The effect of thyroid replacement in dogs with suboptimal thyroid function on owner-directed aggression: A randomized, double-blind, placebo-controlled clinical trial

Abstract

The efficacy of thyroid hormone replacement therapy (THRT) as treatment for owner-directed aggression in client-owned dogs with borderline low thyroid hormone levels was evaluated by means of a 6-week-long, parallel design, double-blind placebo-controlled study. The designation of “borderline hypothyroid” was made if the dog's free normal thyroxine (T4) value was frankly low or in the bottom 20th percentile of the normal range and either total T4, total triiodothyronine (T3), or free T3 was frankly low or in the bottom 30th percentile of the normal range. The presence of thyroid autoantibodies also qualified a dog for enrollment. Owners recorded the number of aggressive episodes directed toward family members on a daily basis for 8 weeks (2-week baseline phase and 6-week study phase). Twenty-nine dogs completed the study; 14 in a treatment group and 15 in a placebo group. The median number of aggressive episodes per day decreased significantly from baseline in both treated and placebo group dogs in weeks 1-2, 3-4, 5-6, and week 6 (treatment, χ² = 24.8, P < 0.001; placebo, χ² = 20.2, P < 0.001), however the median frequency of aggression was significantly lower in the treatment group (1.21 episodes/day) than in the placebo group (1.71 episodes/day) during week 6 of the study (χ² = 4.047, P = 0.044). Three thyroxine-treated dogs had borderline-low thyroid levels on the final day of the study (day 42). When aggression frequency was compared between the treatment and placebo groups after the removal of 3 thyroxine-treated dogs, the treatment group did not have a significantly lower aggression frequency than the placebo group during week 6 (Kruskal–Wallis statistic: χ² = 3.035, n = 26, P = 0.08). The authors discuss the role of thyroid hormones in the regulation of aggression and other cognitive issues and provide rationale for using THRT in dogs exhibiting owner-directed aggression that also have low normal or baseline thyroid hormone levels.

Introduction

It is not surprising that a wide range of behavioral signs and symptoms have been reported in hypothyroid subjects be they human or animal. Even in the early stages of human disease, reduced cognition (Bono et al., 2004) and concentration (Geracioti, 2006) as well as impaired short-term memory (Begin et al., 2008) are reported. Humans may also experience visual and auditory hallucinations, a wide spectrum of fear-based behaviors, mood swings, and aggressive behavior (Denicoff et al., 1990). In hypothyroid dogs, fear-based behaviors (noise and storm phobia, separation anxiety), hyperactivity, poor focus/learning, compulsive behaviors, and aggression (primarily owner-directed or possessive aggression) have been reported (Aronson and Dodds, 2005). Two recent studies, both with relatively small sample size, compared the values of thyroid analytes in dogs with and without behavior problems. In these studies, none of the mean thyroid analytes measured fell outside the normal reference ranges in dogs with or without behavior problems (Carter et al., 2009; Radosta et al., 2011). One retrospective study with a large sample size showed a positive association between aggression and subclinical thyroiditis (defined by elevated thyroglobulin autoantibodies in conjunction with normal thyroxine [T4] and thyroid stimulating hormone [TSH] levels) but not with clinical hypothyroidism (Graham et al., 2003). Clinically, we have also observed a therapeutic effect on anxiety-related and aggressive behavior problems when dogs with borderline or suboptimal thyroid function are treated with a standard dose of thyroxine after 6 weeks or less. Stress often potentiates behavioral issues. Glucocorticoids released in response to stress reduce the pituitary release of TSH in response to thyroid releasing hormone, reduce conversion of T4-T3, and prevent T3 binding to receptors at the cellular level (Re et al., 1976; Cavalieri et al., 1984; Kaptein et al., 1992).

Triiodothyronine (T3) is integral to metabolism throughout the body. In the brain, T4 is converted to T3 and acts directly on neurons. T3 not only can modulate the activity of the neurotransmitters norepinephrine, serotonin, and dopamine and the sensitivity and concentration of their receptors (Bauer et al., 2008) but also can act as a neurotransmitter itself (Dratman and Gordon, 1966). In human psychiatric patients, thyroid hormone supplements are used to augment the effect of antidepressants in refractory euthyroid patients (Cooper-Kazaz et al., 2007) or to accelerate the results of therapy (Altshuler et al., 2001). T3 activity in the brain is affected by many things, including the circulating concentrations of TSH, T3, and T4; availability of unbound hormone; activity of transporters that bind and transport the hormones into cells (Oppenheimer & Schwartz, 1985; Friesema et al., 2006); deiodinase enzymes that convert T4 to the active T3 (Hernandez et al., 2010); and activity of thyroid receptors. The aforementioned last 3 can all be affected by individual genetic variation (Peeters et al., 2006), and it is also true that circulating thyroid levels do not necessarily predict thyroid activity in the brain (Obregon et al., 1984).

So far, documentation of the beneficial effects of treating borderline-low thyroid levels on canine aggression has taken the form of case reports (Dodman et al., 1995; Fatjó et al., 2002). Double-blind placebo-controlled studies are considered the gold standard for determining the efficacy of a particular treatment. In the study reported in the article, we have adopted this approach to evaluate the therapeutic efficacy of thyroxine supplementation for dogs exhibiting owner-directed aggression that also have suboptimal serum thyroid levels.