https://orcid.org/0000-0002-0854-7241
Abstract
The development of new Q waves is an indicator of myocardial infarction following open coronary artery bypass grafting. It is unclear whether this is true for patients with multivessel coronary artery disease (MVD) treated initially with minimally invasive direct coronary artery bypass (MIDCAB). Electrocardiogram (ECG) changes were analyzed in a single center series of twenty-one consecutive patients with MVD undergoing MIDCAB. Six patients exhibited new Q waves in the inferior leads (group 1), 4 patients displayed significant QRS axis changes (group 2), and 11 patients presented with neither (group 3). The QRS axis and new Q waves reversed in most patients undergoing hybrid right coronary artery stenting and persisted in all patients receiving MIDCAB alone. In all cases, significant ECG changes were found to be benign and did not indicate myocardial infarction.
Introduction
Coronary artery bypass grafting (CABG) serves as an effective treatment modality for patients with severe coronary artery disease [1]. In order to evaluate periprocedural complications and predict long-term outcomes, various assessments such as postoperative electrocardiographic (ECG) features, blood tests, and even magnetic resonance imaging have been employed [2–4]. Among these modalities, ECG emerges as the most readily accessible method. Post-CABG ECG changes encompass QRS alterations, ST segment elevation or depression, and the occurrence of postoperative arrhythmias [5–7]. The majority of these postoperative ECG changes have been recognized as transient, indicative of coronary reperfusion [8, 9]. However, the emergence of new Q waves postoperatively is deemed a robust indicator of postprocedural myocardial infarction [10, 11].
Minimally invasive direct coronary artery bypass (MIDCAB) grafting has emerged as an alternative to traditional CABG for patients with isolated proximal left anterior descending coronary artery (LAD) stenosis, and is also employed as part of a hybrid approach in selected cases with multivessel disease (MVD) [12]. In contrast to conventional CABG which generally strives for complete surgical revascularization and often provides multi-vessel bypass, single-vessel MIDCAB exclusively targets the LAD, thereby potentially leading to distinct postoperative ECG alterations and subsequent clinical considerations, particularly concerning the emergence of new Q waves or significant axis changes. However, these changes have not been well characterized.
In this single-center case series, we scrutinized the ECG changes observed in all patients with multivessel coronary disease undergoing MIDCAB, with the aim of elucidating clinical significance and predictors.
Case series
Study population
From June 2022 to January 2024, all patients undergoing MIDCAB at Cedars-Sinai Medical Center with MVD treated initially by MIDCAB were identified. Patients with prior pathological Q waves were excluded. Baseline characteristics, echocardiographic parameters, and left heart catheterization results before MIDCAB were documented. To exclude transient ECG alterations, ECGs obtained both preoperatively, 5–30 days post-MIDCAB and post-percutaneous coronary intervention (PCI) in a hybrid approach, were collected and meticulously analyzed. Based on the presence of new Q waves or significant QRS axis changes post-MIDCAB, all patients were stratified into three groups: group 1 with the presence of new Q waves, regardless of axis change; group 2 with significant axis change; and group 3 without either of these changes. This study received approval from the Institutional Review Board with a waiver of informed consent (study #00003440).
Surgical technique
All MIDCABs were performed using robotic assistance for mammary artery takedown (using the Da Vinci Xi system, Intuitive Surgical, Sunnyvale, CA) and a direct anastomosis via left mini thoracotomy. Routine shunting of the target left anterior descending artery was employed during the anastomosis, and all completed bypasses were assessed using a coronary ultrasonic flow meter (Transonic, Guidant, Indianapolis, IN, USA). All procedures were performed without cardiopulmonary bypass. Hybrid coronary revascularization was conducted by planned, staged PCI for non-LAD lesions following grafting of the LIMA-LAD. Patients underwent stenting with sirolimus-eluting stents 6–14 days post-surgical revascularization in accordance with standard protocols.
ECG recording, measurement, and definitions
All standard 12-lead ECGs were recorded with a sweep speed of 25 mm/s and amplification of 10 mm/mV. Each ECG was observed, measured, and analyzed by two experienced independent observers, who were blinded to the patients’ hospital course, utilizing electronic calipers. QRS amplitudes in leads I, III, aVL, and aVF was measured, and an average value was adopted as the final measurement. The presence or absence of new Q waves was determined independently by the two observers, and any discrepancies were resolved through consensus or by consulting a third observer when necessary. The QRS axis was obtained from the ECG reporting system.
New Q waves were defined as follows: any Q wave in leads V2–V3 ≥ 20 ms or QS complex in leads V2 and V3; or a Q-wave ≥30 ms and ≥ 0.1 mV deep or QS complex in leads I, II, aVL, aVF, or V4–V6 in any two leads of a contiguous lead grouping (I, aVL, V6; V4–V6; II, III, and aVF) [13]. Significant QRS axis change was defined as a QRS axis change >25° [14].
Baseline characteristics
A total of 22 patients with multivessel disease treated with MIDCAB were identified during the study time frame, and one patient was excluded due to prior pathological Q waves. Ultimately, 21 patients (mean age 70.2 ± 8.4 years, 2 females) were included. Among them, 6 patients were identified with new Q waves (group 1), 4 patients exhibited significant QRS axis changes (group 2), and 11 patients displayed neither of these changes (group 3). Within group 1, all patients demonstrated pathological Q waves in leads III and aVF, while none of them had Q waves in lead II. Baseline characteristics are shown in Table 1. None of the patients had reduced left ventricular ejection fraction or wall motion abnormalities.
Baseline characteristics
| No. | Group | Age (years) | Sex | BMI (kg/m2) | Comorbidities | NYHA | LVEF (%) |
|---|---|---|---|---|---|---|---|
| 1 | 1 | 60 | M | 21.6 | – | I | 65 |
| 2 | 1 | 69 | M | 21.8 | – | I | 68 |
| 3 | 1 | 79 | M | 25.4 | AF | II | 55 |
| 4 | 1 | 78 | M | 23.2 | None | II | 65 |
| 5 | 1 | 69 | M | 24.6 | AF, DM | I | 73 |
| 6 | 1 | 66 | F | 40.3 | HTN | I | 55 |
| 7 | 2 | 76 | F | 21.6 | HTN | II | 64 |
| 8 | 2 | 62 | M | 21.3 | HTN, DM | I | 63 |
| 9 | 2 | 66 | M | 24.8 | HTN, DM | II | 66 |
| 10 | 2 | 70 | M | 27.6 | HTN, DM | I | 53 |
| 11 | 3 | 54 | M | 35.1 | HTN,DM | I | 75 |
| 12 | 3 | 68 | M | 29.7 | HTN,DM | I | 62 |
| 13 | 3 | 74 | M | 24.7 | HTN | I | 61 |
| 14 | 3 | 67 | M | 25.5 | HTN, DM | I | 62 |
| 15 | 3 | 77 | M | 26.9 | HTN | II | 55 |
| 16 | 3 | 69 | M | 22.2 | HTN | II | 75 |
| 17 | 3 | 83 | M | 25.3 | HTN, AF | I | 64 |
| 18 | 3 | 54 | M | 28.4 | – | I | 65 |
| 19 | 3 | 85 | M | 28.5 | AF, HTN | II | 57 |
| 20 | 3 | 75 | M | 24.9 | HTN, DM | I | 60 |
| 21 | 3 | 73 | M | 27.9 | DM | I | 60 |
AF, atrial fibrillation; BMI, body mass index; DM, diabetes mellitus; F, female; HTN, hypertension; LVEF, left ventricular ejection fraction; M, male.
Left heart catheterization and procedural results
All patients exhibited severe stenosis in the left main coronary artery or LAD, and MIDCAB was performed successfully. There were no significant periprocedural complications. Two patients had post-procedural pleural effusions, which resolved within 5 days. A preoperative severe RCA stenosis (>50% stenosis) was reported in all patients in groups 1 and 2, whereas it was less prevalent in group 3 (4 patients, 36.4%). Notably, all patients with an absence of severe RCA stenosis were found in group 3. Eight patients and four patients received hybrid percutaneous stent implantation in the RCA and left circumflex artery at Cedars-Sinai Medical Center, respectively (Table 2).
Left heart catheterization and electrocardiographic results
| No. | Group | Coronary artery stenosis (%) | Pre-CAB axis | Post-CAB axis | PCI | Post-PCI axis | |||
|---|---|---|---|---|---|---|---|---|---|
| LM | LAD | LCX | RCA | ||||||
| 1 | 1 | – | 100 | – | 80 | 41 | −18 | RCA | −30 |
| 2 | 1 | – | 100 | – | 90 | 56 | 5 | RCA | 52 |
| 3 | 1 | 50 | 70 | 70 | 100 | 51 | 3 | LCX | 77 |
| 4 | 1 | – | 70 | – | 70 | 60 | −30 | No | – |
| 5 | 1 | – | 80 | 80 | 85 | 9 | −4 | RCA | −5 |
| 6 | 1 | – | 80 | 50 | 80 | 30 | −26 | No | – |
| 7 | 2 | – | 70 | 70 | 90 | 67 | 36 | RCA | 73 |
| 8 | 2 | 50 | 60 | 70 | 80 | 80 | 45 | RCA | 65 |
| 9 | 2 | – | 95 | 50 | 80 | 88 | 38 | No | – |
| 10 | 2 | – | 80 | 70 | 60 | 65 | 15 | No | – |
| 11 | 3 | 95 | 99 | – | 90 | −26 | −30 | RCA | −30 |
| 12 | 3 | 60 | 70 | 90 | 80 | −27 | −21 | RCA | −27 |
| 13 | 3 | – | 90 | 80 | 70 | −26 | −16 | No | – |
| 14 | 3 | – | 100 | – | 80 | −7 | 3 | RCA | 3 |
| 15 | 3 | 50 | 70 | – | 40 | 0 | 6 | No | – |
| 16 | 3 | 80 | 70 | 80 | – | −37 | −18 | LCX | −14 |
| 17 | 3 | – | 85 | 80 | – | 11 | 9 | No | – |
| 18 | 3 | – | 100 | 80 | 40 | 67 | 48 | LCX | 62 |
| 19 | 3 | – | 80 | 80 | 30 | −54 | −66 | LCX | −68 |
| 20 | 3 | – | 90 | 80 | – | −62 | −78 | No | – |
| 21 | 3 | 25 | 90 | – | – | 5 | −20 | No | – |
CAB, coronary artery bypass grafting; LAD, left anterior descending coronary artery; LCX, left circumflex coronary artery; LM, left main coronary artery; PCI, percutaneous coronary intervention; RCA, right coronary artery.
Axis and QRS amplitude changes
Before MIDCAB, the axis in group 1 subjects was more leftward than that of group 2 (41.2° ± 19.1° vs. 75.0° ± 11.0°, P = 0.013). In group 3, patients with significant right coronary artery (RCA) stenosis but without axis change or new Q waves, the primary axis was consistently <0° (−21.5° ± 9.7°). This was not observed in patients in group 1 and group 2 (Table 2, Fig. 1). In patients with severe RCA stenosis, the patient’s primary QRS axis was all above 60° in group 2, consistently between 0° and 60° in group 1, and below 0° in group 3. Following MIDCAB, the QRS axis significantly shifted leftward from 41.2° ± 19.1° to −11.7° ± 15.1° in group 1 (P < .001), and from 75.0° ± 11.0° to 33.5° ± 13.0° in group 2 (P = .003). The QRS axis of group 1 was more leftward than that of group 2 (−11.7° ± 15.1° vs. 33.5° ± 13.0°, P = .001). Pathological Q waves disappeared in two out of three patients following PCI of RCA in group 1 (Fig. 2) and persisted in all patients who did not receive PCI. In the patient where the Q waves persisted following PCI (Patient 1), asymptomatic ST segment elevation was noted in the inferior leads during multiple stent implantations in the RCA, accompanied by the development of new Q waves in lead II subsequent to PCI. In another case with a left coronary artery dominance, Q waves also disappeared after PCI in the left circumflex coronary artery (Table 2). In group 2, the QRS axis returned to a comparable pre-MIDCAB level after PCI in those receiving it (Fig. 3), and persisted in those who did not receive PCI (Table 2). Of note, in the patient initially excluded due to pre-existing pathological Q waves, the disappearance of the Q wave was also observed following PCI.
Primary axis and their changes following MIDCAB. Given a more rightward primary axis in group 2 (>60°), the QRS axis remained positive even after significant axis changes, thereby reducing the likelihood of new Q waves. However, the primary axis in group 1 was less positive (0°–60°), leading to the presence of new Q waves following significantly leftward axis alterations. All patients without significant RCA stenosis were in group 3, characterized by the absence of both new Q waves and significant QRS axis changes (red lines). Notably, among the three remaining patients in group 3 with significant RCA stenosis, a primary axis < 0° was observed. MIDCAB, minimally invasive direct coronary artery bypass; RCA, right coronary artery.
ECG changes in a patient with new Q waves after MIDCAB. (Upper panel) Prior to MIDCAB, the primary QRS axis ranged between 0° and 60°. (Mid panel) Following MIDCAB, enhanced blood flow strengthened anterior wall activation, resulting in a pronounced leftward axis shift to generate new Q waves in leads III and aVF. (Lower panel) Following PCI in the RCA, the QRS axis and amplitude reverted to pre-MIDCAB levels, attributing to increased blood flow in the inferior wall. ECG, electrocardiogram; MIDCAB, minimally invasive direct coronary artery bypass; PCI, percutaneous coronary intervention; RCA, right coronary artery.
ECG changes in a patient with significant QRS axis alteration after MIDCAB. (Upper panel) Prior to MIDCAB, the primary QRS axis exceeded 60°, indicating relatively stronger inferior activation. (Mid panel) Following MIDCAB, enhanced blood flow led to strengthened anterior wall activation, resulting in a marked leftward axis shift, heightened QRS amplitude in lead aVL, and diminished QRS amplitude in leads III and aVF. (Lower panel) Subsequent to PCI in the RCA, the QRS axis and amplitude reverted to pre-MIDCAB levels, attributed to increased blood flow in the inferior wall. Abbreviations as in Fig. 2.
The amplitude of the QRS complex in lead aVL increased after MIDCAB in both group 1 (0.09 mV ± 0.15 mV vs. 0.61 ± 0.16 mV, P < .001) and group 2 (−0.46 mV ± 0.24 mV vs. 0.27 ± 0.14 mV, P = .002). Conversely, it decreased in lead III following MIDCAB in both group 1 (0.21 mV ± 0.24 mV vs. −0.40 ± 0.18 mV, P = .001) and group 2 (1.17 mV ± 0.28 mV vs. 0.18 ± 0.19 mV, P = .016), and in lead aVF following MIDCAB in both group 1 (0.41 mV ± 0.23 mV vs. −0.13 ± 0.22 mV, P = .002) and group 2 (1.44 mV ± 0.13 mV vs. 0.43 ± 0.13 mV, P = .002). No significant change of QRS amplitude was found in lead II following MIDCAB in both group 1 (0.59 mV ± 0.27 mV vs. 0.27 ± 0.30 mV, P = .081) and group 2 (1.45 mV ± 0.41 mV vs. 0.84 ± 0.93 mV, P = .057).
Discussion
Previous studies have indicated that ECG changes, troponin T, and creatine kinase-MB elevation are associated with adverse outcomes after CABG [2, 15–17]. The existence of new Q waves has long been recognized as a robust indicator of myocardial infarction [10, 11]. Additionally, a significant axis change was also found to be associated with myocardial infarction [14]. However, the exact definition of myocardial infarction post CABG remains under debate, as the ECG changes do not necessarily corelate to a worse outcome [18, 19].
MIDCAB is increasingly accepted for patients with LAD stenosis. In this study, patients without significant RCA stenosis did not exhibit new Q waves or significant axis changes post-MIDCAB. However, all patients with either new Q waves or significant QRS axis changes post-MIDCAB had higher-grade RCA stenosis. This phenomenon strongly suggests that RCA stenosis plays a crucial role in ECG changes. While postsurgical measurements of troponin T and creatine kinase-MB were not obtained in all patients, postprocedural MI was ruled out clinically by the lack of any associated signs of MI or patient-reported symptoms.
Our hypothesis for this phenomenon is as follows (Figs 2 and 3): the QRS complex axis is balanced in patients with both LAD stenosis and RCA stenosis, affecting both the anterior and inferior walls. However, after restoring blood supply to the anterior wall through MIDCAB, activation of the anterior wall becomes stronger, resulting in a leftward shift in the QRS complex axis. If the axis reaches the northeast zone, the inferior leads will be negative with new Q waves. However, if the axis remains in the southeast zone, only significant axis changes will be identified. In this study, the primary QRS axis were significantly more leftward in group 1 (0°–60°) than that in group 2 (>60°). This indicates that despite leftward QRS axis changes in both groups after MIDCAB, it is challenging for the QRS axis to shift toward the northeast in patients with a more rightward primary axis (>60°). Consequently, the occurrence of new Q waves becomes less likely in these patients compared to those with a more leftward primary axis (0°–60°). This hypothesis is further supported by the fact that almost all patients experienced disappearance of new Q waves and reversion of the axis following hybrid RCA intervention, even in the patient with pre-existing Q waves, while new Q waves and significant axis deviation persisted in patients that did not receive RCA intervention. An exception was noted in one patient in whom the Q wave persisted and even extended (new Q waves in lead II) following PCI. Notably, in this case during the placement of multiple stents in the RCA the patient developed asymptomatic ST segment elevation, indicating possible injury to the distal segment of the RCA (potentially small thromboemboli).
An additional pertinent observation is the consistent emergence of new Q waves within leads III and aVF across the patient cohort, conspicuously absent in lead II. Theoretically, given their inherently rightward positioning relative to lead II, leads III and aVF are more vulnerable to leftward axis deviation. Hence, the appearance of new Q waves exclusively in leads III and aVF may not invariably signify genuine myocardial infarction as previously thought. The definition of postsurgical myocardial infarction based on ECG criteria alone warrants additional caution, particularly in patients with significant RCA stenosis undergoing MIDCAB. New Q waves and significant QRS axis changes are common after MIDCAB and can be predicted by the primary QRS axis preceding surgery. If the QRS axis before MIDCAB ranges between 60° and 90°, a significant axis change could occur after MIDCAB. Moreover, if the axis before MIDCAB falls between 0° and 60°, new Q waves might appear on the inferior leads. However, the above QRS changes could revert after hybrid RCA stenting. In the absence of significant RCA stenosis or if the axis before MIDCAB is northeast, no significant QRS changes will occur after MIDCAB.
Conflicts of interest
The authors reported no conflicts of interest.
Funding
None declared.
