RED CELL ALLOIMMUNISATION RISK IN PATIENTS WITH DIFFERENT TYPES OF INFECTIONS
(Abstract release date: 05/19/16)
EHA Library. Evers D. 06/12/16; 135327; S833
Dr. Dorothea Evers
Contributions
Contributions
Abstract
Abstract: S833
Type: Oral Presentation
Presentation during EHA21: On Sunday, June 12, 2016 from 08:00 - 08:15
Location: Room H6
Background
Red cell alloantigen exposure can lead to antibody associated morbidity. A valid alloimmunisation prediction score will support identification of high-risk patients, consequently followed by offering them extended matched red cell products.Murine models have suggested inflammation as an important modulator of the humoral response towards red cell antigens.
Aims
We set out to quantify the alloimmunisation risk of various types of infections in humans.
Methods
We performed a multicenter case-control study in a source population of patients who received their first and subsequent red cell transfusions during an eight year follow-up period. Cases, defined as patients developing a first transfusion-induced red cell alloantibody, were each compared with two randomly sampled controls, defined as non-alloimmunised patients receiving at least the same number of red cell units. Logistic regression analysis, stratifying for potential confounders, was used to evaluate the association between red cell alloimmunisation and the presence of various types of infections during a 5-week ‘alloimmunisation risk period’. Tissue-invasive bacterial infections were defined as ‘severe’ or ‘mild’ according to the expected induced systemic inflammatory response accompanying the type of infection. Bacteremiae were subclassified into Gram-positive or Gram-negative bacteremiae according to the causative microorganism. Viraemia and disseminated viral zoster infections were defined as ‘disseminated viral infections', hereby contrasting local viral infections.
Results
Within a cohort of 24,063 newly transfused patients, 505 cases and 1,010 matched controls received a median of eight (interquartile range 4-16) red cell transfusions. During the risk period, 159 cases (31.5%) and 314 controls (31.1%) were diagnosed with at least one infectious episode. Alloimmunisation incidences increased upon red cell antigen exposure during severe, but not mild, bacterial infections (adjusted relative risk (RR) 1.34 with 95% confidence interval (CI) 0.97-1.85; figure). RRs were further pronounced when these bacterial infections were accompanied with long-lasting fever (RR 3.06, CI 1.57-5.96). Disseminated viral disorders demonstrated a RR of 2.41 (CI 0.89-6.53, figure). Surprisingly, bacteraemia with Gram-negative, but not with Gram-positive, microorganisms, coincided with a 2-fold reduction in alloimmunisation incidences (RR 0.58, CI 0.13-1.14, figure 1). Fungal infections, as well as elevated CRP values, and (level of) leucocytosis were not associated with red cell alloimmunisation.
Conclusion
Our findings corroborate animal studies suggesting that red cell transfusions in a setting of severe bacterial or viral infections increase the risk of alloimmunisation, contrasting Gram-negative bacteraemia which substantially reduce alloantibody formation. Data on recent or existing infections should thus be incorporated into an alloimmunisation prediction score, enabling selecting high-risk patients who will benefit most from extended matched red cell transfusions.
Session topic: Transfusion Medicine
Type: Oral Presentation
Presentation during EHA21: On Sunday, June 12, 2016 from 08:00 - 08:15
Location: Room H6
Background
Red cell alloantigen exposure can lead to antibody associated morbidity. A valid alloimmunisation prediction score will support identification of high-risk patients, consequently followed by offering them extended matched red cell products.Murine models have suggested inflammation as an important modulator of the humoral response towards red cell antigens.
Aims
We set out to quantify the alloimmunisation risk of various types of infections in humans.
Methods
We performed a multicenter case-control study in a source population of patients who received their first and subsequent red cell transfusions during an eight year follow-up period. Cases, defined as patients developing a first transfusion-induced red cell alloantibody, were each compared with two randomly sampled controls, defined as non-alloimmunised patients receiving at least the same number of red cell units. Logistic regression analysis, stratifying for potential confounders, was used to evaluate the association between red cell alloimmunisation and the presence of various types of infections during a 5-week ‘alloimmunisation risk period’. Tissue-invasive bacterial infections were defined as ‘severe’ or ‘mild’ according to the expected induced systemic inflammatory response accompanying the type of infection. Bacteremiae were subclassified into Gram-positive or Gram-negative bacteremiae according to the causative microorganism. Viraemia and disseminated viral zoster infections were defined as ‘disseminated viral infections', hereby contrasting local viral infections.
Results
Within a cohort of 24,063 newly transfused patients, 505 cases and 1,010 matched controls received a median of eight (interquartile range 4-16) red cell transfusions. During the risk period, 159 cases (31.5%) and 314 controls (31.1%) were diagnosed with at least one infectious episode. Alloimmunisation incidences increased upon red cell antigen exposure during severe, but not mild, bacterial infections (adjusted relative risk (RR) 1.34 with 95% confidence interval (CI) 0.97-1.85; figure). RRs were further pronounced when these bacterial infections were accompanied with long-lasting fever (RR 3.06, CI 1.57-5.96). Disseminated viral disorders demonstrated a RR of 2.41 (CI 0.89-6.53, figure). Surprisingly, bacteraemia with Gram-negative, but not with Gram-positive, microorganisms, coincided with a 2-fold reduction in alloimmunisation incidences (RR 0.58, CI 0.13-1.14, figure 1). Fungal infections, as well as elevated CRP values, and (level of) leucocytosis were not associated with red cell alloimmunisation.
Conclusion
Our findings corroborate animal studies suggesting that red cell transfusions in a setting of severe bacterial or viral infections increase the risk of alloimmunisation, contrasting Gram-negative bacteraemia which substantially reduce alloantibody formation. Data on recent or existing infections should thus be incorporated into an alloimmunisation prediction score, enabling selecting high-risk patients who will benefit most from extended matched red cell transfusions.
Session topic: Transfusion Medicine
Abstract: S833
Type: Oral Presentation
Presentation during EHA21: On Sunday, June 12, 2016 from 08:00 - 08:15
Location: Room H6
Background
Red cell alloantigen exposure can lead to antibody associated morbidity. A valid alloimmunisation prediction score will support identification of high-risk patients, consequently followed by offering them extended matched red cell products.Murine models have suggested inflammation as an important modulator of the humoral response towards red cell antigens.
Aims
We set out to quantify the alloimmunisation risk of various types of infections in humans.
Methods
We performed a multicenter case-control study in a source population of patients who received their first and subsequent red cell transfusions during an eight year follow-up period. Cases, defined as patients developing a first transfusion-induced red cell alloantibody, were each compared with two randomly sampled controls, defined as non-alloimmunised patients receiving at least the same number of red cell units. Logistic regression analysis, stratifying for potential confounders, was used to evaluate the association between red cell alloimmunisation and the presence of various types of infections during a 5-week ‘alloimmunisation risk period’. Tissue-invasive bacterial infections were defined as ‘severe’ or ‘mild’ according to the expected induced systemic inflammatory response accompanying the type of infection. Bacteremiae were subclassified into Gram-positive or Gram-negative bacteremiae according to the causative microorganism. Viraemia and disseminated viral zoster infections were defined as ‘disseminated viral infections', hereby contrasting local viral infections.
Results
Within a cohort of 24,063 newly transfused patients, 505 cases and 1,010 matched controls received a median of eight (interquartile range 4-16) red cell transfusions. During the risk period, 159 cases (31.5%) and 314 controls (31.1%) were diagnosed with at least one infectious episode. Alloimmunisation incidences increased upon red cell antigen exposure during severe, but not mild, bacterial infections (adjusted relative risk (RR) 1.34 with 95% confidence interval (CI) 0.97-1.85; figure). RRs were further pronounced when these bacterial infections were accompanied with long-lasting fever (RR 3.06, CI 1.57-5.96). Disseminated viral disorders demonstrated a RR of 2.41 (CI 0.89-6.53, figure). Surprisingly, bacteraemia with Gram-negative, but not with Gram-positive, microorganisms, coincided with a 2-fold reduction in alloimmunisation incidences (RR 0.58, CI 0.13-1.14, figure 1). Fungal infections, as well as elevated CRP values, and (level of) leucocytosis were not associated with red cell alloimmunisation.
Conclusion
Our findings corroborate animal studies suggesting that red cell transfusions in a setting of severe bacterial or viral infections increase the risk of alloimmunisation, contrasting Gram-negative bacteraemia which substantially reduce alloantibody formation. Data on recent or existing infections should thus be incorporated into an alloimmunisation prediction score, enabling selecting high-risk patients who will benefit most from extended matched red cell transfusions.
Session topic: Transfusion Medicine
Type: Oral Presentation
Presentation during EHA21: On Sunday, June 12, 2016 from 08:00 - 08:15
Location: Room H6
Background
Red cell alloantigen exposure can lead to antibody associated morbidity. A valid alloimmunisation prediction score will support identification of high-risk patients, consequently followed by offering them extended matched red cell products.Murine models have suggested inflammation as an important modulator of the humoral response towards red cell antigens.
Aims
We set out to quantify the alloimmunisation risk of various types of infections in humans.
Methods
We performed a multicenter case-control study in a source population of patients who received their first and subsequent red cell transfusions during an eight year follow-up period. Cases, defined as patients developing a first transfusion-induced red cell alloantibody, were each compared with two randomly sampled controls, defined as non-alloimmunised patients receiving at least the same number of red cell units. Logistic regression analysis, stratifying for potential confounders, was used to evaluate the association between red cell alloimmunisation and the presence of various types of infections during a 5-week ‘alloimmunisation risk period’. Tissue-invasive bacterial infections were defined as ‘severe’ or ‘mild’ according to the expected induced systemic inflammatory response accompanying the type of infection. Bacteremiae were subclassified into Gram-positive or Gram-negative bacteremiae according to the causative microorganism. Viraemia and disseminated viral zoster infections were defined as ‘disseminated viral infections', hereby contrasting local viral infections.
Results
Within a cohort of 24,063 newly transfused patients, 505 cases and 1,010 matched controls received a median of eight (interquartile range 4-16) red cell transfusions. During the risk period, 159 cases (31.5%) and 314 controls (31.1%) were diagnosed with at least one infectious episode. Alloimmunisation incidences increased upon red cell antigen exposure during severe, but not mild, bacterial infections (adjusted relative risk (RR) 1.34 with 95% confidence interval (CI) 0.97-1.85; figure). RRs were further pronounced when these bacterial infections were accompanied with long-lasting fever (RR 3.06, CI 1.57-5.96). Disseminated viral disorders demonstrated a RR of 2.41 (CI 0.89-6.53, figure). Surprisingly, bacteraemia with Gram-negative, but not with Gram-positive, microorganisms, coincided with a 2-fold reduction in alloimmunisation incidences (RR 0.58, CI 0.13-1.14, figure 1). Fungal infections, as well as elevated CRP values, and (level of) leucocytosis were not associated with red cell alloimmunisation.
Conclusion
Our findings corroborate animal studies suggesting that red cell transfusions in a setting of severe bacterial or viral infections increase the risk of alloimmunisation, contrasting Gram-negative bacteraemia which substantially reduce alloantibody formation. Data on recent or existing infections should thus be incorporated into an alloimmunisation prediction score, enabling selecting high-risk patients who will benefit most from extended matched red cell transfusions.
Session topic: Transfusion Medicine
{{ help_message }}
{{filter}}