Hyperbaric Oxygen for Long COVID? Heart Ultrasounds Offer Limited Support

For people with long COVID symptoms, hyperbaric oxygen therapy modestly improved one parameter of heart function in a small randomized trial.

Left ventricular (LV) global longitudinal strain (GLS) -- an alternative to ejection fraction for the echocardiographic measurement of systolic function -- changed significantly from borderline impaired at baseline to normal after 8 weeks of oxygen therapy (from -17.8% to -20.2%, P=0.0001), reported Marina Leitman, MD, of Shamir Medical Center and Sackler School of Medicine at Tel Aviv University in Israel.

Meanwhile, controls showed no significant change in GLS after their sham sessions (from -17.8% to -19.1%), they noted in a poster presentation at European Association of Cardiovascular Imaging 2023, a scientific congress from the European Society of Cardiology.

"The findings suggest that [hyperbaric oxygen therapy] promotes recovery of cardiac function in patients with post-COVID syndrome. More research is needed to collect long-term results and determine the optimal number of sessions for maximum therapeutic effect," Leitman said in a press release.

"More studies are needed to determine which patients will benefit the most," she added, "but it may be that all long COVID patients should have an assessment of global longitudinal strain and be offered hyperbaric oxygen therapy if heart function is reduced."

Hyperbaric oxygen therapy comprises 100% pure oxygen delivered at high pressure to increase delivery to the body's tissues. Investigators have sought to test the intervention in people with post-COVID syndrome, who may have symptoms such as shortness of breath, fatigue, cough, chest pain, and rapid or irregular heartbeat after SARS-CoV-2 infection. There is limited evidence that hyperbaric oxygen may improve fatigue and cognitive outcomes outside its approved indications for wound care or decompression sickness.

Moreover, the literature tying together LV function and long COVID is itself conflicting.

In one study, there were various parameters of cardiac dysfunction on transthoracic echocardiography flagged 3 to 4 months after hospitalization for COVID -- but not LV GLS. Other work has found echocardiographic pathologies like LV GLS to be common, but unrelated to shortness of breath in long COVID. Yet another group has shown little or no LV dysfunction at all after COVID.

"Without any data on clinical status changes in these patients, it's difficult to know whether hyperbaric oxygen therapy might hold promise in real-world settings," cautioned Nicole Bhave, MD, a cardiologist and imaging specialist at the University of Michigan in Ann Arbor, who was not involved with the study.

James Thomas, MD, a heart valve disease specialist at Northwestern University in Chicago and past president of the American Society of Echocardiography, agreed, adding that "it would be important to know whether more negative GLS was associated with improved functional status. Clearly much more work needs to be done, but the authors are to be congratulated on using novel indices to assess a novel therapy in a poorly understood clinical entity."

The double-blind trial included 60 patients who had had long COVID symptoms for at least 3 months after mild to moderate symptomatic COVID. Participants were randomly split between the hyperbaric oxygen therapy and sham groups.

Each patient underwent five sessions per week over 8 weeks. The intervention group received 100% oxygen through a mask at a pressure of two atmospheres for 90 minutes, with 5-minute air breaks every 20 minutes. Those in the sham group received 21% oxygen by mask at one atmosphere for 90 minutes.

Echocardiography was conducted at baseline and again at the end of the study (1 to 3 weeks after the last treatment session).

All participants had normal LV ejection fractions. Yet compared with a normal GLS value of around 20%, the average baseline GLS was -17.8% across study participants and could be considered reduced in nearly half the patients, according to Leitman and colleagues.

Bhave and Thomas noted that most labs use a GLS cutoff of -18% to indicate normal in clinical practice.

"Given the test-retest variability in strain, we generally look for 15% change in an individual's value to be confident it's real, so I would like to see the individual data from this study, as the overall change was 11.2%," Thomas told MedPage Today.

Comment