Two-Year Surveillance Study of Nosocomial Infections in Patients with Hematological Malignancies: Risk Factors for Antimicrobial Resistance and Their Impact in Mortality

Background: Infections are a leading cause of morbidity and mortality in patients suffering from hematological malignancies, with resistant bacteria being a major contributor. Special precaution is needed in these patients due to immunosuppression and safety measures are warranted to reduce related complications. Surveillance studies can offer a better understanding of risk factors associated with resistant strains and mortality.

Materials and Methods: All patients hospitalized in our department of hematology, hematopoietic stem cell transplantation (HSCT) and post-transplant unit during 2014 and 2015 who presented an infectious episode were included in our study. An infectious episode was defined as a clinically detected infection not necessarily associated with specific bacteria or a positive surveillance culture. A retrospective epidemiological study was conducted using clinical data and positive cultures along with information about resistance. Our goal was to explore risk factors for infections by resistant bacteria and examine the impact of resistance on mortality.

Results: During the two-year period, we identified 439 infectious episodes, 55.4% deriving from the department of hematology and 44.6% from the HSCT and post-transplant unit. Hematological malignancy was diagnosed in 429 patients, with 62.2% suffering from acute leukemia. Prevalence of resistant bacteria in our study population was 37.1%, more frequently isolated in the HSCT and post-transplant unit (52.2% vs 34.7%, p=0.001). Most of them were gram-negative, with the main representative being KPC klebsiella pneumoniae (38.7%), followed by pseudomonas aeruginosa (17.2%). Gram-positive species accounted for 19% of resistant bacteria, including MRSA staphylococcus and VRE enterococcus species. During the first semester of 2014, we identified a burkholderia cepacia outbreak with a total of 13 cases, that was successfully handled and the microorganism was eradicated.

Possible risk factors for the isolation of resistant bacteria were examined. Factors included in our analysis were diagnosis, department of hospitalization, transplantation, prior chemotherapy, the presence of central venous catheter (CVC) and the presence of neutropenia. In multivariate analysis, the only significant risk factor associated with isolation of resistant bacteria was the department of hospitalization (OR 2.57, CI 1.31-5.02, p= 0.006). Duration of neutropenia was not correlated with increased rates of antimicrobial resistance (p=0.91).

Mortality rate was 5.2%, mostly related to infection. Of the 23 patients that died, only six deaths were attributed to refractory disease. However, from the remaining 17, we identified 7 who presented severe immunosuppression after transplantation due to graft versus host disease (GVHD). Mortality was increased in patients with isolation of a resistant strain (73.9% vs 26.1%, p=0.001) but was not related to resistant disease or in combination with GVHD (52.9% vs 50%, p= 0.9). Using multivariate analysis, significant factors predicting mortality were neutropenia, the presence of resistant bacteria and presence of CVC. Other factors included in our analysis were the department of hospitalization, transplantation, diagnosis and preceding chemotherapy. The most powerful factor was the isolation of resistant bacteria (OR 4.08, CI 1.48-11.25, p=0.006) followed by the presence of CVC (OR 3.78, CI 1.14-12.54, p=0.03).

Finally, we compared the impact of resistant bacteria between 2014 and 2015 and we observed a significant reduction in 2015 (57.1% vs 33.9%, p<0.001). We assume that this reduction is the result of more robust safety measures, including isolation of contaminated patients, intensive surveillance cultures in all patients that were routinely hospitalized and personnel discipline according to the guidelines.

Conclusion: From our study we can conclude that patient colonization and possibly personnel behavior play a significant role in spreading of resistant bacteria, as the department of hospitalization was the only significant factor of bacteria isolation. Our two departments are localized in different floors of the same building with different staff on each floor, which makes cross contamination difficult. Resistance has a major impact on mortality, but appropriate measures can change distribution and improve survival in patients with hematological malignancies.