Infectious complications in hematology patients: A clinical focus on prevention

The aim of this thesis was to contribute to infection preventive strategies in hematology patients with prolonged neutropenia. Invasive pulmonary aspergillosis (IPA) develops by pulmonary deposition of conidia. A placebo-controlled trial on the efficacy of prophylactic aerosolized liposomal amphotericin-B (L-AmB) was performed in hematology patients during treatment-related neutropenia. For the intent-to-treat analysis, 18 of 132 placebo patients developed IPA, versus 6 of 139 in L-AmB patients (P=.005). Short-term prophylactic nebulization of liposomal amphotericin B was not associated with decline in pulmonary function or systemic adverse effects, only coughing was significantly more reported. In a study among acute leukemia patients, the mortality rate was 26% and 16%, respectively, in patients with and without IPA (P=.08), with corrected additional IPA-related costs of €15280 (P<.001), which makes IPA prevention both efficacious and cost-effective. Current guidelines advocate administration of 7-14 days of broad-spectrum antibiotics for unexplained ferver (UF) in neutropenic hematology patients. Effective antimicrobial prophylaxis reduces the incidence of gram-negative infections, which may allow shorter treatment. In the Erasmus MC, broad-spectrum antibiotics are initially administered, but discontinued when after 72 hours no bacterial etiology is documented. Among 166 hematology patients on selective antibiotic prophylaxis, 317 febrile episodes were observed; 56% were due to UF. Due to the restrictive antibiotic policy, the mean treatment duration was only 3.7 days. Overall-mortality after neutrophil recovery was 3.6% (6 of 166 patients). No patient died from untreated bacterial infection, suggesting that discontinuation of broad-spectrum antibiotics in this setting is safe, provided that no infectious etiology is documented. Diagnosing catheter-related bloodstream infection (CRBSI) often involves tip culture, using the conventional semi-quantitative roll-plate method. The quantitative sonication technique could have additional value, as it may also detect endoluminal microorganisms. However, in a prospective study of 313 tunnelled catheters from 279 hematology patients, the diagnostic performance in patients with suspected CRBSI was equally limited for both methods. Finally, in a placebo-controlled trial on the efficacy of a daily prophylactic 70%-ethanol lock on the prevention of endoluminal CRBSI in tunneled devices, a 3.6-fold, non-significant, reduction was observed for ethanol-treated hematology patients. Partially, the lack of significance may reflect a lack of power. More patients receiving ethanol discontinued lock-therapy, because of non-severe adverse events.

• View at Worldcat

Free Full Text ( Final Version , 1mb )