The most common cause of death among hemodialysis

Diagnosis of Acute Myocardial Infarction in Hemodialysis Patients With High-Sensitivity Cardiac Troponin T Assay Hua-Lan Huang, MM; Shuai Zhu, MM; Wei-Qing Wang, MM; Xin Nie, MM; Ying-Ying Shi, MD; Yong He, MM; Hao-Lan Song, BM; Qiang Miao, BM; Ping Fu, MD; Lan-Lan Wang, MD; Gui-Xing Li, MM Context. Cardiac troponins have become the gold standard for diagnosing acute myocardial infarction (AMI) in the general population; however, their diagnostic accuracy for hemodialysis (HD) patients presenting with chest pain or dyspnea is uncertain. Objective. To examine the diagnostic accuracy of highsensitivity cardiac troponin T (hs-ctnt) assay for AMI in HD patients. Design. In this prospective study, we enrolled 670 consecutive stable HD patients presenting with chest pain or dyspnea on routine predialysis therapy in the nephrology department. Receiver operating characteristic (ROC) curves were used to examine the diagnostic accuracy of hsctnt levels at enrollment in HD patients presenting with chest pain or dyspnea, and the dynamic change in these levels after 3 hours. Results. Acute myocardial infarction was the adjudicated final diagnosis in 12% of HD patients. Among patients with a final diagnosis other than AMI, 97% had a plasma hs-ctnt concentration above the 99th percentile. At the time of enrollment, the area under the ROC curve of hs-ctnt levels for diagnosis of AMI was 0.68 (95% confidence interval [CI], 0.62 0.74; P,.001) with a cutoff value of 107.7 ng/l; the relative change after 3 hours was 0.90 (95% CI, 0.82 0.96, P,.001) with a cutoff value of 24%, and the absolute change was 0.88 (95% CI, 0.82 0.94, P,.001) with a cutoff value of 32.6 ng/l. The prognostic value for 40-day mortality varied with the magnitude of elevation in hs-ctnt levels. Conclusions. Tracking the dynamic change in hs-ctnt levels during the short term significantly increased this measure s diagnostic accuracy for AMI in HD patients. (Arch Pathol Lab Med. 2016;140:75 80; doi: 10.5858/ arpa.2014-0580-oa) The most common cause of death among hemodialysis (HD) patients is cardiac death predominantly acute myocardial infarction (AMI), which accounts for approximately 50% of all deaths. 1 Rapid and accurate diagnosis of AMI is an urgent clinical need, as delays increase the risk of mortality. 2,3 Delayed exclusion of this diagnosis prolongs hospital length of stay, increases patients anxiety, and carries enormous costs for the health care system. 4 Cardiac troponins have become the gold standard for diagnosing AMI in the general population. 5,6 However, patients with end-stage renal disease (ESRD) have a high incidence of baseline elevation of troponins, 7 9 making the diagnosis of AMI in patients with ESRD more difficult. According to the clinical practice guidelines of the National Academy of Clinical Biochemistry, measurement of cardiac troponin T (ctnt) concentration is warranted for evaluating AMI in the population with renal failure for those who present with symptoms, electrocardiographic findings, or other clinical evidence suggestive of myocardial ischemia. 10 The newest generation of ctnt assays has a 10- to 100- fold reduced limit of detection, below the 99th percentile of a normal reference population, and these assays are being adopted by a growing number of medical institutions. These assays improve early diagnosis of AMI in the general patient population presenting with acute chest pain. 11 14 However, their diagnostic accuracy for HD patients presenting with chest pain or dyspnea is uncertain. Furthermore, because a substantial proportion of HD patients have elevated cardiac troponins in the absence of acute coronary syndrome, a key clinical challenge is to differentiate a clinically relevant rise in cardiac troponin level from its background elevation. We therefore investigated (1) whether the severity of renal disease is associated with increased baseline high-sensitivity troponin T levels and (2) the diagnostic accuracy of high-sensitivity cardiac troponin T (hs-ctnt) levels measured in patients presenting with acute chest pain or dyspnea at enrollment, and the dynamic change in these levels after 3 hours for AMI in HD patients. Accepted for publication April 15, 2015. From the Departments of Laboratory Medicine (Drs Li and Wang L- L; Mss Huang, Wang W-Q, Nie, He, and Song; and Messrs Zhu and Miao) and Nephrology (Drs Shi and Fu), West China Hospital, Sichuan University, Chengdu, China. MATERIALS AND METHODS This study was supported by Roche Diagnostics GmbH in Study Population Shanghai (H1311115). The authors have no relevant financial interest in the products or In the West China Hospital of Sichuan University (Chengdu, companies described in this article. China), for the period between September 2009 and December Reprints: Gui-Xing Li, MM, Department of Laboratory Medicine, 2013, we enrolled HD patients from the nephrology department West China Hospital, Sichuan University, Chengdu 610041, China (e-mail: liguixing2014@163.com). who on routine predialysis therapy showed signs of chest pain or dyspnea. In addition, on the basis of clinical symptoms, clinicians Arch Pathol Lab Med Vol 140, January 2016 High-Sensitivity ctnt Assay for AMI in Hemodialysis Patients Huang et al 75

found these patients to have possible acute coronary syndrome and decided to order troponin T tests. Exclusion criteria were major surgery or trauma within the previous 4 weeks, pregnancy, intravenous drug abuse, age younger than 18 years or older than 85 years, and anemia (hemoglobin level, 10 g/dl). Signed informed consents were obtained from all participants or their guardians, and the study was approved by the Ethics Committee of West China Hospital, Sichuan University. Clinical and Laboratory Data Collection At the time of enrollment, a detailed clinical history was recorded, including age, sex, primary renal diagnosis, blood pressure, smoking history, coronary artery disease history, comorbidity (diabetes mellitus, heart failure, respiratory disease, hypertension, dyslipidemia, or renal anemia), and time undergoing hemodialysis. Conventional ctnt was measured at enrollment and, if needed, the measurement was repeated after 6 to 9 hours with 99th percentiles of 0.01 ng/ml. The decision to order troponin T tests was at the discretion of the physician in charge. High-sensitivity ctnt was measured at enrollment and 3 hours after enrollment with 99th percentiles of 14 ng/l. All assays were performed on the Cobas-E601 analyzer using the electrochemiluminescence immunoassay (Roche Diagnostics, Penzberg, Germany): ctnt with a limit of detection of 0.01 ng/ml, a 99th percentile cutoff point of 0.01 ng/ml, and a coefficient of variation of 10% at 0.03 ng/ml; and hs-ctnt with a limit of detection of 0.003 ng/ml (3 ng/l), a 99th percentile cutoff point of 0.014 ng/ml (14 ng/l), and a coefficient of variation of 10% at 0.013 ng/ml (13 ng/l). High-density lipoprotein cholesterol, low-density lipoprotein cholesterol, triglyceride, total cholesterol, serum calcium, serum phosphate, serum albumin, and serum creatinine concentrations were determined at enrollment via Modular-P800 (Roche Diagnostics). The estimated glomerular filtration rate (egfr) calculated with the creatinine value was obtained by use of the simplified Modification of Diet in Renal Disease formula. 15 Follow-up data were obtained 40 days after enrollment, and the primary end points were all-cause mortality. Adjudication of the Final Diagnosis Attending nephrology physicians were blinded to hs-ctnt results. The diagnosis of acute coronary syndrome was based on the patients available clinical, laboratory, and imaging findings. If acute coronary syndrome was suspected, cardiologists were informed, who then usually hospitalized patients on their intermediate care ward for monitoring and treatment before potential coronary angiography. The final diagnosis was adjudicated by both the nephrology physician and the cardiologist from the time of enrollment to the end of the 40-day follow-up period. A third cardiologist refereed in situations of disagreement. Acute myocardial infarction was adjudicated according to current guidelines. 16 Diagnosis of AMI, either non- ST segment elevation or ST elevation myocardial infarction, required a conventional ctnt above the 99th percentile value, associated with more than one of the following: symptoms of ischemia, new ST-T changes or a new Q wave on the electrocardiogram, and imaging showing new loss of viable myocardium. Data Analysis Continuous variables are presented as means (6SD) or medians (with the interquartile range), and categorical variables as numbers and percentages. For group comparisons, we used Student t test for normally distributed variables and nonparametric Mann-Whitney tests for nonnormally distributed variables. Categorical variables were compared with a v 2 test. The Kruskal-Wallis test was used for multiple comparisons. Correlation between hs-ctnt and renal function was quantified by the Pearson correlation coefficient. Receiver operating characteristic (ROC) curves were constructed to assess the diagnostic accuracy of hs-ctnt measured in HD patients presenting with acute chest pain or dyspnea at enrollment, and its dynamic change after 3 hours, for AMI. The areas under the curve Table 1. Patient Characteristics a Clinical Characteristics, Reference Ranges Total (N ¼ 670) Age, y 56.7 6 17.5 Male sex, No. (%) 372 (55.5) Renal disease, No. Diabetic nephropathy 125 Obstructive uropathy 120 Glomerulonephritis 229 Hypertensive nephropathy 98 Polycystic 52 Chronic pyelonephritis 18 Other 28 Smoker, No. (%) 188 (28.1) Hypertension, No. (%) 582 (86.9) Systolic blood pressure, 90 140 mm Hg 148.9 6 28.1 Diastolic blood pressure, 60 90 mm Hg 86.7 6 18.7 Heart failure, No. (%) 354 (52.8) Diabetes, No. (%) 191 (28.5) Pulmonary infection, No. (%) 323 (48.2) Dyslipidemia, No. (%) 143 (21.3) Prior coronary artery disease, No. (%) 134 (20.0) Triglyceride, 25.7 73.4 mg/dl 130.0 (90.3 177.9) Total cholesterol, 108.3 220.4 mg/dl 146.5 (119.5 175.9) HDL cholesterol,.34.8 mg/dl 71.1 (60.2 91.3) LDL cholesterol,,154.7 mg/dl 56.4 (32.1 73.9) Calcium concentration, 8.4 10.8 mg/dl 8.2 6 1.3 Phosphate concentration, 2.5 4.5 mg/dl 5.3 (4.0 6.7) Serum albumin, 40 55 g/l 33.7 6 7.8 Hs-cTnT,,14 ng/l 74.2 (38.9 133.9) Serum creatinine, 0.60 1.58 mg/dl 8.92 (6.73 11.64) egrf,.90 ml min 1 (1.73 m 2 ) 1 5.75 (4.28 7.53) Time undergoing dialysis, mo 26.7 (12.3 35.2) Follow-up mortality, No. (%) 43 (6.4) Abbreviations: egrf, estimated glomerular filtration rate; HDL, highdensity lipoprotein; Hs-cTnT, high-sensitivity cardiac troponin T; LDL, low-density lipoprotein. a Qualitative data are presented as No. (%). Quantitative data of normal distribution are presented as mean 6 SD (standard deviation). Quantitative data of nonnormal distribution are presented as median (25th and 75th percentiles). (AUCs) of hs-ctnt measured at enrollment and its dynamic change after 3 hours were compared by using the Mann-Whitney U Test. The best cutoff values were determined by the point farthest from the bisector of the ROC curve, plotted by SigmaPlot 10.0 (Systat Software Inc, San Jose, California). To determine the prognostic value of the hs-ctnt assays, we calculated Kaplan-Meier curves, using the logrank test for comparisons and Cox regression. Two-sided statistical tests were used for all analyses, and differences between groups were considered to be significant at P,.05. SPSS for Windows (version 13.0, Chicago, Illinois) was used to perform analyses. RESULTS Patient Characteristics This study included 670 predialysis patients with chronic kidney disease who presented with chest pain or dyspnea and of whom 372 (55.5% of the group) were male; the mean age at enrollment was 56.7 years (Table 1). A total of 80 patients (12% of the cohort) received a diagnosis of AMI. The primary renal diagnoses were diabetic nephropathy (n ¼ 125), obstructive uropathy (n ¼ 120), primary glomerulonephritis (n ¼ 229), hypertensive nephropathy (n ¼ 98), 76 Arch Pathol Lab Med Vol 140, January 2016 High-Sensitivity ctnt Assay for AMI in Hemodialysis Patients Huang et al

Table 2. Baseline Characteristics of Patients Without and With Acute Myocardial Infarction (AMI) a Clinical Characteristics, Reference Ranges Non-AMI (N ¼ 590) AMI (N ¼ 80) P Age, y 55.04 6 17.48 69.05 6 11.46,.001 Male sex, No. (%) 325 (55.08) 47 (58.75).61 Smoker, No. (%) 166 (28.13) 22 (27.5).92 Hypertension, No. (%) 508 (86.1) 74 (92.5).11 SBP, 90 140 mm Hg 144.59 6 23.24 149.51 6 28.69.23 DBP, 60 90 mm Hg 81.18 6 16.21 87.40 6 18.77.01 Heart failure, No. (%) 311 (52.71) 43 (53.75).93 Diabetes, No. (%) 151 (25.59) 40 (50.00),.001 Pulmonary infection, No. (%) 284 (48.14) 39 (48.75).94 Dyslipidemia, No. (%) 129 (21.87) 19 (23.56).71 Prior CAD, No. (%) 110 (18.6) 24 (30.0),.001 Triglyceride, 25.7 73.4 mg/dl 130.1 (90.3 178.8) 123.9 (92.9 173.5).90 Total cholesterol, 108.3 220.4 mg/dl 146.9 (119.5 176.3) 143.1 (116.4 167.8).21 HDL cholesterol,.34.8 mg/dl 72.3 (54.3 95.7) 63.8 (46.9 82.5).04 LDL cholesterol,,154.7 mg/dl 56.4 (32.1 78.4) 58.8 (37.9 71.3).51 Calcium, 8.4 10.8 mg/dl 8.2 6 1.3 8.3 6 1.3.72 Phosphate, 2.5 4.5 mg/dl 5.4 (4.1 6.8) 4.7 (3.6 6.1).01 Serum albumin, 40 55 g/l 33.83 6 7.62 32.99 6 8.72.41 Hs-cTNT,,14 ng/l 69.2 (36.9 121.9) 118.8 (65.7 216.0),.001 Serum creatinine, 0.60 1.58 mg/dl 8.03 (6.16 9.75) 9.21 (6.94 11.88),.001 egrf,.90 ml min 1 (1.73 m 2 ) 1 6.55 (5.14 8.96) 5.60 (4.20 7.39).002 Time undergoing dialysis, mo 18.5 (12.3 25.2) 30.1 (23.6 36.2),.001 Follow-up mortality, No. (%) 33 (5.6%) 10 (12.5%).02 Abbreviations: CAD, coronary artery disease; DBP, diastolic blood pressure; egrf, estimated glomerular filtration rate; HDL, high-density lipoprotein; Hs-cTnT, high-sensitivity cardiac troponin T; LDL, low-density lipoprotein; SBP, systolic blood pressure. a Qualitative data are presented as No. (%). Quantitative data of normal distribution are presented as mean 6 SD (standard deviation). Quantitative data of nonnormal distribution are presented as median (25th and 75th percentiles). All data were collected within 24 hours after admission. autosomal dominant polycystic kidney disease (n ¼ 52), chronic pyelonephritis (n ¼ 18), and unknown (n ¼ 28). Age; diastolic blood pressure; diabetes; prior coronary artery disease; time undergoing dialysis; high-density lipoprotein cholesterol, serum phosphate, hs-ctnt, and serum creatinine levels; and low egfr value were positively associated with AMI. For the other characteristics we analyzed, there was no difference between the AMI and non-ami groups (Table 2). Diagnostic Accuracy of hs-ctnt at Enrollment Patients whose final diagnosis was not AMI (n ¼ 590) had high levels of hs-ctnt at enrollment (69.2 [36.9, 121.9] ng/l) including 572 patients (97%) with a plasma hs-ctnt concentration above the 99th percentile (Figure 1); patients whose final diagnosis was AMI had higher (P,.001) levels of hs-ctnt at enrollment (118.8 [65.7, 216.0] ng/l; Figure 2, A), which were 8.5 (4.7, 15.4) multiples of the 99th percentile for hs-ctnt levels (Figure 2, B). The AUC of hs-ctnt level for diagnosis of AMI was 0.68 (95% confidence interval [CI], 0.62 0.74; P,.001; Figure 3). The ROC-determined optimal cutoff value (107.7 ng/l) to separate AMI from non-ami was greater than 7 times the 99th percentile value for hs-ctnt, with a sensitivity of 58%, specificity of 71%, positive predictive value of 21%, and negative predictive value of 93%. Overall, hs-ctnt assays showed higher sensitivity at the 99th percentile than the best cutoff value (100% versus 58%; P,.001). The increase in sensitivity at the 99th percentile was associated with low specificity (3%) and positive predictive value (12%). Dynamic Changes in hs-ctnt Concentration in AMI Diagnosis Three hours after enrollment, the patients in the AMI group had higher (P,.001) relative and absolute change in hs-ctnt levels (160% [35%, 270%] and 145.3 [34.2, 448.9] ng/l, respectively) than those in the non-ami group (11% [5%, 24%] and 10.4 [4.3, 26.5] ng/l, respectively). The diagnostic accuracies of various cutoff values for relative and absolute change in hs-ctnt are provided in Table 3. According to our ROC analyses (Figure 3), the optimal diagnostic cutoffs for relative and absolute change in hs-ctnt were 24% and 32.6 ng/l, respectively. The relative and absolute change in hs-ctnt levels yielded better diagnostic accuracy for AMI, quantified by AUC, than did hs-ctnt determination at enrollment (0.90 [95% CI, 0.82 0.96; P,.001] versus 0.68 [95% CI, 0.62 0.74; P,.001], P,.001 and 0.88 [95% CI, 0.82 0.94; P,.001] versus 0.68 [95% CI, 0.62 0.74; P,.001], P,.001, respectively). There Figure 1. Distribution of high-sensitivity cardiac troponin T (hs-ctnt) levels at enrollment. The hs-ctnt levels were measured in hemodialysis participants with (n ¼ 80) and without (n ¼ 590) acute myocardial infarction (AMI). Arch Pathol Lab Med Vol 140, January 2016 High-Sensitivity ctnt Assay for AMI in Hemodialysis Patients Huang et al 77

Figure 2. High-sensitivity cardiac troponin (hs-ctnt) levels at enrollment. The boxes indicate the interquartile range (IQR). The dark lines denote the median. The whiskers portray minimum and maximum values. The hs-ctnt levels between acute myocardial infarction (AMI) (n ¼ 80) and non- AMI (n ¼ 590) groups (A), displayed as multiples of the 99th percentile (B). was no difference in diagnostic accuracy between relative and absolute change in hs-ctnt levels 3 hours after enrollment (AUC: [95% CI, 0.82 0.96; P,.001] versus 0.88 [95% CI, 0.82 0.94; P,.001], P ¼.80) (Table 4). Prognostic Value of hs-ctnt Assays At the end of the 40-day follow-up, of patients with a final diagnosis other than AMI (n ¼ 590), 33 (5.6%) had died 9 owing to infection; 8, heart respiratory circulatory failure; 4, ischemic stroke; the remaining patients withdrew from dialysis owing to general debility. None of the patients whose hs-ctnt concentration was less than 14 ng/l at enrollment had died. The risk of death was greater in patients with elevated hs-ctnt (P ¼.02), and the degree of risk was dependent on the magnitude of elevation of hsctnt (Figure 4). We examined the independent determinants of all-cause mortality within 40 days among patients whose final diagnosis was other than AMI, using multivariable logistic regression analyses of age, sex, and comorbidities such as hypertension, diabetes, smoking history, history of coronary artery disease, time undergoing dialysis, and hs-ctnt levels at enrollment. The results showed that age, diabetes, hsctnt levels, history of coronary artery disease, and time undergoing dialysis (hazard ratio: 1.5 (1.0, 1.8), 2.1 (1.2, 3.7), 1.3 (1.0, 1.5), 3.5 (2.0, 4.5), 2.5 (1.0, 4.5); P,.001) were significant independent predictors of death. COMMENT This prospective study involving consecutive patients examined the diagnostic performance of an hs-ctnt assay for predialysis patients with acute chest pain or dyspnea at presentation, as well as dynamic change in hs-ctnt levels 3 hours after presentation, in early diagnosis of AMI. Our results indicate that (1) hs-ctnt concentrations are significantly correlated with renal function parameters such as Figure 3. The area under the receiver-operating characteristic curves of high-sensitivity cardiac troponin T (hs-ctnt) levels, and dynamic changes in these levels, for the diagnosis of acute myocardial infarction (AMI). Figure 4. Prognostic impact of elevation in high-sensitivity cardiac troponin T (hs-ctnt) in patients with final diagnosis other than acute myocardial infarction (AMI). Kaplan-Meier curves showing mortality within 40 days for patients whose final diagnosis was other than AMI (n ¼ 590), comparing patients with levels of sensitive troponins above the optimal cutoff value (as determined by receiver operating characteristic [ROC] curves), above the 99th percentile but below the optimal cutoff value, and below the 99th percentile. 78 Arch Pathol Lab Med Vol 140, January 2016 High-Sensitivity ctnt Assay for AMI in Hemodialysis Patients Huang et al

Table 3. Diagnostic Performance of Dynamic Change in High-Sensitivity Cardiac Troponin T (hs-ctnt) Levels in Acute Myocardial Infarction Diagnosis Relative Change in hs-ctnt Levels 10% 20% 30% 40% 50% 60% Sensitivity, % (95% CI) 97 (82 100) 93 (77 99) 72 (53 87) 59 (39 76) 55 (36 74) 55 (36 74) Specificity, % (95% CI) 41 (34 48) 70 (63 76) 82 (75 87) 92 (87 95) 97 (94 99) 100 (98 100) PPV, % (95% CI) 22 (17 24) 34 (26 41) 40 (26 90) 55 (33 72) 75 (50 92) 92 (90 96) NPV, % (95% CI) 99 (92 100) 98 (94 99) 94 (90 97) 93 (89 96) 100 (75 100) 92 (90 96) Absolute Change in hs-ctnt Levels, ng/l 10 30 50 100 300 500 Sensitivity, % (95% CI) 97 (82 100) 79 (60 92) 69 (49 85) 59 (39 76) 34 (18 54) 21 (8 40) Specificity, % (95% CI) 48 (41 55) 79 (73 84) 88 (82 92) 96 (92 98) 99 (97 100) 100 (98 100) PPV, % (95% CI) 24 (19 27) 39 (27 49) 49 (31 64) 71 (45 86) 85 (50 100) 100 (40 100) NPV, % (95% CI) 99 (93 100) 96 (92 98) 94 (91 97) 93 (90 96) 90 (88 93) 88 (86 90) Abbreviations: CI, confidence interval; NPV, negative predictive value; PPV, positive predictive value. egfr and creatinine, and (2) the dynamic change in hsctnt levels 3 hours after presentation can be used to diagnose AMI in patients receiving chronic HD. By using the 99th percentile of a healthy reference population to define an elevated troponin level, 30% to 85% of asymptomatic HD patients have been reported to have elevated troponin T levels. 17 In our study, using a newer-generation high-sensitivity troponin T assay, the percentage of HD patients with a diagnosis other than AMI who had a baseline level of hs-ctnt above the 99th percentile was 97%. This suggests that increasing the sensitivity of the assay may further increase the prevalence of detected cardiac troponin elevation in the HD population. Clinically, the high incidence of elevated hs-ctnt levels in HD patients challenges the application of the 99th percentile (14 ng/l) as the threshold for AMI diagnosis. Our study showed that the specificity and positive predictive value were only 3% and 12.2%, respectively, when choosing 14 ng/l as the decision limit of hs-ctnt assays to diagnose AMI in HD patients. This finding illustrates the difficulty of using hs-ctnt levels to diagnose AMI in HD patients who present with chest pain or dyspnea. In our study, we raised the threshold of serum hs-ctnt concentration to 107.7 ng/l, the best cutoff value as determined by ROC for diagnosis of AMI in HD patients. However, the diagnostic accuracy was still rather low (AUC: 0.68), with a sensitivity of 58% and a specificity of 71%. A similar result was reported in another study, 18 although patients in this study had renal insufficiency. Wang and Wai-Kei 19 have recommended that a relative change in cardiac troponin levels of 20% or more, with at least 1 value above the 99th percentile, should be used to define AMI in patients undergoing dialysis who present with symptoms suggestive of acute coronary syndrome. Our study was the first to verify this idea, showing that using the relative and absolute change in hs-ctnt levels 3 hours after presentation significantly increases the diagnostic accuracy for AMI in HD patients presenting with chest pain or dyspnea, and it may aid in differentiating AMI from unrelated chest pain or dyspnea. Our study demonstrated that an elevated hs-ctnt level in patients undergoing dialysis with a final diagnosis other than AMI was an independent risk factor for mortality within 40 days, reaffirming the prognostic value of hs-ctnt in patients undergoing dialysis. Our findings extend the results of previous studies 20,21 investigating the short-term mortality of HD subjects with elevated levels of hs-ctnt. Some studies 22 24 have indicated that HD can influence troponin levels, and one limitation of our study was that we only performed analyses on patients in predialysis. In addition, although chest pain is considered to be the cardinal manifestation of coronary ischemia, patients with both predialysis chronic kidney disease 25 and ESRD 26 present with chest pain as their chief concern much less frequently than do patients with normal renal function. In fact, fewer than 50% of patients undergoing dialysis who have AMI present with chest pain. In our study we only evaluated the diagnostic accuracy of hs-ctnt in patients with symptoms. Further studies are needed to establish an Table 4. Diagnostic Performance of Dynamic Change in High-Sensitivity Cardiac Troponin T (hs-ctnt) Levels at ROC-Determined Best Cutoff Value a Optimal Cutoff Value hs-ctnt, ng/l Relative Change in hs-ctnt, % Absolute Change in hs-ctnt, ng/l 107.7 24 32.6 P1 P2 P3 AUC (95% CI) 0.68 (0.62 0.74, 0.90 (0.82 0.96, 0.88 (0.82 0.94,,.001,.001.81 P,.001) P,.001) P,.001) Sensitivity, % (95% CI) 58 (46 68) 89 (73 98) 79 (60 92),.001,.001.01 Specificity, % (95% CI) 71 (67 75) 76 (70 82) 81 (74 86).60.41.21 PPV, % (95% CI) 21 (18 24) 38 (29 48) 41 (28 52),.001,.001.32 NPV, % (95% CI) 93 (88 96) 98 (94 99) 96 (92 98).01.01.60 Abbreviations: AUC, area under the ROC curve; CI, confidence interval; NPV, negative predictive value; PPV, positive predictive value; ROC, receiver operating characteristic. a P1: comparison between hs-ctnt and relative change in hs-ctnt; P2: comparison between hs-ctnt and absolute change in hs-ctnt; P3: comparison between relative and absolute change in hs-ctnt. Arch Pathol Lab Med Vol 140, January 2016 High-Sensitivity ctnt Assay for AMI in Hemodialysis Patients Huang et al 79

efficient method to screen for AMI in asymptomatic patients undergoing dialysis. We gratefully acknowledge the participation of members of the nephrology and emergency departments of West China Medical School (Chengdu, China) in this study. References 1. Herzog CA, Ma JZ, Collins AJ. Poor long-term survival after acute myocardial infarction among patients on long-term dialysis. N Engl J Med. 1998;339(12):799 805. 2. Bassand JP, Hamm CW, Ardissino D, et al. Guidelines for the diagnosis and treatment of non-st-segment elevation acute coronary syndromes. Eur Heart J. 2007;28(13):1598 1660. 3. Antman EM, Cohen M, Bernink PJ, et al. The TIMI risk score for unstable angina/non-st elevation MI: a method for prognostication and therapeutic decision making. JAMA. 2000;284(7):835 842. 4. Forberg JL, Henriksen LS, Edenbrandt L, Ekelund U. Direct hospital costs of chest pain patients attending the emergency department: a retrospective study. BMC Emerg Med. 2006;6:6. 5. Keller T, Zeller T, Peetz D, et al. Sensitive troponin I assay in early diagnosis of acute myocardial infarction. N Engl J Med. 2009;361(9):868 877. 6. Reichlin T, Hochholzer W, Bassetti S, et al. Early diagnosis of myocardial infarction with sensitive cardiac troponin assays. N Engl J Med. 2009;361(9):858 867. 7. Ooi DS, House AA. Cardiac troponin T in hemodialyzed patients. Clin Chem. 1998;44(7):1410 1416. 8. Dierkes J, Domrose U, Westphal S, et al. Cardiac troponin T predicts mortality in patients with end-stage renal disease. Circulation. 2000;102(16): 1964 1969. 9. Deegan PB, Lafferty ME, Blumsohn A, Henderson IS, McGregor E. Prognostic value of troponin T in hemodialysis patients is independent of comorbidity. Kidney Int. 2001;60(6):2399 2405. 10. Morrow DA, Cannon CP, Jesse RL, et al. National Academy of Clinical Biochemistry Laboratory Medicine Practice Guidelines: clinical characteristics and utilization of biochemical markers in acute coronary syndromes. Clin Chem. 2007;53(4):552 574. 11. Januzzi JJ, Bamberg F, Lee H, et al. High-sensitivity troponin T concentrations in acute chest pain patients evaluated with cardiac computed tomography. Circulation. 2010;121(10):1227 1234. 12. Omland T, de Lemos JA, Sabatine MS, et al. A sensitive cardiac troponin T assay in stable coronary artery disease. N Engl J Med. 2009;361(26):2538 2547. 13. Freund Y, Chenevier-Gobeaux C, Bonnet P, et al. High-sensitivity versus conventional troponin in the emergency department for the diagnosis of acute myocardial infarction. Crit Care. 2011;15(3):R147. 14. Zuily S, Chenevier-Gobeaux C, Claessens YE, Wahbi K, Weber S, Meune C. High diagnostic performance of a high-sensitivity cardiac troponin T assay in patients with suspected acute coronary syndrome. Int J Cardiol. 2011;146(1): 115 116. 15. Chenevier-Gobeaux C, Meune C, Freund Y, et al. Influence of age and renal function on high-sensitivity cardiac troponin T diagnostic accuracy for the diagnosis of acute myocardial infarction. Am J Cardiol. 2013;111(12):1701 1707. 16. Thygesen K, Alpert JS, Jaffe AS, et al. Third universal definition of myocardial infarction. J Am Coll Cardiol. 2012;60(16):1581 1598. 17. Apple FS, Murakami MM, Pearce LA, Herzog CA. Predictive value of cardiac troponin I and T for subsequent death in end-stage renal disease. Circulation. 2002;106(23):2941 2945. 18. Pfortmueller CA, Funk GC, Marti G, et al. Diagnostic performance of highsensitive troponin T in patients with renal insufficiency. Am J Cardiol. 2013; 112(12):1968 1972. 19. Wang AY, Wai-Kei LC. The diagnostic utility of cardiac biomarkers in dialysis patients. Semin Dial. 2012;25(4):388 396. 20. Satyan S, Light RP, Agarwal R. Relationships of N-terminal pro-bnatriuretic peptide and cardiac troponin T to left ventricular mass and function and mortality in asymptomatic hemodialysis patients. Am J Kidney Dis. 2007; 50(6):1009 1019. 21. Conway B, McLaughlin M, Sharpe P, Harty J. Use of cardiac troponin T in diagnosis and prognosis of cardiac events in patients on chronic haemodialysis. Nephrol Dial Transplant. 2005;20(12):2759 2764. 22. Lippi G, Tessitore N, Montagnana M, Salvagno GL, Lupo A, Guidi GC. Influence of sampling time and ultrafiltration coefficient of the dialysis membrane on cardiac troponin I and T. Arch Pathol Lab Med. 2008;132(1):72 76. 23. Farkouh ME, Robbins MJ, Zafar MU, et al. Association between troponin I levels and mortality in stable hemodialysis patients. Am J Med. 2003;114(3):224 226. 24. Kumar N, Michelis MF, DeVita MV, Panagopoulos G, Rosenstock JL. Troponin I levels in asymptomatic patients on haemodialysis using a highsensitivity assay. Nephrol Dial Transplant. 2011;26(2):665 670. 25. Sosnov J, Lessard D, Goldberg RJ, Yarzebski J, Gore JM. Differential symptoms of acute myocardial infarction in patients with kidney disease: a community-wide perspective. Am J Kidney Dis. 2006;47(3):378 384. 26. Herzog CA, Littrell K, Arko C, Frederick PD, Blaney M. Clinical characteristics of dialysis patients with acute myocardial infarction in the United States: a collaborative project of the United States Renal Data System and the National Registry of Myocardial Infarction. Circulation. 2007;116(13):1465 1472. CAP16 Abstract Program Submission Dates Announced Abstract and case study submissions to the College of American Pathologists (CAP) 2016 Abstract Program will be accepted beginning on Friday, January 8 through 5 p.m. Central time Friday, March 11, 2016. Accepted submissions will appear on the Archives of Pathology & Laboratory Medicine Web site as a Web-only supplement to the September 2016 issue. The CAP16 meeting will be held from September 25 to 28 in Las Vegas, Nevada. Visit the CAP16 Web site (www.cap.org/cap16) for additional abstract program information as it becomes available. 80 Arch Pathol Lab Med Vol 140, January 2016 High-Sensitivity ctnt Assay for AMI in Hemodialysis Patients Huang et al