As patients began presenting with COVID-19-related pneumonia at The Mount Sinai Hospital, Hooman Poor, MD, noticed a strange phenomenon: they had the hallmarks of acute respiratory distress syndrome (ARDS)—bilateral infiltrates and very low oxygen levels—but the degree of hypoxia they were experiencing was out of proportion with their lung stiffness.
“The first thing that came to mind was diseases that affect the blood vessels of the lungs,” explains Dr. Poor, Assistant Professor of Medicine (Pulmonary, Critical Care and Sleep Medicine) at the Icahn School of Medicine at Mount Sinai and Director of Pulmonary Vascular Disease at the Mount Sinai – National Jewish Health Respiratory Institute. “We had noticed that patients with COVID-19 had a significant amount of clotting in a variety of organs, and there were autopsy reports that demonstrated small blood clots in the lungs, so the physiology was pointing to blood clots, but I had no real evidence to support it.”
Dr. Poor decided to administer thrombolysis to a small number of dying patients as a rescue therapy using tissue plasminogen activator (tPA). Although most of the patients who received tPA died weeks later, he noted significant immediate improvements in their physiology. He also noted that the two patients who received thrombolysis within 24 hours of respiratory failure survived and were subsequently discharged. This suggested that early interventions could potentially have an impact on patient outcomes, but another challenge presented: the patients’ blood clots tended to re-form quickly.
“Extensive studies have demonstrated that concomitant administration of an anticoagulant with thrombolysis is safe,” Dr. Poor says. “We typically do not administer the two at the same time in treating traditional blood clots in the lungs, but we subsequently demonstrated in a few patients with COVID-19 respiratory failure that, for thrombolysis to have a chance to work, we would need to give anticoagulants concomitantly.”
Based on these findings, Dr. Poor partnered with J Mocco, MD, MS, Vice Chair and Professor of Neurosurgery at the Icahn School of Medicine, to launch a multicenter phase 2 randomized controlled trial to look at the efficacy of administering thrombolysis with concomitant blood thinners among patients in the early stages of COVID-19-related respiratory failure. The trial is recruiting patients who require assisted breathing—including by means of ventilators, high-flow nasal cannulae, and noninvasive positive pressure—and who are demonstrating signs of clotting, which will be determined using a D-dimer test. The goal is to recruit 60 patients who will be randomized into two cohorts—one that will receive a one-time dose of 0.25 mg/kg of tenecteplase with a concomitant heparin drip within 48 hours of experiencing respiratory failure and one that will receive a placebo. Patients who have an elevated risk for bleeding, such as those who are older than 75 or who have had a stroke, will be excluded from participating.
“The No. 1 risk we are concerned about when we administer thrombolysis is bleeding in the brain, so having a neurosurgeon such as Dr. Mocco involved in developing the protocols and deciding to use tenecteplase, which is longer-acting and safer than tPA, is extremely helpful,” Dr. Poor says.
Despite the promising outcomes Dr. Poor achieved among patients who received thrombolysis, hard evidence of blood clots in the lungs of patients who presented with COVID-19-related pneumonia remained elusive, particularly given that he observed low rates of pulmonary hypertension and few signs of ventricular dysfunction. “In trying to reconcile this, I thought about the phenomenon we see among patients with cirrhosis of the liver,” Dr. Poor explains. “In addition to portal pulmonary hypertension, a condition where the blood vessels of the lungs are abnormally constricted, they can also concomitantly develop hepatopulmonary syndrome, which results in abnormal vasodilation of blood vessels in the lungs and hypoxemia. But I did not find the missing link I was looking for until I received a call from Alexandra Reynolds, MD.”
An Assistant Professor of Neurosurgery, and Neurology, at the Icahn School of Medicine, and neurointensivist in the Neurocritical Care Unit at The Mount Sinai Hospital, Dr. Reynolds was struggling with a conundrum of her own. Seeking to understand the altered mental state of mechanically ventilated COVID-19 patients, she had been using a robotic transcranial Doppler, the Lucid Robot System by Novasignal, to conduct ultrasounds for blood clots. Finding no signs of microemboli, Dr. Reynolds proceeded to conduct TCD “bubble tests”—in which saline with microbubbles is injected into a patient’s veins—and found, much to her surprise, that 9 of the 11 patients she tested were positive for microbubbles.
“She was asking me to explain what she was seeing, and I almost dropped my coffee mug when she told me,” Dr. Poor says. “Microbubbles in the brain suggest the patients have a patent foramen ovale or vasodilation of the lung capillaries. Given that patent foramen ovale has a prevalence of 20 percent, these findings suggested pulmonary vasodilation, so I said we should explore this further.”
Drs. Poor and Reynolds subsequently conducted a pilot study, “Pulmonary Vascular Dilation Detected by Automated Transcranial Doppler in COVID-19 Pneumonia,” in which 18 mechanically ventilated patients with severe COVID-19 pneumonia underwent TCD with a bubble test. The results, published in the American Journal of Respiratory and Critical Care Medicine, found that 15 patients (83 percent) had detectable microbubbles and that PaO₂/FiO₂ levels were inversely correlated with the number of microbubbles observed.
“That was significant because there had been a previous bubble study using transesophageal echocardiography that detected transpulmonary bubble transit among 26 percent of patients with classic ARDS, which was much lower than what we have been seeing among COVID-19 patients,” Dr. Poor says. “More important, there was no correlation in the previous study between the degree of bubble shunting and oxygenation. The fact that we observed this correlation among our cohort implies that pulmonary vascular dilations are a mechanism of hypoxemia in COVID-19-related respiratory failure but not in typical ARDS.”
Also notable was the fact that the patients who presented with the stiffest lungs had more microbubbles, a phenomenon that surprised Dr. Poor. “I had hypothesized there would be more bubbles among patients whose lungs were less stiff because the issue is mainly vascular, but the patients who had the stiffest lungs had the most bubbles by a wide margin.”
Based on this pilot study, Dr. Poor hypothesizes that there are two processes occurring in parallel among patients with COVID-19-related pneumonia—one that causes inflammation in the lungs and one that causes vasodilation. “The hypothesis has been that there are two distinct phenotypes of COVID-19 pneumonia and respiratory failure: type L, which is an early version of the disease when the lungs are not as stiff and the patient has low oxygen levels because of vascular dysfunction, and type H, which is the more classic ARDS with high elastance and high lung recruitability and occurs later in the disease process,” Dr. Poor says. “Instead, I think the underlying process causing vascular dilatation and dysfunction continues to worsen in the type H phenotype. Ultimately, COVID-19 respiratory failure is ARDS but with a twist—the lung inflammation is accompanied by abnormal pulmonary vasodilation and thrombosis.”
Dr. Poor has launched a multicenter study in partnership with Brown University that will build on these findings and explore another phenomenon he has observed: patients have lower oxygen levels, and are shorter of breath, when sitting up than when lying down.
“The abnormal dilations of the blood vessels of the lungs may be occurring more predominantly at the bases of their lungs when they are sitting up, and thus more blood goes to them, resulting in more shunting,” Dr. Poor says. “We want to see if the number of bubbles correlates with the position the patient is in, but more important, we want to see if the same degree of bubble shunting and transit we observed in our pilot trial occurs in a larger cohort and in patients with earlier and less severe disease. If so, that could be helpful in terms of using these bubbles as a marker of disease severity or as a surrogate endpoint in clinical trials.”
Hooman Poor, MD
Assistant Professor of Medicine (Pulmonary, Critical Care, and Sleep Medicine, and Cardiology)
Alexandra Reynolds, MD
Assistant Professor of Neurosurgery, and Neurology