper LA, et al. Systemic inammatory response syndrome after acute myocar dial infarction complicated by cardiogenic shock. Arch Intern Med. 2005;165(14):164350.

33. Task Force on the management of STseamiotESoC, Steg PG, James SK, Atar D, Badano LP, BlomstromLundqvist C, et al. ESC Guidelines for the management of acute myocardial infarction in patients presenting with STsegment elevation. Eur Heart J. 2012;33(20):2569619.

34. OGara PT, Kushner FG, Ascheim DD, Casey DE Jr, Chung MK, de Lemos JA, et al. ACCF/AHA guideline for the management of STelevation myocardial infarction: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guide lines. Circulation. 2013;127(4):e362425.

35. Rihal CS, Naidu SS, Givertz MM, Szeto WY, Burke JA, Kapur NK, et al. 2015 SCAI/ACC/HFSA/STS Clinical Expert Consensus Statement on the use of percutaneous mechanical circulatory support devices in cardiovascular care: endorsed by the American Heart Association, the Cardiological Society of India, and Sociedad Latino Americana de Cardiologia Interven cion; Affirmation of Value by the Canadian Association of Interventional CardiologyAssociation Canadienne de Cardiologie dintervention. J Am Coll Cardiol. 2015;65(19):e726.

36. Werdan K, Gielen S, Ebelt H, Hochman JS. Mechanical circulatory support in cardiogenic shock. Eur Heart J. 2014;35(3):15667.