- Email: [email protected]
Cardiology electrocardiogram overreads rarely influence patient care outcome
American Journal of Emergency Medicine 32 (2014) 1311–1314
Contents lists available at ScienceDirect
American Journal of Emergency Medicine journal homepage: www.elsevier.com/locate/ajem
Original Contribution
Cardiology electrocardiogram overreads rarely influence patient care outcome Lawrence Proano, MD a,⁎, Andrew Sucov, MD b, Robert Woolard, MD c a b c
Department of Emergency Medicine, Rhode Island Hospital, Providence, RI St. Anne's Hospital, Fall River, MA Texas Tech University, Health Sciences Center, Lubbock, TX
a r t i c l e
i n f o
Article history: Received 17 July 2014 Accepted 28 July 2014
a b s t r a c t Objective: The value of electrocardiogram (ECG) overreads of emergency department (ED) tracings have been questioned in the literature. This review was designed to assess the validity of this criticism. Methods: In this university teaching hospital ED, following the normal quality assurance protocol, each abnormal ECG is reviewed the following day against the corresponding chart; and if the official reading from cardiology is discordant from the initial clinical one, the patient and/or their physician is contacted. If necessary, the patient is instructed to return to the ED or to their private physician's office. This study is a retrospective review of those ECG overreads for a 21-month period, as well as a summary of those patients who required follow-up care. Results: There were 38,490 patients seen with ECGs performed during the study interval. Of these, 16,011 were discharged and 22,479 were admitted from a total patient volume of 117,407. Of those 16,011 patients discharged, follow-up was deemed necessary in 22 patients whose official readings were discordant from the interpretation of the original clinician. Three patients were lost to follow-up (no phone, no address). Review of the tracings and patient/physician follow-up of the 19 remaining patients resulted in a significant change of therapy in 2 patients (admission). The remainder of the abnormal tracings were deemed, after patient or private physician follow-up, to be not significant or to mandate no change in management. Conclusion: Official cardiology overreads seldom affect the clinical outcome of patient care delivered in the ED setting. © 2014 Elsevier Inc. All rights reserved.
1. Introduction The Joint Commission on Accreditation of HealthCare Organizations mandates that hospitals under their jurisdiction provide a mechanism for quality assurance (QA) interpretation of electrocardiograms (ECGs) by qualified physicians with interpretation privileges [1]. The value of such reviews, particularly regarding review of ECGs, remains unresolved. The accuracy of interpretation of ECGs by emergency physicians (EPs) has been the subject of some review and criticism [2–5]. These studies suggest that there is value in QA overreads by cardiologists. However, most of these studies focus on the number of discrepancies or rates of discordance in printed readings of cardiologists vs clinicians, and not on the end point of actual patient outcome. As a result, they also do not address the issue of cost or benefit of these activities. In addition, they do not address the issue of compensation and reimbursement for these activities by cardiologists and EPs, or how this should be allocated on a basis of relative worth. ⁎ Corresponding author. Rhode Island Hospital, Department of Emergency Medicine, 593 Eddy St, Claverick Bldg Room 201, Providence, RI 02903. Tel.: +1 401 444 5826; fax: +1 401 444 2922. E-mail address: [email protected] (L. Proano). http://dx.doi.org/10.1016/j.ajem.2014.07.041 0735-6757/© 2014 Elsevier Inc. All rights reserved.
This retrospective study of 21 months of ECGs obtained on 16,011 patients discharged from the ED, their initial ED interpretation, and final cardiology review was done to assess the efficacy and usefulness of these follow-up measures. These considerations lead to questions regarding cost, benefit, equity in compensation, and ultimate impact on the quality of patient care delivered. 2. Materials and methods This study was performed at a large urban teaching hospital, which is a Level I trauma center, with an annual volume of approximately 75,000. The emergency department (ED) is staffed by PGY 1-4 emergency medicine residents, PGY 2-3 internal medicine housestaff, and PGY 1 OB-GYN residents, who are supervised by board-certified EPs who serve as faculty to the emergency medicine residency program. The faculty provide 24-hour coverage, with double or triple coverage from 7:00 AM until 1:00 AM, for a total of 48 hours of daily coverage. Patients are seen primarily either by attending level EPs or by housestaff in conjunction with oversight and supervision by an attending EP. Electrocardiograms performed on patients seen in the ED are overread by cardiologists on the ECG reading team, usually during the subsequent 24-hour period. These overreads are sent to the ED for reconciliation with the original readings done by the original treating
1312
L. Proano et al. / American Journal of Emergency Medicine 32 (2014) 1311–1314
clinicians. These review and comparison are generally conducted in the morning by an ED clinician who reviews the official interpretations of all ECGs performed during the prior day of patient care. This reconciliation process consists of comparing the official reading by the cardiologist with that of the clinicians (resident or midlevel provider, and supervising attending) who originally saw the patient and interpreted the ECG. Discrepancies between the final official reading and that during the patient’s original visit that are judged by the reviewer to be potentially significant are flagged to initiate callback and/or contact with that patient's private physician. A tracing is not flagged if the patient was admitted (in which case the official overread will be sent to the floor where the patient is being cared for), if there is concordance with the reading noted on the original medical record, or if the changes are of a trivial nature. A log is maintained to track all callbacks and follow-up activities. During the cardiology overread process, in the event of significant, urgent, and life-threatening findings, the cardiologists will call the ED to allow more prompt action rather than wait for the routine process to run its course. However, even with this system, the patient rarely is still in the ED for treatment when the notification is made. A retrospective review was conducted on all those tracings, formal readings, reviews, and callbacks during this 21-month study period. The review was overseen and data checking was performed by the authors (LP). The study was accomplished using the QA log retrospectively reviewing the follow-up performed in cases where patients were called back to the ED or were reassessed by their primary physicians after the ED visit. In cases where patients were called back to the ED and/or admitted to the hospital on the callback visit, these charts were reviewed for outcome and any change in management. In cases where it was deemed that the patient did not need immediate reevaluation, the patient was instructed to follow up with their primary physician. In these cases, the primary care physicians were contacted to ascertain what, if any, changes resulted in this callback. This study was submitted to and approved by the hospital’s Institutional Review Board. 3. Results During the study period, 117,407 patients were seen in the ED. During that same interval, there were 38,490 ECGs performed in the ED. There were approximately 22,479 ECGs obtained in patients who were admitted and 16,011 ECGs obtained in patients who were ultimately discharged from the ED. In the study sample of 16,011 discharged patients during the study interval, patient follow-up was found to be necessary in 22 patients whose official readings were discordant from the interpretation of the original clinician (Table). Three patients were lost to follow-up (no phone, no address). Review of the tracings and patient/physician follow-up of the 19 remaining patients resulted in a significant change of therapy in 2 patients. The remainder of the 17 abnormal tracings were deemed, after patient or private physician follow-up (by phone or in the office), to be not significant after all or to mandate no change in management. The 2 patients who were called back to the ED for reevaluation were admitted, one for potential ischemic changes and the other for new-onset atrial flutter. Although there was a change in management in the sense that these 2 callbacks required admission, there was no ultimate change in outcome, as the workup of these 2 patients revealed no evidence of ischemic cardiac disease, and they sustained no untoward outcomes. Subsequently, there have been no reports of medicolegal cases arising from any missed cardiac events during the study period from the patients whose charts were flagged for review. 4. Discussion Studies have looked at the ECG review process with variable interpretations on its value in patient care and reducing medical error.
Table Summary of recalled patients with abnormal ECGs Case Initial reading # 1 2 3 4 5
6 7 8 9 10 11 12 13 14 15
16
17
18
19 20 21 22
ECG overread
New atrial flutter (patient recalled for admission) Not noted on ED medical record ECG with first-degree AVB, new since last tracing Not noted on ED medical record NSSTT changes on ECG Not noted on ED medical record; Nonspecific ST-T abnormalities; Dx: syncope MI could be contributory Not noted on ED medical record; Compared with ECG of 5/29 Dx: constipation/UTI multifocal ventricular premature complexes have increased Not noted on ED medical record. Diffuse ST-T abnormalities; Dx: acute L/S sprain UTI possible ischemia ECG: NSR @ 79 no ST inc or T Inferior Q waves noted changes nl axis no sig q waves Not noted on ED medical record Left ventricular hypertrophy w/ secondary repolarization abnormalities ECG: no change from 10/97; ? (L) ventricular hypertrophy, Nonspec. Dx: palpitations ST-T abnormalities Not noted on ED medical record Nonspecific ST-T abnormalities,? MI (patient recalled for admission) Not noted on ED medical Atrial fibrillation w/ moderate record; left AMA; Dx: dyspnea ventricular response ECG read as normal; Compared w/ prev., ectopic Dx: esophageal spasm atrial rhythm persists Not noted on ED medical record. Compared with prior ECG of 2/11/98 s/p CABG 1 wk prior to ED visit ST-T abnormalities are newly noted. Not noted on ED medical record Diffuse ST-T changes consist. W/ MI Not noted on ED medical record; ST-segment elevation, distinction Dx: admitted for syncope between myocardial injury and a normal variant requires clinical correlation. Compared w/previous ECG, ST-segment NSR @ 75 with early repol; elevation has increased and borderline no change from old; poor R wave progression is Dx: hypertension newly noted. Not noted on ED medical record. Nonspecific ST-T abnormalities, MI could Dx: hypergly/insulin reaction be contributory. Compared w/ rev. ECG, ST segment elevation is newly noted. ECG documented as nl on ED Possible (R) atrial enlargement, record; Dx: esophageal spasm nondiagnostic Q Waves in inferior region. Significance depends on clinical correlation. Not noted on ED medical record Nonspecific ST-T abnormal, MI could be contributory Not noted on ED medical record; Diffuse T wave abnormalities Dx: myofascial strain consistent w/ischemia Not noted on ED medical record; Decreased T wave in AVL, V5-V6 Dx: acute agitation/dementia Not noted on ED medical record Nonspecific T wave flattening. MI may be contributory. Not noted on ED medical record
AVB, 1st degree AV Block ; NSSTT, Nonspecific ST-T.
Guidelines have been put forth in an effort to improve patient safety and quality of care [6]. Some of the early studies suggested value to this process of cardiology overreads and intimated that the EPs often misread ECGs in the course of caring for ED patients. Jayes et al [3] examined 2320 patients with symptoms suggesting potential cardiac ischemia. In this series, of the tracings officially read as showing abnormal ST segments, the ED physicians read 41% of these as normal. Of the tracings officially read as showing abnormal T waves, the ED physicians read 36% of these as normal. Levy et al [5], as part of an analysis of ED QA review, found that, in 74,760 charts reviewed, 21 patients required callback for ECG misinterpretation, and concluded that such reviews provide for an effective continuing evaluation of patient care in an area with a high volume of patients and a rotating medical staff. However, more recent studies support the concept that, although there may not be complete concordance in interpretation between the original treating physician and the cardiology QA reviewer, the discrepancies found are generally of no clinical import. For example, Kuhn et al [7] did a retrospective review of 400 cases and found 33
L. Proano et al. / American Journal of Emergency Medicine 32 (2014) 1311–1314
(8%) to have discordant readings and abnormal findings on ECGs in patients discharged from the ED. However, in only 2 of these patients were these findings felt to be of clinical significance. Westdrop et al [8] reviewed 1716 ECGs of discharged ED patients, finding a larger disparity of 58% between the reading of the treating EP and the overreading cardiologist, although only 25 (1.5%) were felt to be clinically important (representing missed ischemia, infarct, or abnormal rhythm). Todd et al [9] performed a prospective review of 1000 consecutive ECGs, using a third cardiologist reading as the criterion standard, comparing the concordance of readings of the original EP and the hospital cardiologist overreader. This series found only 8 patients (0.8%) with significantly discordant findings to warrant chart review, with no ultimate change in medical care in any patient. Brady et al [10] performed a retrospective review of ECGs performed in adult chest pain patients presenting to a university teaching hospital over a 3-month period. The gross rate of misinterpretation was 12 (5.9%) of 202, the majority of which were clinically insignificant. Their conclusion was that EPs show a low rate of misinterpretation of ECGs in the setting of chest pain and ST elevation and that the clinical consequences of this misinterpretation are minimal. Brady et al [11] looked further at the question of the ability of EPs to identify the cause of ST-segment elevation, using a questionnaire with a series of 11 ECGs, read by 458 EPs, ranging from PGY 2 through the attending level. The overall rate of correct interpretation of the study ECGs was 95%. A recent similar study of 21,872 ECGs confirmed the contention of Brady et al, finding out of 120 charts flagged for discrepancy checks only 33 cases (0.015%) requiring some type of follow-up [12]. Several other studies, using various methodologies, have reached similar conclusions [13–15,10]. These recent studies make a strong case that, in the majority of cases where there is a discordance between the original and final reading, there is little clinical significance. This may reflect the fact that the ECG as a tool is not very sensitive or specific in terms of detecting acute ischemia. As a diagnostic tool, the ECG provides a specific diagnosis in only about 5% of ED patients with chest pain [2]. Lee et al [16] and Rouan et al [17] found that 1% to 4% of patients with an absolutely normal ECG had a final hospital diagnosis of acute myocardial infarction (AMI) or unstable angina. Furthermore, those with nonspecific ECG abnormalities experienced AMI in 4% of cases. The potential significance of an ECG with even nonspecific findings in patients who are symptom-free, either by spontaneous resolution or by medical intervention, has been documented [18]. Patients with nondiagnostic ECGs who later develop AMI during that hospitalization are often symptom-free on initial presentation. Even the duration of time elapsed from chest pain onset in patients with normal ECGs does not assist greatly in ruling out the possibility of AMI in chest pain patients with single ECG study. Although the negative predictive value is high, it is still not 100% even at 12 hours after the onset of chest discomfort [19]. Despite the limitations of the ECG as a diagnostic tool to evaluate cardiac ischemia, it remains as the primary criterion in decision making regarding administration of thrombolytic therapy [20]. Sharkey et al [21] demonstrated the potential for overliberal interpretation of ECG ST-segment elevations in a series of 93 patients receiving thrombolytics. In this study, 10 patients (11%) who received these agents ultimately were shown to not have sustained a myocardial infarction. Ho et al [22] reported a sensitivity of only 78% for EP interpretation of ECGs on 236 patients who were candidates for thrombolytic therapy. Larson et al [23] assessed the issue of “False-positive cardiac cath lab activations. As evidenced by this study, EPs can read ECGs with a high degree of accuracy; and in the vast majority of cases, overreading of these tracings by cardiologists does not influence patient management or clinical outcome. Even with the QA review system currently in place, 2% to 5% of patients with acute coronary syndromes are inadvertently discharged home after ED visits [24,25]. In addition, a study by Pope et al [26] reviewed the subject of missed AMI in the ED and found that
1313
2.1% of patients with AMI were missed and 2.3% of those with unstable angina. The patients with missed cardiac events tended to be younger and have normal or nonspecific ECGs. This is clearly not a subset of medical error that can be addressed by simply using post hoc interventions. If indeed the criterion standard for ECG interpretation is to be a cardiologist, the only way to apply this standard in a way that it is of clinical value would be for this interpretation to be done contemporaneously with patient care. The multifactorial nature of the acute coronary syndrome is now recognized, and it is clear that the ECG plays only a small component in the diagnosis and risk stratification of such patients [27]. This study supports previous data confirming the relative lack of impact of post hoc interpretations on patient care outcomes. A separate consideration is how reimbursement for ECG interpretation should be allocated or divided among physicians who perform this function. This has been a subject of controversy for EPs for many years [28,29]. The American College of Emergency Physicians (ACEP) has lobbied for support on this contentious issue with Medicare and the Health Care Finance Administration (HCFA), which oversees the Medicare program. ACEP's position to HCFA was that a cardiologist who performs an ECG reading does so at a date, time, and location far removed from the actual patient care, whereas the treating physician of record (the EP) who reads the tracing in real-time receives no compensation. This was argued to be inherently inequitable. Emergency Medicine received support from HCFA in their “Final Rule on X-Ray/ECG Interpretations” that they implemented on January 1, 1996 [30]. In their guidelines, they ruled that Medicare will pay separately for only one interpretation of an ECG furnished to an ED patient. Their stipulation was that, to be considered an official interpretation, it must include a written report, which they distinguish from a simple “review” of the ECG because the review is already included in the ED visit payment. Recognizing the merits of ACEP's argument that online reading was a major variable in determining the value of the interpretation to their subscribers, HCFA noted that, in the event that they receive multiple bills for the same interpretation, they instructed their carriers to cease consideration of physician specialty in determining payment, and pay only for the interpretation and report that directly contributed to the diagnosis and treatment of the individual beneficiary. Preferential payment to a cardiologist would only be made if the interpretation of the procedure was performed contemporaneously with the treatment and diagnosis of the beneficiary. However, HCFA avoided intervening between hospitals, EPs, and cardiologists in settling disputes on this issue. Instead, their ruling simply encouraged hospitals to work with their medical staffs to ensure that only one claim per interpretation is submitted. This study was subject to several limitations. The study was retrospective in nature. The only charts that were reviewed were those that were flagged as part of the QA process. In addition, the criteria for ECG concordance or discrepancy were not explicit; so there could have been some cases with adverse outcomes missed. However, no litigation or risk management issues have subsequently arisen to date on these cases involving chest pain or missed ischemic events. Similarly, there have been no such issues involving cardiac events on the nonreviewed cases for the study period. The review was limited to findings of significant changes in medical therapy and did not address best practices or the concept of what constitutes the highest potential quality care. The ECG callback criteria were not explicit and were subject to the judgment of the reviewing QA clinician. Patients who were admitted to the hospital rather than discharged did not receive follow-up evaluation for what was felt to be minor or undocumented findings because the in-house physicians would receive the same overread report. We therefore cannot make conclusions on the value or lack of value of overreads for admitted patients. Many of the callbacks were for missing or poorly documented readings of the initial tracing.
1314
L. Proano et al. / American Journal of Emergency Medicine 32 (2014) 1311–1314
Finally, the findings may not be applicable to other institutions with different demographics, patient mix, or staffing patterns. This particular institution is an adult-only Level 1 trauma center, with an emergency medicine residency program, with 24-hour board-certified emergency medicine attending coverage. The ECGs generated have preliminary computerized interpretations as they are obtained, whereas other institutions may not use ECG machines with this feature. Future studies might investigate whether real-time interpretation of ECGs by cardiologists while patients are being treated has a positive impact on the quality of delivery of patient care. From this study, we conclude that cardiology overreads seldom affect the clinical outcome for patients treated and released in our ED setting. Post hoc overreads are mostly done long after they are useful to the patient care provided and are unable to decrease the national figure of 2% to 5% of patients with acute coronary syndromes being inadvertently discharged home after ED visits. Continuing efforts should be directed toward more equitable distribution of reimbursement for the interpretations that are performed. Reimbursement for ECG interpretation should be directed to EPs, in accordance with HCFA's “Concurrent Interpretation” guidelines, rather than to cardiologists who perform delayed readings with minimal contemporaneous impact. References [1] 2000 Accreditation manual for hospitals: the official handbook. Oakbrook Terrace, Illinois: Joint Commission on Accreditation of Healthcare Organizations; 2000. [2] Lee TH, Rouan GW, Weisberg MC, Brand DA, Cook EF, Acampora D, et al. Sensitivity of routine clinical criteria for diagnosing myocardial infarction within 24 hours of hospitalization. Ann Intern Med 1987;106:181–6. [3] Jayes RL, Larsen GC, Beshansky JR, D'Agostino RB, Selker HP. Physician electrocardiogram reading in the emergency department: accuracy and effect on triage decisions: findings from a multicenter study. J Gen Intern Med 1992;7:387–92. [4] Zappa MJ, Smith M, Li S. How well do emergency physicians interpret ECGs (abstract). Ann Emerg Med 1991;20:463. [5] Levy R, Goldstein R, Trott A. Approach to quality assurance in an emergency department: a one year review. Ann Emerg Med 1984;13:166–9. [6] Mele PF. The ECG, dilemma: guidelines on improving interpretation. J Healthc Risk Manag 2008;28(2):27–31. http://dx.doi.org/10.1002/jhrm.5600280205. [7] Kuhn M, Morgan MT, Hoffman JR. Quality assurance in the emergency department: evaluation of the ECG review process. Ann Emerg Med 1992;21:10–5. [8] Westdrop EJ, Gratton MC, Watson WA. Emergency department interpretation of electrocardiograms. Ann Emerg Med 1992;21:541–4. [9] Todd KH, Hoffman JR, Morgan MT. Effect of cardiologist ECG review on emergency department practice. Ann Emerg Med 1996;27:16–21. [10] Brady WJ, Perron A, Ullman E. Errors in emergency physician interpretation of STsegment elevation in emergency department chest pain patients. Acad Emerg Med 2000;7:1256–60. [11] Brady WJ, Perron AD, Chan T. Electorcardiographic ST-segment elevation: correct identification of acute myocardial infarction (AMI) and non-AMI syndromes by emergency physicians. Acad Emerg Med 2001;8:349–60.
[12] Santoro JP, Blank FS, Smithline H. Follow-up of discrepancies in x-ray and electrocardiogram interpretations, and positive laboratory results. J Emerg Med 2001;20: 315–9. [13] McCabe JM, Armstrong EJ, Ku I, Kulkarni A, Hoffmayer KS, Bhave PD, et al. Physician accuracy in interpreting potential ST-segment elevation myocardial infarction electrocardiograms. J Am Heart Assoc 2013;2(5):e000268. http://dx. doi.org/10.1161/JAHA.113.000268. [14] Feldman JA, Bernard S, Mitchell P, Rebholz CM. Effects of cardiology review of the electrocardiogram in patients with suspected acute coronary syndromes. Am J Emerg Med 2011;29(3):309–315.e2. http://dx.doi.org/10.1016/j.ajem.2010.09. 023 [Epub 2010 Dec 15]. [15] Eken C, Goksu E, Eray O, Yalcinkaya S. The consistency of emergency physicians' and cardiologists' ECG interpretation and likelihood classification of chest pain patients. Int J Clin Pract 2006;60(10):1194–7 [Epub 2006 Mar 27]. [16] Lee TH, Rouan GW, Weisberg MC, Goldman L. Clinical characteristics and natural history of patients with acute myocardial infarction sent home from the emergency room. Am J Cardiol 1987;60:219–24. [17] Rouan GW, Lee TH, Cook EF, Brand DA, Weisberg MC, Goldman L. Clinical characteristics and outcome of acute myocardial infarction in patients with initially normal or nonspecific electrocardiograms. Am J Cardiol 1989;63:772–6. [18] Brady WB, Roberts D, Morris F. The Nondiagnostic ECG in the chest pain patient: normal and nonspecific initial ECG presentations of acute MI. Am J Emerg Med 1999;17:394–7. [19] Singer AJ, Brogan GRX, Valentine SM, Khan S, Hollander JE. Effect of duration of symptom onset on the negative predictive value of a normal ECG for exclusion of acute myocardial infarction. Ann Emerg Med 1997;29:575–9. [20] Selker HP, Zalenski RJ, Antman EM, Aufderheide TP, Bernard SA, Bonow RO, et al. An evaluation of technologies for identifying acute cardiac ischemia in the emergency department: a report from a national heart attack alert program working group. Ann Emerg Med 1997;29:13–87. [21] Sharkey SW, Berger CR, Brunette DD, Henry TD. Impact of the electrocardiogram on the delivery of thrombolytic therapy for acute myocardial infarction. Am J Cardiol 1994;73:550–3. [22] Ho MT, Kudenchuk PJ, Eisenberg MS. Patient selection for thrombolytic therapy: emergency physicians versus electrocardiographer (abstract). J Am Coll Cardiol 1990;15:192A. [23] Larson David M, Menssen Katie M, Sharkey Scott W, Duval Sue, Schwartz Robert S, Harris James, et al. “False-positive” cardiac catheterization laboratory activation among patients with suspected ST-segment elevation myocardial infarction. JAMA 2007;298(23):2754–60. http://dx.doi.org/10.1001/jama.298.23.2754. [24] Rusnak RA, Stair TO, Hansen K, Fastow JS. Litigation against emergency physicians: common features in cases of missed myocardial infarction. Ann Emerg Med 1989; 18:1029–34. [25] McCarthy BD, Beshansky JR, D'Agostino RB, Selker HP. Missed diagnosis of acute myocardial infarction in the emergency department: results from a multicenter study. Ann Emerg Med 1993;22:579–82. [26] Pope JH, Aufderheide TP, Ruthazer R, Woolard RH, Feldman JA, Beshansky JR, et al. Missed diagnosis of acute cardiac ischemia in the emergency department. N Engl J Med 2000;342:1163–70. [27] Storrow AB, Gibler WB. Chest pain centers: diagnosis of acute coronary syndromes. Ann Emerg Med May 2000;35:449–61. [28] Manag ED. The controversy over billing for ECG/rhythm strip interpretations in the ED. 2012;24(9):105–7. [29] Wu T, Bell MR, Blakeman JR, Edwards I, Mallon WK. Reimbursement for emergency department electrocardiography and radiograph interpretations: what is it worth for the emergency physician. West J Emerg Med 2009;10(3): 178–83. [30] Reference obtainable on internet at. http://www.acep.org/library/index.cfm/id/ 293.htm.
Contents lists available at ScienceDirect
American Journal of Emergency Medicine journal homepage: www.elsevier.com/locate/ajem
Original Contribution
Cardiology electrocardiogram overreads rarely influence patient care outcome Lawrence Proano, MD a,⁎, Andrew Sucov, MD b, Robert Woolard, MD c a b c
Department of Emergency Medicine, Rhode Island Hospital, Providence, RI St. Anne's Hospital, Fall River, MA Texas Tech University, Health Sciences Center, Lubbock, TX
a r t i c l e
i n f o
Article history: Received 17 July 2014 Accepted 28 July 2014
a b s t r a c t Objective: The value of electrocardiogram (ECG) overreads of emergency department (ED) tracings have been questioned in the literature. This review was designed to assess the validity of this criticism. Methods: In this university teaching hospital ED, following the normal quality assurance protocol, each abnormal ECG is reviewed the following day against the corresponding chart; and if the official reading from cardiology is discordant from the initial clinical one, the patient and/or their physician is contacted. If necessary, the patient is instructed to return to the ED or to their private physician's office. This study is a retrospective review of those ECG overreads for a 21-month period, as well as a summary of those patients who required follow-up care. Results: There were 38,490 patients seen with ECGs performed during the study interval. Of these, 16,011 were discharged and 22,479 were admitted from a total patient volume of 117,407. Of those 16,011 patients discharged, follow-up was deemed necessary in 22 patients whose official readings were discordant from the interpretation of the original clinician. Three patients were lost to follow-up (no phone, no address). Review of the tracings and patient/physician follow-up of the 19 remaining patients resulted in a significant change of therapy in 2 patients (admission). The remainder of the abnormal tracings were deemed, after patient or private physician follow-up, to be not significant or to mandate no change in management. Conclusion: Official cardiology overreads seldom affect the clinical outcome of patient care delivered in the ED setting. © 2014 Elsevier Inc. All rights reserved.
1. Introduction The Joint Commission on Accreditation of HealthCare Organizations mandates that hospitals under their jurisdiction provide a mechanism for quality assurance (QA) interpretation of electrocardiograms (ECGs) by qualified physicians with interpretation privileges [1]. The value of such reviews, particularly regarding review of ECGs, remains unresolved. The accuracy of interpretation of ECGs by emergency physicians (EPs) has been the subject of some review and criticism [2–5]. These studies suggest that there is value in QA overreads by cardiologists. However, most of these studies focus on the number of discrepancies or rates of discordance in printed readings of cardiologists vs clinicians, and not on the end point of actual patient outcome. As a result, they also do not address the issue of cost or benefit of these activities. In addition, they do not address the issue of compensation and reimbursement for these activities by cardiologists and EPs, or how this should be allocated on a basis of relative worth. ⁎ Corresponding author. Rhode Island Hospital, Department of Emergency Medicine, 593 Eddy St, Claverick Bldg Room 201, Providence, RI 02903. Tel.: +1 401 444 5826; fax: +1 401 444 2922. E-mail address: [email protected] (L. Proano). http://dx.doi.org/10.1016/j.ajem.2014.07.041 0735-6757/© 2014 Elsevier Inc. All rights reserved.
This retrospective study of 21 months of ECGs obtained on 16,011 patients discharged from the ED, their initial ED interpretation, and final cardiology review was done to assess the efficacy and usefulness of these follow-up measures. These considerations lead to questions regarding cost, benefit, equity in compensation, and ultimate impact on the quality of patient care delivered. 2. Materials and methods This study was performed at a large urban teaching hospital, which is a Level I trauma center, with an annual volume of approximately 75,000. The emergency department (ED) is staffed by PGY 1-4 emergency medicine residents, PGY 2-3 internal medicine housestaff, and PGY 1 OB-GYN residents, who are supervised by board-certified EPs who serve as faculty to the emergency medicine residency program. The faculty provide 24-hour coverage, with double or triple coverage from 7:00 AM until 1:00 AM, for a total of 48 hours of daily coverage. Patients are seen primarily either by attending level EPs or by housestaff in conjunction with oversight and supervision by an attending EP. Electrocardiograms performed on patients seen in the ED are overread by cardiologists on the ECG reading team, usually during the subsequent 24-hour period. These overreads are sent to the ED for reconciliation with the original readings done by the original treating
1312
L. Proano et al. / American Journal of Emergency Medicine 32 (2014) 1311–1314
clinicians. These review and comparison are generally conducted in the morning by an ED clinician who reviews the official interpretations of all ECGs performed during the prior day of patient care. This reconciliation process consists of comparing the official reading by the cardiologist with that of the clinicians (resident or midlevel provider, and supervising attending) who originally saw the patient and interpreted the ECG. Discrepancies between the final official reading and that during the patient’s original visit that are judged by the reviewer to be potentially significant are flagged to initiate callback and/or contact with that patient's private physician. A tracing is not flagged if the patient was admitted (in which case the official overread will be sent to the floor where the patient is being cared for), if there is concordance with the reading noted on the original medical record, or if the changes are of a trivial nature. A log is maintained to track all callbacks and follow-up activities. During the cardiology overread process, in the event of significant, urgent, and life-threatening findings, the cardiologists will call the ED to allow more prompt action rather than wait for the routine process to run its course. However, even with this system, the patient rarely is still in the ED for treatment when the notification is made. A retrospective review was conducted on all those tracings, formal readings, reviews, and callbacks during this 21-month study period. The review was overseen and data checking was performed by the authors (LP). The study was accomplished using the QA log retrospectively reviewing the follow-up performed in cases where patients were called back to the ED or were reassessed by their primary physicians after the ED visit. In cases where patients were called back to the ED and/or admitted to the hospital on the callback visit, these charts were reviewed for outcome and any change in management. In cases where it was deemed that the patient did not need immediate reevaluation, the patient was instructed to follow up with their primary physician. In these cases, the primary care physicians were contacted to ascertain what, if any, changes resulted in this callback. This study was submitted to and approved by the hospital’s Institutional Review Board. 3. Results During the study period, 117,407 patients were seen in the ED. During that same interval, there were 38,490 ECGs performed in the ED. There were approximately 22,479 ECGs obtained in patients who were admitted and 16,011 ECGs obtained in patients who were ultimately discharged from the ED. In the study sample of 16,011 discharged patients during the study interval, patient follow-up was found to be necessary in 22 patients whose official readings were discordant from the interpretation of the original clinician (Table). Three patients were lost to follow-up (no phone, no address). Review of the tracings and patient/physician follow-up of the 19 remaining patients resulted in a significant change of therapy in 2 patients. The remainder of the 17 abnormal tracings were deemed, after patient or private physician follow-up (by phone or in the office), to be not significant after all or to mandate no change in management. The 2 patients who were called back to the ED for reevaluation were admitted, one for potential ischemic changes and the other for new-onset atrial flutter. Although there was a change in management in the sense that these 2 callbacks required admission, there was no ultimate change in outcome, as the workup of these 2 patients revealed no evidence of ischemic cardiac disease, and they sustained no untoward outcomes. Subsequently, there have been no reports of medicolegal cases arising from any missed cardiac events during the study period from the patients whose charts were flagged for review. 4. Discussion Studies have looked at the ECG review process with variable interpretations on its value in patient care and reducing medical error.
Table Summary of recalled patients with abnormal ECGs Case Initial reading # 1 2 3 4 5
6 7 8 9 10 11 12 13 14 15
16
17
18
19 20 21 22
ECG overread
New atrial flutter (patient recalled for admission) Not noted on ED medical record ECG with first-degree AVB, new since last tracing Not noted on ED medical record NSSTT changes on ECG Not noted on ED medical record; Nonspecific ST-T abnormalities; Dx: syncope MI could be contributory Not noted on ED medical record; Compared with ECG of 5/29 Dx: constipation/UTI multifocal ventricular premature complexes have increased Not noted on ED medical record. Diffuse ST-T abnormalities; Dx: acute L/S sprain UTI possible ischemia ECG: NSR @ 79 no ST inc or T Inferior Q waves noted changes nl axis no sig q waves Not noted on ED medical record Left ventricular hypertrophy w/ secondary repolarization abnormalities ECG: no change from 10/97; ? (L) ventricular hypertrophy, Nonspec. Dx: palpitations ST-T abnormalities Not noted on ED medical record Nonspecific ST-T abnormalities,? MI (patient recalled for admission) Not noted on ED medical Atrial fibrillation w/ moderate record; left AMA; Dx: dyspnea ventricular response ECG read as normal; Compared w/ prev., ectopic Dx: esophageal spasm atrial rhythm persists Not noted on ED medical record. Compared with prior ECG of 2/11/98 s/p CABG 1 wk prior to ED visit ST-T abnormalities are newly noted. Not noted on ED medical record Diffuse ST-T changes consist. W/ MI Not noted on ED medical record; ST-segment elevation, distinction Dx: admitted for syncope between myocardial injury and a normal variant requires clinical correlation. Compared w/previous ECG, ST-segment NSR @ 75 with early repol; elevation has increased and borderline no change from old; poor R wave progression is Dx: hypertension newly noted. Not noted on ED medical record. Nonspecific ST-T abnormalities, MI could Dx: hypergly/insulin reaction be contributory. Compared w/ rev. ECG, ST segment elevation is newly noted. ECG documented as nl on ED Possible (R) atrial enlargement, record; Dx: esophageal spasm nondiagnostic Q Waves in inferior region. Significance depends on clinical correlation. Not noted on ED medical record Nonspecific ST-T abnormal, MI could be contributory Not noted on ED medical record; Diffuse T wave abnormalities Dx: myofascial strain consistent w/ischemia Not noted on ED medical record; Decreased T wave in AVL, V5-V6 Dx: acute agitation/dementia Not noted on ED medical record Nonspecific T wave flattening. MI may be contributory. Not noted on ED medical record
AVB, 1st degree AV Block ; NSSTT, Nonspecific ST-T.
Guidelines have been put forth in an effort to improve patient safety and quality of care [6]. Some of the early studies suggested value to this process of cardiology overreads and intimated that the EPs often misread ECGs in the course of caring for ED patients. Jayes et al [3] examined 2320 patients with symptoms suggesting potential cardiac ischemia. In this series, of the tracings officially read as showing abnormal ST segments, the ED physicians read 41% of these as normal. Of the tracings officially read as showing abnormal T waves, the ED physicians read 36% of these as normal. Levy et al [5], as part of an analysis of ED QA review, found that, in 74,760 charts reviewed, 21 patients required callback for ECG misinterpretation, and concluded that such reviews provide for an effective continuing evaluation of patient care in an area with a high volume of patients and a rotating medical staff. However, more recent studies support the concept that, although there may not be complete concordance in interpretation between the original treating physician and the cardiology QA reviewer, the discrepancies found are generally of no clinical import. For example, Kuhn et al [7] did a retrospective review of 400 cases and found 33
L. Proano et al. / American Journal of Emergency Medicine 32 (2014) 1311–1314
(8%) to have discordant readings and abnormal findings on ECGs in patients discharged from the ED. However, in only 2 of these patients were these findings felt to be of clinical significance. Westdrop et al [8] reviewed 1716 ECGs of discharged ED patients, finding a larger disparity of 58% between the reading of the treating EP and the overreading cardiologist, although only 25 (1.5%) were felt to be clinically important (representing missed ischemia, infarct, or abnormal rhythm). Todd et al [9] performed a prospective review of 1000 consecutive ECGs, using a third cardiologist reading as the criterion standard, comparing the concordance of readings of the original EP and the hospital cardiologist overreader. This series found only 8 patients (0.8%) with significantly discordant findings to warrant chart review, with no ultimate change in medical care in any patient. Brady et al [10] performed a retrospective review of ECGs performed in adult chest pain patients presenting to a university teaching hospital over a 3-month period. The gross rate of misinterpretation was 12 (5.9%) of 202, the majority of which were clinically insignificant. Their conclusion was that EPs show a low rate of misinterpretation of ECGs in the setting of chest pain and ST elevation and that the clinical consequences of this misinterpretation are minimal. Brady et al [11] looked further at the question of the ability of EPs to identify the cause of ST-segment elevation, using a questionnaire with a series of 11 ECGs, read by 458 EPs, ranging from PGY 2 through the attending level. The overall rate of correct interpretation of the study ECGs was 95%. A recent similar study of 21,872 ECGs confirmed the contention of Brady et al, finding out of 120 charts flagged for discrepancy checks only 33 cases (0.015%) requiring some type of follow-up [12]. Several other studies, using various methodologies, have reached similar conclusions [13–15,10]. These recent studies make a strong case that, in the majority of cases where there is a discordance between the original and final reading, there is little clinical significance. This may reflect the fact that the ECG as a tool is not very sensitive or specific in terms of detecting acute ischemia. As a diagnostic tool, the ECG provides a specific diagnosis in only about 5% of ED patients with chest pain [2]. Lee et al [16] and Rouan et al [17] found that 1% to 4% of patients with an absolutely normal ECG had a final hospital diagnosis of acute myocardial infarction (AMI) or unstable angina. Furthermore, those with nonspecific ECG abnormalities experienced AMI in 4% of cases. The potential significance of an ECG with even nonspecific findings in patients who are symptom-free, either by spontaneous resolution or by medical intervention, has been documented [18]. Patients with nondiagnostic ECGs who later develop AMI during that hospitalization are often symptom-free on initial presentation. Even the duration of time elapsed from chest pain onset in patients with normal ECGs does not assist greatly in ruling out the possibility of AMI in chest pain patients with single ECG study. Although the negative predictive value is high, it is still not 100% even at 12 hours after the onset of chest discomfort [19]. Despite the limitations of the ECG as a diagnostic tool to evaluate cardiac ischemia, it remains as the primary criterion in decision making regarding administration of thrombolytic therapy [20]. Sharkey et al [21] demonstrated the potential for overliberal interpretation of ECG ST-segment elevations in a series of 93 patients receiving thrombolytics. In this study, 10 patients (11%) who received these agents ultimately were shown to not have sustained a myocardial infarction. Ho et al [22] reported a sensitivity of only 78% for EP interpretation of ECGs on 236 patients who were candidates for thrombolytic therapy. Larson et al [23] assessed the issue of “False-positive cardiac cath lab activations. As evidenced by this study, EPs can read ECGs with a high degree of accuracy; and in the vast majority of cases, overreading of these tracings by cardiologists does not influence patient management or clinical outcome. Even with the QA review system currently in place, 2% to 5% of patients with acute coronary syndromes are inadvertently discharged home after ED visits [24,25]. In addition, a study by Pope et al [26] reviewed the subject of missed AMI in the ED and found that
1313
2.1% of patients with AMI were missed and 2.3% of those with unstable angina. The patients with missed cardiac events tended to be younger and have normal or nonspecific ECGs. This is clearly not a subset of medical error that can be addressed by simply using post hoc interventions. If indeed the criterion standard for ECG interpretation is to be a cardiologist, the only way to apply this standard in a way that it is of clinical value would be for this interpretation to be done contemporaneously with patient care. The multifactorial nature of the acute coronary syndrome is now recognized, and it is clear that the ECG plays only a small component in the diagnosis and risk stratification of such patients [27]. This study supports previous data confirming the relative lack of impact of post hoc interpretations on patient care outcomes. A separate consideration is how reimbursement for ECG interpretation should be allocated or divided among physicians who perform this function. This has been a subject of controversy for EPs for many years [28,29]. The American College of Emergency Physicians (ACEP) has lobbied for support on this contentious issue with Medicare and the Health Care Finance Administration (HCFA), which oversees the Medicare program. ACEP's position to HCFA was that a cardiologist who performs an ECG reading does so at a date, time, and location far removed from the actual patient care, whereas the treating physician of record (the EP) who reads the tracing in real-time receives no compensation. This was argued to be inherently inequitable. Emergency Medicine received support from HCFA in their “Final Rule on X-Ray/ECG Interpretations” that they implemented on January 1, 1996 [30]. In their guidelines, they ruled that Medicare will pay separately for only one interpretation of an ECG furnished to an ED patient. Their stipulation was that, to be considered an official interpretation, it must include a written report, which they distinguish from a simple “review” of the ECG because the review is already included in the ED visit payment. Recognizing the merits of ACEP's argument that online reading was a major variable in determining the value of the interpretation to their subscribers, HCFA noted that, in the event that they receive multiple bills for the same interpretation, they instructed their carriers to cease consideration of physician specialty in determining payment, and pay only for the interpretation and report that directly contributed to the diagnosis and treatment of the individual beneficiary. Preferential payment to a cardiologist would only be made if the interpretation of the procedure was performed contemporaneously with the treatment and diagnosis of the beneficiary. However, HCFA avoided intervening between hospitals, EPs, and cardiologists in settling disputes on this issue. Instead, their ruling simply encouraged hospitals to work with their medical staffs to ensure that only one claim per interpretation is submitted. This study was subject to several limitations. The study was retrospective in nature. The only charts that were reviewed were those that were flagged as part of the QA process. In addition, the criteria for ECG concordance or discrepancy were not explicit; so there could have been some cases with adverse outcomes missed. However, no litigation or risk management issues have subsequently arisen to date on these cases involving chest pain or missed ischemic events. Similarly, there have been no such issues involving cardiac events on the nonreviewed cases for the study period. The review was limited to findings of significant changes in medical therapy and did not address best practices or the concept of what constitutes the highest potential quality care. The ECG callback criteria were not explicit and were subject to the judgment of the reviewing QA clinician. Patients who were admitted to the hospital rather than discharged did not receive follow-up evaluation for what was felt to be minor or undocumented findings because the in-house physicians would receive the same overread report. We therefore cannot make conclusions on the value or lack of value of overreads for admitted patients. Many of the callbacks were for missing or poorly documented readings of the initial tracing.
1314
L. Proano et al. / American Journal of Emergency Medicine 32 (2014) 1311–1314
Finally, the findings may not be applicable to other institutions with different demographics, patient mix, or staffing patterns. This particular institution is an adult-only Level 1 trauma center, with an emergency medicine residency program, with 24-hour board-certified emergency medicine attending coverage. The ECGs generated have preliminary computerized interpretations as they are obtained, whereas other institutions may not use ECG machines with this feature. Future studies might investigate whether real-time interpretation of ECGs by cardiologists while patients are being treated has a positive impact on the quality of delivery of patient care. From this study, we conclude that cardiology overreads seldom affect the clinical outcome for patients treated and released in our ED setting. Post hoc overreads are mostly done long after they are useful to the patient care provided and are unable to decrease the national figure of 2% to 5% of patients with acute coronary syndromes being inadvertently discharged home after ED visits. Continuing efforts should be directed toward more equitable distribution of reimbursement for the interpretations that are performed. Reimbursement for ECG interpretation should be directed to EPs, in accordance with HCFA's “Concurrent Interpretation” guidelines, rather than to cardiologists who perform delayed readings with minimal contemporaneous impact. References [1] 2000 Accreditation manual for hospitals: the official handbook. Oakbrook Terrace, Illinois: Joint Commission on Accreditation of Healthcare Organizations; 2000. [2] Lee TH, Rouan GW, Weisberg MC, Brand DA, Cook EF, Acampora D, et al. Sensitivity of routine clinical criteria for diagnosing myocardial infarction within 24 hours of hospitalization. Ann Intern Med 1987;106:181–6. [3] Jayes RL, Larsen GC, Beshansky JR, D'Agostino RB, Selker HP. Physician electrocardiogram reading in the emergency department: accuracy and effect on triage decisions: findings from a multicenter study. J Gen Intern Med 1992;7:387–92. [4] Zappa MJ, Smith M, Li S. How well do emergency physicians interpret ECGs (abstract). Ann Emerg Med 1991;20:463. [5] Levy R, Goldstein R, Trott A. Approach to quality assurance in an emergency department: a one year review. Ann Emerg Med 1984;13:166–9. [6] Mele PF. The ECG, dilemma: guidelines on improving interpretation. J Healthc Risk Manag 2008;28(2):27–31. http://dx.doi.org/10.1002/jhrm.5600280205. [7] Kuhn M, Morgan MT, Hoffman JR. Quality assurance in the emergency department: evaluation of the ECG review process. Ann Emerg Med 1992;21:10–5. [8] Westdrop EJ, Gratton MC, Watson WA. Emergency department interpretation of electrocardiograms. Ann Emerg Med 1992;21:541–4. [9] Todd KH, Hoffman JR, Morgan MT. Effect of cardiologist ECG review on emergency department practice. Ann Emerg Med 1996;27:16–21. [10] Brady WJ, Perron A, Ullman E. Errors in emergency physician interpretation of STsegment elevation in emergency department chest pain patients. Acad Emerg Med 2000;7:1256–60. [11] Brady WJ, Perron AD, Chan T. Electorcardiographic ST-segment elevation: correct identification of acute myocardial infarction (AMI) and non-AMI syndromes by emergency physicians. Acad Emerg Med 2001;8:349–60.
[12] Santoro JP, Blank FS, Smithline H. Follow-up of discrepancies in x-ray and electrocardiogram interpretations, and positive laboratory results. J Emerg Med 2001;20: 315–9. [13] McCabe JM, Armstrong EJ, Ku I, Kulkarni A, Hoffmayer KS, Bhave PD, et al. Physician accuracy in interpreting potential ST-segment elevation myocardial infarction electrocardiograms. J Am Heart Assoc 2013;2(5):e000268. http://dx. doi.org/10.1161/JAHA.113.000268. [14] Feldman JA, Bernard S, Mitchell P, Rebholz CM. Effects of cardiology review of the electrocardiogram in patients with suspected acute coronary syndromes. Am J Emerg Med 2011;29(3):309–315.e2. http://dx.doi.org/10.1016/j.ajem.2010.09. 023 [Epub 2010 Dec 15]. [15] Eken C, Goksu E, Eray O, Yalcinkaya S. The consistency of emergency physicians' and cardiologists' ECG interpretation and likelihood classification of chest pain patients. Int J Clin Pract 2006;60(10):1194–7 [Epub 2006 Mar 27]. [16] Lee TH, Rouan GW, Weisberg MC, Goldman L. Clinical characteristics and natural history of patients with acute myocardial infarction sent home from the emergency room. Am J Cardiol 1987;60:219–24. [17] Rouan GW, Lee TH, Cook EF, Brand DA, Weisberg MC, Goldman L. Clinical characteristics and outcome of acute myocardial infarction in patients with initially normal or nonspecific electrocardiograms. Am J Cardiol 1989;63:772–6. [18] Brady WB, Roberts D, Morris F. The Nondiagnostic ECG in the chest pain patient: normal and nonspecific initial ECG presentations of acute MI. Am J Emerg Med 1999;17:394–7. [19] Singer AJ, Brogan GRX, Valentine SM, Khan S, Hollander JE. Effect of duration of symptom onset on the negative predictive value of a normal ECG for exclusion of acute myocardial infarction. Ann Emerg Med 1997;29:575–9. [20] Selker HP, Zalenski RJ, Antman EM, Aufderheide TP, Bernard SA, Bonow RO, et al. An evaluation of technologies for identifying acute cardiac ischemia in the emergency department: a report from a national heart attack alert program working group. Ann Emerg Med 1997;29:13–87. [21] Sharkey SW, Berger CR, Brunette DD, Henry TD. Impact of the electrocardiogram on the delivery of thrombolytic therapy for acute myocardial infarction. Am J Cardiol 1994;73:550–3. [22] Ho MT, Kudenchuk PJ, Eisenberg MS. Patient selection for thrombolytic therapy: emergency physicians versus electrocardiographer (abstract). J Am Coll Cardiol 1990;15:192A. [23] Larson David M, Menssen Katie M, Sharkey Scott W, Duval Sue, Schwartz Robert S, Harris James, et al. “False-positive” cardiac catheterization laboratory activation among patients with suspected ST-segment elevation myocardial infarction. JAMA 2007;298(23):2754–60. http://dx.doi.org/10.1001/jama.298.23.2754. [24] Rusnak RA, Stair TO, Hansen K, Fastow JS. Litigation against emergency physicians: common features in cases of missed myocardial infarction. Ann Emerg Med 1989; 18:1029–34. [25] McCarthy BD, Beshansky JR, D'Agostino RB, Selker HP. Missed diagnosis of acute myocardial infarction in the emergency department: results from a multicenter study. Ann Emerg Med 1993;22:579–82. [26] Pope JH, Aufderheide TP, Ruthazer R, Woolard RH, Feldman JA, Beshansky JR, et al. Missed diagnosis of acute cardiac ischemia in the emergency department. N Engl J Med 2000;342:1163–70. [27] Storrow AB, Gibler WB. Chest pain centers: diagnosis of acute coronary syndromes. Ann Emerg Med May 2000;35:449–61. [28] Manag ED. The controversy over billing for ECG/rhythm strip interpretations in the ED. 2012;24(9):105–7. [29] Wu T, Bell MR, Blakeman JR, Edwards I, Mallon WK. Reimbursement for emergency department electrocardiography and radiograph interpretations: what is it worth for the emergency physician. West J Emerg Med 2009;10(3): 178–83. [30] Reference obtainable on internet at. http://www.acep.org/library/index.cfm/id/ 293.htm.