A QUARTER century has passed since the first reports describing use of precordial lead V5to monitor for intraoperative ischemia were published, and over a decade has passed since we documented its sensitivity (75%) using continuous 12-lead monitoring in 100 patients. 1,2Since then, V5has become “a clinical routine.” In this issue, Landesberg et al. present “the next generation,” monitoring a larger cohort (185 patients) undergoing higher-risk surgery (all vascular surgery), for a longer period of time (48–72 h). 3Their results extend our knowledge and add controversy, given their finding that leads V3(75%) and V4(83%) are either equal or more sensitive than V5(75%). They recommend use of V4over V5since its ST-segment is most commonly isoelectric on the baseline electrocardiogram, extrapolating that this makes it more likely to reflect ischemic changes. They also recommend the use of two precordial leads to approach 95% sensitivity to detect ischemia or infarction.
Should this study alter our current clinical practice? Should we “move to the right” in favor of V3or V4and abandon V5? Should we encourage bipartisanship by monitoring two precordial leads (requiring equipment modification)? Or should we take the Libertarian approach by encouraging simplicity in monitoring? I would argue that with the clinical data accumulated over the past 10–15 yr. documenting associations of perioperative tachycardia and ST-segment depression to adverse outcome and beneficial effects of β-blockade, that sophisticated monitoring is considerably less important than adequate prophylaxis and therapy. 4,5However, since it is known that β-blockade cannot ensure suppression of ischemia nor prevention of infarction in all patients, evaluation of the current status of multi-lead monitoring remains worthy of serious consideration. 6
Reference cardiology texts state that subendocardial ischemia induced by demand-related stress is manifested by ST-segment depression in lead V5and does not localize the anatomic site of coronary obstruction. 7Yet even Mason and Likar, the first to use the now universal torso-mounted axial leads during exercise treadmill testing (ETT), reported that V6, not V5, was the most sensitive lead! 8Subsequent investigators have reported varying sensitivity, particularly between V4, V5, and V69–13(table 1).
How can we reconcile these differences? Examination of these studies reveals differences in ST-segment criteria, (including magnitude and timing after the J-point, varying even with the number of leads involved); the time period considered (during exercise, during recovery, perioperatively, etc.); temporal duration (with Landes- berg et al. requiring prolonged duration > 10 min.) 3; mode of analysis (visual vs. computerized); incorporation of other physiologic parameters; and less frequently, but of considerable theoretical interest, normalization based on the height of accompanying R-wave. 14With all of these factors, any simple explanation is likely impossible. Given the common adage that the ST-segment vector during subendocardial ischemia is directed towards the apex of the ventricle (which V5is said to be closest to), it is possible that anthropomorphic factors influencing the position of the heart in the chest such as gender, body habitus, and chest diameter may be important. However, this has not been studied. Complex physiologic approaches using noninvasive body surface mapping have been used, and more recently invasive endo- and epicardial potentials with three-dimensional computer modeling has provided an alternative approach. 15
Why is there not greater interest in the cardiology community to nail down the precise sensitivities? Possible explanations include (1) nearly all stress tests in this country use computerized 12-lead systems, (2) there is strong evidence that a positive response in multiple leads (along with greater magnitude of depression and presence of a downsloping ST-segment) is related to a larger area of myocardium at risk 16,17and (3) despite calls for cost containment, thallium imaging, even after a positive endotracheal tube is very common (as is cardiac catheterization). In clinical practice, the whole of the 12-lead electrocardiogram is clearly greater than the sum of its parts.
Another factor is the growing interest in continuous 12-lead monitoring for patients with acute coronary syndromes (ACS). With transmural ischemia, lead sensitivity is closely associated with the site of coronary occlusion (whether transient or permanent) with leads V2and V3most sensitive for left anterior descending occlusion, and lead III most sensitive for the right coronary artery. In this setting, ST-segment elevation is a nearly universal finding. Circumflex occlusion results in a variable response with primary elevation in posterior precordial leads (i.e. , V7, V8, etc.) or reciprocal ST-depression in other precordial or axial leads. 1A recent multidisciplinary working group recommends leads III, V3, and V5as the most sensitive combination for patients with ACS. 18
Krucoff et al. were the first to make a serious argument for the value of continuous 12-lead electrocardiogram monitoring in ACS patients with the concept of the “12-lead fingerprint,” a unique pattern of leads and ST segment magnitude sensitive to detecting reocclusion after percutaneous transluminal coronary angioplasty (PTCA). 19This approach has been used in major studies of thrombolysis. Many intensive care unit bedside monitors (and telemetry monitors) are now “12-lead ECG capable” with a precordial lead cable and continuous ST segment trending of all 12 leads. Recent American and European ACS Guidelines now acknowledge the utility of this approach but make no firm recommendations for it. 20,21However, this approach includes a high rate of false-positive responses (40%) because of changes in QRS amplitude or vector with positional changes, arrhythmia/pacing artifact and heart-rate–related changes in ST-segment contour. 22In the perioperative setting, the array of catheters, monitors, and drains and the need to mobilize patients quickly are major logistical obstacles. Artifact issues and the economically unfavorable task of investigating episodes are formidable factors.
Aside from the monitoring issues raised by this study, there is important information on perioperative ischemia. As noted in the parent publication, duration of ischemia is a significant predictor of peak cTn-I level, ischemic events associated with infarction are preceded by increased heart rate (32 beats/min). People with diabetes and patients with left ventricular hypertrophy (LVH) are at highest risk. 23Diabetics are already known to be at high risk for adverse outcome. 24Less is known about LVH since many studies excluded these patients because of concerns that the increased QRS voltage may exaggerate the ST-segment response. 14We previously noted that LVH was the strongest preoperative factor multivariately associated with postoperative ischemia. 25Left ventricular hypertrophy is associated with accelerated atherosclerosis, subendocardial ischemia, and adverse long-term outcome. 26But its association with plaque disruption , likely the necessary ingredient for overt morbidity, is suggested by a recent study comparing angiographic results over a 6 months interval. 27The strongest adverse multivariate associations were LV mass and elevated heart rate (> 80 bpm), while the strongest protective association was with chronic β-blocker use.
Integrating the monitoring and clinical data, it seems reasonable that sophisticated monitoring may be of value to people with diabetes and those with LVH. A targeted study in these cohorts of the value of therapeutic intervention guided by multi-lead monitoring (in the setting of concurrent β-blockade) would be most helpful. My clinical observations are that precordial lead placements by physicians and nurses at all levels of training remain imprecise (and are unavoidably affected by surgical factors). Thus, I recommend that clinicians use a “true” V4or V5along with an inferior axial lead, control heart rate and pain, and use β-blockers as tolerated for all patients at risk.