Comparison of epidemiological, clinical and microbiological characteristics of bloodstream infection in children with solid tumours and haematological malignancies.

Bloodstream infection (BSI) is a serious complication in immunocompromised hosts. This study compares epidemiological, clinical and microbiological characteristics of BSI among children with haematological malignancies (HM) and solid tumours (ST). The study was conducted from October 2012 through to November 2015 at a referral hospital for cancer care and included the first BSI episode detected in 210 patients aged 18 years or less. BSI cases were prospectively detected by daily laboratory-based surveillance. The Centers for Disease Control and Prevention definitions for primary or secondary BSI were used. A higher proportion of use of corticosteroids (P = 0.02), chemotherapy (P = 0.01) and antibiotics (P = 0.05) before the BSI diagnosis; as well as of neutropenia (P < 0.001) and mucositis (P < 0.001) at the time of BSI diagnosis was observed in patients with HM than with ST. Previous surgical procedures (P = 0.03), mechanical ventilation (P = 0.01) and bed confinement (P < 0.001) were more frequent among children with ST. The frequency of use of temporary (P = 0.01) and implanted vascular lines (P < 0.01) was significantly higher in children with ST than with HM while the tunnelled line (P = 0.01) use was more frequent in children with HM as compared to ST. Most (n = 181) BSI cases were primary BSI. BSI associated with a tunnelled catheter was more frequent in children with HM (P < 0.01), whereas BSI associated with an implanted (P < 0.01) or temporary central line (P < 0.02) was more common in patients with ST. BSI associated with mucosal barrier injury was more frequent (P = 0.01) in children with HM. Indication for intensive care was more frequent in children (P = 0.05) with ST. Mortality ratio was similar in children with ST and HM, and length of hospital stay after BSI was higher in patients with HM than with ST (median of 19 vs. 13 days; P = 0.02). Infection caused by Gram-negative bacteria (P = 0.04) and polymicrobial infections (P = 0.05) due to Gram-positive cocci plus fungus was more common in patients with HM. These findings suggest that the characteristics of BSI acquisition and mortality can be cancer-specific.

Risk factors for early onset prosthetic valve endocarditis: a case-control study.

BACKGROUND: Early onset prosthetic valve endocarditis (EO-PVE) is an infrequent complication of cardiac valve surgery. It is considered a healthcare-associated infection due to contamination of the prosthesis during the implant or in the early postoperative period. AIM: To evaluate which factors may be related to the acquisition of EO-PVE. METHODS: A nested case-control study was conducted from 2006 to 2016. Cases were patients who had definite prosthetic endocarditis by the modified Duke criteria up to 12 months of heart valve replacement. Cases and controls were matched by age, gender, date and type of surgery. FINDINGS: There were 26 cases and 78 controls, in 2496 valve surgeries. The median incidence of EO-PVE was 1.1%. Risk factors identified during surgery were: use of ≥2 cryoprecipitate units (odds ratio (OR): 5.95; 95% confidence interval (CI): 1.31-27.0) and ≥2 plasma units (OR: 2.73; 95% CI: 1.0-7.5). In the postoperative period, associated factors were bloodstream infection (OR: 14.00; CI: 1.49-131.77), pneumonia (4.38; 1.21-15.84), any infection (4.46; 1.63-12.21), central line for ≥2 weeks (5.33; 2.06-13.78), presence of dialysis catheter (3.22; 1.15-9.03), and new open chest surgery (3.89; 1.28-11.78). Mortality at 12 months was 34.6% in cases and 6.4% in controls (OR: 7.73; CI: 2.3-26.06). CONCLUSION: Cases had more infections, invasive procedures and surgical re-interventions in the early postoperative period, which favoured contamination of the newly implanted prosthesis. A preventive approach, with reinforcement of infection control practices, may curb the incidence of this condition.