Cardiovascular Revascularization Medicine, 2018-07-01, Volume 19, Issue 5, Pages 487-492, Copyright © 2017 Elsevier Inc. Abstract Background The introduction of the highly sensitive troponin (hs-trop) assays into clinical practice has allowed for the more rapid diagnosis or exclusion of type 1 myocardial infarctions (T1MI) by clinicians, in addition type 2 myocardial infarctions (T2MI) are now more frequently detected. Tachyarrhythmias are one of the common causes of T2MI, the medium and long term outcome for this cohort of T2MI is yet to be clarified. Methods Retrospective review of consecutive patients admitted with a diagnosis of either (a) non ST-elevation myocardial infarction (NSTEMI) or (b) tachyarrhythmia was performed. Data were collected on patient demographics and investigations. Patient mortality status was recorded through the Personal Demographics Service (PDS) via NHS Digital. Results A total of 704 patients were eligible for inclusion to the study. 264 patients were included in the study with a final discharge diagnosis of NSTEMI and 440 patients with a final discharge diagnosis of tachyarrhythmia. There was a significantly higher peak troponin in NSTEMI patients compared to the tachyarrhythmia troponin positive group (4552 ng/L vs 571 ng/L, p < 0.001). Mortality was significantly higher in the troponin positive tachyarrhythmia patients than the troponin negative patients (54 vs 34, 26.2% vs 14.5%, log rank p = 0.003), furthermore, the mortality of NSTEMI and troponin positive tachyarrhythmia patients was similar (55 vs 54, 20.8% vs 26.2%, log rank p = 0.416). Only one patient (0.14%) was given a formal diagnosis of T2MI. Conclusions These data suggest that troponin positive tachyarrhythmia is not a benign diagnosis, and has a mortality rate similar to NSTEMI. Formal labeling as T2MI is rare in real life practice. More investigation into the detection and management of T2MI and troponin positive arrhythmia patients is now warranted. Highlights The optimal management strategy for T2MI is unknown. Troponin positive tachyarrhythmia is not a benign diagnosis. The mortality rate of troponin positive tachyarrhythmia is similar to NSTEMI. Formal labeling as T2MI is rare in real life practice. More investigation into the detection and management of T2MI and troponin positive arrhythmia patients is now warranted. 1 Background Early cTn assays were introduced into clinical practice from 1995 onwards and aided clinicians in the diagnosis of acute MI. The value of the early assays to confidently exclude a diagnosis of MI was limited by their lack of sensitivity meaning that a reliable outcome could only be given 10–12 h after the onset of symptoms. Consequently, there has been a drive to develop more sensitive troponin assays that will facilitate earlier exclusion of MI. The new highly sensitive assays are able to detect troponin at much lower concentrations than the previous generation . The universal definition of MI recommends that a troponin assay used to diagnose MI should have a coefficient of variation of ≤ 10% at the threshold concentration representing the 99th percentile upper limit of a normal reference population (ULN) . However, these hs-trop assays detect troponin in > 50% of the general population with some assays able to detect troponin in everyone . Given the number of reasons why hs-trop is detectable apart from Type 1 MI (T1MI), the interpretation of assays results in patients is therefore more challenging than before. According to the third universal definition the classical form or T1MI, is defined as a troponin elevation related to an acute coronary plaque rupture . Type 2 MI (T2MI) is defined as myocardial ischaemia resulting from increased oxygen demand or reduced oxygen supply, most often due to sepsis, anaemia, tachyarrhythmia, hypertension, hypotension, coronary artery spasm and embolism . The evidence for antiplatelet therapy and early revascularisation in the setting of a T1MI is robust and compelling. By contrast, further non-invasive or invasive investigations and management of this type have failed to demonstrate a clinical outcome benefit in T2MI . Furthermore, patients with T2MI have been shown to have a 2-fold higher mortality rate than those with T1MI. With this background, a common dilemma for clinicians is the optimal management of patients presenting with primary tachyarrhythmia and troponin elevation. A simplistic view would be to define troponin rises associated with a primary diagnosis of tachyarrhythmia as evidence of ACS and therefore manage these patients as such. However, many of these patients have not suffered a T1MI and may therefore undergo unnecessary invasive assessment and inappropriate revascularization and pharmacotherapy. The advent of the contemporary hs-trop assay makes this diagnostic dilemma more prevalent. It is well described that troponin levels are often elevated in association with tachyarrhythmias such as atrial fibrillation (AF), supraventricular tachycardia (SVT) and ventricular tachycardia (VT) and, furthermore that many of these patients have unobstructed coronary arteries. However, studies have previously shown that in patients with AF and elevated troponin levels, AF is a risk marker for cardiovascular morbidity and mortality . The selective contribution of the tachyarrhythmia itself, and any concomitant coronary disease, in the interpretation of the troponin result and subsequent management is unclear. The aim of this study was to assess the tracked mortality of a consecutive series of patients presenting with a primary diagnosis of tachyarrhythmia according to whether they were troponin positive or negative. This mortality would be compared to a simultaneous cohort with a primary diagnosis of T1MI. The second aim was to describe how often the diagnosis of T2MI is formally made in routine cardiology practice. 2 Methods 2.1 Patient selection We conducted a retrospective search of patients who were admitted to the University Hospitals Southampton, an 1100 bed general hospital, between April 2014 and December 2015 with a primary diagnosis of AF, SVT, VT or non ST-elevated myocardial infarction (NSTEMI). This was achieved by searching the trust database for patients who had been discharged with a primary diagnosis with the following International Classification of Diseases, version 10 (ICD-10) codes I48(AF), I47.1(SVT), I47.2(VT) and I21.4(NSTEMI) for the study period. This is a coding system predominantly used for reimbursement purposes. These primary diagnoses were exclusive: so that a judgement has been made by the supervising clinical team that this was the dominant cause for hospital admission. Patients were deemed eligible for inclusion into the study according to the following criteria: (a) ≥ 18 years of age, (b) UK residents, (c) had a hs-trop level measured (Beckman Coulter Access AccuTnI + 3 HsTrop assay), and (d) had a final primary diagnosis of AF, SVT, VT or NSTEMI. Patients in the AF, SVT or VT group were excluded if their heart rate on admission was < 100 beats per minute. Patients presenting with a STEMI were also excluded. A total of 2404 patients were identified as having a final primary diagnosis of either AF, SVT, VT or NSTEMI, this final primary diagnosis was made by the responsible clinician for the patient's hospital stay. From this group a total 705 patients were identified as eligible for inclusion to the study after a review of their clinical records. A further patient was excluded due to miscoding of their final diagnosis. The remaining 704 patients were stratified into four primary diagnoses; NSTEMI ( n = 264), AF ( n = 344), SVT ( n = 40), VT ( n = 56) ( Fig. 1 ). No attempt to reinterpret or overrule the original diagnosis was made by the researcher team. Fig. 1 Flow diagram of patients. 2.2 Data collection Data on the patients included in the study were collected through a review of both electronic and paper patient records using a bespoke data collection template, designed by the researchers in Microsoft Excel (version 15.26). Patient demographics, presenting symptoms, traditional risk factors for CAD, heart rate on admission, peak hs-trop reading, left ventricular systolic function, investigations and procedures carried out, and discharge medication. The mortality status of each patient was obtained via NHS Digital through the Personal Demographics Service (PDS) following local ethical and national Confidentiality Advisory Group (CAG) approval. The PDS is the master demographic source for the NHS, which records patient mortality status including date of death where applicable. 2.3 Troponin assay The Beckman Coulter Access AccuTnI + 3 HsTrop assay (Beckman Coulter, Brea, CA, USA) was used to determine the hs-trop level of patients recruited to this study. The manufacturer's 99th percentile cut off for the assay is 40 ng/L. 2.4 Statistical analysis Continuous variables were presented as mean (+/− SD) or as medians and interquartile ranges (IQR). Differences in the continuous variables were evaluated using independent sample t -tests. Categorical variables were compared by using χ 2 tests. Cumulative survival curves were constructed using the Kaplan-Meier method with the log-rank test used to compare the survival curves. For all analyses, two-sided pa values < 0.05 were defined as significant. Statistical analyses were performed using IBM SPSS V.22.0 (SPSS, IBM Coporation, Armonk, New York, USA). Comparisons were made for the following groups: 1) NSTEMI versus tachyarrhythmia 2) Troponin positive tachyarrhythmia versus troponin negative tachyarrhythmia 2.5 Ethical approval The study was sponsored by University Hospital Southampton NHS Foundation Trust Research and Development department, approved by the South East Scotland Research Ethics Committee 01 (REC reference: 16/SS/0194, IRAS project ID:206061) and the CAG (CAG reference: 16/CAG/0146). The study was conducted in accordance with the Declaration of Helsinki. 3 Results 3.1 Observed data Patients included in the study ranged from age 21 to 100 years with a median of 75 years. Three hundred and eighty-eight (55%) patients were male. The patients were divided into one of four primary diagnoses: NSTEMI ( n = 264), AF ( n = 344), VT ( n = 56) or SVT ( n = 40). Of interest, T2MI was listed as a diagnosis in only 1 patient in the whole study population. Table 1 shows a comparison of the baseline demographics of patients in the NSTEMI and tachyarrhythmia group. The latter group has been stratified into troponin positive and negative. Significant differences between NSTEMI and tachyarrhythmia are observed in hypertension (189 vs 255, 72% vs 58%, p < 0.001), hypercholesterolaemia (108 vs 102, 41% vs 23%, p < 0.001), and diabetes mellitus (80 vs 70, 30% vs 16%, p < 0.001) respectively. Of the tachyarrhythmia group ( n = 440), 206 (47%) were troponin positive (ie elevated troponin above the 99th percentile cut off 40 ng/L) and 234 (53%) were troponin negative. Troponin positive tachyarrhythmia patients are older (74.9 Vs 71.1, p = 0.004) and have fewer previous AF diagnoses (168(81%) Vs 206(88%), p = 0.046) than troponin negative tachyarrhythmia patients. Table 1 (SD = Standard deviation, IHD = Ischaemic heart disease, AF = Atrial Fibrillation, SVT = Supraventricular Tachycardia, VT = Ventricular Tachycardia). NSTEMI ( n = 264)Tachyarrhythmia ( n = 440)Tachyarrhythmia Troponin positive ( n = 206)Tachyarrhythmia Troponin negative ( n = 234) Age (years/SD) 73.0/14.33 72.9/13.56 74.9/13.37 71.1/13.53 Gender (male) 161(61%) 227(52%) 116(56%) 110(47%) Gender (female) 103(39%) 213(48%) 91(44%) 124(53%) Hypertension 189(72%) 255(58%) 128(62%) 129(55%) Hypercholesterolaemia 108(41%) 102(23%) 54(26%) 49(21%) Diabetes mellitus 80(30%) 70(16%) 35(17%) 35(15%) IHD 76(28%) 92(21%) 48(23%) 44(19%) Heart failure 61(23%) 121(27%) 66(32%) 56(24%) AF 70(27%) 372(84%) 168(81%) 206(88%) SVT 7(3%) 52(12%) 25(12%) 28(12%) VT 12(4%) 53(12%) 31(15%) 21(9%) Table 2 displays the clinical presentation, investigations and management of NSTEMI patients compared to tachyarrhythmia patients (with further stratification into troponin positive and negative). There are significant differences in the occurrence of chest pain (238 vs 172, 90% vs 39%, p < 0.001) and heart rate (84 vs 147, p < 0.001) between the NSTEMI and tachyarrhythmia groups. In addition, there are significant differences in non-invasive ischaemia testing (29 vs 14, 11% vs 3%, p < 0.001), coronary angiogram (158 vs 40, 60% vs 9%, p < 0.001) and subsequent PCI (90 vs 0, 34.1% vs 0%, p < 0.001). A significant difference in prescribed Dual Antiplatelet Therapy (DAPT) (190 vs 26, 72% vs 6%, p < 0.001) was also seen between the two groups. Table 2 Clinical presentation, investigations and management in NSTEMI and tachyarrhythmia patients. (eGFR = Glomerular Filtration Rate, PCI = Percutaneous Coronary Intervention, NIIT = Non-Invasive Ischaemia Tests, DAPT = Dual Antiplatelet Therapy). NSTEMI ( n = 264)Tachyarrhythmia ( n = 440)Tachyarrhythmia troponin positive (206)Tachyarrhythmia troponin negative (234) Chest pain 238(90%) 172(39%) 101(49%) 73(31%) Heart rate (SD) 84(22.9) 147(33.7) 148(33.4) 145(33.9) eGFR 65.10 64.65 62.12 66.89 LVSD 109(41%) 163(37%) 80(41%) 71(33%) Peak trop (ng/L) 4552 276 571 16 Coronary angiography 158(60%) 40(9%) 35(17%) 3(1%) Lesion 127(48%) 3(8%) 3(9%) 0(0%) PCI 90(34.1%) 0(0%) 0(0%) 0(0%) NIIT 29(11%) 14(3%) 10(5%) 4(2%) DAPT 190(72%) 26(6%) 19(9%) 7(3%) Significantly higher peak troponin levels are seen in the NSTEMI group compared to the tachyarrhythmia troponin positive group (4552 ng/L vs 571 ng/L, p < 0.001). Significant differences are observed in the presence of chest pain (238 vs 94, 90% vs 45.6%, p < 0.001). Additionally, there are significant differences between the number of patients who were referred for non-invasive ischaemia testing (NIIT) (29 vs 10, 11% vs 5%, p = 0.018), coronary angiography (158 vs 35, 60% vs 17%, p < 0.001), and PCI (90 vs 0, 34% vs 0%, p < 0.001) when comparing the NSTEMI and tachyarrhythmia troponin positive groups respectively. In addition, a total of 43 (16.3%) patients discharged with a primary diagnosis of NSTEMI also had a diagnosis of a tachyarrhythmia (AF = 40, VT = 3) as a secondary discharge diagnosis. 36(83.7%) of this group presented with chest pain, 19(44.2%) patients had coronary angiography with subsequent PCI in 7 (16.3%) patients. The troponin positive tachyarrhythmia patients had significantly more cases of chest pain on admission (101 vs 73, 49% vs 31%, p < 0.001), more coronary angiography (35 vs 3, 17% vs 1%, p < 0.001), presence of coronary artery disease (3 vs 0, 9% vs 0%, p < 0.001) and significantly worse eGFR (62.12 vs 66.89, p = 0.013) compared to the troponin negative group. The distribution of peak troponin for each primary diagnosis is displayed as a box plot in Supplementary Fig. A . The boxplot only includes the data of troponin positive patients ( n = 470). 4 Clinical outcomes The mortality data was obtained via NHS Digital through the PDS. Follow-up was at a median of 747 days, with a range of 1–1223 days. Kaplan-Meier survival curves ( Fig. 2 a ) were constructed to compare the mortality between NSTEMI and tachyarrhythmia group, demonstrating no significant difference in mortality between the two groups ( p = 0.401). This is shown in Fig. 2 a. Tachyarrhythmia patients were sub-stratified according to whether they had a raised HsTrop level above the ULN (+ ve) or below the ULN (− ve). Kaplan-Meier survival analysis ( Fig. 2 b) showed a significant difference in mortality between the two groups for the whole follow-up period (54 vs 34, 26.2% Vs 14.5%, p = 0.003). Kaplan-meier survival curves also show no significant difference in mortality over the follow up period ( p = 0.416) between NSTEMI and troponin positive tachyarrhythmia ( Fig. 2 c) but a significantly higher mortality ( p = 0.023) in the NSTEMI group compared to the troponin negative tachyarrhythmia group ( Fig. 2 d). Fig. 2 Kaplan-Meier Survival Curves of NSTEMI and Arrhythmias. Further survival analysis has shown that in the NSTEMI group the absence of chest pain is associated with a higher mortality risk (13 vs 42, 43.3% vs 17.9%, p = 0.01) (see Supplementary Fig. B ). This observation is also seen in the troponin positive tachyarrhythmia patients (42 vs 12, 37.2% vs 12.8%, p < 0.001) (see Supplementary Fig. C ). This is not seen in the troponin negative tachyarrhythmia patients (22 vs 12, 13.7% vs 16.4%, p = 0.603). 5 Discussion This study has described the distribution, management and associated mortality of a consecutive group of patients presenting with a primary diagnosis of tachyarrhythmia, and the impact of hs-trop results on these. A group of NSTEMI patients admitted during the study period is used for comparison. Our main findings were as follows. Firstly, that 47% of patients presenting with an tachyarrhythmia are troponin positive but did not receive a formal diagnosis of either NSTEMI (T1MI) or T2MI. Secondly, that tachyarrhythmia patients with a raised hs-trop level had a low rate of angiography and PCI compared to NSTEMI patients. Third, that the mortality of tachyarrhythmia troponin positive patients is similar to NSTEMI patients, and that this is higher than troponin negative tachyarrhythmia patients. Our findings highlight the challenge that front line staff face in the interpretation of troponin elevation in patients presenting with tachyarrhythmia. This was made clearer because we did not set out to reclassify the diagnoses that had been made in real life practice. Previous studies have also shown that patients presenting with AF and positive troponin to the emergency department ( n = 2898) have a higher mortality compared to patients with normal HsTrop level. More recently, Thelin et al. showed that in a group of AF patients admitted without known coronary artery disease (CAD) ( n = 521) that 9.5% ( n = 49) had troponin levels above the ULN. The investigators went on to show in this cohort that the age-adjusted hazard ratio for all-cause mortality was 3.8(95% CI: 1.7 to 8.5; p = 0.001). However, they could not show a major increased risk of coronary events or death due to ischaemic heart disease, thus suggesting that CAD may not play a significant role in the increased mortality risk seen in patients with AF and troponin rise. In this study a total of 470 (66.8%) of the population are diagnosed as having suffered an MI according to the third universal definition , but only one of the patients were formally diagnosed with a T2MI. In this study the discharge diagnosis made by the supervising clinicians was used to stratify the patients. However, according to the third universal definition, all the patients with a primary diagnosis of tachyarrhythmia who had a troponin rise ( n = 206, 43.8%) in this study can also be categorized as having suffered a T2MI. This is significantly higher than previously reported, including 7.1% from a large Swedish registry and other highly selected population studies where a prevalence of 2–5% is quoted . Importantly, the troponin assay used in the Swedish registry was not highly sensitive. The data from our study are consistent with a study from Shah et al. , using HsTrop, in which 48% of patients were diagnosed with T2MI or myocardial injury. In contrast, a recent study reported that our of patients diagnosed with MI the incidence of T2MI was 19.2%. This contrasts with a formal diagnosis of T2MI in only 1 patient (0.2%) in our study. The measurement of troponin and the finding that the level is elevated, does not appear to have a predictable impact on management in tachyarrhythmia patients according to our observations. This is reflected in the relatively low rate of angiography in this cohort even those with a troponin rise. Whilst the investigators cannot tell accurately why some patients with troponin positive arrhythmia had coronary investigations and treatment, and some did not, it is clear that a formal diagnosis of T2MI is very rare in clinical practice. This is important, because previous studies have shown that T2MI is not appropriately managed using angiography/revascularization. Other studies have demonstrated that the prognosis for T2MI is not benign, and indeed the cohort we observed with troponin positive tachyarrhythmia had a relatively poor prognosis, equivalent to that of NSTEMI, and significantly worse than those with troponin negative tachyarrhythmia. It is unclear from an analysis of the patients in our study exactly how the troponin positive or negative result contributed to the patients' management. This is a potentially important observation because the finding of elevated troponin in tachyarrhythmia patients appears to be associated with a worse survival, yet the clinicians did not have distinct patterns of investigation or management. 6 Study limitations There are several limitations to this study. Firstly, it is observational, although, this has the advantage that we report the actual outcome of a consecutive, real life population. Secondly, we have used the original diagnoses formulated by supervising physicians for our analyses. It is clear that in some cases careful reinterpretation of the data may have led to a different categorization, but our aim was to describe what had actually happened. Thirdly, these are retrospective data, although having tracked mortality as the primary endpoint allows for more robust interpretation of patient outcome. 7 Conclusions These data suggest that troponin positive tachyarrhythmia is not a benign diagnosis, and has a mortality rate similar to NSTEMI. Formal labeling as T2MI is rare in real life practice. More investigation into the detection and management of T2MI and troponin positive tachyarrhythmia patients is now warranted. The following are the supplementary data related to this article. Supplementary Fig. A Box-plot distribution of peak troponin for each primary diagnosis. Supplementary Fig. B and C Kaplan-Meier survival curves for troponin positive patients presenting with and without chest pain. Funding: None. Competing interests: B. Olechowski – none declared M. Mariathas – none declared C. Gemmell – none declared Z. Nicholas – none declared M. Mahmoudi – none declared N. Curzen – unrestricted research grants from: Boston Scientific; Haemonetics; Heartflow; St Jude Medical; Medtronic; speaker fees/consultancy from: Haemonetics, St Jude Medical, Abbot Vascular; Heartflow; Boston Scientific; Travel sponsorship – Biosensors, Abbot, Lilly/D-S; St Jude Medical, Medtronic. All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.