Purpose To determine chest CT changes 6 months and 12 months after the onset of coronavirus disease 2019 (COVID-19) in patients with diabetes or hyperglycemia and the risk factors for these residual lung abnormalities. Methods In total, 141 COVID-19 patients were assigned to group 1 (diabetes), group 2 (secondary hyperglycemia) or group 3 (controls). Initial and six- and twelve-month follow-up computed tomography (CT) scans were performed 16 days, 175 days and 351 days after symptom onset, respectively. CT findings and clinical and peak laboratory parameters were collected and compared. Univariable and multivariable logistic regression analyses were performed to identify the independent predictors for the presence of residual lung abnormalities at the 6-month follow-up exam. Seven variables (age; the presence of acute respiratory distress syndrome; the duration of hospitalization; the peak levels of lactate dehydrogenase (LDH) and C-reactive protein; and the initial total CT score) were chosen in the final multivariable models. Results At the six-month follow-up, abnormalities were still observed on chest CT in 77/141 (54.6%) patients. Reticular patterns (40/141, 28.4%) and ground-glass opacities (GGOs) (29/141, 20.6%) were the most common CT abnormalities on the follow-up CT scans. Patients in Groups 1 and 2 had significantly higher incidences of residual lung abnormalities than those in Group 3 (65.4% and 58.3%, respectively vs. 36.6%; p < 0.05). Twelve months after disease onset, the chest CT changes persisted in 13/25 (52.0%) patients. A duration of hospitalization > 20 days (OR: 5.630, 95% CI: 1.394–22.744, p = 0.015), an LDH level ≥ 317 U/L (OR: 7.020, 95% CI: 1.032–47.743, p = 0.046) and a total CT score > 15 (OR: 9.919, 95% CI: 1.378–71.415, p = 0.023) were independent predictors of residual pulmonary abnormalities in patients with diabetes or secondary hyperglycemia. Conclusions A considerable proportion of surviving COVID-19 patients with diabetes or secondary hyperglycemia had residual pulmonary abnormalities six months after disease onset, and we found evidence of persistent chest CT changes at the one-year follow-up. Residual lung abnormalities were associated with longer hospital stays, higher peak LDH levels and higher initial total CT scores.

Background: Previous studies have found that impaired global myocardial systolic strain is associated with cardiovascular events in T2DM patients. However, the effect of hypertension (HT) on left atrial (LA), right ventricular (RV) and left ventricular (LV) myocardial deformation in hypertensive T2DM patients has not been fully studied by cardiac magnetic resonance feature tracking (CMR-FT). Our aim was to assess LA, RV and LV strain in T2DM patients with T2DM-HT and without hypertension using CMR-FT and to determine the underlying relationships with clinical parameters. Methods: A total of 27 T2DM patients, 23 T2DM-HT patients and 31 controls were studied. LA, LV and RV strain was evaluated using CMR-FT. The clinical and biochemical parameters of the patients were collected. Results: The T2DM patients had reduced LA global circumferential strain (LAGCS), radial strain (LAGRS), longitudinal strain (LAGLS) and right ventricular longitudinal strain (RVGLS) compared with the controls (LAGCS: 27.2 ± 2.1% vs 33.5 ± 2.4%; LAGRS: