Hypothyroidism and Renal Function in Patients with Systolic Heart Failure

Abstract

The extent to which hypothyroidism affects renal function in patients with heart failure remains incompletely explored, despite the known adverse prognostic implications of renal dysfunction in these patients.

In a pilot retrospective study, we evaluated 75 patients (age, ≥18 yr) with left ventricular ejection fractions <0.40. Forty-five patients had normal thyroid function (thyroid-stimulating hormone [TSH], 0.35–5.5 μIU/mL) and 30 had hypothyroidism. The group with hypothyroidism was subdivided into 17 patients who had controlled hypothyroidism (TSH, 0.35–5.5 μIU/mL) and 13 who had uncontrolled hypothyroidism (TSH, >5.5 μIU/mL). Renal function, measured in terms of glomerular filtration rate, was analyzed once in each patient, and the populations were statistically compared, with P <0.05 conferring statistical significance.

Baseline characteristics in all groups were similar. Mean glomerular filtration rate was better in patients with normal thyroid function than those with hypothyroidism (75.45 ± 31.48 vs 63.95 ± 21.43 mL/min/1.73 m²; P=0.032). There was no significant difference between patients with controlled hypothyroidism (66.89 ± 24.18 mL/min/1.73 m²) and those with normal thyroid function (P=0.131). In patients with uncontrolled hypothyroidism, mean glomerular filtration rate (60.2 ± 17.4 mL/min/1.73 m²) was significantly worse than in patients with normal thyroid function (P=0.015).

We found that heart-failure patients with insufficiently treated hypothyroidism have worse renal function than do patients whose thyroid function is normal or whose hypothyroidism is effectively treated. Larger studies will be needed in order to evaluate this conclusion further. We recommend that hypothyroidism in heart-failure patients be strictly controlled, lest it affect prognosis adversely.

Heart failure is a major public health problem. In the United States, approximately 5 million people have heart failure, and an estimated 550,000 new cases will be diagnosed annually.¹ Its prevalence in persons older than age 65 is nearly 10 in 1,000,² and the overall number of deaths exceeds 260,000 annually.³ Several factors contribute to an adverse prognosis for heart-failure patients.^4–9 Among these, renal dysfunction has been strongly correlated with poor outcome.^7,9–12 Hypothyroidism can promote renal dysfunction,^13–16 and it is possible that hypothyroidism worsens renal function in heart-failure patients. We sought to relate the presence and absence of hypothyroidism to renal function in patients with systolic heart failure. We hypothesized that renal function, measured in terms of glomerular filtration rate (GFR), was lower in patients with hypothyroidism.

Patients and Methods

Inclusion Criteria. In this pilot retrospective study at a university heart-failure clinic, we evaluated 75 patients (age, ≥18 yr) who each had a left ventricular ejection fraction lower than 0.40 on echocardiography. The patients were divided on the basis of their thyroid function into 2 groups: 45 patients who had normal thyroid function (the NT group), with thyroid-stimulating hormone (TSH) levels of 0.35 to 5.5 μIU/mL; and 30 patients with a diagnosis of hypothyroidism (the HT group). The HT population comprised 2 subgroups: 17 patients with controlled hypothyroidism (the CH subgroup), whose condition was effectively treated by means of replacement therapy (TSH, 0.35–5.5 μIU/mL); and 13 patients whose hypothyroidism was either inadequately controlled or uncontrolled (the UH subgroup) (TSH, >5.5 μIU/mL).

In the entire study population, baseline characteristics, risk factors, and New York Heart Association functional class did not differ significantly (Table I). Although each patient had a low left ventricular ejection fraction, functional class itself was not a criterion for inclusion in the study.

TABLE I. Baseline Characteristics in Each Thyroid-Function Group

We excluded recently hospitalized patients, persons who had recently undergone an interventional procedure, pregnant or lactating women, persons who were taking an investigational agent or amiodarone, and anyone who had ever undergone transplantation of any organ. Individuals who had diabetes mellitus were excluded if their microalbumin–creatinine ratio exceeded 3.5 mg/mmol or if their hemoglobin A1c level exceeded 7%.

Laboratory Analysis. Laboratory results were obtained within 1 year of the qualifying echocardiogram. Patients abstained from vigorous exercise before undergoing urinalysis. Blood was drawn from each patient in the outpatient clinic and was analyzed by our institution's High-Volume Laboratory in accordance with the National Committee for Clinical Laboratory Standards. The Bayer ADVIA Centaur® (Siemens Medical Solutions USA, Inc.; Malvern, Pa) was used for the analysis of sodium, potassium, creatinine, and blood urea nitrogen; the VITROS® 5.1 FS chemistry system (Cardinal Health, Inc.; Dublin, Oh) for TSH; and the Bayer DCA 2000 analyzer for hemoglobin A1c. We determined renal function (GFR) by using the Modification of Diet in Renal Disease equation: GFR (mL/min/1.73 m²) = 186 × [plasma creatinine]^−1.154 × [age]^−0.203 × [0.742 if female] × [1.212 if black] (conventional units).

Statistical Analysis

Baseline characteristics were analyzed with use of SPSS statistical software, version 14 (SPSS Inc.; Chicago, Ill). The c ² test was used in the analysis of nominal and ordinal data, and the analysis-of-variance Scheffé post hoc test was used for continuous variables for multiple comparisons. The GFR and TSH levels were compared via the 1-tailed, unequal-variance Student t test with use of Microsoft Excel. Results were expressed as mean ± SD. A P value less than 0.05 was considered statistically significant.

Results

Tables II and III show the mean GFR results in each group and the comparative statistical significance between the groups. The mean GFR in the NT group was significantly better than that in the HT group (P=0.032). The mean GFR in the NT group was significantly higher than that in the UH subgroup (P=0.015) but did not differ significantly in comparison with the CH subgroup (P=0.131). When the CH and UH subgroups were compared, there was no statistically significant difference in mean GFR (P=0.193), although the UH subgroup trended toward a lower GFR. Figure 1 illustrates the GFR levels in the groups and subgroups.

TABLE II. Mean Levels for GFR and TSH in the Groups and Subgroups

TABLE III. Comparative Statistical Significance of the GFR and TSH Results

Fig. 1 Renal function (expressed by glomerular filtration rate) in the normal-thyroid, hypothyroidism, controlled-hypothyroidism, and uncontrolled-hypothyroidism populations. P <0.05 was considered statistically significant. + = not statistically significant; * = statistically significant

As shown in Tables II and III, the mean TSH level of 2.11 ± 1.37 μIU/mL in the NT group was significantly better than the level of 3.97 ± 5.2 μIU/mL in the HT group (P=0.032). The mean TSH level of 15.36 ± 3.88 μIU/mL in the UH subgroup was significantly worse than that in the NT group (P=0.015). The mean TSH level of 2.04 ± 1.43 μIU/mL in the CH subgroup was not significantly different from that in the NT group (P=0.131).

Discussion

The present study suggests that uncontrolled hypothyroidism can contribute to renal dysfunction in patients with systolic heart failure—a finding that might have important prognostic and therapeutic implications. The GFR in NT patients was better than that in HT patients, but not significantly different from the GFR in the CH subgroup, wherein patients were receiving adequate thyroid supplementation. The insufficiently treated UH patients tended to have a lower GFR than did the adequately supplemented CH patients.

In noting the relationship between heart failure and renal dysfunction, many investigators have shown that heart-failure patients with renal dysfunction require more hospitalizations and have a higher all-cause mortality rate.^10,17,18 Depending upon the extent of renal disease and its rate of progression, the prognosis worsens.^10,18 Poor renal blood flow (from low cardiac output) is a cause of renal dysfunction in these patients,¹⁹ and with right-heart failure, venous congestion can also contribute.

Although thyroid dysfunction can also play an important role in renal insufficiency,^13–16 we are unaware of any study that has linked thyroid and renal dysfunction in heart-failure patients. The mechanisms by which hypothyroidism affects the kidneys are not completely understood. Evidence suggests that alterations in sodium transport can lead to increased natriuresis in hypothyroidism.^15,20,21 Studies of changes in renal blood flow in the presence of hypothyroidism have produced conflicting results, which may be due to study design.²⁰ Even minimal thyroid dysfunction can increase the mortality rate of heart failure. A recent meta-analysis²² revealed an association between death and subclinical hypothyroidism in heart-failure patients, but the reason behind the adverse prognosis is unknown. Another explanation may be drawn from an animal study by Tang and colleagues,²³ who found that heart failure developed because of the loss of small blood vessels in hypothyroidism. Loss of vascular tissue in the kidneys may also be a mechanism of renal dysfunction in patients who have hypothyroidism.

Our study was limited by its small population, which precluded a more robust statistical analysis of the data. Regardless, the suggested association between hypothyroidism and declining GFR merits further investigation, because the connection may have prognostic significance for heart-failure patients. Further and larger studies are needed in order to evaluate this hypothesis. In the meantime, we recommend that clinicians closely follow their heart-failure patients and ensure supplementation that will overcome hypothyroidism.