Premium
Haemoglobin responses to transfusion in severe iron deficiency anaemia: potential impact of gastrointestinal disorders
Author(s) -
Bosch X.,
Montori E.,
GuerraGarcía M.,
CostaRodríguez J.,
Quintanilla M. H.,
TolosaChapasian P. E.,
Moreno P.,
Guasch N.,
LópezSoto A.
Publication year - 2017
Publication title -
vox sanguinis
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.68
H-Index - 83
eISSN - 1423-0410
pISSN - 0042-9007
DOI - 10.1111/vox.12491
Subject(s) - medicine , confidence interval , blood transfusion , prospective cohort study , iron deficiency , population , gastroenterology , cohort , confounding , gastrointestinal bleeding , hemoglobin , anemia , environmental health
Background and Objectives Red blood cell ( RBC ) transfusion may be justified in iron deficiency anaemia ( IDA ) when an increase in oxygen delivery is needed, as sometimes occurs in subjects with haemoglobin <8·0 mg/ dL , serious comorbidities or at risk of cardiovascular instability. Earlier investigations showed that some patients with severe IDA requiring transfusion had lower than expected post‐transfusion haemoglobin levels with poorer clinical outcomes than other patients. After hypothesizing that haemoglobin responses to transfusion were different and that the underlying gastrointestinal ( GI ) disorders causing IDA could be a confounder explaining this association, these responses were analysed in a prospective cohort of IDA adults referred for outpatient GI investigation. Materials and Methods Transfused patients with proven IDA , baseline haemoglobin at referral <9·0 g/ dL and no extraintestinal bleeding were eligible. To assess a homogeneous population, only GI disorders known to cause occult bleeding were considered. Haemoglobin increments per 100 mL of RBC s were investigated. Results In total, 2818 patients were enrolled over 10·5 years. On multivariable regression, diffuse angiodysplasias and GI cancer independently predicted for reduced increments in post‐transfusion haemoglobin [adjusted regression coefficients: −0·082 (95% confidence interval, −0·093 to −0·072) and −0·073 (95% confidence interval, −0·081 to −0·066), respectively, P < 0·001 in both]. Haemoglobin responses in the remaining bleeding disorders were adequate and agreed with the principle that one RBC unit increases the haemoglobin an average of 1 g/ dL . Conclusion The potential differential impact of GI disorders on changes in haemoglobin levels after RBC transfusion could be useful for transfusing physicians, especially for diagnostic purposes.