Effect of Chronic and Acute Thyroid Hormone Reduction on Perioperative Outcome

Abstract

Depressed serum thyroid hormone concentrations from either preexisting chronic hypothyroidism or induced by physiologic stress may affect clinical outcome in the perioperative period. This article discusses the various forms of thyroid hormone, the effect of hypothyroidism on patients undergoing surgery, the general role of thyroid hormone supplementation in the perioperative period, and the use of triiodothyronine as a possible inotropic agent in the setting of cardiac surgery. In addition, we review alterations in thyroid hormone associated with brain death and the possible role of thyroid hormone supplementation in the setting of organ harvest procedures. Thyroid Hormone Pharmacology/Pharmacokinetics The thyroid gland produces two types of thyroid hormone, triiodothyronine (T3) and tetraiodothyronine (T4) [1]. T4 is a prohormone synthesized from tyrosine and represents 80% of the body's thyroid hormone production. T3, the most biologically active form of thyroid hormone (five times more biologically active than T4), is produced either directly from tyrosine metabolism or from conversion from T4 in peripheral tissues (Figure 1). Of note, only 35%-40% of T4 is converted to T3 by monodeiodination, while 50% of T4 is converted to the biologically inert moiety 3 prime,3,5-triiodothyronine (rT3) [2]. Systemic illness, physiologic stress, and drugs may inhibit the peripheral conversion of T (4) to T3 and induce a hypothyroid-like state often referred to as the "euthyroid sick syndrome" [3]. Two distinct deiodinases located in liver, kidney, and the central nervous system metabolize T3 and rT3 to inactive compounds [4]. Half-lives for endogenously or exogenously administered T3 and T4 are 1.5 and 7 days, respectively [1]. T3 and T4 are both highly protein bound to albumin, thyroid-binding prealbumin, and thyroid-binding globulin, with only 0.2% of T3 and 0.3% of T4 freely circulating unbound and biologically active. The regulation of serum thyroid hormone levels is complex and relies primarily on negative feedback mechanisms involving the thyroid-stimulating hormone. Table 1 lists normal ranges for serum thyroid hormones.Figure 1: Feedback mechanism of thyroid hormone synthesis. T3 = triiodothyronine; rT3 = 5 prime,3 prime-3 triiodothyronine; T4 = tetraiodothyronine; TRH = thyrotropin-releasing hormone; TSH = thyroid-stimulating hormone; 3 prime-3-T2 = 3 prime-3diiodothyronine. Reproduced with permission from [1].Table 1: Reference Ranges for Serum Thyroid Hormones in Human AdultsThyroid Hormone Pharmacodynamics Thyroid hormone has physiologic effects on many organ systems, with greatest impact on the cardiovascular system. Insights into the pharmacodynamic effects of thyroid hormone stem primarily from studies investigating patients with chronic thyroid disease. Cardiovascular changes are often the earliest clinical manifestation of abnormal thyroid hormone levels. In contrast to tachycardia, vasodilation, and the generalized hyperdynamic state seen with chronic hyperthyroidism, patients with hypothyroidism may be asymptomatic or manifest the most severe form of disease, called myxedema coma. Myxedema coma is characterized by coma, loss of deep tendon reflexes, and ultimately, cardiovascular collapse and death. Chronic hypothyroidism results in bradycardia, elevated systemic vascular resistance, hypertension, decreased myocardial contractility, decreased stroke volume, increased circulating catecholamines, and decreased cardiac output (Table 2). Systemic hypertension, particularly diastolic hypertension, occurs in approximately 15% of hypothyroid patients, compared with 5.5% of age-matched controls [5]. Diastolic hypertension results directly from decreased thyroid hormone levels with resultant increased systemic vascular resistance, as well as increased levels of circulating catecholamines.Table 2: Cardiovascular Manifestations of Hypothyroidism and HyperthyroidismSystolic and diastolic myocardial function are both impaired in chronic hypothyroidism, with congestive heart failure occasionally occurring in hypothyroid patients in the absence of underlying heart disease [6]. Of note, myocardial depression associated with severe hypothyroidism may be refractory to the administration of catecholamines. Echocardiography studies of hypothyroid-associated myocardial function demonstrate that indices of systolic function (prolonged preejection period of systole, isovolemic contraction time, left ventricular ejection time) and diastolic function are prolonged in patients with hypothyroidism [7-9], correlate with serum thyroid hormone levels [7], and are reversible with thyroid hormone replacement therapy [7]. Interestingly, patients with hypothyroidism are predisposed to pericardial effusion and may have a higher incidence of atherosclerotic heart disease [10]. The electrocardiogram may reveal bradycardia, low voltage, and prolonged P-R, Q-R-S, and Q-T intervals [6,11,12]; low voltage is not merely a reflection of possible pericardial fluid because it persists after pericardiocentesis and most likely reflects primary myocardial abnormalities [6,13]. Conduction abnormalities may predispose patients to ventricular tachycardia, particularly torsades de pointe. Effects of chronic hypothyroidism on organs other than the cardiovascular system are less well described. Pulmonary function in general is not markedly affected by chronic hypothyroidism. Thyroid hormone appears to be necessary for the normal production of surfactant [14], which is consistent with the observation that cells responsible for surfactant production, i.e., type II pneumocytes, possess thyroid hormone receptors [15]. Hypothyroid-induced alterations in surfactant production may have more clinical relevance in the setting of lung injury or during sepsis [14], settings in which decreased serum thyroid levels have been shown to be associated with worse pulmonary function. Consistent with this observation, the administration of thyroid hormone to patients with acute respiratory distress syndrome improves lung compliance [14]. Furthermore, chronic hypothyroidism is associated with pleural effusions, which may further impair pulmonary function. Ventilatory drive in response to hypoxia and hypercapnia is reduced in patients with severe hypothyroidism [16]. Basal metabolic rate is decreased by approximately 50%, which may provide a mechanism for the hypothermia reported in some hypothyroid patients [17-19]. Several case reports suggest that patients with hypothyroidism may be more sensitive to the sedative and respiratory depressant effects of anesthetics [18,20-24]. Despite this clinical impression, results from animal models suggest that minimum alveolar anesthetic concentration is not significantly altered in hypothyroid patients [25,26]. Patients with Chronic Hypothyroidism Presenting for Surgery Most patients with hypothyroidism are receiving thyroid hormone replacement therapy and are euthyroid (based on clinical and laboratory evidence) at the time of surgery. These patients do not carry an increased risk of perioperative morbidity and do not require special treatment other than continuation of their chronic thyroid hormone replacement. Due to the long half-life of T4 (7 days), receiving T4 on the morning of surgery can be considered optional; however, it may be prudent for patients receiving T3 to take their usual dose on the day of surgery due to its relatively short half-life (1.5 days) [1]. Patients with myxedema coma, those who manifest any severe clinical symptom of chronic hypothyroidism, or those with markedly decreased serum T3 and T4 levels are all at increased risk of having complications during the perioperative period. Perioperative morbidity in these patients relates to cardiovascular depression refractory to catecholamine administration, hypothermia, airway difficulties due to generalized edema, and aspiration due to delayed gastric emptying and depressed mental status. Elective surgery in symptomatic hypothyroid patients should be postponed until patients are rendered euthyroid. Patients with myxedema coma may be treated with either 200-400 micro g of intravenous levothyroxine (T4) followed by 100 micro g daily, or 10-25 micro g of intravenous triiodothyronine (T3) every 8 h [4]. Co-existing adrenal insufficiency in patients with severe hypothyroidism may be manifested by hypotension, weight loss, muscle weakness, and abdominal or flank pain, and should be treated, if necessary, with cortisol [4]. In contrast to euthyroid or severely hypothyroid patients, controversy exists regarding the perioperative management of patients with mild clinical or laboratory evidence of hypothyroidism presenting for elective surgery. No large, well designed clinical trials have proven that postponing surgery in this setting to allow for thyroid replacement therapy improves perioperative outcome. The greatest controversy surrounds the role of preoperative thyroid hormone replacement in patients with coronary or valvular disease. Patients with hypothyroidism and coronary artery disease may have an imbalance between myocardial oxygen consumption and the supply leading to myocardial ischemia. Theoretically, myocardial ischemia might be provoked or worsened after thyroid hormone replacement if myocardial oxygen delivery is fixed and thyroid hormone-mediated inotropic and chronotropic effects increased myocardial oxygen demand. Inducing myocardial ischemia with thyroid replacement has been anecdotally reported [27]; however, large controlled studies have not demonstrated a detrimental effect on overall outcome in this setting. In fact, in a classic study by Keating et al. [28], anginal symptoms improved in a full one third of patients with hypothyroidism and angina after initiation of thyroid replacement therapy; in this series of primarily elderly hypothyroid individuals, fewer than 1% of treated patients developed new-onset angina as a result of replacement therapy [28]. Moreover, in the setting of myocardial infarction, experimental and clinical evidence suggests that untreated hypothyroidism worsens recovery [29,30]. One such study showed a 36% increase in infarct size in hypothyroid dogs compared with euthyroid controls after ischemic insult; ventricular dysrhythmias were also more severe in the hypothyroid group [30]. Taken together, these studies do not suggest that hypothyroidism is protective in patients with cardiovascular disease. Based on these data, it may be prudent to carefully initiate thyroid hormone replacement in hypothyroid patients with coronary artery disease prior to cardiac and noncardiac surgery, although clinicians should be vigilant for signs of myocardial ischemia. Possible Use of Thyroid Hormone as an Inotropic Agent Myocardial dysfunction is common in cardiac surgical patients. Most heart surgery patients have stunned myocardium after cardiopulmonary bypass, which recovers if sufficient inotropic support is provided in the perioperative period [31,32]. Interest in T3 as a potential inotropic agent during cardiac surgery developed from reports that serum T3 levels decrease markedly after cardiopulmonary bypass. Although there is some controversy regarding the magnitude and direction of these changes [33-42], two recent studies of patients undergoing cardiac surgery (combined, n = 353) confirm a significant decrease in the biologically active forms of thyroid hormone associated with cardiopulmonary bypass [43,44]. Decreases in T3 and T4 levels during cardiac surgery are illustrated in Figure 2 and are significant even after correction for hemodilution [44]. Further support for a possible role of thyroid hormone supplementation in high-risk surgical patients comes from the observation that higher serum T4 levels are associated with better outcome in critically ill patients [36,45-51]. It is not known, however, whether acute decreases in thyroid hormone levels seen in surgical and intensive care patients are similar in etiology and physiologic effects to those in patients with chronic hypothyroidism.Figure 2: Serum thyroid hormone concentrations reported from a recent T3 interventional trial in patients during cardiac surgery. Reproduced with permission from [44].An acute inotropic effect of intravenous T3 in the setting of cardiac surgery has been shown in several animal models [29,52-58]. In these animal studies, however, T3 administration exhibited an inotropic effect most consistently after severe global myocardial ischemia [52,53]. This finding suggests that T3 may work via a mechanism of action different from the commonly used inotropic drugs, such as phosphodiesterase inhibitors and beta-adrenergic receptor agonists, since these agents also have inotropic effects in tissue that has not been ischemic [29,54,58]. Furthermore, the inotropic effect observed in T3-treated animals was not immediate and titratable, which also suggests a unique mechanism of inotropic action. Possible mechanisms of T3-mediated inotropic effects on myocytes include augmentation of active ion transport, synthesis of proteins, and sensitization of beta-adrenergic receptor pathways [59-61]. Data in human and animal models suggest that intravenous T3 administration may have mild vasodilating effects, although clinically significant effects in well controlled clinical trials have not been reported thus far [43,44]. Four randomized and blind clinical trials examining the effects of intravenous T3 in cardiac surgery patients have been reported [40,42-44]. Two initial studies had contradictory results and are not widely accepted due to small sample size [40,42] and varied surgical procedures [40]. In one of these studies, Novitzky and colleagues [43] reported dramatic inotropic effects (as measured by reductions in inotropic requirements) in 12 patients undergoing cardiac surgery who received T3. Two recently published larger clinical trials, in contrast, were unable to confirm these dramatic results [43,44]; therefore findings from these trials will now be examined in more detail. Klemperer et al. [43] randomized 142 patients undergoing coronary artery bypass graft surgery with left ventricular ejection fractions of less than 0.4 to either placebo or intravenous T3. T3 administration was initiated upon removal of aortic cross-clamp and was administered as a bolus (0.8 micro g/kg) followed by 0.12 micro g [centered dot] kg-1 [centered dot] hr-1 infusion for 6 h, then weaned over 5 h. No difference in the need for adjunctive therapy with inotropic drugs was observed between T3 and placebo groups. A positive outcome cited in this study was increased cardiac index (2.97 +/- 0.72 vs 2.67 +/- 0.61 L [centered dot] min-1 [centered dot] m2, mean +/- SD) and reduction in systemic vascular resistance in the group receiving T3 compared with placebo. However, since inotropic drugs were titrated to a cardiac index of 2.1 L [centered dot] min (-1) [centered dot] m2 or more in this study, the clinical significance of the reported difference in cardiac index remains unclear. Furthermore, small increases in cardiac index might result from T3-induced reduced afterload rather than from direct inotropic effects. In a separate study, Bennett-Guerrero et al. [44] randomized 211 patients undergoing coronary artery bypass graft surgery to either placebo, dopamine infusion (5 micro g [centered dot] kg-1 [centered dot] min-1), or T3 given as a bolus (0.8 micro g/kg) at the time of aortic cross-clamp removal, followed by a 0.12-micro g [centered dot] kg-1 [centered dot] h-1 infusion for 6 h, then weaned over 6 h. Inclusion criteria in this study targeted patients at high risk of requiring inotropic drugs, defined as at least one of the following: age >or=to65 yr, left ventricular ejection fraction <or=to0.40, or cardiac reoperation. In the Bennett-Guerrero study, T3 administration prevented decreases in serum thyroid hormone concentrations associated with cardiopulmonary bypass (Figure 2). In the dopamine group, heart rate was increased, epicardial pacing requirements were decreased, and a trend toward decreased inotropic drug requirement was evident, whereas patients receiving T3 demonstrated neither a decreased requirement for adjunctive inotropic drugs nor an increase in heart rate. In contrast to tachycardia commonly associated with chronic thyroid hormone elevation, acute intravenous T3 administration resulted in stable heart rate, which suggests that acute and chronic elevations in serum thyroid hormone concentrations may have different end-organ effects. Both clinical trials presented demonstrate that T3 (in the dose used) has no dramatic effects on myocardial performance in cardiac surgery patients at increased risk of postcardiopulmonary bypass ventricular dysfunction. Both studies enrolled patients at high risk of requiring inotropic agents, demonstrated by the high percentage of patients in the placebo groups requiring adjunctive inotropic agents (45%-56%) [43,44]. Although supranormal serum T3 concentrations were achieved in both studies, it is unclear whether still larger doses of T3 might provide more direct inotropic effects. Moreover, based on animal data, trials involving patient populations with a greater ischemic insult (e.g., cardiac transplantation, preoperative intraaortic balloon counterpulsation, ongoing perioperative myocardial ischemia, ischemic cardiomyopathy) or severely depressed ventricular function (e.g., left ventricular ejection fraction less than 25%) may be more likely to demonstrate inotropic effects or an improvement in outcome with intraoperative intravenous T (3) administration. Thyroid Hormone Supplementation in Potential Organ Donors Anesthesiologists are often involved in caring for brain-dead patients in the intensive care unit or in the operating room during harvesting of organs for transplantation. Brain death is associated with an increase in catecholamine levels, a process accompanied by an acute increase in systemic vascular resistance [62]. Animal models of head trauma reveal cardiac effects, such as contraction band necrosis, which can be prevented by surgical sympathectomy prior to induced intracranial injury [63,64]; administration of intravenous T3 to the donor animals resulted in improvement in myocardial function in a pig model of brain death [65]. Consistent with this T3-mediated improvement of myocardial function in a pig model, T3 administration was shown to partially reverse a change from aerobic to anaerobic metabolism in baboons after brain death [66]. After cardiac transplantation in humans, Novitzky et al. [67] reported an improvement in donor heart function in organs obtained from brain-dead donors who had received therapy with T3, cortisol, and insulin. This study, however, was neither randomized nor blind, and it used a historical control group. Novitzky et al. [67] also reported a series of 70 cases in which both organ donors and recipients were administered T3[68]. This series reported a low incidence of poor posttransplant cardiac function; however, the lack of a control group limits its findings. Therefore, despite anecdotal reports of efficacy, no randomized, blind controlled trial has been published that supports the routine use of thyroid hormone supplementation in organ donors and/or recipients. Conclusion Chronic severe reductions in thyroid hormone have significant end organ effects and may increase perioperative morbidity if not corrected prior to surgery. In contrast, it is unclear whether mild hypothyroidism worsens perioperative outcome and warrants preoperative correction. Thyroid hormones decrease during cardiopulmonary bypass and critical illness; however, the clinical significance of this change remains unclear. Hence, the use of thyroid hormone as an inotropic agent and vasodilator cannot be recommended for routine cardiac surgery based on the results of recent trials. Nevertheless, sufficient theoretical rationale and positive animal data exist to warrant further well designed human clinical trials investigating the possible role of thyroid hormone supplementation in other patient groups.