A predictive model of herpes zoster after allogeneic hematopoietic stem cell transplantation: VZV reactivation following antiviral prophylaxis discontinuation.

Herpes zoster (HZ) refers to the rash appearing on dermatomes due to varicella zoster virus (VZV) reactivation. The incidence of HZ is significantly higher in allogeneic hematopoietic stem cell transplantation (allo-HSCT) recipients than in non-HSCT recipients. Although acyclovir prophylaxis is routinely administered to every allo-HSCT recipient for 1 year after transplantation, some individuals eventually develop late-onset HZ after completing prophylaxis. Little information is known about the clinical features of HZ after prophylactic antiviral treatment discontinuation, and an effective predictive model of late-onset HZ needs to be established. A total of 3366 patients who had received allo-HSCT from 2012 to 2017 were included in our study, among whom 201 developed HZ after 1 year (late-onset HZ). We designed a nested case-control study to identify potential predictors of late-onset HZ. Finally, we established a predictive model using binary logistic regression analysis. Age (p < .001), use of immunosuppressants at +1 year (p < .001), CD4-CD8 ratio at +1 year (p < .001), certain mental disorders (depression, anxiety, insomnia and adjustment disorder) (p < .001), engraftment time of neutrophils (p < .001), and CD8+ cell count at +30 days (p < .001) were independent predictors of late-onset HZ. A risk grading system was established based on regression coefficients. Discrimination and calibration analysis indicated that the model had good performance. We also identified several predictive factors of the incidence of HZ-related complications. This is the first scoring system for predicting the incidence of late-onset HZ after allo-HSCT. This model can be applied to identify individuals at high risk of late-onset HZ in the early period after receiving allo-HSCT.

Comparative genomics analysis of Acinetobacter baumannii multi-drug resistant and drug sensitive strains in China.

The incidence of multidrug-resistant Acinetobacter baumannii has posed a major challenge for clinical treatment. There is still a significant gap in understanding the mechanism causing multi-drug resistance (MDR). In this study, the genomes of 10 drug sensitive and 10 multi-drug resistant A.baumannii strains isolated from a hospital in China were sequenced and compared. The antibiotic resistance genes, virulence factors were determined and CRIPSR-Cas system along with prophages were detected. The results showed that MDR strains are significantly different from the drug sensitive strains in the CARD entries, patterns of sequences matching up to plasmids, VFDB entries and CRISPR-Cas system. MDR strains contain unique CARD items related to antibiotic resistance which are absent in sensitive strains. Furthermore, sequences from genomes of MDR strains can match up with plasmids from more diversified bacteria genera compared to drug sensitive strains. MDR strains also contain a lower level of CRISPR genes and larger amount of prophages, along with higher levels of spacer sequences. These findings provide new experimental evidences for the study of the antibiotic resistance mechanism of A. baumannii.