Study underscores long-term impact of COVID-19

Preliminary findings from an investigation of longer-term effects of COVID-19 show more than half experienced breathlessness, fatigue, anxiety and depression two to three months after being discharged from the hospital. The C-MORE study, by a team of scientists at the University of Oxford, followed 58 patients with moderate-to-severe laboratory-confirmed COVID-19 who were admitted to Oxford University Hospitals NHS Foundation Trust between March and May 2020, as well as 30 uninfected matched controls. The subjects underwent MRIs of the brain, lungs, heart, liver and kidneys, spirometry to assess lung function, the six-minute walk test, cardiopulmonary exercise test, plus assessments of their cognitive and mental health and quality of life. Two to three months post-discharge, 64% of the patients still were dogged by breathlessness and 55% had major fatigue. In MRI scans, tissue signal abnormalities were observed the lungs of 60% of COVID-19 patients. Moreover, 29% of patients showed kidney abnormalities, 26% had irregularities in the heart and 10% in the liver. The organ abnormalities were seen across COVID-19 patients, regardless of disease severity when they were admitted. MRI also found change in brain tissue, leading to impaired cognitive performance. Patients reported more anxiety, depression and a significant loss in quality of life vs. the controls. The study was published online Oct. 18, 2020, as a preprint on MedRxiv.

CT scan enhancement via deep learning

A  team of scientists led by Ge Wang, of Rensselaer Polytechnic Institute, has described how a physics-based deep learning algorithm can be applied to a conventional, monoenergetic computed tomography (CT) scan to produce images that would typically require dual-energy CT (DECT) to produce. Conventional CT scans make it possible to see the shape of bodily tissues, but they don’t provide enough information the composition of those tissues to distinguish between subtle substructures in soft tissue and vasculature. Higher-level, DECT addresses this problem by gathering two datasets to produce images of both tissue shape and composition; however, this approach subjects patients to a higher dose of radiation and requires additional hardware. In their work, published Oct. 19, 2020, in Patterns, the team used utilized images generated by DECT to train a model and showed that it was able to produce high-quality approximations with a relative error of less than 20% using single-spectrum CT data. “Since the deep network does not require any hardware modification on a conventional CT scanner, this ML [machine learning] method is highly cost-effective, avoiding geometrical mismatches of projection datasets collected at two energy spectra and inaccurate energy spectral correspondence in the image-based material decomposition, and the hardware cost associated with a DECT scanner,” the authors wrote.

Caution in screening for large fetus size

A predefined literature search has led a team of researchers in the U.K. to conclude that universal third-term ultrasonic screening in pregnancy could help to predict fetal macrosomia, a larger than average fetus, but is not strongly predictive of the risk of associated complications, such as shoulder dystocia. A baby who is diagnosed as having fetal macrosomia weighs more than 4,000 grams, regardless of gestational age. About 9% of babies worldwide fit that definition and are at risk for shoulder dystocia, which occurs when their anterior shoulder gets caught above the mother’s pelvic bone. Stress on the shoulder can result in clavicle fracture or nerve damage between the spinal cord and the shoulder, arm and hand. Evidence has shown that early-term induction of labor could help to reduce rates of shoulder dystocia. The team’s search included studies where ultrasound was performed as part of universal screening and those that included low- and mixed-risk pregnancies. They found that two ultrasound markers, estimated fetal weight (EFW) and abdominal circumference, could predict the majority of macrosomic infants at birth, with high diagnostic performance. However, EFW only predicted 1 in 5 cases of shoulder dystocia. The team recommended “caution prior to introducing universal third-trimester screening for macrosomia, as it would increase the rates of intervention, with potential iatrogenic harm, without clear evidence that it would reduce neonatal morbidity.” The team reported their findings in the Oct. 13, 2020, issue of PLOS Medicine.