© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Transplant Infectious Disease, ISSN 1398-2273
Invasive mold infections in lung and heart-lung transplant recipients: Stanford University experience R. Vazquez, M.C. Vazquez-Guillamet, J. Suarez, J. Mooney, J.G. Montoya, G.S. Dhillon. Invasive mold infections in lung and heart-lung transplant recipients: Stanford University experience. Transpl Infect Dis 2015: 17: 259–266. All rights reserved Abstract: Background. Recipients of lung transplantation (LT) and heart-lung transplantation (HLT) are at increased risk of infection, including invasive mold infections (IMIs). The clinical presentation, radiographic correlates, and outcomes of Aspergillus and nonAspergillus IMIs in this population have not been well documented. Methods. LT and HLT recipients diagnosed with IMIs between 1990 and 2012 were identified using the Stanford Translational Research Integrated Database Environment and Stanford LT and HLT clinical database. Recipient clinical and radiographic characteristics were obtained via retrospective review of medical records and compared between Aspergillus and non-Aspergillus mold recipients. Risk factors for mortality were identified using multivariate logistic regression analysis. Results. During the study period, 87 (14%) transplant recipients were diagnosed with IMIs. Aspergillus species were isolated in 63 (72%) and non-Aspergillus molds in 24 (28%) recipients. No significant difference was seen in presenting symptoms or radiographic findings between Aspergillus and non-Aspergillus mold recipients. Median time to diagnosis was 363 days in the Aspergillus group and 419 days in the non-Aspergillus group, with dissemination occurring only within the non-Aspergillus group (12.5%). Overall 90-day and 1year mortality following IMI was 24% and 44%. One-year mortality was increased in the non-Aspergillus group (39.5% vs. 60.5%, P = 0.03). Conclusions. There is significant overlap in risk factors, presentation, and radiographic patterns in IMI in LT or HLT recipients. Non-Aspergillus molds were more likely to present late, with disseminated disease, and portend increased 1-year mortality.
Lung (LT) and heart-lung transplantation (HLT) are accepted therapies for patients with advanced pulmonary and cardio-pulmonary disease. Since the initial HLT in 1981, outcomes after LT have gradually improved. However, most of these improvements are attributable to improved surgical techniques and perioperative care. The infection-related mortality remains high and provides a potential target for improving patients’ outcomes (1, 2). LT and HLT recipients, as compared to other solid organ transplant (SOT) patients, are at higher risk of infectious complications. The possible reasons include
R. Vazquez1, M.C. VazquezGuillamet1, J. Suarez2, J. Mooney3, J.G. Montoya3, G.S. Dhillon3 1
Department of Medicine, University of New Mexico, Albuquerque, New Mexico, USA, 2Universidad de la Sabana, Bogota, Colombia, 3Department of Medicine, Stanford University School of Medicine, Stanford, California, USA
Key words: invasive mold infections; lung transplant; heart-lung transplant; Aspergillus; non-Aspergillus Correspondence to: Rodrigo Vazquez Guillamet, MD, University of New Mexico School of Medicine, 1 University of New Mexico Avenue, MSC 10-5550, Albuquerque, NM 87131, USA Tel: 203 522 9106 Fax: 505 272 8700 E-mail: [email protected]
Received 7 July 2014, revised 20 October 2014, accepted for publication 19 January 2015 DOI: 10.1111/tid.12362 Transpl Infect Dis 2015: 17: 259–266
higher intensity and duration of immunosuppression, exposure of the allograft to environment, lack of bronchial arterial circulation, depressed cough reflex, and airway anatomical abnormalities. The incidence of fungal infections, especially invasive mold infections (IMIs), is second highest in LT recipients among all SOT patients (3). Non-Aspergillus molds are emerging as important pathogens in SOT, especially among LT recipients (4, 5). The risk factors and outcomes of IMIs in SOT have been described previously (6–8), although these studies included few LT or HLT recipients. The studies
259
Vazquez et al: Lung transplant and invasive mold infections
describing the clinical and radiographic presentations of invasive aspergillosis in LT and HLT recipients have been limited by small number of subjects (9, 10), leaving a gap in understanding the clinical presentation, radiographic findings, and evolution of IMIs in this population. Since the first successful HLT, more than 630 LTs and HLTs have been performed at Stanford University. We report our cumulative experience in IMIs in LT and HLT recipients, and explore differences between Aspergillus species (Aspergillus group) and non-Aspergillus molds (non-Aspergillus group).
Materials and methods The Stanford University investigational review board exempted this study from formal review. The Stanford Translational Research Integrated Database Environment (STRIDE) (11) and the clinical LT database were used to identify LT and HLT recipients with positive cultures or histopathology for molds between January 1, 1990 and January 30, 2012. STRIDE is a research and development project at Stanford University to create a standards-based informatics platform supporting clinical and translational research. The European Organization for Research and Treatment of Cancer/Mycoses Study Group (EORTC/ MSG) criteria (12) were applied retrospectively to all the patients identified by STRIDE and the clinical LT database. Those with proven or probable infection were included in the final analysis. For patients with probable infection, other etiologies were excluded. Based upon their immunosuppressed status, the LT and HLT recipients were deemed to have met the host criteria. The radiological criteria were expanded to include any radiographic abnormality. This decision was based on the exploratory nature of our study, the breadth of radiographic findings described in previous LT IMI case series, and a recent publication reporting similar mortality for patients with probable IMIs whether they met EORTC/MSG radiological criteria or not (13). In case of multiple transplant procedures, the date of the first surgery was used to calculate time to infection. All of the discrete episodes of IMIs were recorded, but only the initial infections were used in the analyses. The infections were defined as disseminated either if at least 2 non-contiguous sites were involved, or if there was central nervous system infection. The infections were classified as “early” if they occurred within 90 days of transplant. All others were recorded as late infections.
260
Recipient demographic and clinical variables collected included age, gender, ethnicity, underlying diagnosis, transplant date and type, presenting symptoms, microbiologic data, radiographic chest computed tomography (CT) findings, infection site, and risk factors for IMIs. Absolute neutrophil count, use of immunosuppressive drugs, use of antifungal prophylaxis, cytomegalovirus (CMV) status, infection or disease according to established definitions (14), and acute rejection treatment were included as potential risk factors. Treatment data consisted of antifungal regimens, surgical resections, and changes in immunosuppression. All-cause mortality at 90 days and 1 year from the diagnosis of IMI was recorded. Standardized immunosuppression and fungal opportunistic infection prophylaxis protocols changed over the study period and are presented in Figure 1. Rejection was diagnosed according to consensus criteria at the time of evaluation (15). The data are reported as percentages or means standard deviation. Categorical data were compared using v2-test or Fisher’s exact test, and continuous variables with unpaired t-test. Univariate and multivariate logistic regression analyses were used to determine mortality risk factors at 90 days and 1 year. P-value
Invasive mold infections in lung and heart-lung transplant recipients: Stanford University experience R. Vazquez, M.C. Vazquez-Guillamet, J. Suarez, J. Mooney, J.G. Montoya, G.S. Dhillon. Invasive mold infections in lung and heart-lung transplant recipients: Stanford University experience. Transpl Infect Dis 2015: 17: 259–266. All rights reserved Abstract: Background. Recipients of lung transplantation (LT) and heart-lung transplantation (HLT) are at increased risk of infection, including invasive mold infections (IMIs). The clinical presentation, radiographic correlates, and outcomes of Aspergillus and nonAspergillus IMIs in this population have not been well documented. Methods. LT and HLT recipients diagnosed with IMIs between 1990 and 2012 were identified using the Stanford Translational Research Integrated Database Environment and Stanford LT and HLT clinical database. Recipient clinical and radiographic characteristics were obtained via retrospective review of medical records and compared between Aspergillus and non-Aspergillus mold recipients. Risk factors for mortality were identified using multivariate logistic regression analysis. Results. During the study period, 87 (14%) transplant recipients were diagnosed with IMIs. Aspergillus species were isolated in 63 (72%) and non-Aspergillus molds in 24 (28%) recipients. No significant difference was seen in presenting symptoms or radiographic findings between Aspergillus and non-Aspergillus mold recipients. Median time to diagnosis was 363 days in the Aspergillus group and 419 days in the non-Aspergillus group, with dissemination occurring only within the non-Aspergillus group (12.5%). Overall 90-day and 1year mortality following IMI was 24% and 44%. One-year mortality was increased in the non-Aspergillus group (39.5% vs. 60.5%, P = 0.03). Conclusions. There is significant overlap in risk factors, presentation, and radiographic patterns in IMI in LT or HLT recipients. Non-Aspergillus molds were more likely to present late, with disseminated disease, and portend increased 1-year mortality.
Lung (LT) and heart-lung transplantation (HLT) are accepted therapies for patients with advanced pulmonary and cardio-pulmonary disease. Since the initial HLT in 1981, outcomes after LT have gradually improved. However, most of these improvements are attributable to improved surgical techniques and perioperative care. The infection-related mortality remains high and provides a potential target for improving patients’ outcomes (1, 2). LT and HLT recipients, as compared to other solid organ transplant (SOT) patients, are at higher risk of infectious complications. The possible reasons include
R. Vazquez1, M.C. VazquezGuillamet1, J. Suarez2, J. Mooney3, J.G. Montoya3, G.S. Dhillon3 1
Department of Medicine, University of New Mexico, Albuquerque, New Mexico, USA, 2Universidad de la Sabana, Bogota, Colombia, 3Department of Medicine, Stanford University School of Medicine, Stanford, California, USA
Key words: invasive mold infections; lung transplant; heart-lung transplant; Aspergillus; non-Aspergillus Correspondence to: Rodrigo Vazquez Guillamet, MD, University of New Mexico School of Medicine, 1 University of New Mexico Avenue, MSC 10-5550, Albuquerque, NM 87131, USA Tel: 203 522 9106 Fax: 505 272 8700 E-mail: [email protected]
Received 7 July 2014, revised 20 October 2014, accepted for publication 19 January 2015 DOI: 10.1111/tid.12362 Transpl Infect Dis 2015: 17: 259–266
higher intensity and duration of immunosuppression, exposure of the allograft to environment, lack of bronchial arterial circulation, depressed cough reflex, and airway anatomical abnormalities. The incidence of fungal infections, especially invasive mold infections (IMIs), is second highest in LT recipients among all SOT patients (3). Non-Aspergillus molds are emerging as important pathogens in SOT, especially among LT recipients (4, 5). The risk factors and outcomes of IMIs in SOT have been described previously (6–8), although these studies included few LT or HLT recipients. The studies
259
Vazquez et al: Lung transplant and invasive mold infections
describing the clinical and radiographic presentations of invasive aspergillosis in LT and HLT recipients have been limited by small number of subjects (9, 10), leaving a gap in understanding the clinical presentation, radiographic findings, and evolution of IMIs in this population. Since the first successful HLT, more than 630 LTs and HLTs have been performed at Stanford University. We report our cumulative experience in IMIs in LT and HLT recipients, and explore differences between Aspergillus species (Aspergillus group) and non-Aspergillus molds (non-Aspergillus group).
Materials and methods The Stanford University investigational review board exempted this study from formal review. The Stanford Translational Research Integrated Database Environment (STRIDE) (11) and the clinical LT database were used to identify LT and HLT recipients with positive cultures or histopathology for molds between January 1, 1990 and January 30, 2012. STRIDE is a research and development project at Stanford University to create a standards-based informatics platform supporting clinical and translational research. The European Organization for Research and Treatment of Cancer/Mycoses Study Group (EORTC/ MSG) criteria (12) were applied retrospectively to all the patients identified by STRIDE and the clinical LT database. Those with proven or probable infection were included in the final analysis. For patients with probable infection, other etiologies were excluded. Based upon their immunosuppressed status, the LT and HLT recipients were deemed to have met the host criteria. The radiological criteria were expanded to include any radiographic abnormality. This decision was based on the exploratory nature of our study, the breadth of radiographic findings described in previous LT IMI case series, and a recent publication reporting similar mortality for patients with probable IMIs whether they met EORTC/MSG radiological criteria or not (13). In case of multiple transplant procedures, the date of the first surgery was used to calculate time to infection. All of the discrete episodes of IMIs were recorded, but only the initial infections were used in the analyses. The infections were defined as disseminated either if at least 2 non-contiguous sites were involved, or if there was central nervous system infection. The infections were classified as “early” if they occurred within 90 days of transplant. All others were recorded as late infections.
260
Recipient demographic and clinical variables collected included age, gender, ethnicity, underlying diagnosis, transplant date and type, presenting symptoms, microbiologic data, radiographic chest computed tomography (CT) findings, infection site, and risk factors for IMIs. Absolute neutrophil count, use of immunosuppressive drugs, use of antifungal prophylaxis, cytomegalovirus (CMV) status, infection or disease according to established definitions (14), and acute rejection treatment were included as potential risk factors. Treatment data consisted of antifungal regimens, surgical resections, and changes in immunosuppression. All-cause mortality at 90 days and 1 year from the diagnosis of IMI was recorded. Standardized immunosuppression and fungal opportunistic infection prophylaxis protocols changed over the study period and are presented in Figure 1. Rejection was diagnosed according to consensus criteria at the time of evaluation (15). The data are reported as percentages or means standard deviation. Categorical data were compared using v2-test or Fisher’s exact test, and continuous variables with unpaired t-test. Univariate and multivariate logistic regression analyses were used to determine mortality risk factors at 90 days and 1 year. P-value
