Open access Protocol Study protocol for an observational cohort evaluating incidence and clinical relevance of perioperative elevation of high-­sensitivity troponin I and N-­ terminal pro-­brain natriuretic peptide in patients undergoing lung resection María Alonso ‍ ‍,1,2 Ekaterine Popova ‍ ‍,3,4 Ascencion Martin-­Grande,5 Javier Pérez-­Vélez,6 Juan Carlos Trujillo,7 Luis Gajate,5 Marcos de Miguel,1 Anna González-­Tallada,1 Elisabeth Martínez-­Téllez,7 Esther Cladellas-­Gutierrez,7 Georgina Planas,7 Angélica de Pablo,5 Diego Parise,5 Angel Candela-­Toha,5 Miriam de Nadal1,2 To cite: Alonso M, Popova E, ABSTRACT Martin-­Grande A, et al. Study Introduction Myocardial injury after non-­cardiac surgery STRENGTHS AND LIMITATIONS OF THIS STUDY protocol for an observational has been defined as myocardial injury due to ischaemia, with ⇒ This is the first study to integrate serial measure- cohort evaluating incidence ment of N-­ terminal pro-­brain natriuretic peptide or without additional symptoms or ECG changes occurring and clinical relevance of (NT-­proBNP) and high-­sensitivity troponin I (hs-­TnI) perioperative elevation of during or within 30 days after non-­cardiac surgery and mainly diagnosed based on elevated postoperative cardiac troponin biomarkers, more sensitive and specific for early high-­sensitivity troponin I and N-­terminal pro-­brain natriuretic (cTn) values. In patients undergoing thoracic surgery for lung detection of cardiovascular complications after lung peptide in patients undergoing resection, only postoperative cTn elevations are seemingly resection. lung resection. BMJ Open not enough as an independent predictor of cardiovascular ⇒ The hs-­TnI assay used in our study is more sensitive 2022;12:e063778. doi:10.1136/ complications. After lung resection, troponin elevations and accurate to detect low concentrations, contrib- bmjopen-2022-063778 may be regulated by mechanisms other than myocardial uting to a probable increase in detection of myo- ischaemia. The combination of perioperative natriuretic peptide cardial injury after non-­cardiac surgery (MINS) than ► Prepublication history and additional supplemental material measurement together with high-­sensitivity cTns may help to detected with conventional cardiac troponin assays. for this paper are available identify changes in ventricular function during thoracic surgery. ⇒ Determining the clinical significance of hs-­TnI ele- online. To view these files, Integrating both cardiac biomarkers may improve the predictive vation (due to ischaemic or non-­ischaemic causes) please visit the journal online value for cardiovascular complications after lung resection. We may be crucial in the development of diagnostic (http://dx.doi.org/10.1136/​ designed our cohort study to evaluate perioperative elevation and therapeutic protocols, aimed at improving MINS bmjopen-2022-063778). of both high-­sensitivity troponin I (hs-­TnI) and N-­terminal pro-­ prognosis in patients undergoing thoracic surgery brain natriuretic peptide (NT-­proBNP) in patients undergoing for lung resection. Received 11 April 2022 ⇒ The results of our study may be difficult to compare Accepted 21 November 2022 lung resection and to establish a risk score for major with previously published works, due to differences cardiovascular postoperative complications. in methodology, especially when using both a hs-­TnI Methods and analysis We will conduct a prospective, assay, together with NT-­proBNP. multicentre, observational cohort study, including 345 ⇒ Our protocol is designed as a short-­term (30 days patients undergoing elective thoracic surgery for lung follow-­up) study, therefore, long-­term studies for resection. Cardiac biomarkers such as hs-­TnI and better understanding of NT-­proBNP and hs-­TnI ef- NT-­proBNP will be measured preoperatively and at ficiency will be still needed. postoperatively on days 1 and 2. We will calculate a risk © Author(s) (or their score for major cardiovascular postoperative complications employer(s)) 2022. Re-­use based on both biomarkers’ perioperative changes. All permitted under CC BY-­NC. No commercial re-­use. See rights patients will be followed up for 30 days after surgery. INTRODUCTION and permissions. Published by Ethics and dissemination All participating centres were Myocardial injury is usually presented as acute BMJ. approved by the Ethics Research Committee. Written informed coronary syndrome, manifested as myocar- For numbered affiliations see consent is required for all patients before inclusion. Results will dial infarction (MI) or unstable angina, and end of article. be disseminated through publication in peer-­reviewed journals associated with symptoms as chest pain or and presentations at national or international conference shortness of breath.1 However, perioperative Correspondence to meetings. Dr Ekaterine Popova; myocardial injury after non-­cardiac surgery Trial registration number NCT04749212. ​EPopova@​santpau.​cat (MINS), apparently triggered by myocardial Alonso M, et al. BMJ Open 2022;12:e063778. doi:10.1136/bmjopen-2022-063778 1 Open access oxygen supply–demand mismatch and mostly silent, is an earlier time and in lower concentrations,16 therefore presented as an isolated cardiac troponin (cTn) elevation considered more precise prognostic tools. There is only without signs or symptoms or ECG changes.2 Therefore, one available study, published by our group, which used a MINS term derived from a large observational study and high-­sensitivity troponin I (hs-­TnI) assay (ADVIA Centaur is defined as myocardial injury due to ischaemia occur- Siemens analyser) in high-­risk thoracic surgery patients, ring during or within 30 days after non-­cardiac surgery where the overall incidence of MINS (defined as at least and mainly diagnosed based on elevated cTn values.3 The one elevated value of hs-­TnI ≥0.04 ng/mL at the first two incidence of MINS is quite elevated, especially in patients postoperative days) was 27%.4 In this study, no signifi- undergoing thoracic surgery, and is usually related to cant association between MINS and 30-­day mortality was several additional factors: smoking status, type and length observed. Therefore, the elevated incidence of MINS of the surgery, pericardial incision or intraoperative use of after thoracic surgery, the independent relationship with vasoactive drugs.4 Based on available evidence postopera- the extent of lung resection, as well as the fact that MINS tive cTn elevation is an independent predictor of 30-­day was not associated with greater short-­term mortality, may mortality in high-­ risk non-­ cardiac surgery patients.5 suggest nonischaemic causes for the hs-­TnI elevation in Without cTn measurements, many ischaemic episodes thoracic surgery patients. may go missed and underdiagnosed. Routine perioper- ative monitoring of cardiac biomarkers is highly recom- Natriuretic peptides as cardiac biomarkers mended6 and supported by recent guidelines in patients Natriuretic peptides are vasodilator hormones responsible with high cardiovascular risk.5 7 for regulation of blood pressure and volume homoeo- stasis. B-­type natriuretic peptide (BNP) and its inactive MINS in thoracic surgery cleavage product, N-­ terminal fragment (NT-­ proBNP), In patients undergoing thoracic surgery serial cTn moni- are produced and secreted by ventricular myocytes in toring seems reasonable, considering the frequent use of response to ventricular wall stretching, myocardial isch- high doses of analgesics in postoperative settings. Serial aemia and endocrine modulation through other neuro-­ cTn monitoring, previously indicated only by cardiac hormones such as norepinephrine, glucocorticoids or symptoms or ECG changes, would help identify subclin- proinflammatory cytokines.17 18 Both NT-­proBNP and BNP ical myocardial injuries with prognostic significance in are useful for detection in plasma, but NT-­proBNP has a patients susceptible to cardiovascular complications. In longer plasma half-­life and therefore persists at a higher patients undergoing thoracotomy and lung surgery, cTn concentration when compared with BNP. NT-­proBNP is has been proven a reliable biomarker for myocardial accepted to be biochemically more stable than BNP,19 injury detection.8 Although cTn elevations are mainly thereby fewer requirements are needed to protect its associated with intrapericardial resection and pneumo- degradation. Moreover, scaling varies among BNP assays nectomy, the available evidence about incidence varies while NT-­proBNP assays are already standardised, and greatly (0%–49%). The routine use of cTn monitoring there is no simple conversion factor to compare BNP and in patients undergoing thoracic surgery is not yet imple- NT proBNP levels.20 mented, and its predictive value is still unclear.5 9–15 The Natriuretic peptides are commonly used for the diag- table 1 summarises available studies using different cTn nosis and management of congestive heart failure.21 assay measurements in thoracic surgery patients. However, studies in several types of acute coronary During the last years, most hospitals have replaced syndromes such as ST-­ segment myocardial infarction, conventional cTn assay with the new generation high-­ non-­ ST elevation myocardial infarction or unstable sensitivity cTn assays which are more accurate and sensi- angina, have shown that elevated levels of natriuretic tive to detect myocardial ischaemia, being more precise at peptides are independently associated with adverse Table 1 Summary of studies including thoracic surgery patients with cardiac troponin measurements: cut-­off values and incidence of troponin elevations Study/authors Patients (n) Cardiac troponin (cTn) assay Cut-­off value Incidence of cTn elevation 9 Vikenes et al 24 cTn I, cTn T >0.2 սg/L 0% Lim et al10 41 cTn I >0.04 սg/L 7% Lucreziotti et al11 50 cTn I >0.06 սg/L 20% 12 Muley et al 64 cTn I >0.32 ng/mL 14% VISION group (2012)5 376 cTn T >0.03 ng/mL 8.7% Hua et al13 491 cTn I >0.04 ng/mL 16% Puelacher et al14 219 hs*-­Tn T ≥0.14 ng/mL increase from preoperative value 24% 15 Uchoa and Caramelli 151 cTn I ≥0.16 ng/mL 49% González-­Tallada et al4 177 hs*-­Tn I ≥0.04 ng/mL 27% *High-­sensitivity cardiac troponin. 2 Alonso M, et al. BMJ Open 2022;12:e063778. doi:10.1136/bmjopen-2022-063778 Open access outcomes, especially mortality22 and could be used as hs-­TnI elevations in patients undergoing lung resection. prognostic markers. Our study will also evaluate: (1) NT-­ proBNP and/or hs-­TnI elevation’s association with major cardiovascular Natriuretic peptides in the perioperative setting complications (non-­fatal cardiac arrest, acute myocardial Nowadays, there is growing evidence that BNP measure- infarction, angina, congestive heart failure, new clini- ment is an independent predictor of perioperative and cally significant atrial fibrillation, cardiovascular death); long-­term complications related to non-­cardiac surgical (2) perioperative variables associated with NT-­ proBNP procedures and is also independently associated with and hs-­ cTnT elevations; and (3) predictive value of the composite outcome of death and non-­fatal MI at 30 NT-­proBNP and hs-­TnI for cardiovascular complications. and 180 postoperative days.23 Recent meta-­analysis of 18 studies including at least 2000 patients examined the role of postoperative natriuretic peptides as predictors of METHODS AND ANALYSIS postoperative complications. Patients with postoperative Study design BNP ≥245 pg/mL or NT-­proBNP ≥718 pg/mL had signifi- The current study is a prospective, multicentre, obser- cantly greater risk for 30-­day mortality or non-­fatal MI vational cohort including 345 patients undergoing (OR 4.5, 95% CI 2.75 to 7.4), mortality (OR 4.5, 95% CI elective thoracic surgery for lung resection. The study 2.29 to 7.69) and cardiac failure (OR 18.5, 95% CI 4.55 will be carried out in three University Spanish hospitals to 75.29).20 performing the largest number of thoracic surgeries The results of nested VISION substudy in 10 402 patients per year. The study protocol adheres to the STROBE undergoing non-­cardiac surgery, showed a strong associ- Statements checklist (online supplemental file 1), and ation between NT-­proBNP elevation, vascular death and is approved by the Ethics Research Committees of each MINS within 30 days. Results of this cohort showed that participating hospital. Written informed consent will patients with NT-­proBNP values <200 pg/mL presented a be obtained from all patients before inclusion (online risk of death or MI of ≤3%, compared with patients with supplemental file 2). The study is registered at C ​ linical- NT-­proBNP values between 200 and <1500 pg/mL, where Trials.​gov. risk of death or MI was 7.9%.24 Furthermore, authors concluded that NT-­proBNP may be a helpful tool in iden- Study population tifying patients at higher risk of postoperative cardiac We will include patients aged ≥45 years old, scheduled for events and who may benefit from perioperative troponin elective thoracic surgery for lung resection (pneumonec- monitoring. tomy, lobectomy, bilobectomy or segmentectomy) under general anaesthesia. Patients will be excluded if they meet Natriuretic peptides in thoracic surgery one of the following criteria: (1) undergoing urgent, Both BNP and NT-­proBNP have been reported to reflect emergent or non-­thoracic surgery; (2) presence of symp- postoperative right ventricular dysfunction,25 and to be toms related with infection or sepsis and (3) documented an effective predictor of atrial fibrillation,26–30 as well as a history of severe heart failure and/or an ejection fraction high-­quality predictor of postoperative cardiopulmonary less than 30%. complications in elderly patients with lung resection by cancer.18 31 Furthermore, some studies tried to establish Patient recruitment thresholds of NT-­ proBNP (>30 pg/mL) for postopera- Research personnel will screen elective thoracic surgery tive complications in elderly patients.31 32 A meta-­analysis patients daily, to identify eligible candidates. Potentially of five studies including 742 elective thoracic surgery eligible patients will be approached and invited to partic- patients, showed that elevated preoperative NT-­proBNP ipate in the study. Written informed consent form will levels were associated with an OR 3.13 (95% CI 1.38 to be obtained before surgery. Patients will be managed 7.12; I2=87%) for postoperative atrial fibrillation.30 Based following the routine anaesthetic and surgical protocols on this evidence, a recent study conducted by Amar et al of each hospital. including 635 patients developed a brain peptide predic- tion model for atrial fibrillation after thoracic surgery.33 Study procedures Another recent study conducted by Cagini et al with Biomarkers’ measurements 294 patients undergoing elective pulmonary resection, All the three participating centres will use the same immu- showed BNP levels measured at 24 hours after surgery to noassay system (Atellica IM Siemens) for measuring both be an independent predictor of postoperative complica- cardiac biomarkers (hs-­TnI and NT-­proBNP) considering tions, with an association between BNP elevation and an institutional thresholds to define the elevations. increased risk for adverse cardiopulmonary events.18 In Both biomarkers will be measured in each patient at conclusion, it seems plausible that perioperative natri- three time points: preoperatively and at day 1 and day 2 uretic peptide measurement may help identify changes after surgery. Both biomarkers will be added to the routine in ventricular function during thoracic surgery. preoperative or postoperative blood tests, avoiding the We designed our cohort study to evaluate the inci- need for extra blood samples. In patients with postopera- dence and magnitude of perioperative NT-­proBNP and tive elevated hs-­TnI levels and/or presence of ischaemic Alonso M, et al. BMJ Open 2022;12:e063778. doi:10.1136/bmjopen-2022-063778 3 Open access Figure 1 Study organigram. hs-­TnI, high-­sensitivity troponin I; NT-­proBNP, N-­terminal pro-­brain natriuretic peptide. symptoms, an ECG will be performed and non-­ischaemic Variables causes of hs-­TnI elevation will be excluded. All patients 1. Preoperative variables: (a) Age; (b) gender; (c) history will be closely monitored after surgical procedure and for of ischaemic heart disease: angina, myocardial infarc- 2 days postoperatively. (figure 1 - Study organigram). tion or acute coronary syndrome, segmental cardiac For patients with confirmed MINS, a cardiologist eval- wall motion abnormality on echocardiography or a uation will be performed. Treatment will be at clinician’s segmental fixed defect of radionuclide imaging, posi- discretion considering the patient’s cardiovascular risk tive radionuclide exercise, echocardiographic exercise factors and available guidelines recommendations.7 or pharmacological cardiovascular stress test demon- We will try to provide clinical and practical threshold strating cardiac ischaemia and coronary angiographic values based on guidelines recommendations, which in or computer tomography evidence of a atherosclerot- turn determine institutional laboratory threshold values. ic stenosis ≥50% of the diameter of any coronary ar- In case of NT-­ proBNP, two different threshold values tery; (d) history of cardiac arrest; (e) congestive heart will be provided: ≥125 pg/mL (for outpatients) used in failure; (f) peripheral vascular disease; (g) stroke or preoperative setting and ≥300 pg/mL (for patients in the transient ischaemic attack; (h) deep venous thrombo- emergency department)34 used in postoperative setting. sis or pulmonary embolism; (i) diabetes mellitus; (j) In contrast, only one institutional threshold value for hypertension; (k) obstructive sleep apnoea; (l) chron- hs-­TnI (<45 ng/L) will be provided, used only in postop- ic obstructive pulmonary disease; (m) active cancer; erative setting.1 (n) history or current smoking; (o) chronic kidney disease; and (p) abnormal ECG prior to surgery (left Follow-up ventricular hypertrophy, left bundle branch block, ST-­ All recruited patients will be followed up by researchers T abnormalities). during hospitalisation until discharge. We will further 2. Intraoperative variables: (a) Type of surgical resection: perform a telephone follow-­ up on the 30th day after surgery to evaluate major cardiovascular events including pneumonectomy, bilobectomy, lobectomy, segmen- non-­fatal cardiac arrest, acute myocardial infarction, tectomy; (b) surgical approach: open, video-­assisted cardiac revascularisation, congestive heart failure, new thoracic surgery (VATS), VATS-­converted to open; (c) clinically significant atrial fibrillation, cardiovascular side of surgery (right/left); (d) duration of the proce- death and stroke (online supplemental material file 3). dure; (e) type of anaesthesia (volatile or intravenous); (f) use of regional analgesic techniques (thoracic epi- Outcomes dural or paravertebral block); (g) intraoperative hypo- Primary outcome: Perioperative hs-­TnI and NT-­proBNP tension (≥20 mm Hg decrease from basal mean arterial elevation in patients scheduled for elective lung resec- pressure or a 20% change in mean arterial pressure) tion. Major cardiovascular complications at 30-­days: non-­ and its duration (minutes); and (h) intraoperative fatal cardiac arrest, acute myocardial infarction, cardiac blood loss and transfusion requirements (number of revascularisation, congestive heart failure, new clinically Red Blood Cells (RBC)). significant atrial fibrillation, cardiovascular death and 3. Postoperative variables: (a) Hypotension (≥20 mm stroke. Hg decrease from basal mean arterial pressure or Secondary outcome: MINS. a 20% change in mean arterial pressure) and its 4 Alonso M, et al. BMJ Open 2022;12:e063778. doi:10.1136/bmjopen-2022-063778 Open access duration (minutes) during Post-­anaesthesia care unit Table 2 Sample size calculation (PACU)/Intensive care unit (ICU) and stay; (b) ma- Sample size and precision estimates for primary endpoints based on jor revascularisation, congestive heart failure, new previous studies clinically significant atrial fibrillation, cardiovascular Incidence Sample size Precision 95% CI death, stroke; (c) pulmonary complications (atelecta- hs-­TnI 28% 310 5% 23% to 33% sis, pulmonary oedema, pneumonia, acute respiratory NT-­proBNP 22% 264 5% 17% to 27% distress syndrome). NT-­proBNP 22% 310 4.61% 17.4% to 26.6% Data collection and data management hs-­TnI, high-­sensitivity cardiac troponin; NT-­proBNP, N-­terminal pro-­brain natriuretic We will record data from electronic clinical records peptide. of each hospital (baseline, operative assessment and hospital discharge). One month after surgery data will Statistical analysis be collected by telephone follow-­up. If patients (or rela- All variables will be reviewed to prevent inconsistencies, tives) indicate that they have experienced any of the main and a flow chart will be designed to describe withdraws outcomes, relevant source documents will be obtained. from the study. Continuous variables will be presented An electronic case report form (eCRF) will be designed as the mean and SD or the median and IQR and cate- in a secure online database (www.clinapsis.com) to record gorical variables as the count and percentage (%). The all the protocol-­required information reported for each univariate relationship between each factor and the patient to preserve and maintain quality and integrity cardiovascular complications will be analysed using an of all the data. All clinical records, source documents, independent samples t-­test or the Mann-­Whitney U test, follow-­up visit logs and CRFs will be locked in the appro- as appropriate, for the continuous variables and the χ2 priate study files in each site. Only study team members test or Fisher’'s exact test for dichotomous categorical will have access to protected health information. Privacy variables. The OR and the 95% CI for all exposure factors and confidentiality of the obtained data will be ensured will also be calculated. Multivariable logistic regression according to Spanish Organic Law 1/1999 of the 3rd analyses will be performed to identify the independent of December regarding the protection of personal risk factors of major cardiovascular complications and data. Moreover, a specific database will be generated, the predictive performance of identified variables will with a special identification number as a unique iden- be evaluated using receiver-­ operating characteristic tifier for each patient, to preserve the confidentiality. curve analysis. In order to calculate the adjusted OR as a All completed data will be uploaded and stored in the measure of association between hs-­TnI and NT-­proBNP electronic database previously mentioned with password elevation and cardiovascular postoperative complica- protected access. tions, a score table will be created depending on both biomarkers’ perioperative changes. The table 3 defines Statistical considerations the independent or predictive variable comparisons of Sample size baseline characteristics among the groups categorised by The primary aim of the study is the incidence of periop- the cut-­off values. Comparisons of baseline characteristics erative hs-­ TnI and NT-­ proBNP elevation in patients among the groups (categorised by the cut-­off values) will scheduled for elective lung resection. The threshold be assessed by one-­way analysis of variance or the Kruskal-­ to define perioperative elevation of hs-­TnI is ≥45 ng/L Wallis test for continuous variables and χ2 test for categor- (higher than upper reference percentile (upper refer- ical variables. P values less than 0.05 will be considered ence limit (URL))), whereas the threshold to define significant. perioperative elevation of NT -proBNP is ≥125 pg/m Prognostic thresholds for the prediction of major (higher than institutional value for outpatients). We cardiovascular complications should be associated with a calculate the sample size according to the results of the high enough OR result in a post-­test probability sufficient previous studies in patients undergoing lung resection, to change the perception of clinical risk. We will perform considering an incidence for hs-­TnI elevation of 28%4 regression analysis using the post-­hoc biomarkers thresh- and an incidence of 22% for NT-­proBNP elevation,27 olds and convert continuous measurement into a dichot- assuming a similar incidence in our study. The sample omous variable to calculate unadjusted OR for the size calculation is based on a formula that derives directly prediction on postoperative complications. from the method to calculate the CI for estimating a mean of a quantitative variable assuming normal distri- Study organisation bution. Sample size and precision estimates for primary The study will be coordinated by principal investi- endpoints are presented in table 2. We assume that 310 gator and each hospital will have one lead investigator. patients will provide a precision (half width of 95% CI) Research personnel of all participating hospitals will be estimate of 23% to 33% for hs-­TnI elevation and 17.4% responsible for the organisation of the study in the local to 26.6% for NT-­proBNP elevation at a 95% confidence settings, completing the study database and ensuring level. Assuming a drop-­out rate of 10%, a total sample high data quality. The research group of the study is size of 345 subjects is planned. made up of multidisciplinary clinical investigators from Alonso M, et al. BMJ Open 2022;12:e063778. doi:10.1136/bmjopen-2022-063778 5 Open access associated with the study and the Ethic Committees will Table 3 Independent or predictive variable comparisons of baseline characteristics among the groups categorised by have access to the clinical data. All obtained data will be the cut-­off values stored on a secure online database with a special identifi- cation number to preserve strict confidentiality. Biomarker analysis Cut-­off value Score Preoperative NT-­proBNP ≥125 pg/mL 1 Dissemination policy Postoperative NT-­proBNP ≥300 pg/mL 1 The results obtained from this study will be presented at Postoperative hs-­TnI ≥45 ng/L 1 national or international conferences and will be consid- ered for publication in peer-­reviewed scientific journals. *Possible score value groups: 0, 1, 2, 3. The knowledge dissemination plan includes traditional hs-­TnI, high-­sensitivity troponin I; NT-­proBNP, N-­terminal pro-­brain natriuretic peptide. modes of dissemination (ie, publication in a policy-­driving journal, national/international conference presenta- tions), as well as engagement of influential medical organisations. Broader dissemination will be performed anaesthesiology, thoracic surgery, clinical epidemiology by public websites and Twitter accounts of participating and biochemistry departments, with large experience in hospitals. In addition, dissemination will be conducted in the perioperative setting. The study structure includes the Spanish Association of Anaesthesiology as well as in Adjudication Committee, Steering Committee, Oper- the international network of European Society of Anaes- ations Committee and Monitoring Committee. The thesiology and Intensive Care. Adjudication Committee is composed by clinicians with considerable expertise in perioperative outcomes and will adjudicate all important clinical events. The Steering Committee will supervise the whole project, and the DISCUSSION Operations Committee will ensure optimal running of Executive summary the study. Since our study is designed as an observational In contrast to other non-­cardiac surgeries, troponin eleva- cohort a Monitoring Committee will be composed by tion seems more common after lung resection. Despite researchers with substantial experience and knowledge the importance of systematic perioperative high-­sensitivity in perioperative medicine and by internal monitoring cardiac troponin measurements in patients undergoing analysis, which will ensure high data quality, verifying the lung resection, available data is still not compelling which consistency and accuracy of data entries in the eCRF and raises major interest. Since troponin elevation after other protocol-­related documents, as well as verify the thoracic surgery may be regulated by mechanisms other adherence to the protocol and the completeness. than myocardial ischaemia (eg, elevation of right ventric- ular afterload), introduction of systematic perioperative Patient and public involvement NT-­proBNP measurements together with hs-­TnI may help No patients were involved in proposing the research ques- identify changes in myocardial function. Therefore, we tion or the outcome measures, nor were they involved in suggest that integrating both cardiac biomarkers (hs-­TnI the design or implementation of the study. If patients and NT-­proBNP) may better predict postoperative cardio- develop MINS, they will be advised to seek cardiologist vascular complications in this group of patients. Our data consultation. If they are already followed-­up by a cardiol- will provide a valuable overview on the pathophysiology ogist, we will advise to schedule an earlier appointment of perioperative myocardial injury, crucial for optimising and to inform doctors about their new condition. There treatment and improving prognosis in patients under- are no plans to disseminate the results of the research to going lung resection. study participants. Our study in the context of previous research Ethics considerations The present study stands out for being the first up to Ethics approval for this study, with protocol ID: date in which hs-­TnI and NT-­proBNP measurements are PR(AG)621/2020, was provided by the Ethics Research simultaneously extracted from a considerable sample of Committees of three participant hospitals: University patients undergoing thoracic surgery and are directly Hospital Vall d’Hebron, Barcelona, Spain (M Navarro-­ correlated with cardiovascular complications occurring Sebastian, MD, 15 January 2021), University Hospital in the immediate postoperative period. Santa Creu I Sant Pau, Barcelona, Spain (M Alonso-­ We believe that the combination of both cardiac Martinez, MD, 14 April 2021) and University Hospital biomarkers may improve the predictive value for cardio- Ramon y Cajal, Madrid, Spain (M Angeles Gálvez-­ vascular complications after lung resection as well as Múgica, MD, 14 June 2021). The study has been designed potentially improving the management and prognosis of according to the Declaration of Helsinki, as a statement these patients. of ethical principles for medical research. Furthermore, determining the diagnostic threshold Research personnel with good clinical practice will values of each of these cardiac biomarkers for risk stratifi- obtain written informed consent for each patient who cation may contribute not only to reducing the morbidity agrees to participate in the study. Only the researchers and mortality of patients undergoing non-­cardiac surgery 6 Alonso M, et al. BMJ Open 2022;12:e063778. doi:10.1136/bmjopen-2022-063778 Open access 7 but also in reducing their hospital stay and in conse- Department of Thoracic Surgery, Hospital de la Santa Creu i Sant Pau, Barcelona, quence the associated costs. Spain In addition, the multiple variables collected, such as Twitter María Alonso @maralonsos patients’ age, previous medical history or medication; as Acknowledgements We wish to acknowledge the research working group: well as the surgical course itself may prove useful in iden- Dina Butron Carmona, Daniela Garrido and Melissa Heifez for their contribution tifying predictive factors for postoperative complications. to our study. MA is a PhD candidate in Surgery and Morphological Sciences, Universitat Autònoma. Barcelona, Spain. EP is funded by a research contract (SLT017/20/000089) supported by the Department of Health of the Generalitat de Implications for practice and research Catalunya. The existence of a direct relationship between elevated Contributors MdN (principal investigator), MA and EP wrote the first draft of the NT-­proBNP and hs-­TnI and the appearance of cardiovas- protocol manuscript. MdN, MA, EP and MdM planned the conceptualisation and the cular complications could mean that their routine deter- design of the study, and the protocol. AM-­G, JP-­V, JCT, LG, AG-­T, EM-­T, EC-­G, GP, mination in high cardiovascular risk patients should be AdP, DP and AC-­T contributed to the design and implementation of the protocol. included in current protocols for the better management All authors provided critical revisions to the manuscript before approving the final version. MdN obtained funding for this work. of patients undergoing thoracic surgery. Funding This work was supported by a research grant number: ‘PI20/00154’ from the Instituto de Salud Carlos III (co-­funded by the European Regional Development Study’s strengths and limitations Fund ‘Una manera de hacer Europa’). Our study has several limitations. First, our protocol Competing interests None declared. is designed as a short-­ term (30-­ day follow-­up) study, Patient and public involvement Patients and/or the public were not involved in therefore long-­term studies for better understanding of the design, or conduct, or reporting, or dissemination plans of this research. efficiency of hs-­TnI and NT-­proBNP systematic measure- Patient consent for publication Consent obtained directly from patient(s). ments in thoracic surgery patients are still needed. Provenance and peer review Not commissioned; externally peer reviewed. Second, results of our study may be difficult to compare Data availability statement Any data required to support the protocol can be with previously published results, because of differences supplied upon reasonable request. The data used in the present study is part of a in methodology when using a hs-­TnI assay, together with larger data set. The data not used for this manuscript will be employed in future NT-­proBNP. Third, the use of hs-­TnI, would only be appli- manuscripts. Technical appendix, statistical code and data set available from the cable at centres using the same assay. Fourth, elevation of Dryad repository. hs-­TnI above the 99th percentile URL has become widely Supplemental material This content has been supplied by the author(s). It has used as an indicator for myocardial injury. Based on guide- not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-­reviewed. Any opinions or recommendations discussed are solely those line recommendations, we have chosen this definition to of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and evaluate the incidence and magnitude of hs-­cTnI eleva- responsibility arising from any reliance placed on the content. Where the content tion in patients after lung resection. However, further includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, studies are needed to establish prognostic cut-­off values terminology, drug names and drug dosages), and is not responsible for any error for major cardiovascular complications in these patients. and/or omissions arising from translation and adaptation or otherwise. Our study also has some strengths. First, to our knowl- Open access This is an open access article distributed in accordance with the edge, this is the first study integrating systematic measure- Creative Commons Attribution Non Commercial (CC BY-­NC 4.0) license, which ment of hs-­ TnI and NT-­ proBNP, preoperatively and permits others to distribute, remix, adapt, build upon this work non-­commercially, postoperatively, which is strongly recommended by recent and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use European Society of Cardiology (ESC)/European Society is non-­commercial. See: http://creativecommons.org/licenses/by-nc/4.0/. of Anaesthesiology and Intensive Care (ESAIC) guide- lines.7 Second, the hs-­TnI assay used in our study is more ORCID iDs María Alonso http://orcid.org/0000-0001-6053-4870 sensitive and accurate in detecting low concentrations, Ekaterine Popova http://orcid.org/0000-0002-8781-9873 contributing to an overdetection of MINS than would be detected by conventional cTn assays. Third, determining the clinical significance of hs-­TnI and NT-­proBNP eleva- tions (due to ischaemic or non-­ischaemic causes) may REFERENCES be crucial in the development of diagnostic and thera- 1 Thygesen K, Alpert JS, Jaffe AS, et al. Fourth universal definition of peutic protocols, aimed at improving MINS prognosis in myocardial infarction (2018). Circulation 2018;138:e618–51. 2 Ruetzler K, Khanna AK, Sessler DI. Myocardial injury after noncardiac patients undergoing thoracic surgery for lung resection. surgery: preoperative, intraoperative, and postoperative aspects, implications, and directions. Anesth Analg 2020;131:173–86. Author affiliations 3 Botto F, Alonso-­Coello P, Chan MTV, et al. Myocardial injury after 1 noncardiac surgery: a large, international, prospective cohort study Department of Anaesthesia and Intensive Care, Hospital Vall d'Hebron, Barcelona, establishing diagnostic criteria, characteristics, predictors, and 30-­ Spain day outcomes. Anesthesiology 2014;120:564–78. 2 Department of Surgery, Universitat Autònoma de Barcelona, Barcelona, Spain 4 González-­Tallada A, Borrell-­Vega J, Coronado C, et al. Myocardial 3 IIB SANT PAU, Institut d'Investigació Biomèdica Sant Pau, Barcelona, Spain injury after noncardiac surgery: incidence, predictive factors, and 4 Centro Cochrane Iberoamericano, Barcelona, Spain outcome in high-­risk patients undergoing thoracic surgery: an 5 Department of Anaesthesia and Intensive Care, Hospital Ramon y Cajal, Madrid, observational study. J Cardiothorac Vasc Anesth 2020;34:426–32. 5 Devereaux PJ, Biccard BM, et al, Writing Committee for the VISION Spain Study Investigators. Association of postoperative high-­sensitivity 6 Department of Thoracic Surgery and Lung Transplants, Hospital Vall d'Hebron, troponin levels with myocardial injury and 30-­day mortality among Barcelona, Spain patients undergoing noncardiac surgery. JAMA 2017;317:1642–51. Alonso M, et al. BMJ Open 2022;12:e063778. doi:10.1136/bmjopen-2022-063778 7 Open access 6 Puelacher C, Bollen Pinto B, Mills NL, et al. Expert consensus on 21 Wettersten N, Maisel AS. Biomarkers for heart failure: an update for peri-­operative myocardial injury screening in noncardiac surgery: a practitioners of internal medicine. Am J Med 2016;129:560–7. literature review. Eur J Anaesthesiol 2021;38:600–8. 1. 22 Zethelius B, Berglund L, Sundström J, et al. Use of multiple 7 Halvorsen S, Mehilli J, Cassese S. 2022 ESC guidelines on biomarkers to improve the prediction of death from cardiovascular cardiovascular assessment and management of patients undergoing causes. N Engl J Med 2008;358:2107–16. non-­cardiac surgery. Eur Heart J 2022. 23 Biccard BM, Devereaux PJ, Rodseth RN. Cardiac biomarkers in the 8 Keshava HB, Boffa DJ. Cardiovascular complications following prediction of risk in the non-­cardiac surgery setting. Anaesthesia thoracic surgery. Thorac Surg Clin 2015;25:371–92. 2014;69:484–93. 9 Vikenes K, Andersen KS, Farstad M, et al. Temporal pattern of 24 Duceppe E, Patel A, Chan MTV, et al. Preoperative N-­terminal cardiac troponin I after thoracotomy and lung surgery. Int J Cardiol pro-­B-­type natriuretic peptide and cardiovascular events 2004;96:403–7. after noncardiac surgery: a cohort study. Ann Intern Med 10 Lim E, Li Choy L, Flaks L, et al. Detected troponin elevation is 2020;172:96–104. associated with high early mortality after lung resection for cancer. 25 Tayama K, Takamori S, Mitsuoka M, et al. Natriuretic peptides after J Cardiothorac Surg 2006;1:37. pulmonary resection. Ann Thorac Surg 2002;73:1582–6. 11 Lucreziotti S, Conforti S, et al. Elevaciones de la troponina I 26 Hoksch B, Fahrner R, Alexander Schmid R. Procalcitonin and brain cardiaca tras la cirugía torácica. Incidencia y correlaciones con natriuretic peptide as parameters in the postoperative course of las características clínicas basales, la proteína C reactiva y los patients with major pulmonary resection. Interact Cardiovasc Thorac parámetros perioperatorios [Cardiac troponin-­I elevations after Surg 2007;6:155–9. thoracic surgery Incidence and correlations with baseline clinical 27 Cardinale D, Colombo A, Sandri MT, et al. Increased perioperative characteristics, C-­reactive protein and perioperative parameters]. N-­terminal pro-­B-­type natriuretic peptide levels predict atrial Rev Esp Cardiol 2007;60:1159–66. fibrillation after thoracic surgery for lung cancer. Circulation 12 Muley T, Kurz M, Männle C, et al. Comparison of serum cardiac 2007;115:1339–44. specific biomarker release after non-­cardiac thoracic surgery. Clin 28 Nojiri T, Maeda H, Takeuchi Y, et al. Predictive value of B-­type Lab 2011;57:925–32. natriuretic peptide for postoperative atrial fibrillation following 13 Hua A, Pattenden H, Leung M, et al. Early cardiology assessment pulmonary resection for lung cancer. Eur J Cardiothorac Surg and intervention reduces mortality following myocardial injury after 2010;37:787–91. non-­cardiac surgery (MINS). J Thorac Dis 2016;8:920–4. 29 Amar D, Zhang H, Shi W, et al. Brain natriuretic peptide and risk of 14 Puelacher C, Lurati Buse G, Seeberger D, et al. Perioperative atrial fibrillation after thoracic surgery. J Thorac Cardiovasc Surg myocardial injury after noncardiac surgery: incidence, mortality, and characterization. Circulation 2018;137:1221–32. 2012;144:1249–53. 15 Uchoa RB, Caramelli B. Troponin I as a mortality marker after lung 30 Simmers D, Potgieter D, Ryan L, et al. The use of preoperative resection surgery - a prospective cohort study. BMC Anesthesiol B-­type natriuretic peptide as a predictor of atrial fibrillation after 2020;20:118. thoracic surgery: systematic review and meta-­analysis. 16 Reichlin T, Hochholzer W, Bassetti S, et al. Early diagnosis of J Cardiothorac Vasc Anesth 2015;29:389–95. myocardial infarction with sensitive cardiac troponin assays. N Engl J 31 Nojiri T, Inoue M, Yamamoto K, et al. B-­Type natriuretic peptide as a Med 2009;361:858–67. predictor of postoperative cardiopulmonary complications in elderly 17 Cagini L, Capozzi R, Tassi V, et al. Fluid and electrolyte balance after patients undergoing pulmonary resection for lung cancer. Ann Thorac major thoracic surgery by bioimpedance and endocrine evaluation. Surg 2011;92:1051–5. Eur J Cardiothorac Surg 2011;40:e71–6. 32 Nojiri T, Inoue M, Shintani Y, et al. B-­Type natriuretic peptide-­guided 18 Cagini L, Andolfi M, Leli C, et al. B-­Type natriuretic peptide following risk assessment for postoperative complications in lung cancer thoracic surgery: a predictor of postoperative cardiopulmonary surgery. World J Surg 2015;39:1092–8. complications. Eur J Cardiothorac Surg 2014;46:e74–80. 33 Amar D, Zhang H, Tan KS, et al. A brain natriuretic peptide-­ 19 Potter LR, Abbey-­Hosch S, Dickey DM. Natriuretic peptides, their based prediction model for atrial fibrillation after thoracic surgery: receptors, and cyclic guanosine monophosphate-­dependent development and internal validation. J Thorac Cardiovasc Surg signaling functions. Endocr Rev 2006;27:47–72. 2019;157:2493–9. 20 Rørth R, Jhund PS, Yilmaz MB, et al. Comparison of BNP and NT-­ 34 Pascual-­Figal DA, Casademont J, Lobos JM, et al. Consensus proBNP in patients with heart failure and reduced ejection fraction. document and recommendations on the use of natriuretic peptides in Circ Heart Fail 2020;13:e006541. clinical practice. Rev Clin Esp 2016;216:313–22. 8 Alonso M, et al. BMJ Open 2022;12:e063778. doi:10.1136/bmjopen-2022-063778