Hrev_master Abstract Early recognition of myocardial infarction (MI) remains a challenge, especially in patients presenting with non-ST-segment elevation MI. Heart-type fatty acid binding protein (hFABP) has shown to be a more sensitive marker for myocardial necrosis as compared to non-high sensitive troponins (4th generation and older). However, since high sensitive troponin (hs TnT) assays are available, it is questionable whether hFABP still has value as a diagnostic tool for MI. A systematic search was conducted in Medline, Embase and Cochrane. After selecting the articles, they were assessed for risk of bias according to the QUADAS-2 criteria. Negative predictive value, positive predictive value, sensitivity and specificity were extracted or calculated if possible. Nine stud- ies met the inclusion criteria. Overall, hs TnT showed higher sen- sitivity than hFABP, while hFABP had higher specificity. In patients presenting within 3 hours after onset of symptoms, sensi- tivity is low for both hFABP and hs TnT (19-63% vs 25-55%, respectively), while specificity seems higher for hFABP than for hs TnT (70-99% vs 57-86%, respectively). In these patients, the area under the curve for hs TnT is equal or better than that for hFABP (0.67-0.92 for hs TnT vs 0.69-0.85 for hFABP). The addition of hFABP to hs TnT did not improve sensitivity and specificity. This systematic review finds no advantage for using hFABP over hs TnT in either early presenting patients or overall. Furthermore, no advantage was found if a combination of hFABP and hs TnT was used for the diagnosis of MI. Introduction It is important to diagnose myocardial infarction (MI) as soon as possible, to improve the prognosis by timely intervention and reduce the time in the emergency department. However, early diagnosis of MI remains a challenge, especially in patients with non-ST-segment elevation MI (NSTEMI) or in patients presenting early after onset of symptoms. According to the European Society of Cardiology guidelines, MI is diagnosed by an increase of a cardiac biomarker, with at least one value above the 99th percentile of the upper reference limit, in combination with signs of ischemia. These can either be clinical symptoms, ECG changes or abnormalities seen on imaging.1 High sensitive cardiac troponin is the preferable cardiac biomarker to diagnose MI, because it has a higher sensitivity and specificity than other laboratory tests, like creatine kinase (CK), and its MB isoenzyme (CK-MB).2 Although diagnostic sensitivity in the first hours after symptom onset has improved since the introduction of high sensitive assays, guidelines still recommend to redo the test after 1 or 3 hours if a negative result is found for patients present- ing within 6 hours after onset of symptoms.1 However, most patients present within 2-3 hours of symptom onset.3 Heart-type fatty acid binding protein (hFABP) has been pro- posed as an addition to or replacement for CK-MB and cardiac tro- ponin, because of a higher sensitivity for myocardial damage in patients presenting early after symptom onset compared to non- high sensitive troponins.4 Many of the previously published reviews did not look specifically at hFABP in comparison to high- sensitive troponins. The aim of this systematic review is to assess whether hFABP can be a replacement or addition to high-sensitive troponin T (hs TnT) for the diagnosis of MI in general practice, and more specifically in patients presenting within 3 hours after symp- tom onset. Both hs TnT and high sensitive troponin I assays are available. Due to the heterogeneity between the different troponin I assays, we have chosen to compare hFABP to hs TnT only. Emergency Care Journal 2019; volume 15:8049 Correspondence: Daniel M.F. Claassens, Department of Cardiology, St. Antonius Hospital, Koekoekslaan 1, 3435CM Nieuwegein, The Netherlands. Tel.: +31.0.88.3201248 - Fax: +31.0.88.3207069. E-mail: d.claassens@antoniusziekenhuis.nl Key words: Heart-type fatty acid binding protein; hFABP; High-sensi- tive troponin T; Acute coronary syndrome; Myocardial infarction. Contributions: DMFC, DOC, literature search, collecting data, analyz- ing and manuscript writing; TOB literature search, manuscript writing; DRPPCY, JMB manuscript writing. Conflict of interest: Dr. Ten Berg reports grants and personal fees from AstraZeneca, Eli Lilly, Daiichi Sankyo, The Medicines Company, BMS, Pfizer, Bayer, Boehringer-Ingelheim, outside the submitted work. All other authors declare no potential conflict of interest. Funding: none. Received for publication: 23 January 2019. Accepted for publication: 13 March 2019. This work is licensed under a Creative Commons Attribution 4.0 License (by-nc 4.0). ©Copyright: the Author(s), 2019 Licensee PAGEPress, Italy Emergency Care Journal 2019; 15:8049 doi:10.4081/ecj.2019.8049 [Emergency Care Journal 2019; 15:8049] [page 51] Is there a place for heart-type fatty acid binding protein in the era of high-sensitive cardiac troponin T for the diagnosis of acute myocardial infarction? A systematic review Daniel M.F. Claassens,1 Djura O. Coers,2 Thomas O. Bergmeijer,1 Dean R.P.P. Chan Pin Yin,1 Jurriën M. Ten Berg1 1Department of Cardiology, St. Antonius Hospital, Nieuwegein; 2Department of Primary Care and Elderly Care, Amsterdam Public Health Institute, Amsterdam UMC, The Netherlands Non -co mmerc ial us e o nly Methods of research Search strategy A systematic literature search was conducted using Medline, Embase and Cochrane databases of all literature until the 19th of March 2018. The search was constructed by combining high sen- sitive troponin t and heart type fatty acid binding protein and myocardial infarction (see the Appendix for full search). After removing duplicates, all publications were independently screened by the first two authors on title and abstract, using predetermined in- and exclusion criteria. The inclusion criteria were: diagnostic cohort study; inclusion of patients ≥18 years suspected of an acute coronary syndrome (ACS); testing of both hFABP and hs TnT prior to the diagnosis; cut-off value for hs TnT>14ng/L; and NSTEMI or a combined outcome of NSTEMI and STEMI as out- come variable. We excluded studies in which ACS, including unstable angina pectoris, was used as the primary outcome vari- able, studies which were not available in English or Dutch, were performed in the primary care setting, were non-human studies or if no full text version was available. We did not discriminate between studies using lab tests or point-of-care tests for determin- ing hFABP or hs TnT. The remaining publications were screened for full text. The references of the included publications were screened in Web of Science and Scopus to check for additional relevant publications. Any disagreement was resolved by discussion between the first three authors. Assessment for risk of bias All included publications were independently screened by the first three authors (DMFC, DOC, TOB) for the risk of bias, based on predefined criteria. These criteria are a modified version of the QUADAS-2 criteria for diagnostic research.5 We assessed whether there was more than 10% missing data, if there was standardization for how both the determinant and the outcome parameter were determined, if the blood samples for hFABP and hs TnT were col- lected at the same time, if there was a pre-specified cut-off value for hFABP and if the outcome assessor was blinded for the results of the hFABP test. Furthermore, we assessed whether there was a risk of bias in the inclusion process. Exclusion of STEMI patients was allowed, since it is not standard practice to test for cardiac tro- ponin in these patients. Any disagreement was resolved by discus- sion between the first two authors and the third author. For each item, one point was awarded if specific criteria were met. Studies scoring six or seven points were assessed as having a low risk of bias, studies scoring four or five points were assessed as having a medium risk of bias and studies scoring less than four points were assessed as having a high risk of bias. Data extraction and analysis Data regarding prior probability, positive and negative predic- tive value (PPV and NPV), sensitivity and specificity and their 95% confidence intervals were extracted from the full text article. If these parameters were not presented, they were calculated using the available data using the statistical program ‘R’ (CRAN project) or the area under the curve (AUC) was extracted from the article if this was available. We choose not to perform a meta-analysis, due to differences in cut-off values and outcome parameters in the selected studies. We also analysed the combination of hFABP and hs TnT and compared hFABP with hs TnT in patients presenting within 3 hours after onset of symptoms. Results Search strategy Our search yielded 232 unique publications, which were sub- sequently screened on inclusion and exclusion criteria, using the title and abstract. Of these, 35 publications were screened on full text and finally 9 publications were adjudicated for risk of bias (Figure 1). An overview of the included publications is presented in Table 1.6-14 Risk of bias assessment Table 2 shows the risk of bias assessment.6-14 All publications were either prospective or retrospective cohort studies. One of the studies was adjudicated as having a high risk of bias, one of the studies was adjudicated as having a medium risk of bias and all other studies as having a low risk of bias. Review Figure 1. Literature search results. Hs TnT, high-sensitive tro- ponin T; hFABP, heart-type fatty acid binding protein; ACS, acute coronary syndrome. [page 52] [Emergency Care Journal 2019; 15:8049] Non -co mmerc ial us e o nly [Emergency Care Journal 2019; 15:8049] [page 53] Data analysis Results of the studies are shown in Table 3.6-12,14 The last col- umn specifies whether NSTEMI or MI was used as the outcome parameter. In two of the studies the NPV was significantly higher for hs TnT as compared to hFABP, while there were no significant differences in the PPV. Specificity of hFABP was significantly bet- ter in two studies, while the sensitivity of hs TnT was significantly better in one study. Overall, NPV and sensitivity seemed to be higher for hs TnT, while PPV and specifity seemed to be higher for hFABP. Table 4 shows the results from the different studies on early presenters.9,11-13 Only the study from Schoenenberger et al. focused specifically on early presenting patients.13 Studies reporting an AUC have equal results or showed an advantage for hs TnT over hFABP. Studies reporting a PPV showed an advantage for hFABP. However, no statistical significance was either found or reported in these studies. No difference in NPV was found. Review Table 1. Overview of included studies. Authors Year of Study Number of Male (%) Age Time from hFABP Outcome Prevalence publication design patients (N) symptom cut-off parameter of outcome onset to value (%) presentation (µg/L) (hours) Cappellini et al.6 2013 Retrospective cohort 67 67 74 (mean) NA 3.49 NSTEMI 33 Eggers et al.7 2012 Retrospective cohort 360 66 67 (mean) <8 5.8 NSTEMI 36 Gami et al.8 2015 Prospective cohort 88 NA NA <6 5.09 MI 39 Inoue et al.9 2011 Retrospective cohort 432 73 67 (median) <24 6.2 NSTEMI 9 Kellens et al.*10 2016 Prospective cohort 203 81 63 (NA) <24 5.3 MI 63 Kitamura et al.11 2013 Prospective cohort 85 78 67 (median) <24 6.2 MI 55 Reiter et al.12 2013 Prospective cohort 1037 67 64 (median) <12 4.2 NSTEMI 16 Schoenenberger et al.13 2016 Prospective cohort 105 70 61 (mean) <1 4.0 NSTEMI 32 Willemsen et al.14 2015 Prospective cohort 202 78 62 (median) <24 4.0 MI 17 hFABP, heart-type fatty acid binding protein; NA, not available; NSTEMI, non-ST-elevated myocardial infarction; MI, myocardial infarction. *Point-of-care test used for heart-type fatty acid binding protein. Table 2. Risk of bias analysis. Authors Missing Patient Standardization of Pre-specified Blood sample Outcome Blinding for Risk of data selection determinant cut-off value collection standardization hFABP bias Cappellini et al.6 + - + + + + + Low Eggers et al.7 - + + + + + + Low Gami et al.8 NA - + - + + NA High Inoue et al.9 NA - + + + + + Medium Kellens et al.*10 + + + + + + NA Low Kitamura et al.11 + - + + + + + Low Reiter et al.12 + - + + + + + Low Schoenenberger et al.13 + - + + + + + Low Willemsen et al.14 + + + - + + + Low Legend Missing data + ≤10% missing data - >10% missing data Patient selection + No exclusion or only STEMI patients excluded - Other patients excluded Standardization of determinant + Standardized protocol to test for hs TnT and hFABP - No standardized protocol to test for hs TnT and hFABP Pre-specified cut-off value + pre-specified cut-off value for hFABP - no pre-specified cut-off value for hFABP Blood sample collection + Blood sample collected at the same time - Blood sample not collected at the same time Outcome standardization + Standardized method to determine outcome - No standardized method to determine outcome Blinding for hFABP +Outcome assessor blinded for hFABP result - Outcome assessor not blinded for hFABP result Low risk of bias: 6 or 7 +, medium risk of bias 4 or 5 +, high risk of bias <4 +. hFABP, heart-type fatty acid binding protein; hs TnT, high-sensitive troponin T; NA, not available; STEMI, ST-elevated myocardial infarction. *Point-of-care test used for heart-type fatty acid binding protein. Non -co mmerc ial us e o nly [page 54] [Emergency Care Journal 2019; 15:8049] Table 5 shows the results when hs TnT and hFABP were com- bined to diagnose MI.7,8,10,12,13 Compared to hs TnT alone, the stud- ies from Eggers et al. and Schoenenberger et al. found no signifi- cant differences, whilst the study from Reiter et al. found a better AUC for hs TnT alone. Gami et al. found a higher PPV and speci- ficity whilst Kellens et al. found a slightly higher sensitivity and NPV.8,10 Discussion and Conclusions The aim of this systematic review was to study the value of hFABP as a marker for myocardial necrosis, alone or added to hs TnT. hFABP is known to rise very early after myocardial necrosis occurs (as soon as 30 minutes after onset of symptoms) and peaks in the blood after approximately 6-8 hours.15 hFABP was reported to be more specific to myocardial tissue compared to other bio- markers, like myoglobin and CK-MB.16 Compared to non-high sensitive cardiac troponins, hFABP has shown promising results for the diagnosis of MI, especially early after symptom onset.4 Multiple studies report higher sensitivity for the combination of troponin and hFABP compared to troponin alone.17-19 However, these studies and most published reviews have compared hFABP with non-high sensitive troponins instead of the (more accurate and now widely used) hs TnT.20 Though many of the studies in this review used a combined outcome of both NTEMI and STEMI, the latter group is not rou- tinely tested, since the ST-segment elevation on the ECG usually is enough to confirm the diagnosis of MI. In patients without these clear ECG changes, the ideal test for myocardial damage combines Review Table 3. Heart-type fatty acid binding protein vs high-sensitive troponin T in patients suspected of myocardial infarction. Authors NPV (%) PPV(%) Specificity (%) Sensitivity (%) Outcome (95%CI) (95%CI) (95%CI) (95%CI) parameter hFABP hs TnT hFABP hs TnT hFABP hs TnT hFABP hs TnT Cappellini et al.6 100 (76-100) 88 (69-97) 44 (30-60) 49 (32-66) 39 (25-55) 56 (40-71) 100 (80-100) 85 (61-96) NSTEMI Eggers et al.7 74 (69-79) 87a (81-90) 81 (69-90) 63 (55-71) 95 (91-97) 75 (69-80) 39 (31-48) 79a (71-86) NSTEMI Gami et al.8 90 (79-97) 94 (81-99) 82 (66-93) 61 (47-75) 88a (77-96) 62 (49-76) 85 (69-95) 94 (80-99) MI Inoue et al.9 77 86 60 54 78 61 59 83 NSTEMI Kellens et al.*10 52 61 85 82 84 73 54 72 MI Kitamura et al.11 69 (52-81) 69 (52-83) 91 (76-98) 76 (61-87) 92 (79-98) 71 (54-85) 66 (51-79) 74 (60-86) MI Reiter et al.12 94 (92-95) 98a (97-99) 41 (35-47) 42 (37-47) 80 (77-82) 77 (74-80) 72 (65-89) 93 (88-96) NSTEMI Willemsen et al.14 94 93 36 34 71 71 78 73 MI NPV, negative predictive value; CI, confidence interval; PPV, positive predictive value; hFABP, heart-type fatty acid binding protein; hs TnT, high-sensitive troponin T; NSTEMI, non-ST elevated myocardial infarction; MI, myocardial infarction. *Point-of-care test used for heart-type fatty acid binding protein. aSignificant difference between hFABP and hs TnT. Table 4. Heart-type fatty acid binding protein vs high-sensitive troponin T in early presenters (<3 hours since symptom onset). Authors Time to NPV (%) PPV (%) Specificity (%) Sensitivity (%) AUC presentation (95%CI) (95%CI) (95%CI) (95%CI) (95%CI) (hours) hFABP hs TnT hFABP hs TnT hFABP hs TnT hFABP hs TnT hFABP hs TnT Inoue et al.9 3 NA NA NA NA NA NA NA NA 0.69 (0.58-0.81) 0.67 (0.56-0.78) Kitamura et al.11 2 57 (34-77) 40 (19-64) 86 (42-100) 40 (12-74) 93 (66-100) 57 (29-82) 38 (15-65) 25 (7-52) NA NA Reiter et al.12 3 NA NA NA NA NA NA NA NA 0.85 (0.82-88) 0.92a (0.89-0.94) Schoenenberger et al.13 1 AUC: 2 90 91 89 50 99 86 50 63 0.81 (0.75-0.87) 0.90 (0.86-0.94) NPV, negative predictive value; CI, confidence interval; PPV, positive predictive value; AUC, area under the curve; hFABP, heart-type fatty acid binding protein; hs TnT, high-sensitive troponin T. aSignificant difference between hFABP and hs TnT. Table 5. Combination of heart-type fatty acid binding protein and high-sensitive troponin T for the diagnosis of myocardial infarction. Authors NPV (%) PPV (%) Specificity (%) Sensitivity (%) AUC (95%CI) (95%CI) (95%CI) (95%CI) (95%CI) hs TnT hs TnT and hFABP Eggers et al.7 87 (82-91) 63 (55-70) 75 (69-80) 80 (72-86) NA NA Gami et al.8 100 85 89 100 NA NA Kellens et al.*10 70 83 71 82 NA NA Reiter et al.12 NA NA NA NA 0.94a (0.92-0.95) 0.88 (0.86-0.90) Schoenenberger et al.13 NA NA NA NA 0.85 (0.76-0.95) 0.92 (0.84-0.99) NPV, negative predictive value; CI, confidence interval; PPV, positive predictive value; AUC, area under the curve; hs TnT, high-sensitive troponin T; hFABP, heart-type fatty acid binding protein. *Point-of-care test used for heart-type fatty acid binding protein. aSignificant difference. Non -co mmerc ial us e o nly [Emergency Care Journal 2019; 15:8049] [page 55] high sensitivity with high specificity to quickly discriminate between patients who can be send home safely and patients that should be admitted and treated for MI. Our review shows that the sensitivity of hs TnT was found to be higher in six of the eight studies.7-12 However, only in two of studies the difference was sig- nificant. In the studies from Cappellini et al. and Willemsen et al., sensitivity of hFABP was found to be higher than that of hs TnT (100% vs 85% and 78% vs 73% respectively).6,14 This higher sen- sitivity can be explained by the low cut-off value that was used (3.49µg/L and 4.0µg/L respectively). Earlier studies have shown that the median of hFABP for patients without MI was 3.9µg/L (1.7-7.9 (interquartile range)).4 Consequently, this high sensitivity comes with a low specificity, especially in the study from Cappellini et al., of only 39%.6 This is contrary to the other studies, in which a higher specificity for hFABP is found.7-12 Since the risk of death is almost twofold when the diagnosis of MI is missed, high sensitivity is thought to be of more importance than specifici- ty in this patient group.21 In patients presenting early after symptom onset, the results of the two biomarkers are comparable. Although hFABP seems to be a valuable tool for early diagnosis, its sensitivity, even in the early hours, is not better than that of hs TnT. It must be noted that there is a wide variation in both sensitivity and specificity between dif- ferent studies. However, with a sensitivity of just 50-60%, hFABP cannot replace hs TnT assays as a standalone test for the diagnosis of a MI. This finding is in line with a systematic review by Bruins Slot et al. in 2010, comparing hFABP with other biomarkers like CK, CK-MB and non-high sensitive troponins, although in this review almost half of the included studies used POC tests and/or different definitions for MI.22 The three studies that reported sensitivity and specificity for the combination of high sensitive troponin T assays and hFABP all show similar or increased specificity for the combined tests.7,8,10 This is in contrast with our expectations, since in different reviews sensitivity increases when combined tests are performed, at the cost of a loss in specificity.17,23 For example, in a recently pub- lished article by Young et al., looking specifically at combining ECG with hs TnT and hFABP for the diagnosis of acute MI, sensi- tivity increased from 94.8% (ECG or hs TnT) to 97.2% (ECG or hs TnT or hFABP), while specificity decreased from 69.6% to 43.6%.24 The studies in our review provided no clear explanation for this discrepancy. The comparison of hFABP to hs TnT is hampered by a major methodological problem, because hs TnT are used as the gold stan- dard for the diagnosis of MI.1 Most of the studies we included for our review tried to solve this problem by using multiple independ- ent physicians and all medical data available (including ECGs, lab results and coronary angiograms) to confirm the diagnosis of MI. However, it will be very difficult for any cardiac biomarker to out- perform hs TnT. With this limitation in mind, we found no con- vincing evidence that hFABP, alone or in combination with hs TnT, has a better diagnostic performance compared to hs TnT, if hs TnT is easily available. An important advantage for hFABP is the availability as point- of-care (POC) test, which can be used when classic laboratory tests like hs TnT are not available, for instance in the pre-hospital set- ting.10,14,25-27 However, in a review by Bruins Slot et al. published in 2013, NPV was too low to safely exclude MI diagnosis when using the hFABP POC test, although many studies were of poor quality.25 Also, a recently published study by Bank et al. found that the used POC test was inferior to the hs TnT test.26 However, newer POC tests have been introduced since, for instance from FABPulous BV (Maastricht, the Netherlands), which was used in the studies by Kellens et al. included in our review. Furthermore, POC TnT tests are being developed and first results of the newer generation tests have recently been published.28,29 Even though the limit of detection in the newer generation POC tests is less than half of that of the older generation, it is still more than tenfold higher than that of the high sensitive troponins (40ng/L and 3ng/L, respectively).29 Further research is necessary to study the value of POC tests in the pre-hospital setting. Concluding, no convincing evidence was found to support the use of hFABP instead of, or in combination with hs TnT for the diagnosis of myocardial infarction, neither in early presenting patients nor in patients presenting after more than 3 hours after onset of symptoms. Limitations There are a few limitations to this review to mention. First, dif- ferent cut-off levels were used for hFABP, which has major influ- ence on NPV, PPV, specificity and sensitivity, leading to difficulty in comparing the different studies. Second, there was a large vari- ation in the prevalence of NSTEMI and MI between the different studies (9-63%). A higher prevalence leads to a higher PPV and a lower NPV and a different population does not only effect the prevalence, but it can also affect the sensitivity and specificity of a diagnostic test.30,31 This variation can therefore have a major influ- ence on the results. Last, only one study specifically investigated patients presenting early after onset of symptoms. In all other stud- ies this was only a sub-group analysis. References 1. Roffi M, Patrono C, Collet JP, et al. 2015 ESC Guidelines for the management of acute coronary syndromes in patients pre- senting without persistent ST-segment elevation. Eur Heart J 2016;37:267-315. 2. Thygesen K, Mair J, Giannitsis E, et al. How to use high-sen- sitivity cardiac troponins in acute cardiac care. Eur Heart J 2012;33:2252-7. 3. Goodacre S, Bradburn M, Cross E, et al. The RATPAC (ran- domised assessment of treatment using panel assay of cardiac markers) trial: a randomised controlled trial of point-of-care cardiac markers in the emergency department. Health Technol Assess 2011;15:1-102. 4. McCann CJ, Glover BM, Menown IB, et al. Novel biomarkers in early diagnosis of acute myocardial infarction compared with cardiac troponin T. Eur Heart J 2008;29:2843-50. 5. Whiting PF, Rutjes AWS, Westwood ME, et al. QUADAS-2: A revised tool for the quality assessment of diagnostic accuracy studies. Ann Intern Med 2011;155:529-36. 6. Cappellini F, Da Molin S, Signorini S, et al. Heart-type fatty acid-binding protein may exclude acute myocardial infarction on admission to emergency department for chest pain. Acute Card Care 2013;15:83-7. 7. Eggers KM, Venge P, Lindahl B. High-sensitive cardiac tro- ponin T outperforms novel diagnostic biomarkers in patients with acute chest pain. Clin Chim Acta 2012;413:1135-40. 8. Gami BN, Patel DS, Haridas N, et al. Utility of heart-type fatty acid binding protein as a new biochemical marker for the early diagnosis of acute coronary syndrome. J Clin Diagn Res 2015;9:BC22-4. 9. Inoue K, Suwa S, Ohta H, et al. Heart fatty acid-binding pro- tein offers similar diagnostic performance to high-sensitive Review Non -co mmerc ial us e o nly [page 56] [Emergency Care Journal 2019; 15:8049] troponin T in emergency room patients presenting with chest pain. Circ J 2011;75:2813-20. 10. Kellens S, Verbrugge FH, Vanmechelen M, et al. Point-of-care heart-type fatty acid binding protein versus high-sensitivity troponin T testing in emergency patients at high risk for acute coronary syndrome. Eur Heart J Acute Cardiovasc Care 2016;5:177-84. 11. Kitamura M, Hata N, Takayama T, et al. High-sensitivity car- diac troponin T for earlier diagnosis of acute myocardial infarction in patients with initially negative troponin T test- comparison between cardiac markers. J Cardiol 2013;62:336- 42. 12. Reiter M, Twerenbold R, Reichlin T, et al. Heart-type fatty acid-binding protein in the early diagnosis of acute myocardial infarction. Heart 2013;99:708-14. 13. Schoenenberger AW, Stallone F, Walz B, et al. Incremental value of heart-type fatty acid-binding protein in suspected acute myocardial infarction early after symptom onset. Eur Heart J Acute Cardiovasc Care 2016;5:185-92. 14. Willemsen RT, Van Severen E, Vandervoort PM, et al. Heart- type fatty acid binding protein (H-FABP) in patients in an emergency department setting, suspected of acute coronary syndrome: optimal cut-off point, diagnostic value and future opportunities in primary care. Eur J Gen Pract 2015;21:156-63. 15. Kleine AH, Glatz JF, Van Nieuwenhoven GJ, et al. Release of heart fatty acid-binding protein into plasma after acute myocardial infarction. Mol Cell Biochem 1992;116:155-62. 16. Glatz JF, Van der Vusse GJ, Simoons ML, et al. Fatty acid- binding protein and the early detection of acute myocardial infarction. Clin Chim Acta 1998;272:87-92. 17. Lippi G, Mattiuzzi C, Cervellin G. Critical review and meta- analysis of the combination of heart-type fatty acid binding protein (H-FABP) and troponin for early diagnosis of acute myocardial infarction. Clin Biochem 2013;46:26-30. 18. McMahon CG, Lamont JV, Curtin E, et al. Diagnostic accura- cy of H-FABP for the early diagnosis of AMI. Am J Emerg Med 2012;30:267-74. 19. Body R, McDowell G, Carley S, et al. A FABP-ulous ‘rule out’ strategy? Heart fatty acid binding protein and troponin for rapid exclusion of AMI. Resuscitation 2011;82:1041-6. 20. Roche Diagnostics. Troponin T hs (package insert). Mannheim: Roche Diagnostics GmbH; 2013. 21. Pope HJ, Aufderheide TP, Ruthazer R, et al. Missed diagnoses of acute cardiac ischemia in the emergency department. N Engl J Med 2000;342:1163-70. 22. Bruins Slot MH, Reitsma JB, Rutten FH, et al. Heart-type fatty acid-binding protein in the early diagnosis of acute myocardial infarction: a systematic review and meta-analysis. Heart 2010;96:1957-63. 23. Carroll C, Al Khalaf M, Stevens JW, et al. Heart-type fatty acid binding protein as an early marker for myocardial infarction: systematic review and meta-analysis. Emerg Med J 2013;30:280-6. 24. Young JM, Pickering JW, George PM, et al. Heart fatty acid binding protein and cardiac troponin: development of an opti- mal rule-out strategy for acute myocardial infarction. BMC Emerg Med 2016;16:34. 25. Bruins Slot MH, Van der Heijden GJ, Stelpstra SD, et al. Point- of-care tests in suspected acute myocardial infarction: a sys- tematic review. Int J Cardiol 2013;168:5355-62. 26. Bank IE, Dekkers MS, Hoes AW, et al. Suspected acute coro- nary syndrome in the emergency room: Limited added value of heart type fatty acid binding protein point of care or ELISA tests: The FAME-ER (Fatty Acid binding protein in Myocardial infarction Evaluation in the Emergency Room) study. Eur Heart J Acute Cardiovasc Care 2016;5:364-74. 27. Gerede DM, Güleç S, Kiliçkap M, et al. Comparison of quali- tative measurement of heart-type fatty acid binding protein with other cardiac biomarkers as an early diagnostic marker in the diagnosis of non-ST segment elevation myocardial infarc- tion. Cardiovasc J Afr 2015;26:204-9. 28. Slagman A, Von Recum J, Möckel M, et al. Diagnostic per- formance of a high-sensitive troponin T assay and a troponin T point of care assay in the clinical routine of an emergency department: a clinical cohort study. Int J Cardiol 2017;230:454-60. 29. Jungbauer C, Hupf J, Giannitsis E, et al. Analytical and clinical validation of a point-of-care cardiac troponin T test with an improved detection limit. Clin Lab 2017;63:633-45. 30. Leeflang MM, Rutjes AW, Reitsma JB et al. Variation of a test’s sensitivity and specificity with disease prevalence. CMAJ 2013;185:E537-44. 31. Knottnerus JA, Leffers P. The influence of referral patterns on the characteristics of diagnostic tests. J Clin Epidemiol 1992;45:1143-54. Review Non -co mmerc ial us e o nly