Anita Modi1, Lisa Rybicki2, Navneet S Majhail3, Sherif Beniameen Mossad1. 1. Department of Infectious Diseases, Cleveland Clinic, Cleveland, OH, USA. 2. Department of Quantitative Health Sciences, Cleveland Clinic, Cleveland, OH, USA. 3. Department of Hematology and Oncology, Blood and Marrow Transplant Program, Cleveland Clinic, Cleveland, OH, USA.
Abstract
BACKGROUND: Infections are the most common cause of non-relapse mortality in adult allogeneic hematopoietic stem cell transplant (allo HSCT) recipients. Acute gastrointestinal graft-vs-host disease (GI GVHD) often leads to friable mucosa as well as treatment interventions which can increase risk of infection. Our primary objective was to describe the relationship between increasing grades of acute GI GVHD and incidence of bloodstream infections (BSI). METHODS: We reviewed 441 adults who underwent allo HSCT from 2011 to 2017 and were clinically diagnosed with GI GVHD, non-GI GVHD, or no GVHD based on the modified Glucksberg scale within the first 100 days of transplantation. The maximum grades of acute GI GVHD and non-GI GVHD were used in the analysis. BSI was defined based on the presence of a blood culture positive for bacteria or fungi and treatment with antibiotics. The incidence of BSI within the first 180 days of transplantation was estimated with the cumulative incidence method. Fine and Gray regression was used to assess association between clinical grade of acute GI GVHD and BSI risk, adjusting for grade of non-GI GVHD and for other significant baseline patient risk factors for BSI identified by multivariable analysis. Results are shown as hazard ratio (HR) and 95% confidence interval (CI). A similar analysis was conducted in 130 patients with histologic grade of acute GI GVHD. RESULTS: Overall BSI incidence by day 180 was 32%; gram-negative bacilli were the predominant organisms, followed by gram-positive cocci and fungi. Patients with grade 4 acute GI GVHD had higher risk of BSI as compared to patients with no GI GVHD (HR 2.98, CI 1.65-5.37, P < .001), while those with grade 3 acute GI GVHD had similar BSI risk (HR 0.89, CI 0.36-2.21, P = .81). Grade of GI GVHD had no association with risk of non-BSI. Results were similar in patients with histologic grade acute GI GVHD. Patients who developed BSI or non-BSI had significantly higher overall mortality risk compared to those without infectious complications (HR 2.52, CI 1.92-3.31, P < .001 for BSI; HR 1.60, CI 1.20-2.13, P = .001 for non-BSI). CONCLUSIONS: Grade 4 acute GI GVHD is associated with a higher risk of BSI, which is in turn associated with a higher risk of overall mortality in this population. Understanding the relationships between acute GI GVHD, BSI, and overall mortality can guide future treatment strategies for adult allo HSCT recipients.
BACKGROUND: Infections are the most common cause of non-relapse mortality in adult allogeneic hematopoietic stem cell transplant (allo HSCT) recipients. Acute gastrointestinal graft-vs-host disease (GI GVHD) often leads to friable mucosa as well as treatment interventions which can increase risk of infection. Our primary objective was to describe the relationship between increasing grades of acute GI GVHD and incidence of bloodstream infections (BSI). METHODS: We reviewed 441 adults who underwent allo HSCT from 2011 to 2017 and were clinically diagnosed with GI GVHD, non-GI GVHD, or no GVHD based on the modified Glucksberg scale within the first 100 days of transplantation. The maximum grades of acute GI GVHD and non-GI GVHD were used in the analysis. BSI was defined based on the presence of a blood culture positive for bacteria or fungi and treatment with antibiotics. The incidence of BSI within the first 180 days of transplantation was estimated with the cumulative incidence method. Fine and Gray regression was used to assess association between clinical grade of acute GI GVHD and BSI risk, adjusting for grade of non-GI GVHD and for other significant baseline patient risk factors for BSI identified by multivariable analysis. Results are shown as hazard ratio (HR) and 95% confidence interval (CI). A similar analysis was conducted in 130 patients with histologic grade of acute GI GVHD. RESULTS: Overall BSI incidence by day 180 was 32%; gram-negative bacilli were the predominant organisms, followed by gram-positive cocci and fungi. Patients with grade 4 acute GI GVHD had higher risk of BSI as compared to patients with no GI GVHD (HR 2.98, CI 1.65-5.37, P < .001), while those with grade 3 acute GI GVHD had similar BSI risk (HR 0.89, CI 0.36-2.21, P = .81). Grade of GI GVHD had no association with risk of non-BSI. Results were similar in patients with histologic grade acute GI GVHD. Patients who developed BSI or non-BSI had significantly higher overall mortality risk compared to those without infectious complications (HR 2.52, CI 1.92-3.31, P < .001 for BSI; HR 1.60, CI 1.20-2.13, P = .001 for non-BSI). CONCLUSIONS: Grade 4 acute GI GVHD is associated with a higher risk of BSI, which is in turn associated with a higher risk of overall mortality in this population. Understanding the relationships between acute GI GVHD, BSI, and overall mortality can guide future treatment strategies for adult allo HSCT recipients.