Blood transfusion linked to acute kidney injury in patients with ACS undergoing PCI
Click Here to Manage Email Alerts
Among patients with ACS undergoing PCI, blood transfusion appears to be strongly linked to acute kidney injury, even when used in patients with preprocedural anemia who later experience a postprocedural bleeding event, according to recent findings.
In the retrospective cohort study, researchers evaluated 1,756,864 patients enrolled in the National Cardiovascular Data Registry CathPCI Registry, the largest PCI registry in the United States. Patients included in the present study were admitted to the hospital for ACS and underwent PCI between July 1, 2009, and June 30, 2014.
The researchers designated acute kidney injury (AKI) as the study’s primary outcome; AKI was defined as a post-PCI increase in serum creatinine of at least 0.5 mg/dL or at least 25% above baseline kidney values. The CathPCI data collection form was used to define transfusion, bleeding complications and in-hospital events.The researchers stratified patients by the occurrence of AKI after PCI and also by preprocedural hemoglobin (≤ 10 g/dL; > 10 to ≤ 13 g/dL; > 13 to ≤ 15 g/dL; and > 15 g/dL). Additionally, patients were grouped by changes in hemoglobin before and after PCI as follows:
- group A: no reduction in hemoglobin or reduction < 1g/dL;
- group B: reduction from 1 g/dL to < 2 g/dL;
- group C: reduced hemoglobin from 2 g/dL to < 3 g/dL; and
- group D: hemoglobin reduction ≥ 3 g/dL.
The researchers determined the relationships between preprocedural hemoglobin, transfusion and AKI, and reported adjusted ORs for the overall population and for specified subgroups.
According to the researchers, of the overall patient sample, 157,386 patients (9%) developed AKI. The following characteristics were more prevalent in patients who developed AKI: older age (66 vs. 64 years; P < .01), female sex (43.2% women with AKI vs. 31.9% women with no AKI; P < .01), and comorbid conditions such as diabetes (46.1% vs. 35.3%; P < .01) and hypertension (85.1% vs. 80.9%; P < .01).
Patients with AKI also more frequently had congestive HF within 2 weeks of hospital presentation (18.4% vs. 9.1%; P < .01).
Transfusion and AKI
Blood transfusion was administered to 38,626 (2.2%) patients in the overall cohort. Patients with low baseline hemoglobin (≤ 10 g/dL; 17.9%) and patients with bleeding (35.8%) were more likely to undergo blood transfusion. However, a small proportion of patients received blood transfusion despite normal baseline hemoglobin and a lack of known bleeding events (0.2%). AKI occurred in 35.1% of the overall cohort who were given blood transfusion vs. 8.4% of patients who did not receive transfusion (adjusted OR = 4.87; 95% CI, 4.71-5.04).
The researchers discovered a stepwise, significant correlation between the severity of hemoglobin drop and AKI in the study population. AKI occurred as follows in the strata of hemoglobin change: 7.9% in group A (the reference group); 8.4% in group B (adjusted OR = 1.22; 95% CI, 1.2-1.24); 10.8% in group C (adjusted OR = 1.71; 95% CI, 1.67-1.75) and 19.4% in group D (adjusted OR = 3.55; 95% CI, 3.43-3.67).
Hemoglobin and AKI
The researchers observed a relationship between lower baseline hemoglobin and higher incidence of AKI (≤ 10 g/dL, 19%; > 10 to ≤ 13 g/dL, 11.1%; > 13 to ≤ 15 g/dL, 7.6%; > 15 g/dL, 6.6%; P for trend < .01). Across all baseline hemoglobin levels, those who had a blood transfusion, regardless of whether they had a bleeding event, were more likely to have AKI than those who did not have a blood transfusion.
“In this retrospective study, we identified an independent association between blood transfusion and [AKI] in [patients with ACS] undergoing [PCI],” the researchers wrote. “This association was significant even among patients with anemia at baseline and who did subsequently sustain a bleeding event, suggesting that a restrictive blood transfusion policy needs to be further investigated for its potential to improve the safety of [PCIs].” – by Jennifer Byrne
Disclosure: The researchers report no relevant financial disclosures.