Medical Policies - Medicine

Electrocardiographic Body Surface Mapping


Effective Date:10-15-2017



Electrocardiographic body surface mapping is considered experimental, investigational and/or unproven for the diagnosis or management of cardiac disorders, including acute coronary syndrome.


Electrocardiographic (ECG) body surface mapping (BSM) consists of an 80-lead disposable electrode array in the form of a vest and includes a conducting gel that is applied to the patient’s chest and back. The vest can be affixed to the patient in less than 5 minutes. This system displays clinical data in 3 forms; a colorimetric 3-D torso image, an 80-lead single beat view, and the 12-lead ECG. The colorimetric torso images are said to allow the practitioner to rapidly scan the heart for significant abnormalities.

Currently, in patients presenting to the emergency department with symptoms suggestive of myocardial ischemia, a standard 12-lead ECG is obtained. In the presence of ST segment elevation on the ECG, personnel are activated to respond in a timely manner to open a presumed coronary artery occlusion, either by mechanical means through balloon angioplasty, or medically, through intravenous thrombolytic drugs. The 12-lead ECG has a specificity of 94%, leading to relatively few erroneous interventions. However, the sensitivity is approximately 50%. These patients may be further stratified by scoring systems and time-sensitive cardiac enzymes, which may require up to 24 hours of monitored observation.

BSM is being considered as a method to assist in the rapid identification of patients who would benefit from earlier coronary artery intervention than is achieved utilizing current standard of care. The negative predictive value of the test, which has the potential to identify patients who do not require further evaluation with serial cardiac enzymes and clinical observation, is not currently receiving attention as a research topic.

Regulatory Status

In March 2002, the device PRIME ECG® (Verathon, Bothell, WA) was cleared for marketing by the United States (U.S.) Food and Drug Administration (FDA) through the 510(k) process. The FDA determined that the device was substantially equivalent to existing devices for use in recording of ECG signals on the body surface. As of July 2014, neither the PRIME ECG device nor its successor, the Heartscape™ 3D ECG System are being marketed in the U.S. Product code: DPS.


This policy review was originally created in 2007 and was regularly updated with searches of the

MEDLINE database. The most recent literature review was performed through August 21, 2016. Following is a summary of the key literature to date.

Assessment of a diagnostic technology focuses on the following parameters:

a) Technical performance;

b) Diagnostic accuracy (sensitivity, specificity, positive predictive value [PPV], negative predictive value [NPV]) in relevant clinical populations; and

c) Clinical utility (i.e., demonstration that the diagnostic information can be used to improve patient outcomes).

Technical Performance

The investigation of additional leads in electrocardiography (ECG) and body surface mapping (BSM) is not new. Patterns of electric potentials in normal subjects have been established, and the significance of abnormal signals has been explored over past decades. (1,2)

A 2006 publication describes the use of the 80-lead technique in the evaluation of patients with chest pain in the emergency department. (3) The authors comment that use of this approach has been hampered by slow acquisition time and the complexity of interpretation but that technologic advances are overcoming these limitations. However, they add that the future of BSM in emergency medicine is unclear and that more research is needed to define its benefits and limitations.

In 2007, Lefebvre and Hoekstra described the improvements in technical performance and ease of use in recent modifications to BSM technologies. A standardized vest improves lead placement, and changes to software direct clinicians’ attention to locations on the body mapping that may be significant, possibly reducing the amount of training needed. (4)

Diagnostic Accuracy

For patients with suspected ischemia, does electrocardiographic BSM improve the accuracy of diagnosis for acute myocardial infarction and/or acute coronary syndrome (ACS), compared with standard 12-lead ECG?

In June 2012, the Agency for Healthcare Research and Quality (AHRQ) published a technology assessment on the diagnostic utility of ECG-based signal analysis technologies for patients at low to intermediate risk of coronary artery disease (CAD). (5) Findings of the updated review were summarized in a 2013 publication by Leisy et al. (6) The AHRQ literature review focused on studies evaluating U.S. Food and Drug Administration (FDA) approved or cleared devices that are commercially available in the U.S. and can feasibly be used in most medical facilities. The 2012 assessment combined data from 10 studies on the PRIME ECG that involved patients with chest pain. Six of these were published by the same research group in Northern Ireland. The studies from Northern Ireland may have included a patient population that was higher risk than average, because some patients were treated in mobile cardiac care centers. Using a bivariate, random-effects model, the summary estimate for sensitivity was 71.1% (95% confidence interval [CI], 45.6% to 87.8%), and for specificity was 90.2% (95% CI, 83.2% to 94.4%). The summary estimate for the positive likelihood ratio was 6.3 (95% CI, 3.3 to 12.1), and the summary negative likelihood ratio negative was 0.30 (95% CI, 0.16 to 0.56). These combined summary estimates were compared with estimates compiled from 10 studies reporting 12-lead ECG performance. (Eight of these 10 studies were also included in the analysis of PRIME ECG sensitivity, previously described.) The pooled sensitivity was 43.1% (95% CI, 25.8% to 62.2%) and the pooled specificity was 94.4% (95% CI, 88.4% to 97.4%). The difference in sensitivity between the PRIME ECG and the 12-lead ECG were not significantly different (p<0.078). In addition, differences in specificity did not differ significantly (p<0.234).

Many of the individual studies have shown higher sensitivity for BSM and some have shown lower specificity. For example, in a retrospective study conducted at 4 centers, Ornato et al. reviewed the cardiac enzyme-confirmed cases of acute myocardial infarction (AMI) against results of 12-lead ECG and BSM. (7) Due to a change in standard practice during the study, AMI was defined by either elevated troponin or heart-specific creatinine kinase (CK-MB). Of 647 patients, 58 (8.9%) were not analyzed due to lack of enzyme data. Sensitivity comparison between BSM and 12-lead ECG in the CK-MB group favored BSM (100% vs 72.7%, respectively, p=0.031; n=364), and also in the troponin group (92.9% vs 60.7%, respectively, p=0.022; n=225). Specificity for BSM was not significantly different from 12-lead ECG in either group (96.5% vs 97.1 and 94.9 vs 96.4, both respectively).

A 2013 study from Northern Ireland retrospectively reported on 645 consecutive patients with sudden out of hospital cardiac arrest initially attended by a mobile cardiac care unit. (8) Eighty patients survived initial resuscitation, and 59 of these underwent ECG BSM and 12-lead ECG analysis by the physician leading the mobile unit. Twenty-four of the patients died prehospital, and 35 were admitted to the hospital and underwent coronary angiography. Twenty-six of the 35 patients (75%) who received angiography had acute occlusion of a main coronary artery. ECG BSM post resuscitation showed ST segment elevation in 23 of 35 patients (66%) and had had 88% sensitivity and 100% specificity for diagnosing acute coronary occlusion in these 35 patients. In contrast, the combination of either ST elevation myocardial infarction (STEMI) or ST segment depression on 12-lead ECG had a sensitivity of 46% and specificity of 100% for diagnosing acute coronary occlusion. A 2008 retrospective study from Northern Ireland included 755 patients presenting to the Emergency Department, mobile cardiac care, or in hospital with symptoms of ischemic chest pain. (9) Each patient’s clinical course was guided by standard American College of Cardiology 12-lead ST segment criteria and subsequent cardiac enzymes, if electrocardiographically negative. A cardiologist blinded to the clinical details measured BSM retrospectively. AMI was defined by elevated cardiac troponin levels. The standard 12-lead electrocardiograph demonstrated a sensitivity of 45% and a specificity of 92% for detecting troponin positive ischemia. When non-ST electrographic changes were permitted as part of the criteria for AMI, sensitivity increased (51%-68%), but specificity decreased (71%-89%). In this study, BSM performed with a sensitivity of 76% and specificity of 92%.

Fermann et al. found very different performance characteristics of BSM in comparison with 12-lead ECG from other studies. (10) A convenience sample of 150 patients with chest pain presenting to the ED had BSM measured within 30 minutes of the standard ECG. Emergency physicians, who had been trained in BSM, interpreted both the BSM and the ECGs at the time of presentation. Both were stored electronically for review by a BSM expert over read; after the study had ended, a convenience sample of 135 BSMs were over read. Of 43 patients, 10 (23.3%) judged to have normal BSM by the emergency physicians were found to have abnormal findings or frank infarction by the expert interpreter. Overall correlation between the emergency physicians and expert reviewer was only fair (correlation coefficient κ=0.627; 95% CI, 0.530 to 0.724). Sensitivity of both standard ECG and BSM were low at 10.5 (95% CI, 1.8 to 34.5) and 15.8 (95% CI, 4.2 to 40.5), respectively. This low sensitivity likely reflects the spectrum of patients in the study. Specificities were also comparable between the 2 groups at 90.1 (95% CI, 83.3 to 94.4) and 86.3 (95% CI, 78.9 to 91.4), respectively.

In 2010, O’Neil et al. published results from a secondary analysis of the Optimal Cardiovascular Diagnostic Evaluation Enabling Faster Treatment of Myocardial Infarction (OCCULT-MI) trial. (11) A multicenter (10-site), prospective, observational study, the OCCULT-MI trial enrolled 1830 subjects presenting to the ED with moderate- to high-risk chest pain. Patients were simultaneously tested with 12-lead and 80-lead ECGs, with clinicians able to access the 12-lead results only. The patients were treated by standard care based on 12-lead result or clinical suspicion. Off-site clinicians, who were not involved in the patients’ care, reviewed the 80-lead ECG and made a diagnostic determination, validated through multiple reviewers.

In this publication, 12-lead ECG was compared with 80-lead ECG mapping for detecting high-risk ECG abnormalities. Patients diagnosed with STEMI by 12-lead ECG (n=91), and patients with missing data (n=255) were excluded from the analysis on specificity and sensitivity. When detecting myocardial infarction (MI) and acute coronary syndrome (ACS), the 80-lead ECG mapping sensitivity was significantly higher than the 12-lead ECG for MI (19.4% vs 10.7%, p=0.0014) and for ACS (12.3% vs 7.1%, p=0.001). The authors attributed these low sensitivity rates to the exclusion of STEMI patients in this analysis. Specificity for the 80-lead ECG mapping was significantly lower than the 12-lead ECG for MI (93.9% vs 96.4%, p=0.001) and for ACS (93.7% vs 96.4%, p=0.001). Positive and negative predictive values and negative and positive likelihood ratios were not statistically different between the 12-lead and 80-lead groups. The 80-lead ECG mapping resulted in the identification of 18 additional MI patients and 21 additional ACS patients who could potentially have benefited from more aggressive treatment. However, the 80-lead ECG mapping results were not incorporated into treatment decision making, and thus, no conclusions can be made from this study on the impact of this technology on patient outcomes. Also, the authors did not explore the impact of decreased specificity, and increased false-positive rates, on patient outcomes. Other limitations of this study include lack of enrollment of low-risk ED patients and the lack of power to detect differences in ACS diagnosis.

In 2012, Daly et al. also compared 12-lead ECG with 80-lead ECG mapping in a retrospective review of 2810 consecutive patients admitted with ischemic-type chest pain. (12) All patients included in the study had coronary angiography and cardiac troponin levels during admission. The analysis was confined to patients with significant left main stem (LMS) coronary stenosis (>70%), which was found in 116 (4.1%) patients. Of these 116 patients with LMS coronary stenosis, 92 (79%) had AMI, diagnosed when cardiac troponin levels were 0.03 μg/L or higher. BSM was found to be more sensitive for diagnosing AMI in patients with LMS coronary stenosis compared with 12-lead ECG. BSM detected STEMI in 85/92 patients for an 88% sensitivity, 83% specificity, 95% PPV, and 65% NPV. Twelve-lead ECG (using Minnesota 9-2 criteria) detected STEMI in 13 patients (11%), for a 12% sensitivity and 92% specificity. The C statistic for the diagnosis of AMI in patients with LMS stenosis by 12-lead ECG was 0.580 (95% CI, 0.460 to 0.701, p=0.088) compared with 0.800 (95% CI, 0.720 to 0.881; p<0.001) using physician interpretation of BSM or 0.792 (95% CI, 0.690 to 0.894, p<0.001) using the “PRIME ECG®” algorithm.

A small 2014 study from the U.K. evaluated the 80-lead ECG mapping system (PRME-ECG) along with internally-developed software to create a BSM Delta map. (13) The study included 49 patients who presented to the ED with cardiac-sounding chest pain. Using the final diagnosis of ACS as the reference standard, the sensitivity and specificity of the BSM Delta map for diagnosing ACS were 71% (22/31) and 78% (14/18), respectively. This compares with a sensitivity of 67% (21/31) and specificity of 55% (10/18) when 12-lead ECG was used. The authors did not analyze whether differences in diagnostic accuracy were statistically significant. Moreover, the BSM Delta mapping software, an important part of the diagnostic process in this study, may not be available outside of the European research setting.

Section Summary: Diagnostic Accuracy

Numerous published studies compare the accuracy of BSM with standard 12-lead electrocardiography for the diagnosis of ACS. These studies are mostly retrospective and did not enroll the ideal clinical populations (i.e., consecutive patients presenting with clinical signs/symptoms of ischemia). They also tended to compare the accuracy of BSM alone with 12-lead ECG alone. This is less clinically relevant because 12-lead ECG is not used alone to diagnose ACS, but rather is combined with the clinical presentation and results of cardiac enzymes. The 2012 AHRQ technology assessment did not find a statistically significant difference in the diagnostic accuracy of BSM compared with a standard 12-lead ECG. Among the individual studies, the difference in sensitivity is variable, and there is uncertainty around whether there is higher sensitivity that is clinically significant. The specificity of BSM may be lower than 12-lead ECG, as some studies report lower specificity but others do not. Because of the uncertainty in the sensitivity and specificity in the available studies, it is not possible to estimate the tradeoff between additional cases of ACS detected and false positive results leading to further unnecessary testing. Further prospective studies are needed that include relevant clinical populations and that compare the incremental value of BMS when used as part of the overall diagnostic workup for ACS.

Clinical Utility

Does ECG BSM lead to changes in management that improve health outcomes?

The 2012 AHRQ assessment, noted earlier, (5) did not identify any studies in patients at low to intermediate risk of CAD that provided evidence on the question of whether findings from ECG-based technologies other than the standard 12-lead ECG had an impact on patient management decisions or health outcomes. One study, the OCCULT-MI trial, was identified that addressed the issue of patient outcomes in a population of patients at moderate to high risk for CAD. This study is discussed earlier in the section on diagnostic accuracy. Primary results of the OCCULT-MI trial were published in 2009 by Hoekstra et al. (11, 14) Primary outcome in the OCCULT-MI trial was door-to-sheath time in 12-lead STEMI patients versus door-to-sheath time in patients with ST elevations noted on 80-lead testing. Secondary outcomes were clinical outcomes at 30 days and angiographic data. Of the 1830 subjects, 91 had a discharge diagnosis of STEMI, 84 of whom underwent cardiac catheterization with a mean door-to-sheath time of 54 minutes. Twenty-five subjects (1.4% of the study population) met criteria for ST elevation in the 80-lead alone, 14 of whom underwent cardiac catheterization with a mean door-to-sheath time of 1002 minutes (estimated treatment difference, 881; 95% CI, 181 to 1079 minutes, respectively). Neither 30-day clinical outcomes, nor adverse events, differed significantly in the identified at-risk groups. These 25 patients were in addition to the 91 STEMI patients identified on 12-lead, leading the authors to conclude that the additional leads identified 27.5% more AMI patients than 12-lead alone (25/91).

An editorial accompanying the publication of the OCCULT-MI trial acknowledged the limitation of 12-lead ECG in identifying patients with AMI. However, a distinction was made between those patients for whom it is well established that early intervention is beneficial (i.e., STEMI on standard 12-lead ECG) and those for whom BSM is positive but 12-lead is not. It is not known whether these patients benefit from early intervention. The editorial suggested that the patients identified thusly are more similar to the non-ST elevation myocardial infarction (NSTEMI) patients based on peak troponin levels found in the Hoekstra et al. study (14) and that identification of these patients should not lead to a change in treatment. (15)

Section Summary: Clinical Utility

There are no studies that demonstrate how BSM can be used to change clinical management in ways that improve health outcomes. Indirect evidence suggests that BSM might be used in a subset of patients presenting with suspected ACS to reduce the time to diagnosis and thereby provide revascularization more expediently. Whether this strategy improves outcomes has yet to be demonstrated. To demonstrate clinical utility, the ideal study design is a randomized controlled trial in which patients are randomized to BSM or standard 12-lead ECG and patients are followed for changes in management and clinical outcomes.

Ongoing and Unpublished Clinical Trials

A search of in June 2015 did not identify any ongoing or unpublished trials that would likely influence this review.

Practice Guidelines and Position Statements

The American College of Cardiology Foundation guidelines for electrocardiography standardization and interpretation recognize that while the studies of body surface maps from large electrode arrays have provided useful information about localization of ECG information on the thorax, at this time their complexity precludes their use as a substitute for the standard 12-lead ECG for routine recording purposes. (16)

Summary of Evidence

The evidence for use of electrocardiographic (ECG) body surface mapping (BSM) in patients with suspected or confirmed cardiac disorders includes a number of studies on the association between ECG BSM and acute myocardial infarction. Relevant outcomes are overall survival, disease-specific survival, test performance, and morbid events. No prospective trials using BSM to guide treatment have been conducted. Results of published studies have been variable and an Agency for Healthcare Research and Quality review did not find statistically significant differences in the diagnostic accuracy of BSM and 12-lead ECG. Under ideal conditions, it is possible that BSM has a higher sensitivity than 12-lead ECG alone for acute coronary events. However, the data also suggest that the specificity may be lower, highlighting concerns regarding false-positive results. In clinical practice, patients with symptoms suspicious for ischemia are not diagnosed with 12-lead ECG alone but in combination with clinical presentation and serial cardiac enzymes. There is no evidence demonstrating that electrocardiographic BSM leads to changes in management that improve health outcomes. The evidence is insufficient to determine the effect of the technology on health outcomes.


Each benefit plan, summary plan description or contract defines which services are covered, which services are excluded, and which services are subject to dollar caps or other limitations, conditions or exclusions. Members and their providers have the responsibility for consulting the member's benefit plan, summary plan description or contract to determine if there are any exclusions or other benefit limitations applicable to this service or supply. If there is a discrepancy between a Medical Policy and a member's benefit plan, summary plan description or contract, the benefit plan, summary plan description or contract will govern.



Disclaimer for coding information on Medical Policies

Procedure and diagnosis codes on Medical Policy documents are included only as a general reference tool for each policy. They may not be all-inclusive.

The presence or absence of procedure, service, supply, device or diagnosis codes in a Medical Policy document has no relevance for determination of benefit coverage for members or reimbursement for providers. Only the written coverage position in a medical policy should be used for such determinations.

Benefit coverage determinations based on written Medical Policy coverage positions must include review of the member’s benefit contract or Summary Plan Description (SPD) for defined coverage vs. non-coverage, benefit exclusions, and benefit limitations such as dollar or duration caps.


The following codes may be applicable to this Medical policy and may not be all inclusive.

CPT Codes

0178T, 0179T, 0180T



ICD-9 Diagnosis Codes

Refer to the ICD-9-CM manual

ICD-9 Procedure Codes

Refer to the ICD-9-CM manual

ICD-10 Diagnosis Codes

Refer to the ICD-10-CM manual

ICD-10 Procedure Codes

Refer to the ICD-10-CM manual

Medicare Coverage:

The information contained in this section is for informational purposes only. HCSC makes no representation as to the accuracy of this information. It is not to be used for claims adjudication for HCSC Plans.

The Centers for Medicare and Medicaid Services (CMS) does not have a national Medicare coverage position. Coverage may be subject to local carrier discretion.

A national coverage position for Medicare may have been developed since this medical policy document was written. See Medicare's National Coverage at <>.


1. Gulrajani RM. The forward and inverse problems of electrocardiography. IEEE Eng Med Biol Mag. Sep-Oct 1998; 17(5):84-101, 122. PMID 9770610

2. Thivierge M, Gulrajani RM, Savard P. Effects of rotational myocardial anisotropy in forward potential computations with equivalent heart dipoles. Ann Biomed Eng. May-Jun 1997; 25(3):477-498. PMID 9146803

3. Self WH, Mattu A, Martin M, et al. Body surface mapping in the ED evaluation of the patient with chest pain: use of the 80-lead electrocardiogram system. Am J Emerg Med. Jan 2006; 24(1):87-112. PMID 16338516

4. Lefebvre C, Hoekstra J. Early detection and diagnosis of acute myocardial infarction: the potential for improved care with next-generation, user-friendly electrocardiographic body surface mapping. Am J Emerg Med. Nov 2007; 25(9):1063-1072. PMID 18022503

5. Coeytaux RM, Leisy PJ, Wagner GS, et al. Systematic review of ECG-based signal analysis technologies for evaluating patients with acute coronary syndrome. Agency for Healthcare Research and Quality Technology Assessment Report. Project ID: CRDD0311. June 2012; Available at <> Accessed September 12, 2016.

6. Leisy PJ, Coeytaux RR, Wagner GS, et al. ECG-based signal analysis technologies for evaluating patients with acute coronary syndrome: a systematic review. J Electrocardiol. Mar-Apr 2013;46(2):92-97. PMID 23273746

7. Ornato JP, Menown IB, Peberdy MA, et al. Body surface mapping vs 12-lead electrocardiography to detect ST elevation myocardial infarction. Am J Emerg Med. Sep 2009; 27(7):779-784. PMID 19683104

8. Daly MJ, Finlay DD, Scott PJ, et al. Pre-hospital body surface potential mapping improves early diagnosis of acute coronary artery occlusion in patients with ventricular fibrillation and cardiac arrest. Resuscitation. Jan 2013; 84(1):37-41. PMID 22986067

9. Owens C, McClelland A, Walsh S, et al. Comparison of value of leads from body surface maps to 12-lead electrocardiogram for diagnosis of acute myocardial infarction. Am J Cardiol. Aug 1 2008; 102(3):257-265. PMID 18638583

10. Fermann GJ, Lindsell CJ, O'Neil BJ, et al. Performance of a body surface mapping system using emergency physician real-time interpretation. Am J Emerg Med. Sep 2009; 27(7):816-822. PMID 19683110

11. O'Neil BJ, Hoekstra J, Pride YB, et al. Incremental benefit of 80-lead electrocardiogram body surface mapping over the 12-lead electrocardiogram in the detection of acute coronary syndromes in patients without ST-elevation myocardial infarction: Results from the Optimal Cardiovascular Diagnostic Evaluation Enabling Faster Treatment of Myocardial Infarction (OCCULT MI) trial. Acad Emerg Med. Sep 2010; 17(9):932-939. PMID 20836773

12. Daly MJ, Adgey JA, Harbinson MT. Improved detection of acute myocardial infarction in patients with chest pain and significant left main stem coronary stenosis. QJM. Feb 2012; 105(2):127-135. PMID 21890878

13. Zeb M, Mahmoudi M, Garty F, et al. Detection of regional myocardial ischaemia by a novel 80-electrode body surface Delta map in patients presenting to the emergency department with cardiac-sounding chest pain. Eur J Emerg Med. Apr 2014; 21(2):89-97. PMID 23883775

14. Hoekstra JW, O'Neill BJ, Pride YB, et al. Acute detection of ST-elevation myocardial infarction missed on standard 12-Lead ECG with a novel 80-lead real-time digital body surface map: primary results from the multicenter OCCULT MI trial. Ann Emerg Med. Dec 2009; 54(6):779-788 e771. PMID 19766352

15. Hollander JE. The 80-lead ECG: more expensive NSTEMI or Occult STEMI. Ann Emerg Med. Dec 2009; 54(6):789-790. PMID 19766356

16. Kligfield P, Gettes LS, Bailey JJ, et al. Recommendations for the standardization and interpretation of the electrocardiogram: part I: the electrocardiogram and its technology a scientific statement from the American Heart Association Electrocardiography and Arrhythmias Committee, Council on Clinical Cardiology; the American College of Cardiology Foundation; and the Heart Rhythm Society endorsed by the International Society for Computerized Electrocardiology. J Am Coll Cardiol. Mar 13 2007; 49(10):1109-1127. PMID 17349896

17. Electrocardiographic body surface mapping. Chicago, Illinois: Blue Cross Blue Shield Association Medical Policy Reference Manual (2015 August) Medicine 2.02.23.

Policy History:

Date Reason
10/15/2017 Reviewed. No changes.
11/1/2016 Document updated with literature review. Coverage unchanged.
10/15/2015 Reviewed. No changes.
4/15/2014 Document updated with literature review. Coverage unchanged. CPT/HCPCS code(s) updated.
4/15/2013 Document updated with literature review. Coverage unchanged.
6/15/2011 Document updated with literature review. Coverage unchanged. Description and rationale substantially revised.
9/15/2009 Policy reviewed with literature review; no changes to coverage statement.
7/1/2007 New medical document.

Archived Document(s):

Back to Top