OBJECTIVES: To compare microbial agents causing first episodes of bacteraemia in patients with and without haematological malignancies, and to examine mortality within 90 days of follow-up. METHODS: We conducted a population-based study in North Jutland County, Denmark including patients (> or = 15 years) registered in the County Bacteraemia Registry with their first episode of bacteraemia during the period 1992-2002. Diseases prior to the bacteraemia were identified in the Hospital Discharge Registry. Mortality was determined through the Civil Registration System. We estimated adjusted mortality rate ratios (MRRs) using Cox's regression analysis. RESULTS: We included 7456 patients with a first episode of bacteraemia; of these, 444 (6%) had a haematological malignancy. Haematological patients had almost the same proportions of gram-negative (49% vs. 50%) and gram-positive (37% vs. 39%) bacteraemia as other patients. Cumulative 30-day mortality in patients with haematological malignancies was 32% vs. 20% in patients without any malignancy, yielding an adjusted MRR of 1.6 (95% CI 1.3-2.0). For patients with non-haematological malignancies the adjusted MRR was 1.5 (95% CI 1.4-1.7). CONCLUSION: Despite a similar distribution of the microbial agents causing bacteraemia, patients with haematological malignancies and also patients with non-haematological malignancies had increased 30-day mortality of bacteraemia when compared with patients without any malignancy.
OBJECTIVES: To compare microbial agents causing first episodes of bacteraemia in patients with and without haematological malignancies, and to examine mortality within 90 days of follow-up. METHODS: We conducted a population-based study in North Jutland County, Denmark including patients (> or = 15 years) registered in the County Bacteraemia Registry with their first episode of bacteraemia during the period 1992-2002. Diseases prior to the bacteraemia were identified in the Hospital Discharge Registry. Mortality was determined through the Civil Registration System. We estimated adjusted mortality rate ratios (MRRs) using Cox's regression analysis. RESULTS: We included 7456 patients with a first episode of bacteraemia; of these, 444 (6%) had a haematological malignancy. Haematological patients had almost the same proportions of gram-negative (49% vs. 50%) and gram-positive (37% vs. 39%) bacteraemia as other patients. Cumulative 30-day mortality in patients with haematological malignancies was 32% vs. 20% in patients without any malignancy, yielding an adjusted MRR of 1.6 (95% CI 1.3-2.0). For patients with non-haematological malignancies the adjusted MRR was 1.5 (95% CI 1.4-1.7). CONCLUSION: Despite a similar distribution of the microbial agents causing bacteraemia, patients with haematological malignancies and also patients with non-haematological malignancies had increased 30-day mortality of bacteraemia when compared with patients without any malignancy.
Authors: Michael R Clausen; Matthew J Maurer; Sinna Pilgaard Ulrichsen; Thomas S Larsen; Bodil Himmelstrup; Dorthe Rønnov-Jessen; Brian K Link; Andrew L Feldman; Susan L Slager; Grzegorz S Nowakowski; Carrie A Thompson; Per Trøllund Pedersen; Jakob Madsen; Robert S Pedersen; Jette Sønderskov Gørløv; James R Cerhan; Mette Nørgaard; Francesco D'Amore Journal: Clin Epidemiol Date: 2019-11-14 Impact factor: 4.790