Michael T. Koller, Manfred D. Seeberger and Miodrag Filipovic
Transcrição
Michael T. Koller, Manfred D. Seeberger and Miodrag Filipovic
Troponin T and Brain Natriuretic Peptide after On-Pump Cardiac Surgery: Prognostic Impact on 12-Month Mortality and Major Cardiac Events After Adjustment for Postoperative Complications Giovanna A. L. Lurati Buse, Daniel Bolliger, Esther Seeberger, Jorge Kasper, Martin Grapow, Michael T. Koller, Manfred D. Seeberger and Miodrag Filipovic Circulation. published online August 6, 2014; Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231 Copyright © 2014 American Heart Association, Inc. All rights reserved. Print ISSN: 0009-7322. Online ISSN: 1524-4539 The online version of this article, along with updated information and services, is located on the World Wide Web at: http://circ.ahajournals.org/content/early/2014/08/06/CIRCULATIONAHA.113.007253 Data Supplement (unedited) at: http://circ.ahajournals.org/content/suppl/2014/08/06/CIRCULATIONAHA.113.007253.DC1.html Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in Circulation can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for which permission is being requested is located, click Request Permissions in the middle column of the Web page under Services. Further information about this process is available in the Permissions and Rights Question and Answer document. Reprints: Information about reprints can be found online at: http://www.lww.com/reprints Subscriptions: Information about subscribing to Circulation is online at: http://circ.ahajournals.org//subscriptions/ Downloaded from http://circ.ahajournals.org/ by guest on August 10, 2014 DOI: 10.1161/CIRCULATIONAHA.113.007253 Troponin T and Brain Natriuretic Peptide after On-Pump Cardiac Surgery: Prognostic Impact on 12-Month Mortality and Major Cardiac Events After Adjustment for Postoperative Complications Running title: Lurati Buse et al.; Biomarkers and Outcome after Cardiac Surgery Giovanna A.L. Lurati Buse, MD, MSc1; Daniel Bolliger, MD1; Esther Seeberger, RN1; Jorge Kasper, MD1; Martin Grapow, MD2; Michael T. Koller, MD3; Manfred D. Seeberger, MD1; Miodrag Filipovic, MD1,4 1 Dept for Anesthesia, Surgical Intensive Care, Prehospital Emergency Medicine and Pain Therapy, Th her erap apy, ap y, U University n veersit ni i y Hospital of Basel, Basel, Sw Swi Switzerland; itzerland; 2Ca Cardiac ard rdiacc Su Surgery urgery Dept, University Hospital Hosspital sp off Ba B Basel, sell, se l, Basel, Bas asel el,, Switzerland; el Swit Sw ittze zerl rlan rl a d; 3Ba an Basel Base sell In se Inst Institute tittutee fo forr Cl Clin Clinical nic icall E Epidemiology pide pi demi de miol olog ol ogyy & Bi og Bios Biostatistics, osta os tati ta tist ti stic st i s U Un University iversiity of of Basel, Bassel,, Basel, Baasel, Switzerland; Swit Sw itze zeerllandd; 4Ane Anesthesiology esthessio est iolo logy lo gy y Dep Dept, ept, Kan Kantonsspital anttonnssspital an sp l S St. t. G Gallen, allenn, St. Stt. Ga Gall Gallen, l en ll en, Sw Swit Switzerland itze it zerl ze rllan andd Address for Correspondence: Giovanna A.L. Lurati Buse, MD, MSc Anesthesiology University Hospital Basel CH-4031 Basel Switzerland Tel: +41-61-328-6427 Fax: +41-61-265-7320 E-mail: [email protected] Journal Subject Codes: Cardiovascular (CV) surgery:[36] CV surgery: coronary artery disease, Cardiovascular (CV) surgery:[35] CV surgery: aortic and vascular disease 1 Downloaded from http://circ.ahajournals.org/ by guest on August 10, 2014 DOI: 10.1161/CIRCULATIONAHA.113.007253 Abstract Background—The independent predictive value of troponin T (TNT) after on-pump cardiac surgery was established in several studies. However, adjustment was limited to preoperative risk factors without considering perioperative complications. Data on the prognostic value of postoperative B-type natriuretic peptide (BNP) are scarce. Our aim was to assess independent value of TNT and BNP to predict 12-month outcome after cardiac surgery after adjustment considering preoperative risk estimates and postoperative complications and to report risk stratification gains when considering the EuroSCORE combined with postoperative biomarkers. Methods and Results—This prospective cohort study includedd consecutive patients patie iennts nts undergoing unnde derg rgoi rg oing ng on-pump cardiac surgery between 2007 and 2010. We evaluated postoperative TNT and BNP, the he EuroSCORE, EurroSC Eu roSC SCOR O E, and and n postoperative complications complicatio ons n as as predictors off adverse adv dvver ersse events using Cox regression. egrres e sion. The Th he primary prim prim imaary endpoint en ndp dpoi oint nt was wa death deeathh or or major maajoor adverse adver dverrsee cardiac cardiac diac events eveent ntss (MACE) (MAC (MAC ACE) E) within wit ithi hinn 1 hi year ye ar after aft f er surgery. sur urge gerry. ry Wee calculated cal alcu cula late la teed the the net net reclassification recclas re assi as s fic si ficatiion index ind ndex ex (NRI) (NR NRI) I ooff TN I) TNT T an aand d BNP BNP in n event. addition to the th he EuroSCORE. Euro Eu roSC ro SCOR SC ORE. OR E We E. We enrolled enro en r llled 1559 ro 155 5599 patients; p ti pa t ent n s; of of whom whom 176 176 (11.3%) (11 11.3 .3%) .3 %) ssuffered uffe uf fere fe r d an even nt The adjusted hazard ratio (HR) of TNT >0.8 ȝg/L was 2.13 (95% CI, 1.47-3.15), of BNP >790 ng/L 2.44 (95% CI, 1.65-3.62). The NRI of the addition of TNT and BNP to the EuroSCORE was 0.276 (95% CI, 0.195-0.348). Conclusions—Postoperative TNT and BNP are strong predictors of 1-year events after on-pump cardiac surgery independent of preoperative risk factors and postoperative complications. Updating the preoperative EuroSCORE risk with postoperative TNT and BNP after surgery allows for improved prediction of 1-year death or MACE. Key words: prognosis, mortality, troponin, brain natriuretic peptide, surgery 2 Downloaded from http://circ.ahajournals.org/ by guest on August 10, 2014 DOI: 10.1161/CIRCULATIONAHA.113.007253 Introduction Surgical risk stratification models,1 including the EuroSCORE,2, 3 exclusively consider pre- and intraoperative risk factors. Although useful for healthcare planning and for preoperative riskbenefit assessments, this view may prove fragmentary postoperatively, as it disregards perioperative myocardial ischemic events4, 5 or postoperative complications6-8 despite their association with mortality. Therefore, there is a need for tools that allow for early postoperative risk re-assessment to guide management decisions (eg, transfer to step-down units and postdischarge follow-up schedules). Troponin T (TNT) has an established prognostic value for short-term9-11 and mid-term outcome12-14 after cardiac surgery. However, previous studies on the predictive vvalue alu ue of troponin roponin9-11 have focused on preoperative risk factors and have not accounted for postoperative co omp mpli liccati li cati tion o s (e on eg, g, iinfections, nfections, acute kidney failu ure r ). Inhospital co omp m liica cattions, ti however, may complications (eg, failure). complications, exert ex xert errt a relevant nt prognostic proogn gnos osti ticc change ti ch hange angee by by being beeing associated asssocciaated d with wit ithh both both h eelevated leva le vaate tedd ca cardiac ard rdia iacc bi biom biomarkers om marrker kers postoperatively post po stop st oper op e at er ativ ivel iv elyy an andd m mid-term idd-te d-term m ooutcome. utco ut come co mee. Several Severa raal studies s ud st udie i s have ie h ve described ha desscr criibe bed the the as asso association soci so c at ci atio ionn be io between etw twee eenn pr ee preoperative preop oper op erat er ativ at ivee na iv natriuretic atr triu iure iu reti re t c peptides ti and outcomes after cardiac surgery.15, 16 Evidence supporting the prognostic value of postoperative concentrations, in contrast, is scarce17-22 and has been generated mostly in small studies17, 21, 22 underpowered for multiple adjustment. Moreover, the prognostic gain of a risk stratification based on the EuroSCORE alone vs. a risk re-stratification based the EuroSCORE with the addition of early postoperative cardiac biomarker concentrations (reclassification) is insufficiently known. The aim of this cohort study was first, to test the hypothesis that postoperative TNT and BNP concentrations are predictive of 1-year outcome in adults undergoing on-pump cardiac 3 Downloaded from http://circ.ahajournals.org/ by guest on August 10, 2014 DOI: 10.1161/CIRCULATIONAHA.113.007253 surgery independent of preoperative risk factors and postoperative complications, and second, to estimate reclassification of extended models that include postoperative TNT and BNP information in addition to the EuroSCORE,2, 3, 23 an established and clinically widely used preoperative risk stratification tool. Patients and Methods Study Design and Participants We conducted a single-center prospective cohort of consecutive adult patients who underwent on-pump cardiac surgery (ie, including isolated CABG, single and multiple valvular procedures, and combined CABG and valvularr surgery) at the University Hospital of Basel, Switzerland Swi witz tzer erla land la nd between January 2007 and January 2010. Patients were included independent of preoperative TN NT. T. We We excluded e clud ex ded patients from the current analysis anaalyssis if they under e weent off-pump er off-pump surgery or TNT. underwent proc ced e ures requiring req qui uiriing deep deeep hypothermic hyppoth hy poth her ermi m c circulatory mi circul ulaatorry arre reest. st Th Thee st tud udyy w ass ap pprrov oved ed bby y th thee loca lo ocaal procedures arrest. study was approved local ethics et thi hics css bboard, oard oa rd,, an rd andd all all pa pparticipants art rttic icip pan a ts pprovided rovi ovided w written riitt tten en n iinformed nfoorme medd co me cons consent. nssen nt. peri riop ri oper op erat er a iv at ivee an nd in iintensive tens te n iv ns ive ca care re m anag an agem ag emeent em nt,, in incl c ud udin i g su in surg rgic rg ical ic al ttechniques, echn ec hniq hn iq que u s, choice All pe perioperative and management, including surgical and dosage of catecholamines for cardiopulmonary bypass weaning and in the intensive care unit (ICU) coagulation management, blood transfusion triggers, duration of ICU stay, and investigations for and treatment of postoperative complications were at the discretion of the attending clinicians. Trained research personnel prospectively collected detailed data on perioperative patient treatment, surgical techniques, and inhospital course and conducted followup. Inhospital complications were not adjudicated but were extracted from the hospital charts as diagnosed by the attending physicians with the exception of acute kidney injury (AKI) classification.24 Its detection is based on highest postoperative creatinine concentration measured 4 Downloaded from http://circ.ahajournals.org/ by guest on August 10, 2014 DOI: 10.1161/CIRCULATIONAHA.113.007253 during hospitalization independent of its mention in the hospital charts. For the categories risk, injury, and failure,24 the AKI classification did not require any minimal duration for creatinine value elevations above the various cut-offs. Cardiac Marker Analyses TNT (Elecsys, Roche Diagnostics, F. Hoffmann-La Roche Ltd., Basel, Switzerland) and brain natriuretic peptide (BNP) concentrations (ECLIA, Roche Diagnostics, F. Hoffmann-La Roche Ltd., Basel, Switzerland) were measured routinely at 6 am on the first and second postoperative day in all patients. Analyses were performed by the central hospital laboratory, which is subjected to practice-certified regulatory quality controls. The 99th percentile and 10% coefficient of variation of TNT lies at <0.01 ȝg/L and at 0.035 ȝg/L, respectivel respectively. ly. y. T The hee 995th 5thh 5t percentile and 6.7% coefficient of variation of BNP were at 135 ng/L and at 180 ng/L, respectively. esppec ecti tive ti veely ly. Clinical C lin nic i al Endpo Endpoint poin i t Definitions Defiini De niti tion ti onss and on an nd Endpoint Endpoi oiint Ascertainment Assce scerta ta ainm inment nt The primary endpoint was predefined Th he pr prim i arry en im end dpoint dpo oint w as pr rede rede defi fine fi nedd as tthe ne he ccomposite ompo om posi po s te te ooff ddeath eat athh fr at from om aany nyy ccause ause ause s or tthe he ooccurrence ccurrre ccur renc n e nc of non-fatal major maj ajor orr adverse adv dver errsee cardiac car ardi diiac events eve v nt ntss (MACE) (MAC (M ACE) AC E) within wit i hi hinn one one year ye r after afte af terr surgery. te surg su r er rg e y. y MACE MAC A E included myocardial infarction, cardiac arrest, need for subsequent surgical or percutaneous coronary intervention, and congestive heart failure requiring hospitalization. Follow-up data were obtained by a questionnaire mailed to all patients one year after surgery. Patients or next of kin were asked to report the occurrence and date of any hospitalization within the first year after surgery and/or the patient’s vital status. If the questionnaire was not returned, research personnel contacted the patient by telephone. If unable to contact a patient, we asked his general practitioner to provide the health or vital status information. A trained research nurse blinded to the patients’ data conducted the interviews. If 5 Downloaded from http://circ.ahajournals.org/ by guest on August 10, 2014 DOI: 10.1161/CIRCULATIONAHA.113.007253 the questionnaire or the interview indicated any hospitalization, the research team contacted the hospital or the patient’s general practitioner to obtain appropriate documentation. Based on this information, the occurrence of an endpoint was determined by two independent adjudicators blinded to inhospital TNT and BNP data. Statistical Analysis Continuous data are presented as mean with standard deviation (SD) or as median with the 1st and 3rd quartiles (Q1-Q3) as appropriate. Categorical data are presented as absolute numbers and percentages. Baseline differences between patients with events compared to event-free patients were calculated with the Student’s t, Mann-Whitney U Test, or Pearson’s Ȥ2 tests as appropriate. All analyses were based on the higher TNT and BNP value measured at 6 am m oon n postoperative day 1 and 2. We explored the association between increasing concentrations of TNT TN NT and and BNP BNP for fo or the th composite outcome by calculating cal allcu c lating the interval interrva v l likelihood liike keli l hood ratio over 10percentile perc cen e tile groups gro oup upss and andd pplotted lo ott tted ed it. it. year Wee calculated cal alcu cu ulaate tedd the th he association asso soci so ciat ci atio at ionn between io betw betw tweeen een cardiac c rdi ca rdiac iac biomarkers biiom omar arrke k rs and and n outcome ouutco utco ome at at 1 ye earr iin n a Cox Co regression egression model mod odel e with el wit i h forced f rcced entry. fo enntr try. y. The The dependent dep pen ende dent de nt variable var aria iabl ia blee was bl was the the time time to to the th he composite comp co mposite mp endpoint. We modelled cardiac biomarkers linearly, nonlinearly and dichotomized. For dichotomization, we defined the optimal cut-off using receiver operating characteristic (ROC) curves of each biomarker for the composite outcome at 1 year under assuming equal weight for sensitivity and specificity. We predefined the logistic EuroSCORE23(continuous per % increase), sepsis, sternal infection (without sepsis), respiratory infections (pneumonia, ventilator associated pneumonia, or purulent tracheobronchitis without sepsis), AKI classification,24 as additional independent variables. The explanatory variables either were acute events (e.g. postoperative sepsis), or apply to a single point in time (logistic EuroSCORE). Therefore, we opted to model 6 Downloaded from http://circ.ahajournals.org/ by guest on August 10, 2014 DOI: 10.1161/CIRCULATIONAHA.113.007253 them as static variables rather than time-dependent. To avoid model overfitting, we decided to limit the number of independent variables. The selection of postoperative variables to include into the model was guided by the following principles: 1) the concept that early complications affecting the cardiovascular system (eg, postoperative cardiogenic or hemorrhagic shock) are reflected by TNT and BNP concentrations); 2) the prevalence of the risk factor; and 3) the prognostic challenge. The decision not to include postoperative ECMO-dependency exemplifies this approach: first, hemodynamic instability requiring ECMO results in elevated TNT and BNP concentrations (ie, ECMO is not as much a confounder, but rather a link of the pathophysiological chain leading to elevated cardiac biomarkers). Second, the requirement for ECMO was exceedingly rare in our cohort. Third, ECMO-dependency is such a sstrong trron o g in indi dica di c to ca tor indicator of poor outcome that clinicians hardly need to rely on cardiac biomarkers to restratify such pa atiien ents ts. ts patients. We assessed assses esseed th the mo mode dells de ls uusing s ng si ng tthe he A kaike in ka nform rmat rm atio i n cr io cri iteerio ionn (A AIC C),, tthe hee H arrrel’ ar l’ss C l’ models Akaike information criterion (AIC), Harrel’s C-nde dex, x total x, tot otal all model mod odeel chi-square chi hi-s -sqquar qua e (Wald (W Wal aldd test) teest st)) , and and adjusted adju ad just ju s ed ed R squared. squuar a ed. ed. We assessed ass sses essseed ca cal li rat libr atiion ion off index, calibration he final model mode deel by plotting plo ott t in i g the the observed obse ob serv se r ed and rv and the the predicted pred pr ed dic i te tedd co comp mpos mp o it itee ev even ents en ts.. We eevaluated ts valu va luated the composite events. collinearity using the variance inflation factor (VIF). We internally validated the model by bootstrapping with 1000 samples. We generated 4 risk stratification strata (5%, >5% to 10%, >10% to 15%, and >15%) according to the EuroSCORE, the full model, and a model including the EuroSCORE and early cardiac biomakers information. We then assessed the net reclassification index (NRI) for EuroSCORE vs. the full model and for EuroSCORE vs. EuroSCORE and TNT plus BNP. We calculated the NRI in patients with events, in those without events, and the overall NRI.25A secondary Cox regression model with the same independent variables was fitted for 30-day mortality and MACE. We generated a third model by including 7 Downloaded from http://circ.ahajournals.org/ by guest on August 10, 2014 DOI: 10.1161/CIRCULATIONAHA.113.007253 the variables recently added to the EuroSCORE II26 (ie, CCS4, NYHA class, urgency of the procedure in 3 categories [urgent, emergent, and salvage], insulin-dependent diabetes, creatinine clearance, and the new procedure scores) as independent variables in our model. The focus of this analysis were any on-pump cardiac surgery procedures, however we also analyzed the distribution of troponin and BNP concentrations following specific procedures. Further, we explored the potential interaction in the association between cardiac biomarkers and outcome by the type of procedure. We opted to test for interaction within the subgroups “isolated CABG” vs. “other procedures” to maintain power for the interaction analysis, rather than define subgroups by specific types of non-CABG procedures. Statics were calculated by SPSS 20, R 3.1.0., and by Microsoft Excel 201 010. 0. W 2010. Wee considered the null-hypothesis refuted if the two-sided P value was < 0.05. Sa amp mple le S izee iz Sample Size F orr th tthee 36-mon onth th eenrollment nroollm nr men entt peri pperiod, eriod od, we od we ant tic iciipatted d a ssample amplee si amp siz ze off 15 11500 00 ppatients. 00 attie ient ntts. s A ssum umin um ing ng a For 36-month anticipated size Assuming 1 1-ye year ar event eve vent nt rate rat atee of of 12%, 12% 2%,,14 we we expected expe ex pect pe cteed ed up up to 180 180 0 events, eveentts, s, aallowing llo lo owi wing ng g ffor orr a sstatistically taati t st stic icaally ally rrobust obustt obus 1-year o el bbased od ased as e oon n up u tto o 18 vvariables. aria ar iabl ia b es bl e .27 multivariate m model Results Descriptive Analysis Of the 1713 consecutive patients who underwent cardiac surgery in our institution between 2007 and 2010, 1569 fulfilled the eligibility criteria. In 10 patients (0.6%), no TNT values were available, of whom 8 (0.5%) had died prior to the first sampling time (6 am on postoperative day 1) (Figure 1). Follow-up was completed in 1545 patients (99.1%); the remaining 14 patients were censored at last contact date. One hundred-fifteen patients (7.3%) underwent cardiac 8 Downloaded from http://circ.ahajournals.org/ by guest on August 10, 2014 DOI: 10.1161/CIRCULATIONAHA.113.007253 surgery within 7 days of an acute coronary syndrome with detectable preoperative TNT concentrations. Their median preoperative TNT was 0.28 ug/L (Q1-Q3, 0.12-0.68 ug/L). BNP concentrations were available in 1364 patients (86.9%). In patients missing POD 1 or POD2 samples, we assumed the available concentration to represent the higher concentration. Within the 12-month postoperative follow-up, 176 patients (11.3%) suffered the composite endpoint event, including 103 deaths (6.6%). We recorded 86 MACE (5.5%): 17 acute coronary syndromes (1.1%), 14 cardiac arrests (0.9%), 20 percutaneous coronary interventions (1.3%), 6 subsequent CABG (0.4%), and 29 acute congestive heart failures (1.9%) requiring hospitalization. Eighty-three events (5.3%) occurred within 30 days of surgery, of which there were 58 deaths (3.7%). Compared to patients with uneventful follow-up, follow w-u -up, p those tho hose se who who suffered uffered an event had a higher EuroSCORE, higher TNT, higher BNP, and had presented more postoperative complications BNP were po ost stop opeerat op erat ativ ivee co iv omp mpli l cations (Table 1; Figure 22a-c). a c). Troponin and aan nd BN NP we w re higher after procedures proc ced e ures other oth herr than thhan han isolated issol olat ated at ed CABG CAB ABG G (Table (Tablle 2). Association As Asso soci so ciat ci atio io on between betw betw ween een Postoperative Po ost stop perrat atiive ve TNT TNT and and d BNP BNP P with wit ith h 1-Year 1-Ye 1Y ar Ye ar Events E ent Ev ents ts and and Reclassification Rec ecllasssiffic icaatio at o n ratio event Figure 3 plots plo otss tthe he llikelihood i el ik e ih i oood ra rati tioo of TN ti TNT T an andd BN BNP P fo forr th thee 111-year year ye a ccomposite omp mpos osit os i e ev it even entt by increasingg en concentrations (groups defined by 10-percentiles). The AUC of the ROC curve of TNT was 0.688 (95% CI, 0.640-0.735) with an optimal cut-off at 0.8 μg/L. The corresponding values for BNP were 0.714 (95% CI, 0.670-0.759) and 790 ng/L. The median duration from surgery to event was 34 days (Q1-Q3, 8.25-114.75) in the whole population. In patients with elevated cardiac biomarkers, the composite event occurred after a median of 22 days (Q1-Q3, 6-90). Patients without elevated cardiac markers suffered their event after a median of 87 days (Q1-Q3, 22.75-157) (P<0.001). Table 3 reports models’ informations. Concentrations of both TNT and BNP that above 9 Downloaded from http://circ.ahajournals.org/ by guest on August 10, 2014 DOI: 10.1161/CIRCULATIONAHA.113.007253 the cut-off within the first two postoperative days were independently associated with the composite endpoint at 1 year after on-pump cardiac surgery after adjustment by the EuroSCORE, postoperative infectious complications, and AKI (Table 4). VIF were not suggestive of collinearity (VIF <2 for all variables). Figure 4 shows the calibration results at various risk levels. The overall NRI for 1-year events with the addition of TNT and BNP information to the EuroSCORE vs. the EuroSCORE alone was 0.276 (95% CI, 0.195-0.348). The addition of biomarkers performed well for detecting increased risk (NRI, 0.696 [95% CI, 0.620-0.762]) but poorly for decreased risk (NRI, -0.420 [95% CI, -0.448 to -0.0393]). The same pattern was present for the final model. The overall NRI of the final model was 0.426 (95% CI, CI, 00.347-0.493) .34 34734 7 0. 70.49 493)) with a NRI for higher risk amounting to 0.734 (95% CI, 0.660-0.797) and -0.308 (95% CI, 0.335 0..3335 to -0.283) -0. 0.28 0. 283)) ffor 28 or decreased risk (Supplemental or (Supplement nttal Tables Tables 1-4). The models mode mo d lss fitted de fit itttedd to predict predi redict c 30-day ct 30-day events eventss and eve and with with h the thhe he addition add ddit itio ionn of EuroSCORE EuroS uroS SCO CORE R II, RE II, tthe hee variables Wee di vari va riab ri able ab l s showed le shhow owed ed d similar sim imiilaar results res esuults (Supplemental (Su Supp pp pleeme menttal a Tables Tabl ables les 5-6). 5-66).. W 5didd not not fi find nd aany ny eevidence vidden dence nce off significant ignificant interaction in nte tera r ct ra ctio io on inn the thhe association a so as soci c at ci atio ionn between betw be twee tw eenn TNT ee TNT (p=0.153) (p=0 (p =0 0.1 . 53 53)) and an nd BNP BNP (p=0.785) ( =0 (p 0.7 .785 85)) and 85 outcome by the type of procedure (isolated CABG vs other procedures). Discussion Main Results This cohort study showed that the association between an increase in TNT concentration within the first 2 postoperative days and 1-year all-cause mortality and MACE after cardiac surgery persisted after adjustment considering preoperative as well as postoperative confounders. Further, we detected a significant independent association between postoperative BNP 10 Downloaded from http://circ.ahajournals.org/ by guest on August 10, 2014 DOI: 10.1161/CIRCULATIONAHA.113.007253 concentrations and 1-year mortality and MACE. Extending the EuroSCORE to include postoperative TNT and BNP information allowed for improved the risk stratification in every fourth patient. The time-window of approximately 1 month between events and elevated biomarkers suggests that measurement of cardiac biomarkers flags high-risk patients early (i.e., at a time point when management optimization holds promise for outcome improvement). Preoperative cardiac surgery risk stratification models,1 including the EuroSCORE,2, 3 that exclusively consider pre- and intraoperative risk factors are the most useful for healthcare planning and for preoperative risk-benefit assessments. However, there is a need for reevaluation and risk reassessment after the surgical procedure itself. The data generated in this cohort study uggest that postoperative TNT and BNP concentrations within the first 2 postop perat a iv at ve da days ys aare re suggest postoperative helpful tools for risk reassessment after cardiac surgery. They are non-invasive, easily obtained, an nd allow allo al low lo w for for early eaarl rlyy flagging f agging of patients at riskk for fl or death death and MACE MAC CE within wiith thiin the next (few) and m onnth(s). nt This is oobservational bsserrva vati tiion nal sstudy tudy tu dy y ddid id d nnot ot aaddress dd dresss tthe hee qquestion uesstion ho ue ow to o ooptimize ptim pt imiize im ize th thee m annage nageemen month(s). how management off ppatients atie at ient ie ntss wi with th iincreased nccre reaaseed ed rrisk issk aafter fter ft er ccardiac arrdiiac ssurgery. urge ur gerry. ge ry. P oteent ntiiall ap appr proa pr oach oa ch hess m ight ig h iinclude ht nclludde m nc orre Potential approaches might more tringent inhospital inho hoosp spit ital al m onitor on orin or in ng in step-down ste tep--do down wn uunits n ts ni t oorr sy yst stem emaati em ticc te ttelemetry leme le metr me tryy mo tr moni nito ni tori to ring ri ng on the stringent monitoring systematic monitoring regular ward, dedicated post-cardiac surgery outpatients clinics, or virtual wards. Comparison with Previous Studies Several studies evaluated the association between troponin elevations after cardiac surgery and short-9, 10 and long-term adverse events.12-14 Adabag and colleagues9 reported on a concentrationresponse relationship between troponin I release and short-term mortality (OR, 1.3 [95% CI 1.11.5] per 0.05 μg/L increase in troponin I levels in valvular patients and 1.4 [95% CI 1.2-1.6] in CABG patients). Similarly, Nesher and colleagues10 described a concentration-dependent association of TNT and short-term MACE. Further, their analysis revealed an association 11 Downloaded from http://circ.ahajournals.org/ by guest on August 10, 2014 DOI: 10.1161/CIRCULATIONAHA.113.007253 between a TNT cut-off concentration of 0.8 μg/L and short-term mortality. In our data, the optimal cut-off for the prediction of 1-year mortality or MACE was also 0.8 μg/L. Croal and colleagues12 showed an OR of 1.10 (95% CI, 1.03-1.18) per 10 μg/L troponin I increase for one-year mortality. Our group previously described in next to 750 patients (enrolled in 2005-2006) an AUC 0.78 (95% CI, 0.7-0.86) using TNT for 12-month mortality and an independent association between elevated TNT and 12-month mortality (OR, 3.83 [95% CI, 2.10-7.12]) after adjustment using the EuroSCORE.14 Previous studies (including our own data), however, disregarded potential confounding by postoperative complications and did not assess BNP concentrations. Further, they did not quantify the prognostic gain compared to preoperative isk stratification tools in terms off reclassification (eg, the EuroSCORE). risk Fox and colleagues19 assessed the association between postoperative BNP and 5-year mo mort rtal rt alit al ityy in 11183 it 1883 patients pa undergoing elective C ABG. In contras AB st to our our data, data, they did not detect mortality CABG. contrast an any ny si ssignificant gnifican nt independent inddepe in peend den nt association asso as soci ciiat atio ionn bbetween etw ween po postop postoperative opperat erattiv ve BN BNP P an and nd lo long-term ong ng-t -ter erm m mo m mortality rtal rt alit al ityy (HR (H off 1.62, 1.662, 95% 95% CI, CI,, 0.71–3.68). 0.7 .71– 1– –3.68 68). ) This Thi hiss fact fact might mig ighht be be related reelaateed to their theeir choice cho oic icee off a longer lon onge gerr follow-up ge follow fo ow-u ow -upp duration. It iiss pl plau plausible au usi sibl blee th bl tthat at over ove verr a 5-year 5 yeear period, 5per erio iod, io d, in in an a elderly elde derl rlyy population rl popuula po lati t on with ti wit ithh a high high burden of risk factors, subsequent cardiac and noncardiac events might have diluted the strength of the association between early postoperative BNP concentrations and all-cause death. This hypothesis is supported first, by the report of the association between congestive heart failure episodes and heart failure mortality in the same population,20 and second, by the significant independent association between postoperative BNP and poor physical function within a shorter follow-up duration in a subgroup of the same patients.18 Three smaller cohort studies (<400 patients in total)17, 21, 22 described higher postoperative BNP concentrations in patients who died after cardiac surgery. However, they were underpowered to conduct multivariable adjustment. 12 Downloaded from http://circ.ahajournals.org/ by guest on August 10, 2014 DOI: 10.1161/CIRCULATIONAHA.113.007253 Strength and Limitations The strength of our study stems from first, the inclusion of a broad, unselected population of consecutive patients undergoing on-pump cardiac surgery; second, the minimal number of missing TNT data (0.6%); third, achieving >99% follow-up completeness; fourth, the blinded event adjudication by two independent readers with extensive knowledge and experience of the perioperative cardiac surgery setting (cardiac anesthesiologists and intensive care physicians); fifth, the high data quality with extensive data consistency checks and quality controls; and finally, our analysis plan that considers not only preoperative risk factors but also postoperative complications in the prediction model given their role as confounders and with the quantification of reclassification by cardiac biomarkers. We are aware of some limitations of our study. First, we assessed the concentration of biomarkers all independent bi iom mar arkkers kers iin n al ll patients pa at 6 am on postoperative postoperatiive v dday ay 1, independ n en nd nt of the time of surgery. Therefore, thee samples hours after T heerefore, er sam amplles of of postoperative post po stop st oper eraativ er ativ ve dayy 1 were re collected col ollleect cted ed between betweeen 122 and and 220 0 ho hour u s af ur fter t er surgery after surgery. Blood urg rger eryy and er and the the samples sampl sam mples les of postoperative posto tope perrati pe ratiivee day day ay 2 between beetw weeen 36 and and 44 44 ho hhours urss af fte terr su surg rgeery. rg y. B loo lo od sample collection conforms clinical ample colle ecttio ionn at a ffixed i ed ix d ppoint oint oi n ooff th nt thee da day, y hhowever, y, owev ow ever ev e , co er conf nfor nf o ms too cl or clin inic in ical ic al rroutine ou uti tine ne rrather a her than too at sample collection at a defined time after surgery. A further disadvantage of our set time for blood sample collection is the fact that cardiac biomarkers were not measured in the few patients who died before 6 am on the first postoperative morning. Second, in contrast to the standardized approach to cardiac biomarkers sampling, the study protocol did not mandate any screening standard for infectious or renal complications; rather, their detection relied on diagnostic work-up ordered by the attending physicians. Therefore, we cannot exclude that some patients might have been misclassified. However, this approach mimicked clinical routine, and as such it better complied with the pragmatic approach 13 Downloaded from http://circ.ahajournals.org/ by guest on August 10, 2014 DOI: 10.1161/CIRCULATIONAHA.113.007253 of our study question. Further, the large sample size can be expected to have mitigated this issue. Third, follow-up assessment based on mailed questionnaires may be subjected to non-response bias. However, telephone interviews with patients and/or their general practitioners allowed us to achieve almost complete follow-up (>99%). Our follow-up assessment may also have been biased by misclassification of outcomes, but the probability of forgetting a hospitalization seemed remote. Moreover, misclassification bias could be expected to be nondifferential as the patients were not aware of their exposure status (ie, TNT and BNP readings). Fourth, the choice of equal weight for sensitivity and specificity to define the optimal cutoff point resulted in the fact that at the proposed cut-off value sensitivity was higher than specificity. pecificity. This in turn led to a good reclassification in patients with increased risk, riisk sk,, whereas wher wh erea er eass risk ea ri down-rating was poor. Fifth, focused only. Thielmann Fift Fi fth ft h, wee fo h, foc cused on postoperative TNT cconcentrations onncentrations on nly l . Th Thi ie ielmann and colleagues28 showed howed ow w a prognostic progn gnoostticc vvalue alu luee of eelevated lu levvat le vated ted ppreoperative reopper perative ve tro troponin oponi poniin I co concentrations oncen enttra tratio ons ns ffor or sshort-term hort ho rtt-t - erm e rm mortality. However, the presenting acute coronary mo ort r al alit i y. it y H owever, owe er, th he hi highh bbaseline asel as elin el inee ri in risk sk ooff ppatients a ients at nt pr pres eseenti es enti ting ng a rrecent e entt ac ec acut u e co ut oro onaary ry syndrome yndrome and/or and ndd/o /orr emergent emer em errge g ntt surgery sur u ge gery r is ry is already alre al read re adyy considered ad co ons nsid ider id ered ed d by by preoperative preo pr eope eo peera rati tivve risk ti risk s stratification str trat a if at ific ication ic (EuroSCORE).2 Further, fewer than 8% of the included patients demonstrated detectable TNT concentrations before surgery. We also did not measure preoperative of BNP. Multiple studies15, 16, 19 established the significant independent association between preoperative BNP concentrations and outcome after cardiac surgery. Sixth, for adjustment by preoperative risk estimation we used the EuroSCORE that was initially generated to predict short-term outcome. However, its prognostic value was validated for long-term events.23 Further, the independent prognostic impact of cardiac biomarkers was maintained when restricting the analysis to short-term mortality and MACE. An additional 14 Downloaded from http://circ.ahajournals.org/ by guest on August 10, 2014 DOI: 10.1161/CIRCULATIONAHA.113.007253 limitation of the EuroSCORE is its overestimation of mortality, particularly with higher scores.1,29 Seventh, a new version of the EuroSCORE (II) was proposed after completing enrollment.26 We had information on the variables newly included in the model with the exception of pulmonary arterial pressure, which was collected as pulmonary hypertension (>60 mm Hg) instead of the categories (31-55 mm Hg and >55 mm Hg) as per the EuroSCORE II. Therefore, we were not able to calculate the EuroSCORE II in our patients. However, the addition of the new variables included in the EuroSCORE II (eg, CCS 4, NYHA, creatinine clearance) into our model did not affect the association between TNT and BNP with the 30-day or 1-year endpoint. Further, the improvement in predictive performance of the E uroS o CO oS CORE RE IIII is EuroSCORE till not fully established.1, 29-31 still Eigh Ei ghth gh th, the th th he sample s mple size in the subgroups was sa was nott sufficient sufficien nt to test tes estt for interaction in the Eighth, association asso oci c ation be between etw weeen ca card cardiac rd dia iacc ma m markers rk ker erss aand nd outcome ouutccomee within with hin n specific spe peci cifi fiic surgical surggic su ical al procedures. prooce cedu dure du rees. H However, owev ow ever, ev wee did did not not find fin indd any any evidence ev vid den encee for for interaction int nter eraacti acti tion on inn th tthee subg ssubgroup ubg bgro roup ro up iisolated s late so lateed CA CABG BG G vvss ot other the her pro pprocedures. roced oced dur ures e. Finally, Finall lly, y, 12% 12% of of postoperative p st po stop oper op errat ativ iv ve BNP BNP concentrations conc co ncen nc entr en t at tr a ion onss we on were re m missing. isssi sing ng.. Ho ng Howe However, weve we ver, ve r, w with i h a total ooff it over 1300 patients, this cohort still represents one of the largest to address this topic. Conclusion Postoperative TNT and BNP elevations are independently and strongly associated with 1-year mortality and MACE after on-pump cardiac surgery. The extension of the EuroSCORE, a preoperative risk stratification tool, with postoperative TNT and BNP concentrations allowed for a better prediction of 1-year all-cause mortality and the occurrence of MACE after on-pump cardiac surgery. 15 Downloaded from http://circ.ahajournals.org/ by guest on August 10, 2014 DOI: 10.1161/CIRCULATIONAHA.113.007253 Acknowledgments: The authors thank Salome Dell-Kuster, MD, MSc for statistical advice, Tanja Schmidt, MD, Claudia Werner, RN, and Anne-Michelle Droux, RN, for data collection, Domenik Zwahlen for data management, and Allison Dwileski, BS, for her expert editorial assistance. Funding Sources: This study was supported by a research Grant of the Swiss National Science Foundation (SNF Number 3200B0_121943/1), the European Association of Cardiothoracic Anaesthesiologists, and by the Foundation for Research and Education, Department for Anesthesia, Surgical Intensive Care, Prehospital Emergency Medicine and Pain Therapy, Prehospital Emergency Medicine and Pain Therapy, University Hospital Basel, Switzerland. Conflict of Interest Disclosures: None. References: 1. N illss sson on JJ,, Al A gots go t so ts s n L, Hoglund P, Luhrs C, Br Bran a dt J. Comparison n ooff 19 pre-operative risk Nilsson Algotsson Brandt tra rati tiifica fic tiionn m odel od elss in n ope pen-he hear a t su ssurgery. rgerry. rg y Eu ur Hea H eart J. J 20 2006;2 27:8 7:8677-87 8744. stratification models open-heart Eur Heart 2006;27:867-874. 2. N 2. ashef SA, R oquess F ich hel P auduche heau E he eme mesh shoow S sh alamo m n R. E mo urrop opea eaan ssystem ysttem m Nashef Roques F,, Mi Michel P,, Ga Gauducheau E,, Le Lemeshow S,, S Salamon European fo or ca card diac op oper eraatiive rrisk iskk eva vaalu uat atio ionn (E Euros urosccore core). ). Eur Eur J Cardiothorac Card Ca rdioth rd th horac Surg. Surg urg. 19 999 99;;16:9--133. for cardiac operative evaluation (Euroscore). 1999;16:9-13. 33.. R oque oq uess F, ue F, M iche ic hell P he olds ol dsto ds tone to ne A R, Na Nash shef sh ef S A. T he llogistic ogis og isti is ticc Eu ti Euro rosc ro scor sc oree. Eu or Eurr Heart Hear He artt JJ.. ar Roques Michel P,, G Goldstone AR, Nashef SA. The Euroscore. 2003;24:8811-88 8822. 88 2003;24:881-882. 4. Gensini GF, Fusi C, Conti AA, Calamai GC, Montesi GF, Galanti G, Noferi D, Carbonetto F, Palmarini MF, Abbate R, Vaccari M. Cardiac troponin I and Q-wave perioperative myocardial infarction after coronary artery bypass surgery. Crit Care Med. 1998;26:1986-1990. 5. Quaini E, Colombo T, Russo C, Vitali E, Pellegrini A. Hospital morbidity and mortality after myocardial revascularisation surgery: Current changes in risk factors. Eur J Cardiothorac Surg. 1995;9:275-282. 6. Kollef MH, Sharpless L, Vlasnik J, Pasque C, Murphy D, Fraser VJ. The impact of nosocomial infections on patient outcomes following cardiac surgery. Chest. 1997;112:666-675. 7. Kubota H, Miyata H, Motomura N, Ono M, Takamoto S, Harii K, Oura N, Hirabayashi S, Kyo S. Deep sternal wound infection after cardiac surgery. J Cardiothorac Surg. 2013;8:132. 8. Tolpin DA, Collard CD, Lee VV, Virani SS, Allison PM, Elayda MA, Pan W. Subclinical changes in serum creatinine and mortality after coronary artery bypass grafting. J Thorac 16 Downloaded from http://circ.ahajournals.org/ by guest on August 10, 2014 DOI: 10.1161/CIRCULATIONAHA.113.007253 Cardiovasc Surg. 2012;143:682-688 e681. 9. Adabag AS, Rector T, Mithani S, Harmala J, Ward HB, Kelly RF, Nguyen JT, McFalls EO, Bloomfield HE. Prognostic significance of elevated cardiac troponin I after heart surgery. Ann Thorac Surg. 2007;83:1744-1750. 10. Nesher N, Alghamdi AA, Singh SK, Sever JY, Christakis GT, Goldman BS, Cohen GN, Moussa F, Fremes SE. Troponin after cardiac surgery: A predictor or a phenomenon? Ann Thorac Surg. 2008;85:1348-1354. 11. Sellgren A, Nilsson F, Jeppsson A. The relationship between ASAT, CKMB, troponin-T and mortality after cardiac surgery. Scand Cardiovasc J. 2007:1-5. 12. Croal BL, Hillis GS, Gibson PH, Fazal MT, El-Shafei H, Gibson G, Jeffrey RR, Buchan KG, West D, Cuthbertson BH. Relationship between postoperative cardiac troponin I levels and outcome of cardiac surgery. Circulation. 2006;114:1468-1475. The 13. Lurati Buse GA, Koller MT, Grapow M, Bolliger D, Seeberger M, Filipovic M.. T he Eur prognostic value of troponin release after adult cardiac surgery - a meta-analysis. s. Eu ur J Cardiothorac Surg. 2010;37:399-406. GA,, Koller MT, Grapow M, Bruni CM, Kasper J, Seeberger MD, Filipovic M. 14. Lurati Buse GA G 12-month 12 2-m mon onth th outcome out utcome me after after cardiac surgery: Prediction Prediccti tion on n by troponin T in combination co ombination with the European system cardiac operative Ann 2009;88:1806-1812. Eu uro ope p an sys ysste tem m fo fforr ca ard dia iacc op oper erat er a iv at ivee ri risk sk eevaluation. valu va luaatiion. A nn Thorac Tho ora racc Surg. Surg Su rg.. 200 rg 009; 9;88 9; 88:1 88 :180 8066 18812 612. 15. Fox AA, Shernan SK, Collard CD,, L Liu Aranki SF, DeSantis P,, B Body 15 5. F ox AA A, Sh S ernnan SK K, C ollar ll rd CD C iu KY KY, A ran anki ki S F, D eS San ntis SM SM,, JJarolim arroli r ol m P od dy SC SC. Preoperative natriuretic peptide independent predictor Preo Pr eope eo pera pe rati tive ve bb-type -typ -t ypee na yp atrriu iure r ti ticc pe pept ptiide pt ide iss aass in nde d penndeent ent pr pred ed dic icto to or of vventricular en ntr tric iccullar a ddysfunction ysfu ys unccti tion on n aand n nd mortality coronary bypass mo ort rtal alit ity af afterr pr pprimary im mar ary co oro r na nary ry aartery rtery by bypa pass s ggrafting. raftin ng. J Thorac Thor Th oracc Cardiovasc Car ardi d ovvas ascc Surg. Su 2008;136:452-461. 2008;136:452 52 2-4 461 1. 16. Lurati Buse GA, Koller MT, Burkhart C, Seeberger MD, Filipovic M. The predictive value of preoperative natriuretic peptide concentrations in adults undergoing surgery: A systematic review and meta-analysis. Anesth Analg. 2011;112:1019-1033. 17. Crescenzi G, Landoni G, Bignami E, Belloni I, Biselli C, Rosica C, Guarracino F, Marino G, Zangrillo A. N-terminal b-natriuretic peptide after coronary artery bypass graft surgery. J Cardiothorac Vasc Anesth. 2009;23:147-150. 18. Fox AA, Marcantonio ER, Collard CD, Thoma M, Perry TE, Shernan SK, Muehlschlegel JD, Body SC. Increased peak postoperative b-type natriuretic peptide predicts decreased longer-term physical function after primary coronary artery bypass graft surgery. Anesthesiology. 2011;114:807-816. 19. Fox AA, Muehlschlegel JD, Body SC, Shernan SK, Liu KY, Perry TE, Aranki SF, Cook EF, Marcantonio ER, Collard CD. Comparison of the utility of preoperative versus postoperative btype natriuretic peptide for predicting hospital length of stay and mortality after primary 17 Downloaded from http://circ.ahajournals.org/ by guest on August 10, 2014 DOI: 10.1161/CIRCULATIONAHA.113.007253 coronary artery bypass grafting. Anesthesiology. 2010;112:842-851. 20. Fox AA, Nascimben L, Body SC, Collard CD, Mitani AA, Liu KY, Muehlschlegel JD, Shernan SK, Marcantonio ER. Increased perioperative b-type natriuretic peptide associates with heart failure hospitalization or heart failure death after coronary artery bypass graft surgery. Anesthesiology. 2013;119:284-294. 21. Provenchere S, Berroeta C, Reynaud C, Baron G, Poirier I, Desmonts JM, Iung B, Dehoux M, Philip I, Benessiano J. Plasma brain natriuretic peptide and cardiac troponin I concentrations after adult cardiac surgery: Association with postoperative cardiac dysfunction and 1-year mortality. Crit Care Med. 2006;34:995-1000. 22. Suttner S, Boldt J, Lang K, Rohm KD, Piper SN, Mayer J. Association of N-terminal probrain natriuretic peptide and cardiac troponin T with in-hospital cardiac events in elderly patients undergoing coronary artery surgery. Eur J Anaesthesiol. 2008;25:834-841. 23. Toumpoulis IK, Anagnostopoulos CE, DeRose JJ, Swistel DG. European system for cardiac operative risk evaluation predicts long-term survival in patients with coronary artery arter erry bypass byypa p ss grafting. Eur J Cardiothorac Surg. 2004;25:51-58. w. 24. Bellomo R, Ronco C, Kellum JA, Mehta RL, Palevsky P, Acute Dialysis Quality Initiative w Acute renal failure - definition, outcome measures,, animal models, fluid therapy and information technology Thee second international consensus ech hno nolo logy lo gy needs: needs dss: Th T cons nssen e sus conferencee off tthe he aacute cute dialysis quality initiative Critic Care. nitia iaati tive ((ADQI) AD DQI QI)) gr ggroup. oup. ou p Cr p. Crit itic ic C a e. 22004;8:R204-212. ar 004; 00 4;;8: 8:R2 R2004--2122. 25. Pencina MJ,, D D'Agostino RB, Sr., D'Agostino RB, Vasan RS. Evaluating the 25 5. P encina MJ M 'Aggosstino no oR B, Sr. r.., D'A D 'Agoosttino R B, JJr., r.,, Va r. V sann R S. E valu aluatiing th he aadded ddeed predictive new marker: the and beyond. pred pr edic ed icti ic t ve aability ti bili bi lity ty y ooff a ne ew ma mark rk ker er:: From From aarea rea un re uunder der th der he ro rocc curve cu urv ve to rreclassification ecllass ssif i iccat if atio ionn an io nd be bey yonnd. nd 2008;27:157-172; 207-112. Stat at Med. Med. d 20 2008 8;2 ; 7: 7:15 1577-17 1722; ddiscussion iscu is ussion 20 2077-11 112. 2. 26. Na Nashef SA, Roques F, S Sharples LD, Ni Nilsson Smith C, Go Goldstone 26 Nash shef ef S A R oque oq uess F harp ha rple less LD Nils lsso sonn JJ, S mith mi th C Gold ldst ston onee AR, AR Lockowandt Lock Lo ckow owan andt dt U. U Euroscore II. Eur J Cardiothorac Surg. 2012;41:734-744; discussion 744-735. 27. Bagley SC, White H, Golomb BA. Logistic regression in the medical literature: Standards for use and reporting, with particular attention to one medical domain. J Clin Epidemiol. 2001;54:979-985. 28. Thielmann M, Massoudy P, Neuhauser M, Knipp S, Kamler M, Piotrowski J, Mann K, Jakob H. Prognostic value of preoperative cardiac troponin I in patients with non-ST-segment elevation acute coronary syndromes undergoing coronary artery bypass surgery. Chest. 2005;128:35263536. 29. Howell NJ, Head SJ, Freemantle N, van der Meulen TA, Senanayake E, Menon A, Kappetein AP, Pagano D. The new Euroscore II does not improve prediction of mortality in high-risk patients undergoing cardiac surgery: A collaborative analysis of two european centres. Eur J Cardiothorac Surg. 2013;44:1006-1011; discussion 1011. 18 Downloaded from http://circ.ahajournals.org/ by guest on August 10, 2014 DOI: 10.1161/CIRCULATIONAHA.113.007253 30. Chalmers J, Pullan M, Fabri B, McShane J, Shaw M, Mediratta N, Poullis M. Validation of Euroscore II in a modern cohort of patients undergoing cardiac surgery. Eur J Cardiothorac Surg. 2013;43:688-694. 31. Grant SW, Hickey GL, Dimarakis I, Trivedi U, Bryan A, Treasure T, Cooper G, Pagano D, Buchan I, Bridgewater B. How does Euroscore II perform in UK cardiac surgery; an analysis of 23 740 patients from the society for cardiothoracic surgery in Great Britain and Ireland National Database. Heart. 2012;98:1568-1572. 19 Downloaded from http://circ.ahajournals.org/ by guest on August 10, 2014 DOI: 10.1161/CIRCULATIONAHA.113.007253 Table 1. Patient Characteristics, Surgical and Postoperative Data All Patients n=1559 Preoperative data Age (years), mean (SD) 67 (10) Male 1151 (73.8) Recent ACS (30 days) 234 (15.0) Left main stem stenosis >50% 345 (22.2) Acute endocarditis 32 (2.1) Pulmonary hypertension 60 mm Hg 86 (5.5) Preoperative EF 60 (49-65) Arterial hypertension 1092 (70.1) Diabetes (insulin or OAD) 359 (23.1) BMI (kg/m2), mean (SD) 27.8 (10.8) History of TIA/CVI 137 (8.8) History of PVD 195 (12.5) Preoperative GRF (ml/min) 82 (61-106) COPD under treatment 87 (5.6) Logistic EuroSCORE 2.25 (1.42-3.80) Additive EuroSCORE 5 (3-7) Procedural data Isolated 778 (50.0) solated CABG Isolated 196 (12.6)) sollat ated ed aortic aor orti ticc valve v lv va ve replacement replacement CA ABG and and d aortic aor orticc va valv lve surg gery 3) CABG valve surgery 176 (11.3) CABG CAB BG and 2-valve 2-v -val allve ve procedures pro oce cedu durees du 3 ((2.2) 2.2) 2. 2 2) CABG 34 Emergent er 46 (3.0) (3 3.0 0) Emergent surgeryy 46 Po osttop o erativ ve data ta Postoperative Trop Tr op pon o in T P OD 1 ((μg/L) μg/L μg /L)) 00.45 0. 45 5 ((0.25-0.90) 0.25 0. 2 -0 25 -0.9 .90) .9 0) Troponin POD Trop Tr opon op onin on in T P OD 2 ((μg/L) μg/L μg /L)) /L 0.40 ((0.19-0.96) 0.40 0.19 0.19 19-0 0.9 .96) 96) Troponin POD (ng/ (n g/L) g/ L L) 43 33 (277-794) (277 (2 77-7 77 -794 -7 94)) 94 BNP POD 1 (ng/L) 433 O 2 (ng/L) ( / ) (326 812) BNP POD 498 (326-812) ECMO 6 (0.4) IABP 55 (3.5) Postoperative acute kidney 141 (9.0) injury/failure Inhospital sepsis 33 (2.1) Inhospital respiratory infection* 73 (4.7) Sternal infection 49 (3.1) ICU LOS (days) 2 (2-3) Hospital LOS (days) 11 (9-16) No Events n=1383 Events n=176 P Value 67 (11) 1031 (74.5) 199 (14.4) 291 (21.1) 28 (2.0) 67 (4.9) 60 (50-65) 965 (69.9) 306 (22.2) 27.6 (9.1) 122 (8.8) 157 (11.4) 83 (63-107) 72 (5.2) 2.14 (1.41-3.44) 5 (3-7) 71 (10) 120 (68.2) 35 (19.9) 54 (30.7) 4 (2.3) 19 (10.8) 54 (40-60) 127 (72.2) 53 (30.1) 32.7 (27.1) 15 (8.5) 38 (21.6) 71 (58-93) 15 (8.6) 6) (1.87-6.96) 3.68 (1.8 877-6. 6.96 6. 96)) 96 (4-9) 7 (4 4-9 9) 0.64 0.07 0.054 0.04 0.827 0.001 0.012 0.533 0.019 0.103 0.985 <0.001 0.012 0.068 0.0 .06 68 <0.001 <0.0 <0 .0 001 0 <0.001 <0.0 <0 001 0 695 (50.3) 187 (13.5) 153 (11.1 1) (11.1) 331 1 (2 ((2.3) .3 3) 37 (2.7) (2.7) (2.7 37 83 (47.2) 9 (5.1) 23 (13.1) (13.1) 3 (1 (1.8 .8)) .8 (1.8) (5.11) (5 9 (5.1) 0.429 0.002 0.434 00.668 0.66 .66 668 0.07 0. 72 0.072 00.42 0. 42 ((0.25-0.82) 0 25 0. 25-0 -0 0.882) 0.37 ((0.18-0.82) 0.37 0.18 0.18 18-0 -0 0.8 .82) 82) 417 (259-722) (259 (2 59-7 59 -7 7222) 417 4 0 (313-724) (313 24) 470 1 (0.07) 28 (2.0) 00.87 . 7 (0 .8 .40400-1 1.92 92 2) (0.40-1.92) 00.83 .83 .8 83 (0 .31.3 31 1.93 1.93 93)) (0.31-1.93) 870 87 0 (4 483 83-1 - 36 -1 367) 7 (483-1367) 892 (440 1 06) (440-1506) 5 (2.8) 27 (15.3) <0.0 <0 .001 0 01 <0.001 <0 0.0 .001 001 <0.001 <0.001 0 001 <0.001 <0.001 <0.001 89 (6.5) 52 (29.5) <0.001 15 (1.1) 43 (3.1) 34 (2.5) 2 (2-2) 11 (9-15) 18 (10.2) 30 (17.0) 15 (8.5) 4 (2-7) 15.5 (10-31) <0.001 <0.001 <0.001 <0.001 <0.001 Number (%) unless stated otherwise, continuous variables are reported as median (Q1-Q3) unless stated otherwise. *Respiratory infections included pneumonia, ventilator associated pneumonia, or purulent tracheobronchitis. ACS, acute coronary syndrome; BMI, body mass index; BNP, B-type natriuretic peptide; CABG, coronary artery bypass graft; COPD, chronic obstructive pulmonary disease; CVI, cerebrovascular insult; ECMO, extra-corporal membrane oxygenation; EF, ejection fraction; GFR, glomerular filtration rate (Cockcroft-Gault estimate); IABP, intra-aortic balloon pump; ICU, intensive care unit; LOS, length of stay; OAD, oral antidiabetics; POD, postoperative day; PVD, peripheral vascular disease; Q1-Q3, first to third quartile; SD, standard deviation; TIA, transient ischemic attack. 20 Downloaded from http://circ.ahajournals.org/ by guest on August 10, 2014 DOI: 10.1161/CIRCULATIONAHA.113.007253 Table 2. Troponin T and BNP concentrations by type of procedure. Higher TNT POD 1 and 2 at 6 am (μg/L) TNT POD 1 (μg/L) TNT POD 2 (μg/L) Higher BNP POD 1 and 2 at 6 am (ng/L) BNP POD 1 (ng/L) BNP POD 2 (ng/L) Isolated CABG Other procedures 0.38 (0.20-0.71) 0.37 (0.15-0.90) 0.31 (0.2-0.78) 400 (259-705) 369 (233-661) 434 (285-766) 0.55 (0.31-1.07) 0.54 (0.25-1.04) 0.46 (0.32-1.05) 532 (353-926) 493 (328-900) 541 (377-835) Difference of the median (95%CI) 0.17 (0.12-0.21) 0.17 (0.12-0.22) 0.16 (0.11-0.20) 132 (75-189) 124 (79-169) 107 (38-176) p-value <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 Median (Q1 (Q1-Q3); Q3); BNP: brain natriuretic peptide; POD: postoperative day; TNT: troponin T; pp-values vallue uess ca cal calculatated lcul ulattat ateed using Mann-Whitney-U test 21 Downloaded from http://circ.ahajournals.org/ by guest on August 10, 2014 DOI: 10.1161/CIRCULATIONAHA.113.007253 Table 3. Cox-regression Model Information Model Univariate Models Peak TNT, linear Peak TNT, non-linear Peak TNT, binary (0.8) BNP, linear BNP, non-linear BNP, binary (790) Models with BNP and TNT BNP binary, TNT binary BNP non-linear, TNT non-linear Multivariable Models Euroscore Eu uro rosc scor sc oree linear, or line li n arr, postop po complications, co omp mpli licatiionns, T TNT NT bbinary, inar in ary, y y, BNP binary (fi BNP (final fina nal mo na model) mode del) l) Euroscore Eu uro osc scor o e linear, liine near ar,, postop postop post opp complications, comp co mp pli lica cati ca tioons, ti ons, T TNT NT nonlinear, nonl no nlin nl inea in earr, B ea BNP NP nnonlinear onli on line li near ne ar AIC C-index Total model Chi-Square (Wald-test) 2516.6 2480.1 2486 2211.51 2173.88 2178.91 0.689 0.689 0.666 0.707 0.707 0.674 96.19 99.24 80.11 104.95 97.1 83.97 0.04 0.07 0.063 0.046 0.081 0.074 2145.979 0.721 0.729 (original) 0.7258 (corrected, 150 bootstrap samples) 117.91 0.103 123.22 00.103 .1103 217.377 217.37 21 3777 0.155 0. 210.6961 210. 21 0 69961 00.147 .1147 2150.397 2098.685 2098 20 98.6 .685 85 2105.984 21005.9 21 05 984 0.761 0.7661 ((original) original) 0.757 57 (corrected, (cor (c rrect cted ed,, 15 1150 0 bootstrap boootstrapp samples) samp samp mple les) s)) 0.7594 0.75 0. 7594 75 94 (original)0.7531 (orrigi (o gina gi naal))0.7 0.7531 7531 3 (corrected, (corre (c ect cted ed, 15 1500 bootstrap boot bo o st ot s ra rapp samples) samp sa mple mp lees) 22 Downloaded from http://circ.ahajournals.org/ by guest on August 10, 2014 Adjusted R squared DOI: 10.1161/CIRCULATIONAHA.113.007253 Table 4. Cox-regression Model for Composite Endpoint within 1 Year after Surgery. HR Logistic EuroSCORE TNT 0.8 ȝg/L BNP 790 ng/L Inhospital sepsis Inhospital respiratory infection Sternal infection Acute kidney injury Model Derivation 95% CI P Value Model Validation (1000 Bootstrap Samples) HR 95% CI P Value 1.029 2.128 2.445 2.064 0.999 1.483 1.705 1.159 1.061 3.053 3.508 3.676 0.062 0.000 0.000 0.014 1.029 2.128 2.445 2.064 0.996 1.472 1.647 1.151 1.060 3.154 3.616 4.118 0.059 0.001 0.001 0.021 2.270 1.260 1.310 1.433 0.702 1.120 3.595 2.262 1.534 0.000 0.440 0.001 2.270 1.260 1.310 1.340 0.610 1.089 3.613 2.437 1.559 0.003 0.515 0.005 Logistic EuroSCORE entered continuously per % increase. Respiratory infections included pneumonia, ventilator associated pneumonia, or purulent tracheobronchitis. Acute kidney injury included the following categories: a) Risk: serum erum creatinine (SCr) increased by 1.5 or glomerular filtration rate (GFR) decreased >25%; 25%; b) Injury: Inj njury: y: SCr SCr increased SCr ncreased by 2 or GFR decreased >50%; c) Failure: SCr increased by 3 or GFR decreased 75% orr S Cr 4 4 mg/dL mg/d mg /dL /d L or acute cute rise 0.5 mg/dL; d) Loss: persistent acute renal failure: complete loss of kidney function >4 >4 weeks week we ek ks24 TNT, troponin BNP, brain natriuretic peptide; CI, confidence interval; HR, hazard ratio; LR, likelihood ratio; T NT T, tr tropon oniin T Figure F igu gure u Legends: Legen nds ds: CABG, coronary bypass Figure 1. Flow Flo ow chart ch har artt of patient pat atie ient ie n inclusion. nt inc n lu usion on.. CA on CABG BG,, co BG coro r na nary ry aartery rter rt eryy by er bypa pass pa ss ggraft; r ft ra ft; CP CPB, B, cardiopulmonary bypass; DHCA, deep hypothermic circulatory arrest; MACE, major adverse cardiac events. Figure 2a-c. KaplanMeier curves and survival table stratified by TNT (a), BNP (b), and postoperative complications (c). Figure 3. Likelihood ratio of the 1-year composite by increasing TNT and BNP concentrations. Figure 4. Calibration of the final model by risk groups. 23 Downloaded from http://circ.ahajournals.org/ by guest on August 10, 2014 Figure 1 Downloaded from http://circ.ahajournals.org/ by guest on August 10, 2014 Figure 2A Downloaded from http://circ.ahajournals.org/ by guest on August 10, 2014 Figure 2B Downloaded from http://circ.ahajournals.org/ by guest on August 10, 2014 Figure 2C Downloaded from http://circ.ahajournals.org/ by guest on August 10, 2014 Figure 3 Downloaded from http://circ.ahajournals.org/ by guest on August 10, 2014 Figure 4 Downloaded from http://circ.ahajournals.org/ by guest on August 10, 2014 SUPPLEMENTAL MATERIAL Reclassification Tables Supplemental Table 1. Reclassification by Using Biomarker Information in Patients with 1-year Events EuroSCORE and Cardiac Biomarker Predicted Predicted Predicted Predicted ≤0.05 >0.05-0.1 >0.1-0.15 >0.15 Predicted ≤0.05 27 4 34 42 Predicted >0.05-0.1 0 5 1 23 Predicted >0.1-0.15 0 3 0 10 Predicted >0.15 0 0 1 8 EuroSCORE Supplemental Table 2. Reclassification by Using Biomarker Information in Patients without 1-year Events EuroSCORE and Cardiac Biomarker Predicted Predicted Predicted <=0.05 Predicted >0.05-0.1 >0.1-0.15 >0.15 601 75 265 87 Predicted >0.05-0.1 0 59 10 61 Predicted >0.1-0.15 0 8 0 22 Predicted >0.15 0 2 3 13 Predicted ≤0.05 EuroSCORE Supplemental Table 3. Reclassification by Using Biomarker and Postoperative Complication Information (Full Model) in Patients with 1-year Events Full Model Predicted Predicted Predicted Predicted ≤0.05 >0.05-0.1 >0.1-0.15 >0.15 Predicted ≤0.05 21 17 18 51 Predicted >0.05-0.1 0 3 7 19 Predicted >0.1-0.15 0 3 2 8 Predicted >0.15 0 0 1 8 EuroSCORE Supplemental Table 4. Reclassification by Using Biomarker and Postoperative Complication Information (Full Model) in Patients without 1-year Events Full Model Predicted Predicted Predicted Predicted ≤0.05 >0.05-0.1 >0.1-0.15 >0.15 Predicted ≤0.05 587 208 111 122 Predicted >0.05-0.1 26 27 37 40 Predicted >0.1-0.15 0 8 5 17 Predicted >0.15 0 5 2 11 EuroSCORE Supplemental Table 5. Cox-Regression Model for Composite Endpoint within 30 Days after Surgery Hazard Ratio 95% Confidence Interval P Value Logistic EuroSCORE 1.039 0.998 1.082 0.062 TNT ≥0.8 μg/L 2.964 1.695 5.185 <0.001 BNP ≥790 ng/L 2.664 1.540 4.609 <0.001 Inhospital sepsis 2.079 0.974 4.440 0.059 infection 1.382 0.704 2.715 0.347 Sternal infection 1.159 0.493 2.726 0.735 Acute kidney injury 1.301 1.048 1.614 0.017 Inhospital respiratory BNP, brain natriuretic protein; TNT, troponin T Supplemental Table 6. Independent Association between Cardiac Biomarkers and 1-year and 30-day Events in the full Model and in the Model including EuroSCORE II Variables TNT ≥0.8 μg/L 1-year Composite 30-day Composite Event Event HR (95% CI) HR (95% CI) 2.13 2.96 (1.48-3.05) (1.69-5.18) 2.45 2.66 (1.70-3.51) (1.54-4.61) 2.18 2.70 (1.49-3.19) (1.48-4.92) 2.28 2.12 (1.55-3.37) (1.17-3.82) Full model BNP≥790 ng/L Full model + TNT ≥0.8 μg/L EuroSCORE II variables BNP ≥790 ng/L *EuroSCORE II variables: CCS4, NYHA class, urgency of the procedure in 3 categories (urgent, emergent, and salvage), insulin-dependent diabetes, creatinine clearance, and the procedure weights according to the EuroSCORE II ; Full model variables: Logistic EuroSCORE, TNT, BNP, and postoperative sepsis, pneumonia, sternal infection, acute kidney injury