The differences of clinical characteristics and outcomes between imported and local patients of COVID-19 in Hunan: a two-center retrospective study

All patients were diagnosed according to the guidelines of the National Health Commission of the PRC [11], all suspected patients were tested with nasal/pharyngeal swabs by reverse transcription polymerase chain reaction (RT-PCR) immediately. If the nucleic acid test was positive, patients would be registered as confirmed.

We collected the basic information (gender, age, epidemiological history, underlying diseases, etc.), clinical characteristics and signs (fever, respiratory symptoms, gastrointestinal symptoms, respiratory frequency, fatigue, etc.), laboratory data (blood biochemistry, coagulation function, blood gas analysis, liver and kidney function assessment, infection-related biomarkers, etc.), imaging results (ground-glass opacity, consolidation, involved scope, pleural effusion, etc.), and treatment measures (respiratory support, antiviral drugs, antibiotics, hormones, etc.), from the hospital medical records. The data were as of March 2nd, 2020. If important data was missing or clarification was needed, we obtained the data by communicating with the attending clinicians. For imaging data, we used two commercial multi-detector CT scanners (LightSpeed, GE Medical Systems, USA; SOMATOM, Siemens Medical, Germany) to perform chest CT scans in all patients. CT images were collected during one or two breath holds. Imaging features include the type and distribution of lesions, the number of lesions and the lobes involved. All CT images were read by two radiologists with more than 10 years' experience, and if differences arose, a third radiologist was added for analysis.

Before the data was collected, the attending clinicians had classified all patients into four categories: mild, moderate, severe and critical types, according to the guidelines of the National Health Commission of the PRC [11]. We followed the attending clinicians' classification.

The primary outcome was death. The secondary outcomes included the incidence of serious complications, utilization rate of mechanical ventilation and the duration to negative RT-PCR tests.

Fever is defined as the temperature equal or higher than 37.3℃, slight fever is 37.3–38.0 ℃, and high fever is > 39.0 ℃. Anemia is defined as hemoglobin lower than 120 g/L. Hypoalbuminemia is the level of albumin lower than 40.0 g/L. The level of serum sodium lower than 135 mmol/L is defined as hyponatremia.

Oxygen therapy: Including nasal cannula oxygen therapy, high-flow nasal cannula oxygen therapy, non-invasive positive pressure ventilation and invasive positive pressure ventilation. All the patients were administered oxygen therapy, except the patients did not accept. The indication for setting up extracorporeal membrane oxygenation (ECMO): patient with invasive positive pressure ventilation, (1) with a FiO₂ > 90%, the oxygenation index is less than 80 mmHg, lasts for more than 3–4 h; (2) airway platform pressure ≥ 35 cmH₂O [11].

Confirmed standard [11]:

For continuous variables, we expressed them as a mean with standard deviation (SD) or median with interquartile range (IQR). We assessed the differences by two-sample T-test or Mann–Whitney test. Categorical variables were expressed as counts (%), and differences were tested by χ² or Fisher exact tests, if appropriate. For the laboratory results, we also evaluated whether the values were beyond the normal range. We used SPSS (version 26.0) for all analyses, p < 0.1 was considered statistically significant, due to the small sample size.

In our study, 74 (43.7%) patients had a clear history of living in Wuhan before the onset of the disease, including Wuhan locals and those working, studying, or traveling in Wuhan. These patients who were infected with the virus in Wuhan were regarded as imported patients, while the rest were regarded as local Hunan patients. There were no medical staff in our patient population.

The median age of all patients was 45-years (IQR 34.5 to 55). The median age of imported patients was 45-years, and local patients was 43-years. The age of the patients was mainly between 19 and 49-years (67.6% of imported patients and 58.9% of local patients). The proportion of local patients over 50-years old (41%) was slightly higher than that of imported patients (32.5%). There was no significant difference in gender between the two groups. The male-to-female ratio of all patients was close to 1:1, which was consistent with the study of Wu et al. [12].

The abnormal physical signs on pulmonary auscultation were largely crackles or mixed crackles and stridor. The proportion of abnormal pulmonary signs was significantly higher in imported patients (10.9% vs 4.2%). Only three of the imported patients had mixed crackles and stridor compared with none of the local patients.

Among all the patients, 31.4% had at least one comorbidity. Compared with local patients, imported patients had a higher proportion of underling comorbidities (35.8% vs 25.7%), no statistical significance was found. The top three comorbidities were hypertension (8.1% vs 13.7%), diabetes (5.4% vs 9.5%), and cardiovascular and cerebrovascular diseases (2.7% vs 8.4%) in each group.

In this study, cough (86.5% vs 75.8%), fever (78.4% vs 68.4%) and fatigue (both 43.2%) were still the three most common clinical symptoms of COVID-19 pneumonia in both groups. However, the proportion of gastrointestinal symptoms (such as nausea, vomiting, diarrhea, etc.) was also close to one third (29.7% vs 32.6%), which is not as low as the previous study [7]. However, no statistical significance was observed between the two groups. In addition, the median symptom onset-visit interval of both imported and local patients was 3-days, with no significant difference.

On admission, radiographs demonstrated that 130 of the 166 patients (78.3% vs 78.1%) had ground-glass opacity (GGO), and 21 of the 166 patients (13.7% vs 11.8%) had mixed GGO and consolidation. One patient had consolidation only. The proportion of bilateral pneumonia was higher in imported patients (84.9% vs 79.6%) while the proportion of unilateral pneumonia was higher in local patients (8.2% vs 14.0%). However, no significant difference was observed.

When treated for 5–7 days after admission, the proportion of patients with only GGO decreased (54.2% vs 46.2%), while the proportion of mixed GGO and consolidation patients increased (29.2% vs 35.2%) in both groups, which conforms to the natural variation of lung imaging performance.

Leukopenia accounted for 22.6% (25.7% vs 20.2%), lymphocytopenia 47.6% (48.6% vs 46.8%), anemia 24.4% (20.3% vs 27.7%), thrombocytopenia 14.3% (16.2% vs 12.8%), but there was no significant difference between the two groups at admission.

The results of blood gas analysis in the two groups were similar. Alkalosis was present in 68.6% of the patients, while 28.3% of the patients had a decrease in PaO₂ and 37.1% had a decrease in PaCO₂. No significant difference was found.

Most patients had hypoalbuminemia (56.2%), of which the albumin level of imported patients (median 37.87 g/L, 62.2%) is lower than that of local patients (median 39.64 g/L, 51.6%), but there are no significant difference. The prevalence of abnormal serum ALT was 11.8% and AST was 10.7%, while that of CK and CK-MB was less than 10%, but there was still no significant difference.

There was a significant difference in hyponatremia 25.7% vs 11.6% (p = 0.025). The median serum potassium and median serum sodium of imported patients (3.95 mmol/L and 137.0 mmol/L respectively) were significantly lower than those of local patients (4.15 mmol/L and 137.8 mmol/L), while the proportion of hyperglycemia was significantly higher than that of local patients (50% vs 34%, p < 0.1).

In addition, C-reactive protein (CRP) was elevated in more than half of the patients (58.1% vs 50.5%). The median CRP of imported patients was 12.71 mg/L, while the median CRP of local patients was 9.92 mg/L, but there was no statistical difference. Surprisingly, only three patients had increased procalcitonin.

During hospitalization, 18 patients had severe complications, such as acute respiratory distress syndrome (ARDS), shock, acute kidney injury (AKI), and multiple organ dysfunction syndrome (MODS). The incidence of ARDS in imported patients (12.2%) was higher than that in local patients (7.4%). Shock only occurred in imported patients (2.7%). Twenty patients were admitted to the ICU, of which 13.5% were imported patients and 10.5% were local patients. Oxygen therapy was administered in about 95% of patients in both groups. There were two patients treated with ECMO and one patient was treated with continuous renal replacement therapy, all were imported patients. No statistical difference was observed.

All patients received antiviral therapy during hospitalization, with about 25% receiving monotherapy and 75% receiving combination therapy. The most used antiviral drugs were Lopinavir/Ritonavir, interferon and Arbidol. The patients received monotherapy were using Lopinavir/Ritonavir or Arbidol, and the patients received combination therapy were using Lopinavir/Ritonavir and Arbidol or interferon. The proportion of monotherapy or combination therapy between two groups don’t have significantly difference.

The use of antibiotics, glucocorticoids, and gamma globulins was significantly higher in the imported patients than in local patients (60.8% vs 44.2%, 36.5% vs 17.9%, 36.5% vs 17.9%, respectively), p < 0.05. The antibiotics included Moxifloxacin, Cefoperazone, Piperacillin, and Meropenem.

As of March 2nd, 2020, a total of 154 patients were discharged, one died (imported case), and 14 (1 imported case and 13 local patients) were still in hospital. Since Wuhan was closed on 23rd January, there were no more imported patients after the 5th of February, therefore, the patients admitted to the hospital at the later stages were local patients. The median length of hospital stay was 13-days.