Hypothyroidism is a common disease that is thought to have a heritable basis in some breeds of dogs.[1-3] Because many purebred dogs undergo purposeful breeding, any effect of hypothyroidism on reproduction has considerable clinical importance. Although a wide variety of abnormalities, including menstrual irregularities, infertility, abortion, stillbirth, premature birth, and low birth weight, occur in hypothyroid women, little is known about the effects of thyroid hormone deficiency on reproductive performance in bitches. Infertility, prolonged interestrus interval, abortion, and stillbirth have been reported in a colony of Borzoi bitches with lymphocytic thyroiditis and hypothyroidism. Other reports of reproductive abnormalities in bitches with naturally occurring hypothyroidism have been poorly documented.[6-8]
Other studies in dogs have failed to show a detrimental effect of hypothyroidism on fertility. Reproductive abnormalities are rarely reported in retrospective studies of canine hypothyroidism and the number of bitches included in these studies is small.[9-11] Thyroid function was not correlated with reproductive performance in Greyhounds. Recently, we reported preliminary results of the effects of experimentally induced hypothyroidism of 19 weeks duration on reproduction. Prolonged whelping, decreased viability of pups, increased periparturient mortality, and decreased birth weight were found in hypothyroid bitches, whereas fertility, cycle duration, and pregnancy rates were unaffected. The purpose of this report is to evaluate reproduction after more prolonged hypothyroidism in the same bitches. We hypothesized that prolonged hypothyroidism (approximately 1 year duration) would result in infertility, and that fertility would be restored after thyroid hormone replacement treatment.
- Top of page
- Materials and Methods
Serum total T4 concentrations were <5 nmol/L before and after TSH administration in all dogs in the hypothyroid group and post-TSH total T4 concentrations were >35 nmol/L in all dogs in the control group at each testing period. Clinical signs of hypothyroidism, including some combination of weight gain, thin hair coat, alopecia, seborrhea, weakness, and lethargy were present in all hypothyroid dogs. Mean ± SD body weight of the control and hypothyroid groups were 9.8 ± 1.13 and 9.7 ± 0.91 kg, respectively, before induction of hypothyroidism, 10.8 ± 1.20 and 11.6 ± 0.98 kg, respectively, at the 1st breeding, and 10.1 ± 1.28 and 12.0 ± 0.61 kg, respectively, at the 2nd breeding. The 1st breeding took place a mean of 20.9 ± 4.0 weeks and the 2nd breeding 56 ± 7.6 weeks, respectively, after radioiodine administration. Results of the 1st breeding have been previously reported, and only comparisons between the 1st and subsequent breedings are described here. The 3rd breeding occurred 139.6 ± 9.4 weeks after radioiodine, after levothyroxine administration for a 37 ± 14 weeks in the hypothyroid group. The mean duration of hypothyroidism before levothyroxine administration was 102.3 ± 8.1 weeks. The mean serum total T4 and TSH concentrations of dogs receiving levothyroxine supplementation during the 3 months immediately before breeding were 47.4 ± 12.8 nmol/L and 0.17 ± 0.23 ng/mL, respectively. All dogs had resolution of all clinical signs of hypothyroidism including alopecia, seborrhea, weight gain, activity, strength, and alertness. Body weight decreased in the hypothyroid group from 12.8 ± 1.44 kg before initiating levothyroxine supplementation to 10.3 ± 0.90 kg (P < .0001) immediately before the final breeding whereas the weight in the control group did not differ during the same time period (10.6 ± 1.30 and 10.8 ± 1.18 kg, respectively). The mean initial levothyroxine dosage was 0.016 ± 0.002 mg/kg every 24 hours and was increased in 3 dogs, with a final dosage of 0.020 ± 0.006 mg/kg in all hypothyroid dogs.
The interestrus interval (Table 1) was not significantly different between groups at any time period or between times within a group. Breeding behavior was not affected by hypothyroidism at any time period (Table 1). Despite the finding that only 4 of 8 hypothyroid and all 6 control bitches became pregnant at the second breeding, there was no difference in pregnancy rate between groups (P = .0699). After supplementation with levothyroxine in the hypothyroid bitches, all 6 dogs bred in that group became pregnant (including 3 of the 4 that were infertile at the second breeding) and no difference in pregnancy rate was noted between TH and control dogs at the third breeding (Table 1). However, hypothyroid dogs were less likely to become pregnant at the 2nd breeding compared with the first (P = .0455), but there was no difference between the 2nd and 3rd breedings within the UT and TH groups (P = .0833). There was no difference in the prevalence of pregnancy within the control group between the 3 breedings.
Table 1. Characteristics associated with reproduction in healthy control, hypothyroid, and levothyroxine-treated hypothyroid bitches and health outcomes of their offspring.
|Time Period||Breeding 1||Breeding 2||Breeding 3|
|Group||Control (n = 9)||Untreated Hypothyroid (n = 9)||Control (n = 6)||Untreated Hypothyroid (n = 8)||Control (n = 6)||Levothyroxine-Treated Hypothyroid (n = 6)|
|Interestrus interval (days)||260 ± 118||256 47||217 ± 44|
|Gestation duration (days)||63 ± 1.3||63 ± 1.5||62 ± 0.8|
|Breeding behavior||3.0 ± 0.0||2.8 ± 0.41||2.7 ± 0.82|
|Contraction duration (min)||11 ± 12||17 ± 18||31 ± 37|
|Contraction strength||4 (2–5) (4,5; 1st, 3rd quartile)|
3 (2–5) (3,4; 1st, 3rd quartile)
|4 (2–5) (4,4; 1st, 3rd quartile)|
4 (2,5) (3,5; 1st, 3rd quartile)
|5 (3–5) (4,5; 1st, 3rd quartile)|
5 (2–5) (3,5; 1st, 3rd quartile)
|Litter size||5.4 ± 1.9||4.7 ± 0.8||4.8 ± 2.9|
|Interval between whelping pups (min)||42 ± 25||32 ± 19||47 ± 40|
|Birth weight (kg)||0.68 ± 0.10||0.29 ± 0.04||0.31 ± 0.04|
|Total APGAR score||9 (7–10), (9, 9; 1st, 3rd quartile)|
9 (0–10) (8,9; 1st, 3rd quartile)
|9 (7–10) (9, 10; 1st, 3rd quartile)|
8.5 (0–10) (0, 10; 1st, 3rd quartile)
|9 (7–10) (8, 10; 1st, 3rd quartile)|
9 (7–10) (9, 9; 1st, 3rd quartile)
|Periparturient mortality (%)||0 (0–0), (0, 0; 1st, 3rd quartile)|
14.3 (0, 100) (0, 25; 1st, 3rd quartile)
|0 (0–0) (0,0; 1st, 3rd quartile)|
55.7 (33.3–100) (36.7, 85.7; 1st, 3rd quartile)
|0 (0–20) (0, 10; 1st, 3rd quartile)|
0 (0–0) (0, 0; 1st, 3rd quartile)
Because only 4 of 8 hypothyroid bitches became pregnant during the 2nd breeding, comparisons were limited to 4 hypothyroid and 6 control dogs. During the 3rd breeding period, 1 bitch in the control group did not become pregnant and another control dog was euthanized 46 days after ovulation because of acute glaucoma unresponsive to medical management. The euthanized dog had pregnancy documented on ultrasound examination with viable fetuses observed from days 21–35, but fetal resorption was noted at day 42 after ovulation. This dog was considered pregnant for statistical purposes. Therefore, comparisons of gestation duration and data derived at whelping were limited to 6 TH and 4 control dogs at the final breeding.
Gestation duration was not different between UT and TH dogs or control dogs at any of the time periods (Table 1). There also was no difference in gestation duration within UT and TH dogs or control dogs when compared among breedings. At the second breeding, 1 hypothyroid dog had 2 suspected sites of fetal resorption on ultrasound examination at 28 days after ovulation and a single viable fetus. The bitch whelped 1 pup that was stillborn after stage II labor had occurred for 8.5 hours. Fetal resorption sites were identified on ultrasound examination of 1 hypothyroid (2 sites) and 1 control (1 site) dog during the 2nd breeding. In dogs that whelped normal pups during the final breeding, a single fetal resorption site was identified in 1 dog in both the hypothyroid and control groups. Duration of contractions during whelping was significantly different (P = .03) between UT and control dogs at the 2nd breeding, but not in TH dogs at the 3rd breeding (Table 1). There was no difference in contraction duration within each group among the 3 breeding times. The interval between delivery of pups was not different between or within groups at any time. Scores for contraction strength were not different between the UT and HT dogs and control dogs at the 2nd or third breeding, although it was recorded in only 3 of the 4 and 2 of the 6 hypothyroid dogs that whelped at the 2nd and third breeding, respectively.
The litter size whelped by UT or TH dogs was not different from those of the control group at the 2nd (P = .0691) or 3rd (P = .9459) breeding. There was no difference in viability scores of live pups between UT or TH dogs and control dogs in the 2nd or 3rd breeding. There was no difference in viability scores within the control or the UT or TH groups among any of the time periods. The number of pups that were stillborn or died in the periparturient period was higher in the UT group compared with controls at the 2nd breeding (P = .024), but not between TH and control dogs at the 3rd breeding (P = .167; Table 1). At least 1 pup died in the periparturient period from each of the 4 UT dogs that became pregnant, whereas no death was reported at the same breeding in the control group. Birth weight was significantly less (P = .0016) in pups born to hypothyroid dams compared with controls on the 2nd breeding. Levothyroxine supplementation resolved the difference in birth weight between the groups on the 3rd breeding.
There were no significant differences in serum progesterone concentrations at any week between the control and UT or TH groups during both the 2nd and 3rd breedings (Figs 1 and 2). There was no difference in serum progesterone concentration at any week in UT or HT dogs comparing the 2nd and 3rd breedings. A lower progesterone concentration on the third breeding at week 1 after ovulation (P = .0181) was noted in controls in the 2nd and 3rd breedings.
Figure 1. Mean ± standard deviation serum progesterone concentrations in control (diamonds) and untreated hypothyroid (triangles) 56 ± 7.6 weeks after radioiodine administration in the hypothyroid group. Time 0 represents the date of ovulation.
Download figure to PowerPoint
Figure 2. Mean ± standard deviation serum progesterone concentrations in control (diamonds) and treated hypothyroid (squares) after 37 ± 14 weeks of levothyroxine administration in the hypothyroid group. Time 0 represents the date of ovulation.
Download figure to PowerPoint
- Top of page
- Materials and Methods
Results of this study provide evidence that prolonged experimentally induced hypothyroidism in bitches has detrimental effects on offspring and may interfere with fertility. Periparturient mortality was higher and birth weight lower in pups born to hypothyroid dogs than controls after short-term and more prolonged hypothyroidism. Although the effect of hypothyroidism on pregnancy was not statistically significant, only 4 of 8 untreated hypothyroid dogs became pregnant, and 1 of the pregnant dogs whelped a single stillborn pup. It is likely that the small number of dogs, resulting primarily from reduction of the control group, prevented sufficient power for statistical significance to be reached. This is supported by the finding that hypothyroid dogs were significantly less likely to become pregnant after disease of 1 year's duration compared with 19 weeks when all hypothyroid dogs became pregnant and whelped live pups. This finding also suggests that, if hypothyroidism affects fertility, it only does so after the disease has been present for more than 5 months. The dogs used in this study were obtained after they had documented normal pregnancies for 2 or 3 consecutive breedings to decrease any nonexperimental influence on fertility. However, the results of this study must be considered inconclusive with regard to the effect of hypothyroidism on fertility. The deleterious effects of hypothyroidism on reproduction are reversible based on all effects on offspring and return of 100% pregnancy rate during levothyroxine supplementation.
Infertility previously has been attributed to hypothyroidism, but reports of the association have not been well documented.[5, 7, 8] Although an early case series attributed abnormal estrus to hypothyroidism, retrospective studies using more stringent criteria for a diagnosis of hypothyroidism have failed to identify reproductive disorders.[9-11] However, these latter reports have included few intact females. In 2 prospective studies, abnormal thyroid function was not associated with infertility.[12, 18] Failure to document hypothyroidism as a cause of infertility is probably related to the difficulty in definitively diagnosing hypothyroidism, the relatively low incidence of the disease in bitches studied to date, and an inadequate number of studies evaluating spontaneous hypothyroidism in bitches with reproductive disorders. Unfortunately, the study reported here was unable to determine with certainty the effects of hypothyroidism on fertility in the bitch. Other effects of hypothyroidism on reproduction, particularly on gestation duration and litter size, may have been overlooked because of the low power of the study.
Overt hypothyroidism in women can cause infertility because of anovulation.[19, 20] Excessive prolactin secretion in hypothyroidism[21, 22] plays a major role in ovulatory dysfunction in hypothyroidism because of its inhibition of gonadotropin releasing hormone secretion and decrease of pulsatile luteinizing hormone secretion leading to decreased folliculogenesis and decreased ovarian estrogen synthesis.[23-26] In a separate study of the dogs of the present report, excessive prolactin secretion was documented at a mean of 39 weeks after induction of hypothyroidism. However, the lack of difference in the interestrus interval and the similar progesterone concentrations during diestrus between hypothyroid and control dogs are consistent with normal graafian follicle formation, ovulation, and corpus luteum function. Therefore, it is unlikely that ovulation is affected by hypothyroidism in dogs.
Any effect of hypothyroidism on fertility in the present study likely resulted from failure of oocyte fertilization or early embryonic death. It remains unclear how hypothyroidism could impair fertilization. Breeding behavior was not affected by hypothyroidism, and the same 2 male dogs were used for breeding throughout the study, making this factor unlikely to account for the infertility. Thyroid hormone receptors have been identified in oocytes, so it is possible that thyroid hormone deficiency could impair fertilization. Because no fetus or uterine swelling was identified on ultrasound examination on days 21 and 28 after ovulation in the 4 untreated hypothyroid dogs that did not become pregnant, it is likely that implantation did not occur. Implantation and placenta formation are complex processes requiring the coordinated influence of numerous regulators, growth factors, and hormones, such as epidermal growth factor, human placental lactogen, vascular endothelial growth factor, and regulators of apoptosis that are influenced by thyroid hormones.[29, 30] Thyroid hormone receptors are present in the uterus and many different placental cell types, and uterine epithelial cells are morphologically altered by hypothyroidism.[29, 31, 32] It is possible that abnormalities in 1 or more of these factors played a role in the 50% pregnancy rate in hypothyroid dogs of the present study.
The decreased birth weight and increased periparturient mortality noted in the present study are similar to the effects of hypothyroidism in pregnant women. When pregnancy is established in hypothyroid women, numerous adverse outcomes have been reported, including abortion or fetal resorption, stillbirth, premature birth, and low birth weight.[33-36] The detrimental effects of maternal hypothyroidism on the fetus in the present study occurred at the first breeding after induction of hypothyroidism, before the onset of infertility. Therefore, maternal hypothyroidism should be considered a possible cause of stillbirth and low birth weight in dogs.
Menstrual cycle irregularities, including menorrhagia, oligomenorrhea, and amenorrhea, are common in hypothyroid women,[19, 20] and abnormalities of the estrous cycle have been suggested to occur in hypothyroid bitches.[1, 6, 37] In the present study, the interestrus interval was similar in UT, TH, and control dogs, and no hypothyroid dog had a delayed onset of estrus. Subjectively, signs of proestrus and estrus, including proestrual bleeding, were similar in UT, TH, and control dogs. Although the number of dogs in the present study was limited, these results and the fact that other reports have provided few details or have been based primarily on empirical observations suggest that irregular interestrus interval, weak or silent estrus, prolonged or severe proestrual vaginal hemorrhage, and reduced libido are unlikely to be associated with hypothyroidism.[1, 6, 7, 37]
Hypothyroidism was induced approximately 1 year before the second breeding in the current study, with overt clinical signs being present for over 40 weeks. Although it is unlikely that dogs with equivalent clinical signs would routinely be used for breeding, hypothyroidism is a chronic disease with an insidious onset. The experimental model used to create hypothyroidism in the current study induced severe deficiency with an acute onset. The duration of hypothyroidism before diagnosis in dogs with spontaneous disease can be difficult to determine, but a recent study found that clinical signs of hypothyroidism were present for 5 months to 4 years before diagnosis. Therefore, it is feasible that many affected dogs would have hypothyroidism for 1 year or longer before diagnosis.
Treatment of hypothyroidism with levothyroxine resolved the infertility and detrimental fetal effects induced by the hypothyroid state because all hypothyroid dogs became pregnant and had normal litters after levothyroxine supplementation. Once-daily administration of levothyroxine was selected because recent studies support this is an adequate method of treatment.[17, 39] Adequacy of treatment was documented by resolution of clinical signs, normalization of body weight, serum T4 concentrations consistently in the upper 75% of the reference range or higher, and TSH concentrations within the reference range for at least 4 months before breeding.
Although the experimental model used in this study effectively created hypothyroidism, it did not induce thyroid autoimmunity that is responsible for at least 50% of the cases of spontaneous hypothyroidism in dogs. Some studies in women have noted that thyroid autoimmunity is a risk factor for fetal complications, and it is difficult to separate the effects of thyroid autoimmunity from those of hypothyroidism on fetal health because most cases of hypothyroidism in women are caused by autoimmune thyroiditis. However, increased prevalence of infertility and placental abruption has been reported in some studies of women with thyroid autoimmunity.[4, 40, 41] A relationship between the presence of serum antibodies against thyroglobulin was not reported to be associated with reproductive performance in greyhounds, although the prevalence of thyroid autoimmunity was low.
Results of this study show that chronic, severe hypothyroidism in bitches results in low birth weight and increased periparturient mortality early in the course of hypothyroidism and persists. Further investigation on the effect of naturally occurring hypothyroidism on fertility is indicated based on failure to document statistically significant infertility despite a 50% reduction in pregnancy rate. All reproductive abnormalities appear to be reversed by thyroid hormone replacement treatment.