Initial Postoperative Hemoglobin Values and Clinical Outcomes in Transfused Patients Undergoing Noncardiac Surgery

Nicholas D. Will, Daryl J. Kor, Ryan D. Frank, Melissa A. Passe, Timothy J. Weister, Martin D Zielinski, Matthew A. Warner

Research output: Contribution to journalArticle

2 Scopus citations

Abstract

BACKGROUND: Intraoperative red blood cell (RBC) transfusion is common, yet transfusion strategies remain controversial as pretransfusion hemoglobin triggers are difficult to utilize during acute bleeding. Alternatively, postoperative hemoglobin values may provide useful information regarding transfusion practices, though optimal targets remain undefined. METHODS: This is a single-center observational cohort study of adults receiving allogeneic RBCs during noncardiac surgery from 2010 through 2014. Multivariable regression analyses adjusting for patient illness, laboratory derangements, and surgical features were used to assess relationships between initial postoperative hemoglobin values and a primary outcome of hospital-free days. RESULTS: A total of 8060 patients were included. Those with initial postoperative hemoglobin <7.5 or ≥11.5 g/dL had decreased hospital-free days [mean (95% confidence interval [CI]), -1.45 (-2.50 to -0.41) and -0.83 (-1.42 to -0.24), respectively] compared to a reference range of 9.5-10.4 g/dL (overall P value .003). For those with hemoglobin <7.5 g/dL, the odds (95% CI) for secondary outcomes included acute kidney injury (AKI) 1.43 (1.03-1.99), mortality 2.10 (1.18-3.74), and cerebral ischemia 3.12 (1.08-9.01). The odds for postoperative mechanical ventilation with hemoglobin ≥11.5 g/dL were 1.33 (1.07-1.65). Secondary outcome associations were not significant after multiple comparisons adjustment (Bonferroni P < .0056). CONCLUSIONS: In transfused patients, postoperative hemoglobin values between 7.5 and 11.5 g/dL were associated with superior outcomes compared to more extreme values. This range may represent a target for intraoperative transfusions, particularly during active bleeding when pretransfusion hemoglobin thresholds may be impractical or inaccurate. Given similar outcomes within this range, targeting hemoglobin at the lower aspect may be preferable, though prospective validation is warranted.

Original languageEnglish (US)
Pages (from-to)819-829
Number of pages11
JournalAnesthesia and analgesia
Volume129
Issue number3
DOIs
StatePublished - Sep 1 2019

ASJC Scopus subject areas

  • Anesthesiology and Pain Medicine

Fingerprint Dive into the research topics of 'Initial Postoperative Hemoglobin Values and Clinical Outcomes in Transfused Patients Undergoing Noncardiac Surgery'. Together they form a unique fingerprint.

  • Cite this