The Effect of Thyroid Status on the Acute Heart Failure Exacerbation: A Retrospective Study

Presentation Number: SAT-0565
Date of Presentation: June 21st, 2014

Meredith McFarland*1, Qalb Khan1, René J Alvarez Jr.2 and Ajay D Rao2
1Temple University School of Medicine, 2Temple University School of Medicine, Philadelphia, PA



Studies examining the association between thyroid function and all-cause mortality have yielded conflicting results, mostly because of heterogeneity of baseline cardiovascular risk.  A direct link between TSH and cardiac function has been lacking.  A functional thyrotropin receptor has been found to be expressed in ventricular myocytes and has been implicated in TSH-induced brain natriuretic peptide (BNP) secretion, suggesting direct functional effect of TSH on the heart(1).  In a recent NHANES study, hypothyroidism was associated with greater mortality as compared to euthyroidism in those participants with congestive heart failure (CHF)(2).  Hypothyroidism was also associated with greater mortality in Blacks with CHF(2). We hypothesized that thyroid function status would be a predictor of morbidity in patients admitted with acute heart failure exacerbations as measured by BNP, ejection fraction (EF), and hospitalizations in a predominantly non-white population.


We conducted a retrospective study at Temple University Hospital (TUH, Philadelphia, USA).  We identified admissions for heart failure to the TUH Heart Failure Service during the calendar year 2012 with TSH values drawn during the inpatient admission.  Data extracted included race, age, gender, thyroid function, any prior history of hypothyroidism, EF, amiodarone use, creatinine level, brain natriuretic peptide level, and history of drug or alcohol abuse. We also measured the total number of hospitalizations these patients had for an acute heart failure exacerbation from January 2012 to December 2013.


344 admissions for heart failure were examined during the calendar year 2012.  The population was predominantly male (62% male) with mean age of 59±14 years and mostly Black (66% Black, 9% Hispanic, 15% White).  Median ejection fraction was 25% (interquartile range (IQR) 15-45%) and median BNP was 1134 pg/mL (IQR 361-1930 pg/mL).  Median TSH was 1.88 mIU/L (IQR 0.983-3.44 mIU/L) and TSH was above the upper limit of normal of the hospital assay (>5.10 mIU/L) in 15% of the admissions.  Both TSH (r=0.15, p=0.01) and EF (r=-0.34, p<0.01) significantly correlated with BNP.  In those admissions where TSH > 5.10, BNP was significantly higher (1923±185 pg/mL vs 1245±76 pg/mL, p=0.0008).  TSH did not correlate with hospitalizations during the calendar years 2012-2013.  Having a TSH > 5.10 was a significant predictor of BNP (p=0.04), even after accounting for age, gender, ethnicity, # of hospitalizations, and EF.


This retrospective study demonstrates an association between thyroid status and cardiac function in mostly non-white individuals admitted for CHF.  Potential cross-talk between TSH and natriuretic peptides may be a further line of inquiry within these racial/ethnic groups.  Definitive studies examining the role of thyroid supplementation and their effect on BNP in this population is needed.


Nothing to Disclose: MM, QK, RJA Jr., ADR