Rapid Exclusion of COVID Infection With the Artificial Intelligence Electrocardiogram

OBJECTIVE

To rapidly exclude severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection using artificial intelligence applied to the electrocardiogram (ECG).

METHODS

A global, volunteer consortium from 4 continents identified patients with ECGs obtained around the time of polymerase chain reaction-confirmed COVID-19 diagnosis and age- and sex-matched controls from the same sites. Clinical characteristics, polymerase chain reaction results, and raw electrocardiographic data were collected. A convolutional neural network was trained using 26,153 ECGs (33.2% COVID positive), validated with 3826 ECGs (33.3% positive), and tested on 7870 ECGs not included in other sets (32.7% positive). Performance under different prevalence values was tested by adding control ECGs from a single high-volume site.

RESULTS

The area under the curve for detection of acute COVID-19 infection in the test group was 0.767 (95% CI, 0.756 to 0.778; sensitivity, 98%; specificity, 10%; positive predictive value, 37%; negative predictive value, 91%). To more accurately reflect a real-world population, 50,905 normal controls were added to adjust the COVID prevalence to approximately 5% (2657/58,555), resulting in an area under the curve of 0.780 (95% CI, 0.771 to 0.790) with a specificity of 12.1% and a negative predictive value of 99.2%.

CONCLUSION

Infection with SARS-CoV-2 results in electrocardiographic changes that permit the artificial intelligence-enhanced ECG to be used as a rapid screening test with a high negative predictive value (99.2%). This may permit the development of electrocardiography-based tools to rapidly screen individuals for pandemic control.

Randomized, single-blind, placebo-controlled pilot study of catheter-based myocardial gene transfer for therapeutic angiogenesis using left ventricular electromechanical mapping in patients with chronic myocardial ischemia

BACKGROUND

Catheter-based myocardial gene transfer (GTx) has not been previously tested in human subjects. Accordingly, we performed a pilot study to investigate the feasibility and safety of catheter-based myocardial GTx of naked plasmid DNA encoding vascular endothelial growth factor-2 (phVEGF-2) in patients with chronic myocardial ischemia.

METHODS AND RESULTS

A steerable, deflectable 8F catheter incorporating a 27-guage needle was advanced percutaneously to the left ventricular myocardium of 6 patients with chronic myocardial ischemia. Patients were randomized (1:1) to receive phVEGF-2 (total dose, 200 microgram), which was administered as 6 injections into ischemic myocardium (total, 6.0 mL), or placebo (mock procedure). Injections were guided by NOGA left ventricular electromechanical mapping. Patients initially randomized to placebo became eligible for phVEGF-2 GTx if they had no clinical improvement 90 days after their initial procedure. Catheter injections (n=36) caused no changes in heart rate or blood pressure. No sustained ventricular arrhythmias, ECG evidence of infarction, or ventricular perforations were observed. phVEGF-2-transfected patients experienced reduced angina (before versus after GTx, 36.2+/-2.3 versus 3.5+/-1.2 episodes/week) and reduced nitroglycerin consumption (33.8+/-2.3 versus 4.1+/-1.5 tablets/week) for up to 360 days after GTx; reduced ischemia by electromechanical mapping (mean area of ischemia, 10.2+/-3.5 versus 2.8+/-1.6 cm(2), P=0.04); and improved myocardial perfusion by SPECT-sestamibi scanning for up to 90 days after GTx when compared with images obtained after control procedure. Conclusions-This randomized trial of catheter-based phVEGF-2 myocardial GTx provides preliminary indications regarding the feasibility, safety, and potential efficacy of percutaneous myocardial GTx to human left ventricular myocardium.

Gene therapy with vascular endothelial growth factor for inoperable coronary artery disease: anesthetic management and results

Gene transfer for therapeutic angiogenesis represents a novel treatment for medically intractable angina in patients judged not amenable to further conventional revascularization. We describe the anesthetic management of 30 patients with class 3 or 4 angina, enrolled in a Phase 1 clinical trial to assess the safety and bioactivity of direct myocardial gene transfer of naked DNA-encoding vascular endothelial growth factor (phVEGF(165)), as sole therapy for refractory angina. The phVEGF(165) was injected directly into the myocardium through a mini-thoracotomy. All patients had major clinical predictors for adverse perioperative cardiac complications. Fast-track anesthetic management with remifentanil and desflurane, multimodal analgesia, and aggressive hemodynamic control with nitroglycerin and esmolol were used. All patients tolerated anesthesia and surgery without problems. No perioperative myocardial infarction, hemodynamic instability, or ventricular failure occurred. VEGF injections caused no clinically significant changes in cardiovascular function. Mean hospital stay was 3.8 days. There was one late death (5 months postoperative). Twenty-nine of 30 patients experienced reduced angina (56.2 +/- 4.1 episodes/week preoperatively versus 3.8 +/- 1.6 postoperatively, P < 0.0001) and reduced sublingual nitroglycerin consumption (60.1 +/- 4.4 tablets/week preoperatively versus 2.9 +/- 1.1 postoperatively, P < 0.0001).

IMPLICATIONS

Previously revascularized patients now judged "inoperable," continue to present with chronic, recurrent angina. Our study describes the anesthetic considerations and management of such patients treated with a novel approach by using gene therapy to stimulate angiogenesis and improve perfusion to ischemic myocardium.

Left ventricular electromechanical mapping to assess efficacy of phVEGF(165) gene transfer for therapeutic angiogenesis in chronic myocardial ischemia

BACKGROUND

NOGA left ventricular (LV) electromechanical mapping (EMM) can be used to distinguish among infarcted, ischemic, and normal myocardium. We investigated the use of percutaneous LV EMM to assess the efficacy of myocardial gene transfer (GTx) of naked plasmid DNA encoding for vascular endothelial growth factor (phVEGF(165)), administered during surgery by direct myocardial injection in patients with chronic myocardial ischemia.

METHODS AND RESULTS

A total of 13 consecutive patients (8 men, mean age 60.1+/-2. 3 years) with chronic stable angina due to angiographically documented coronary artery disease, all of whom had failed conventional therapy (drugs, PTCA, and/or CABG), were treated with direct myocardial injection of phVEGF(165) via a minithoracotomy. Foci of ischemic myocardium were identified on LV EMM by preserved viability associated with an impairment in linear local shortening. Myocardial viability, defined by mean unipolar and bipolar voltage recordings >/=5 and >/=2 mV, respectively, did not change significantly after GTx. Analysis of linear local shortening in areas of myocardial ischemia, however, disclosed significant improvement after (15.26+/-0.98%) versus before (9.94+/-1.53%, P:=0. 004) phVEGF(165) GTx. The area of ischemic myocardium was consequently reduced from 6.45+/-1.37 cm(2) before GTx to 0.95+/-0. 41 cm(2) after GTx (P:=0.001). These findings corresponded to improved perfusion scores calculated from single-photon emission CT-sestamibi myocardial perfusion scans recorded at rest (7.4+/-2.1 before GTx versus 4.5+/-1.4 after GTx, P:=0.009) and after pharmacological stress (12.8+/-2.7 before GTx versus 8.5+/-1.7 after GTx, P:=0.047).

CONCLUSIONS

The results of EMM constitute objective evidence that phVEGF(165) GTx augments perfusion of ischemic myocardium. These findings, together with reduction in the size of the defects documented at rest by serial single-photon emission CT-sestamibi imaging, suggest that phVEGF(165) GTx may successfully rescue foci of hibernating myocardium.

Intramyocardial gene therapy with naked DNA encoding vascular endothelial growth factor improves collateral flow to ischemic myocardium

Both VEGF protein and VEGF DNA in combination with an adenoviral vector have been shown to enhance collateral formation in a porcine model of chronic myocardial ischemia. We sought to determine whether direct intramyocardial injection of naked DNA encoding for VEGF could similarly improve myocardial perfusion. Initially, 23 nonischemic pigs received either 200 microg of plasmid DNA encoding beta-galactosidase (pCMVbeta, n = 11) or 500 microg of phVEGF165 (n = 12) into four separate sites in the myocardium via a small anterolateral thoracotomy incision in the fourth intercostal space. Two additional groups of pigs received an intramyocardial injection of either phVEGF165 (n = 6) or pCMVbeta (n = 7) 3 to 4 weeks after implantation of an ameroid constrictor around the left circumflex coronary artery. The injections caused no change in heart rate or blood pressure, and no ventricular arrhythmias or histologic evidence of inflammation. VEGF protein was detected by Western blot in VEGF-treated animals, with the strongest bands closest to the injection site. Plasma VEGF concentration (ELISA) increased from 3+/-2 to 27+/-13 pg/ml (p = 0.035) by day 4 after treatment. No increase in VEGF protein was noted in pCMVbeta-treated animals whereas these did stain positive for beta-Gal. Resting myocardial blood flow (colored microspheres) was significantly reduced in the ischemic versus nonischemic territory in control animals (1.07+/-0.05 versus 1.32+/-0.05; p < 0.05) but not VEGF-treated pigs (1.32+/-0.24 versus 1.13+/-0.12; p = NS). Maximal vasodilatation with adenosine significantly increased flow to the ischemic region in VEGF-treated pigs (2.16+/-0.57 versus 1.32+/-0.24; p < 0.05) but not controls (1.31+/-0.05 versus 1.17+/-0.06;p = NS). Collateral filling of the occluded circumflex artery improved in five of six VEGF-treated pigs (mean change in Rentrop score, +1.5). We conclude that direct intramyocardial transfection phVEGF165 is safe and capable of producing sufficient VEGF protein to enhance collateral formation and myocardial perfusion. This approach may offer an alternative therapy for patients with intractable myocardial ischemia not amenable to PTCA or CABG.

Intraoperative multiplane transesophageal echocardiography for guiding direct myocardial gene transfer of vascular endothelial growth factor in patients with refractory angina pectoris

Gene transfer for therapeutic angiogenesis represents a novel treatment for patients with chronic angina refractory to standard medical therapy and not amenable to conventional revascularization. We sought to assess the role of intraoperative multiplane transesophageal echocardiography (MPTEE) in guiding injection of naked DNA encoding vascular endothelial growth factor (VEGF) into the left ventricular (LV) myocardium of patients with refractory angina. After exposing the LV myocardium via a limited lateral thoracotomy, each of 17 patients in this series received 4 separate injections of VEGF DNA into different myocardial sites. Initial injections in the first patient produced intracavitary microbubbles, indicating injection of DNA into the LV chamber. Subsequently, each injection was preceded by a test injection of agitated saline. The absence of microbubbles while visualizing the LV cavity during the test injection verified that the ensuing injection of DNA would not be inadvertently squandered in the LV chamber itself. Intracavitary LV microbubbles were observed by MPTEE in 13 of 64 (20.3%) saline test injections and in 8 of 16 (50.0%) patients in which saline test injection was used, leading to adjustments in needle position. MPTEE imaging detected a previously unknown large, apical left ventricular thrombus in one patient, thereby preventing inadvertent injection of VEGF DNA through the myocardium into the thrombus. Imaging during and after injection verified no deleterious impact on LV function. We conclude that MPTEE is a useful tool for ensuring that myocardial gene therapy performed by direct needle injection results in gene transfer to the LV myocardium.

Gene therapy with vascular endothelial growth factor for inoperable coronary artery disease

BACKGROUND

Patients presenting with medically intractable angina who have undergone previous coronary bypass (CABG) and/or percutaneous revascularization procedures are frequently deemed "inoperable" based on angiographic findings of diffuse distal disease or a lack of available conduits. We initiated a phase I clinical trial to assess the safety and bioactivity of intramyocardial transfection of plasmid DNA encoding for the angiogenic mitogen vascular endothelial growth factor (ph-VEGF165) in such patients.

METHODS

phVEGF165 (125 microg, n = 10; 250 microg, n = 10) was injected directly into the myocardium through a mini left anterior thoracotomy as sole therapy in 20 patients (15 male, 5 female, age 48 to 74 years) with class III or IV angina, reversible ischemia on stress sestamibi scans, and "inoperable" coronary artery disease.

RESULTS

All patients tolerated surgery uneventfully and were extubated on the table. No perioperative myocardial infarction, hemodynamic instability, or change in ventricular function occurred. Mean hospital stay was 3.9 days. There was one late death (4 months). Plasma VEGF protein level increased from 30.6+/-4.1 pg/mL pretreatment to 73.7+/-10.1 pg/mL 14 days posttreatment (p = 0.0002) and returned to baseline by day 90. All 16 patients followed to day 90 reported a reduction in angina (nitroglycerin use/week = 60.2+/-4.9 preop vs 3.5+/-1.6 at 90 days; p<0.0001). Seventy percent (7 of 10) patients were completely angina free at 6 months. A reduction in ischemic defects on single photon emission computerized tomography sestamibi scans was observed in 13 of 17 patients at 60 days (7 of 8 in the 250-microg group). Stress perfusion score decreased from 19.4+/-3.7 at baseline to 15.9+/-3.4 at 60 days (p = 0.025). Angiographic evidence of improved collateral filling of at least one occluded vessel was observed in all patients evaluated at day 60.

CONCLUSIONS

Direct myocardial gene transfer with phVEGF165 via a mini-thoracotomy can be performed safely and may result in significant symptomatic improvement in patients with "inoperable" coronary artery disease.

Gene therapy for myocardial angiogenesis: initial clinical results with direct myocardial injection of phVEGF165 as sole therapy for myocardial ischemia

BACKGROUND

We initiated a phase 1 clinical study to determine the safety and bioactivity of direct myocardial gene transfer of vascular endothelial growth factor (VEGF) as sole therapy for patients with symptomatic myocardial ischemia.

METHODS AND RESULTS

VEGF gene transfer (GTx) was performed in 5 patients (all male, ages 53 to 71) who had failed conventional therapy; these men had angina (determined by angiographically documented coronary artery disease). Naked plasmid DNA encoding VEGF (phVEGF165) was injected directly into the ischemic myocardium via a mini left anterior thoracotomy. Injections caused no changes in heart rate (pre-GTx=75+/-15/min versus post-GTx=80+/-16/min, P=NS), systolic BP (114+/-7 versus 118+/-7 mm Hg, P=NS), or diastolic BP (57+/-2 versus 59+/-2 mm Hg, P=NS). Ventricular arrhythmias were limited to single unifocal premature beats at the moment of injection. Serial ECGs showed no evidence of new myocardial infarction in any patient. Intraoperative blood loss was 0 to 50 cm3, and total chest tube drainage was 110 to 395 cm3. Postoperative cardiac output fell transiently but increased within 24 hours (preanesthesia=4.8+/-0.4 versus postanesthesia=4.1+/-0.3 versus 24 hours postoperative=6. 3+/-0.8, P=0.02). Time to extubation after closure was 18.4+/-1.4 minutes; average postoperative hospital stay was 3.8 days. All patients had significant reduction in angina (nitroglycerin [NTG] use=53.9+/-10.0/wk pre-GTx versus 9.8+/-6.9/wk post-GTx, P<0.03). Postoperative left ventricular ejection fraction (LVEF) was either unchanged (n=3) or improved (n=2, mean increase in LVEF=5%). Objective evidence of reduced ischemia was documented using dobutamine single photon emission computed tomography (SPECT)-sestamibi imaging in all patients. Coronary angiography showed improved Rentrop score in 5 of 5 patients.

CONCLUSIONS

This initial experience with naked gene transfer as sole therapy for myocardial ischemia suggests that direct myocardial injection of naked plasmid DNA, via a minimally invasive chest wall incision, is safe and may lead to reduced symptoms and improved myocardial perfusion in selected patients with chronic myocardial ischemia.

Vascular endothelial growth factor/vascular permeability factor produces nitric oxide-dependent hypotension. Evidence for a maintenance role in quiescent adult endothelium

In vitro studies suggest that vascular endothelial growth factor/vascular permeability factor (VEGF/VPF) may stimulate release of nitric oxide (NO) from endothelial cells. To investigate the hemodynamic consequences of recombinant VEGF/VPF administered in vivo, recombinant human VEGF/VPF was administered as a bolus dose of 500 micrograms to anesthetized (n = 6) or conscious (n = 5) New Zealand White rabbits, as well as anesthetized rabbits with diet-induced hypercholesterolemia (HC; n = 7). Anesthetized Yorkshire farm pigs (no specific dietary pretreatment) were studied before and after receiving 500 micrograms intravenous (IV; n = 5) or intracoronary (IC; n = 5) VEGF/VPF. In anesthetized, normal rabbits, mean arterial pressure (MAP) fell by 20.5 +/- 1.4% (P < .05 versus baseline) within 3 minutes after IV VEGF/VPF. Pretreatment with N omega-nitro-L-arginine caused a significant inhibition of VEGF/VPF-induced hypotension. In conscious, normal rabbits, VEGF/VPF produced a consistent though lesser reduction in MAP. The fall in MAP induced by VEGF/VPF in anesthetized, HC rabbits (21.5 +/- 2.5% from baseline) was no different from that observed in normal anesthetized rabbits. In pigs, both IV and IC administration of VEGF/VPF produced a prompt reduction in MAP. Heart rate increased, while cardiac output, stroke volume, left atrial pressure, and total peripheral resistance all declined to a similar, statistically significant degree in both IV and IC groups. Epicardial echocardiography disclosed neither global nor segmental wall motion abnormalities in response to VEGF/VPF. We conclude that (1) VEGF/VPF-stimulated release of NO, previously suggested in vitro, occurs in vivo; (2) this finding suggests that functional VEGF/VPF receptors are present on quiescent adult endothelium, consistent with a maintenance function for VEGF/VPF, which may include regulation of NO; and (3) the preserved response of HC rabbits suggests that endothelial cell receptors for VEGF/VPF are spared in the setting of hypercholesterolemia.

Determination of cardiac output using acoustic quantification in critically ill patients

The reliability of automated acoustic quantification (AQ) in yielding real-time left ventricular (LV) area, volume, and ejection fraction has been validated. The purpose of this study was to explore the potential of AQ in providing accurate immediate determination of cardiac output in critically ill patients. A total of 48 patients were studied. One group consisted of 37 critically ill patients with indwelling Swan-Ganz catheters. In these patients, cardiac output by AQ, manual tracing of end-diastolic and end-systolic frames of 2-dimensional images, and thermodilution were measured. AQ was also compared with Doppler calculation of flow through the left and right ventricles in 11 additional patients. Adequate data for calculation of cardiac output with AQ were obtained in 78% of critically ill patients. There was an excellent correlation between AQ and off-line manual analysis for LV volumes (r = 0.94 and 0.91 for end-diastole and end-systole), ejection fraction (r = 0.85), and cardiac output (r = 0.93). AQ also correlated well with Doppler analysis (r = 0.97) and thermodilution technique (r = 0.95) in the determination of cardiac output. However, AQ slightly underestimated thermodilution measurements, with rather wide limits of agreement (-0.3 +/- 1.1 liter/min). There was a similar underestimation of cardiac output with manual analysis when compared with thermodilution. Given the absence of significant differences between AQ and manual analysis, this observation suggests that the bias is related to the echocardiographic determination of stroke volume, and not to errors from the automated border detection. It is concluded that AQ, besides providing information on LV volumes and ejection fraction, also can yield rapid measurements of cardiac output in most patients who are acutely ill.

Visualization of left ventricular pseudoaneurysm with panoramic transesophageal echocardiography

The diagnosis and precise delineation of a pseudoaneurysm of the left ventricle have important implications regarding surgical planning and operative repair. A 68-year-old man was admitted to the hospital with the incidental finding of a large pseudoaneurysm of the left ventricle 6 months after a St. Jude mitral valve replacement performed for partial papillary muscle rupture after an acute myocardial infarction. Panoramic transesophageal echocardiography provided a full depiction of the left ventricle and the adjoining pseudoaneurysm, enhancing the preoperative evaluation of the extent of the defect. A successful repair was undertaken with a pericardial patch.

Delineation of pulmonary artery sarcoma with multiplane and panoramic transesophageal echocardiography

A 38-year-old woman came for treatment with massive hemoptysis. A hilar density was observed on a chest radiograph. Transthoracic echocardiography demonstrated an intravascular pulmonary arterial mass that was studied in further detail with multiplane and panoramic transesophageal echocardiography. Surgical resection proved the mass to be a sarcoma arising from the main pulmonary artery. This case report demonstrates the utility of multiplane and panoramic transesophageal echocardiography in the study of intravascular tumors of the thorax.