According to the clinical guidelines on SFTS released by the Ministry of Health of the People's Republic of China in 2010, SFTSV-infected patients were enrolled in our study between May 2016 and September 2016. Individuals who had concurrent hepatitis C virus, hepatitis B virus, or human immunodeficiency virus (HIV)-1 infection; who were positive for Epstein-Barr virus or who met the clinical or biological criteria for bacterial or fungal infection were excluded. All 29 patients (including 2 deceased patients) were admitted to the hospital by approximately days 3–10 after disease onset. To conduct a retrospective analysis, detailed patient data (including their clinical history, physical examination findings, and routine haematological laboratory results) were extracted from the medical records. The basic clinical and laboratory characteristics of the patients with mild and severe cases of SFTSV infection at admission are listed in Table 1.
Characteristics ALL cases (N = 29) Mild (N = 15) Severe (N = 14, including 2 deceased patients) P value Age, years 58 (35-74) 54 (35-70) 61 (40-74) 0.02b Male, sex, n (%) 15(51.7) 9 (60) 6 (42.8) 0.85c Days of hospitalization, days 8 (3-21) 6(4-10) 11 (3-21) 0.003d Duration of fever, days 6 (3-10) 6 (3-8) 6 (3-10) 0.67d Plasma RNA, Log10 3.48 (1.13-6.95) 2.22 (1.13-5.61) 4.14 (2.03-6.95) 0.02d Platelet count, 109/L 48 (11-82) 50 (21-78) 49.5 (11-82) 0.69b Monocyte count, 109/L 0.21 (0.01-1.44) 0.21 (0.01-1.94) 0.19 (0.01-1.44) 0.95b Lymphocyte count, 109/L 0.68 (0.21-2.31) 0.74 (0.31-2.31) 0.70 (0.21-1.96) 0.73b Leukocyte count, 109/L 2.75 (0.75-12.63) 2.75 (0.75-5.9) 5.8 (0.95-12.63) 0.07b aData are presented as the median (range) unless otherwise specified
bAccording to a t test
cAccording to Pearson's Chi square test
dAccording to the Mann–Whitney U-test
Table 1. Differences in clinical and laboratory characteristics between patients with mild and severe cases of SFTSV infection.
After admission to the hospital, all patients with SFTSV infection received standard treatments based on the SFTS treatment guidelines of the Chinese Ministry of Health. We collected blood samples from patients at regular intervals (within 24 h of admission, every three days thereafter and on the day before discharge). The healthy controls consisted of fifteen SFTSV-uninfected blood donors who had no underlying diseases and were matched with the infected patients by sex, age and ethnic background, from whom blood samples were collected at the time of enrolment. Each sample was processed within 4 h after collection. Peripheral blood mononuclear cells (PBMCs) were separated via density gradient centrifugation with Ficoll-Paque Plus (DAKEW Biotech, China) following the manufacturer's instructions.
A total of 2 × 105 PBMCs were collected in tubes and surface-stained with APC-Cy7-labelled CD3 (300318, BioLegend, USA), PE-labelled CD56 (304605, BioLegend) or APC-labelled CD56 (362504, BioLegend), APC-labelled CD16 (302012, BioLegend) or PE-Cy7-labelled CD16 (302016, BioLegend), PE-labelled NKG2D (320806, BioLegend), FITC-labelled NKG2A (130-098-818, Miltenyi Biotec), PE-labelled CD107a (328608, BioLegend) antibodies and eFluor 506-labelled Fixable Viability Dye (65-0866-18, eBioscience, USA) at 4 ℃ for 30 min. To measure intracellular cytokines or intranuclear cytokines, cells were permeabilized with Cytofix/Cytoperm (BD, USA), followed by intracellular staining with FITC-labelled IFN-γ (11-7319-82, eBioscience) and granzyme B (372206, BioLegend) antibodies and intranuclear staining with a PE-labelled Ki-67 antibody (350504, BioLegend). Subsequently, the samples were resuspended in 200 μL of PBS before being run in a FACSCalibur flow cytometer (BD Biosciences). Approximately 5 × 104–1.0 × 105 events per tube were acquired via flow cytometry to determine the phenotype of the circulating NK cells, and analyses were performed with FlowJo software (Tree Star, Ashland, OR, USA).
PBMCs at a concentration of 106 cells/mL were incubated overnight in complete medium at 37 ℃ and 5% CO2. The following day, NK cell effector functions were determined by mixing 0.4 × 106 PBMCs with target cells at a ratio of 10:1 in V-bottom 96-well plates in a final volume of 200 μL. The suspensions were incubated for 5.5 h at 37 ℃ in 5% CO2. To measure the intracellular cytokine IFN-γ, PBMCs were stimulated with PMA (50 ng/mL; Sigma-Aldrich, USA) and ionomycin (1 μg/mL; Sigma-Aldrich). To analyse degranulation (CD107a), PBMCs were collected in tubes without either PMA or ionomycin stimulation. Brefeldin A (00-4506-51, eBioscience) was added at a dilution of 1:1000 following 2 h of co-culturing, after which the expression of CD107a and IFN-γ in CD56dim NK cells was evaluated as previously described (Bryceson et al. 2010).
Total RNA from all clinical patient serum samples was extracted using a viral RNA purification kit, and the viral load of SFTSV RNA copies was determined using a certified real-time PCR kit for the clinical diagnosis of SFTS patients, according to the manufacturer's instructions (SFDA Registration No. 340166, China) (Sun et al. 2012b; Peng et al. 2016; Zhang et al. 2017). This kit targets the small segment of SFTSV with 98.6% sensitivity and 99.1% specificity.
Our group (Xiong et al. 2016) found that the SFTSI of patients at the time of admission (where SFTSI = 5×neurological symptom level + 4×respiratory symptom level + 3×LG10 viral load -2 × LN monocyte %-7) can be used to assess disease severity. In this study, we further showed that patients with an SFTSI greater than 16 were much more likely to die. In the current study, we therefore defined patients with an SFTSI > 16 as "severe" and patients with an SFTSI < 16 as "mild".
Statistical analyses were performed with SPSS software, version 17.0 (SPSS Inc., Chicago, IL, USA), and GraphPad Prism software, version 7.03 (GraphPad Software Inc., San Diego, CA, USA). The median (range), mean ± SD or median ± 95% confidence interval (CI) was used to summarize continuous variables, and absolute values and percentages were used to report dichotomous variables. For the comparison of two or more independent groups, normally distributed data were compared using the two-tailed unpaired t test or multiple t tests; otherwise, the two-tailed Mann–Whitney U test was used. For the comparison of two paired samples, the two-tailed paired t test was applied. Categorical variables in the table were analysed using Pearson's Chi square test. Correlation analysis of two variables was conducted using Pearson's test. For all of the tests, two-tailed P-values of less than 0.05 (95% CI) were considered statistically significant.