Hyperthyroidism is Underdiagnosed and Undertreated in 3336 Patients: An Opportunity for Improvement and Intervention

Artificial intelligence-enabled electrocardiography contributes to hyperthyroidism detection and outcome prediction

BACKGROUND

Hyperthyroidism is frequently under-recognized and leads to heart failure and mortality. Timely identification of high-risk patients is a prerequisite to effective antithyroid therapy. Since the heart is very sensitive to hyperthyroidism and its electrical signature can be demonstrated by electrocardiography, we developed an artificial intelligence model to detect hyperthyroidism by electrocardiography and examined its potential for outcome prediction.

METHODS

The deep learning model was trained using a large dataset of 47,245 electrocardiograms from 33,246 patients at an academic medical center. Patients were included if electrocardiograms and measurements of serum thyroid-stimulating hormone were available that had been obtained within a three day period. Serum thyroid-stimulating hormone and free thyroxine were used to define overt and subclinical hyperthyroidism. We tested the model internally using 14,420 patients and externally using two additional test sets comprising 11,498 and 596 patients, respectively.

RESULTS

The performance of the deep learning model achieves areas under the receiver operating characteristic curves (AUCs) of 0.725-0.761 for hyperthyroidism detection, AUCs of 0.867-0.876 for overt hyperthyroidism, and AUC of 0.631-0.701 for subclinical hyperthyroidism, superior to a traditional features-based machine learning model. Patients identified as hyperthyroidism-positive by the deep learning model have a significantly higher risk (1.97-2.94 fold) of all-cause mortality and new-onset heart failure compared to hyperthyroidism-negative patients. This cardiovascular disease stratification is particularly pronounced in subclinical hyperthyroidism, surpassing that observed in overt hyperthyroidism.

CONCLUSIONS

An innovative algorithm effectively identifies overt and subclinical hyperthyroidism and contributes to cardiovascular risk assessment.

Can a Multidisciplinary Endocrine Clinic be Beneficial for Patients and Surgeons?

BACKGROUND

Multidisciplinary clinics are expected to improve patient care by enhancing efficiency for both patients and care providers. We hypothesized that while these clinics are an efficient use of time for patients, they can limit a surgeon's productivity.

METHODS

A retrospective review was performed for patients evaluated in a Multidisciplinary Endocrine Tumor Clinic (MDETC) and Multidisciplinary Thyroid Cancer Clinic (MDTCC) from 2018 to 2021. Time from evaluation to surgery and prevalence of surgery were evaluated. Patients were compared to those evaluated in a surgeon-only endocrine surgery clinic (ESC) from 2017 to 2021. Chi-square and t-tests were used to test significance.

RESULTS

Patients referred to the ESC underwent surgery more often than those referred to either multidisciplinary clinic (ESC 79.5%, MDETC 24.6%, MDTCC 7%; P < .001) but had a significantly longer delay between appointment and operation (ESC 19.9 days, MDETC 3.3 days, MDTCC 16.4 days; P < .001). Patients had a longer wait from referral to appointment for the MDCs (ESC 22.6 days, MDETC: 44.5, MDTCC 33; P < .05). There was no significant difference in miles traveled by patients to any clinic.

CONCLUSION

Multidisciplinary clinics can provide fewer appointments and faster time to surgery for patients but may lead to longer wait time from referral to appointment and fewer overall surgeries than endocrine surgeon-only clinics.

Postoperative Hypocalcemia in Hyperthyroid Patients: The Parathyroids Aren't Always to Blame

INTRODUCTION

Hypocalcemia is commonly reported after thyroidectomy and has multiple possible etiologies including: parathyroid devascularization, reactive hypoparathyroidism from relative hypercalcemia in thyrotoxicosis, and abrupt reversal of thyrotoxic osteodystrophy. In patients that are actively hyperthyroid and undergoing thyroidectomy, it is not known how many experience hypocalcemia from nonhypoparathyroidism etiologies. Therefore, our aim was to examine the relationship among thyrotoxicosis, hypocalcemia, and hypoparathyroidism.

METHODS

A retrospective review was performed of prospectively-collected data from all patients undergoing thyroidectomy for hyperthyroidism by 4 surgeons from 2016 to 2020. All patients carried a diagnosis of Graves' disease or toxic multinodular goiter. Patient demographics, preoperative medications, laboratory reports, and postoperative medications were reviewed. Hypocalcemia within the first month of surgery despite a normal parathyroid hormone (PTH) level was the primary outcome of interest and was compared between patients with and without thyrotoxicosis. Secondary outcomes were duration of postoperative calcium use and the relationship between preoperative calcium supplementation and postoperative calcium supplementation. Descriptive statistics, Wilcoxon rank-sum, and chi-square tests were used for bivariate analysis, as appropriate.

RESULTS

A total of 191 patients were identified, with mean age of 40.5 y (range 6-86). Most patients were female (80%) and had Graves' disease (80%). At the time of surgery, 116 (61%) had uncontrolled hyperthyroidism (thyrotoxic group, Free Thyroxine >1.64 ng/dL or Free Triiodothyronine > 4.4 ng/dL), with the remaining 75 (39%) considered euthyroid. Postoperative hypocalcemia (calcium < 8.4 mg/dL) developed in 27 (14%), while hypoparathyroidism (PTH < 12 pg/mL) was observed in 39 (26%). Thyrotoxic patients comprised a majority of those with hypocalcemia (n = 22, 81%, P = 0.01) and hypoparathyroidism immediately following surgery (n = 14, 77%, P = 0.04). However, a majority of initially hypocalcemic, thyrotoxic patients had normal PTH values within the first month after surgery (n = 17, 85%), pointing to a potential nonparathyroid etiology. On bivariate analysis, no significant relationship was found for thyrotoxic patients with initial postoperative hypocalcemia (18%) and hypoparathyroidism <1-month after surgery (29%, P = 0.29) or between 1 and 6 mo after surgery (2%, P = 0.24). Of the 19 patients in the nonhypoparathyroidism group, 17 (89%) were off all calcium supplements by 6 mo postop.

CONCLUSIONS

In patients with hyperthyroidism, those in active thyrotoxicosis at time of surgery have a higher rate of postoperative hypocalcemia compared to euthyroid patients. When hypocalcemia lasts >1 mo postoperatively, data from this study suggest that hypoparathyroidism may not be the primary etiology in many of these patients, who typically require calcium supplementation no more than 6 mo postoperatively.

Time to Symptom Resolution After Total Thyroidectomy for Graves' Disease

INTRODUCTION

Total thyroidectomy (TT) has been shown to be a safe and effective treatment for Graves' disease. However, the time course for improvement of symptoms has not been defined.

METHODS

With an institutional review board approval, we prospectively gathered survey data of all patients (n = 79) undergoing TT for Graves' disease at a single institution from 2019 to 2021. After informed consent was obtained, patients completed surveys preoperatively and at 2 wk followed by monthly postoperative visits/phone calls. Patient demographics and survey results were collected and analyzed. Symptom recovery time was evaluated using Kaplan-Meier analysis.

RESULTS

A total of 50 patients completed the survey on postoperative follow-up (response rate 63%). Average age was 38 y (range 12-80 y) and 88% of patients were female. The most common preoperative symptoms were fatigue (90%) and heat/cold intolerance (88%). Tremor (median time to resolution: 1 wk; interquartile range [IQR] 1-3), diarrhea (median 1 wk [IQR 1-3]), and palpitations (median 1 wk [IQR 1-3]) resolved the most rapidly followed by eye symptoms (median 3 wk [IQR 1-6]), heat/cold intolerance (median 3 wk [IQR 3-30]), memory deficits (median 3 wk [IQR 1-undefined]), and fatigue (median: 3 wk [IQR 1-14]). There were no significant differences in time to resolution of symptoms by gender or age (less than versus 40 y and older). Those with uncontrolled Graves' had more severe symptoms but no difference in time to resolution from the euthyroid Graves' patients.

CONCLUSIONS

Many Graves' disease symptoms improve rapidly following TT, with a median time to improvement of less than 1 mo.

Modulating Thyroid Hormone Levels in Adult Mice: Impact on Behavior and Compensatory Brain Changes

Thyroid hormone (TH) perturbation is a common medical problem. Because of substantial public health impact, prior researchers have studied hyper- and hypothyroidism in animal models. Although most prior research focused on in utero and/or developmental effects, changes in circulating TH levels are commonly seen in elderly individuals: approximately 20% of persons older than 80 years have clinically impactful hypothyroidism and up to 5% have clinical hyperthyroidism, with women being more often affected than men. TH disease model methodology in mice have varied but usually focus on a single sex, and the impact(s) of TH perturbation on the adult brain are not well understood. We administered thyroxine to middle-aged (13 to 14 months) male and female mice to model hyperthyroidism and TH-lowering drugs propylthiouracil (PTU) and methimazole, to induce hypothyroidism. These pharmacological agents are used commonly in adult humans. Circulating TH-level changes were observed when thyroxine was dosed at 20 µg/mL in drinking water for two weeks. By contrast, PTU and methimazole did not elicit a consistent reproducible effect until two months of treatment. No substantial changes in TH levels were detected in brain tissues of treated animals; however, pronounced changes in gene expression, specifically for TH-processing transcripts, were observed following the treatment with thyroxine. Our study indicated a robust compensatory mechanism by which the brain tissue/cells minimize the TH fluctuation in CNS by altering gene expression. Neurobehavioral changes were related to the TH perturbation and suggested potential associations between cognitive status and hyper- and hypothyroidism.

Wide provider variation in cost for thyroidectomy: Potential benefits of standardizing practice?

OBJECTIVE(S)

Identifying provider variation in surgical costs could control rising healthcare expenditure and deliver cost-effective care. While these efforts have mostly focused on complex and expensive operations, provider-level variation in costs of thyroidectomy has not been well examined.

METHODS

We retrospectively evaluated 921 consecutive total thyroidectomies performed by 14 surgeons at our institution between September 2011 and July 2016. Data were extracted from the Change Healthcare Performance Analytics Program.

RESULTS

Mean patient age was 47.4 ± 0.5 y, 81% were females, 64.7% were Caucasians, and 18.8% were outpatients. The number of thyroidectomies performed by the 14 surgeons ranged from 4 to 597 (mean = 66). The mean costs per provider varied widely from $4,293 to $15,529 (P < 0.001). The mean length of stay was 1d ± .03 with wide variation among providers (0-6 d). Providers whose hospital cost exceeded the institutional mean demonstrated significantly higher anesthesia fees and lab costs (P < 0.001).

CONCLUSIONS

We found substantial variation in hospital cost among providers for thyroidectomy despite practicing in the same academic institution, with some surgeons spending 3x more for the same operation. Implementing institutional standards of practice could reduce variation and the costs of surgical care.

Hyperthyroidism Symptoms, Management, and Outcomes in Children and Adults Seeking Definitive Surgical Treatment

OBJECTIVE

To identify the difference in presentation, time to treatment, and outcomes between hyperthyroid adults and children referred to surgical evaluation.

BACKGROUND

There is little data on differences in presenting symptoms, time to treatment, and outcomes between adults and children presenting for thyroidectomy for Graves' disease.

METHODS

We retrospectively reviewed records of patients with hyperthyroidism referred for thyroidectomy between January 2016 and April 2017. We divided our cohort into 2 groups based on age, children (age ≤18 years), and adults (age >18), and evaluated the difference in prevalence of symptoms, time from diagnosis, and initiation of antithyroid medications to surgery, and outcomes.

RESULTS

We identified 38 patients (27 adults and 11 children) with data on hyperthyroidism symptoms referred for thyroidectomy. Relative to hyperthyroid adults, children with hyperthyroidism were more likely to present with hoarseness (55% vs 15%, P = 0.01) and difficulty concentrating (45% vs 7%, P = 0.01) at initial presentation. There was no statistically significant difference in prevalence of vision changes, exophthalmos, pretibial myxedema, palpitations, fatigue, temperature intolerance, dysphagia, tremors, or constitutional symptoms. A median of 15 months elapsed from diagnosis to thyroidectomy among adult and 6 months among pediatric patients. Adult and pediatric patients waited a median of 13 and 6 months from initiation of antithyroid medications to thyroidectomy, respectively. There was no significant difference in outcomes.

CONCLUSIONS

Children with hyperthyroidism were more likely to present with hoarseness and difficulty concentrating than adults. Concentration and communication are critical skills in developing children, and early intervention with definitive therapy may improve such symptoms.

Incidence and prevalence of hyperthyroidism: a population-based study in the Piedmont Region, Italy

PURPOSE

Unrecognized and untreated hyperthyroidism leads to serious clinical complications with adverse outcomes for patients and increasing costs for the health care system. Hence, adequate knowledge of the epidemiological features of such condition is desirable to plan effective interventions. The aim of our study was to estimate incidence and prevalence of hyperthyroidism in the mildly iodine-deficient Italian Region of Piedmont.

METHODS

A retrospective cohort study was conducted using Administrative Health Databases of the Piedmont Region, Italy (2012-2018). Hyperthyroidism cases were defined as the subjects who had at least one of the following claims: (i) hospital discharge records with hyperthyroidism diagnosis code; (ii) exemption from co-payment for hyperthyroidism; (iii) prescription of one of the following medications: methimazole, propylthiouracil, or potassium perchlorate.

RESULTS

The overall prevalence was 756 per 100,000 inhabitants [95% CI 748-764], and the overall incidence was 81 per 100,000-person year [95% CI 80-82]. The prevalence and incidence increased with age and were two-fold higher among women than men. Women also showed two distinct peaks in incidence at the age of 30 and 50; after the age of 60, the trend became similar between sexes. With regard to the geographic distribution, an increasing gradient of incidence was observed from the northern to the south-western areas of the Region.

CONCLUSION

This is the first Italian study based on health databases to estimate the incidence and prevalence of hyperthyroidism in the general population. This approach can represent an inexpensive and simple method to monitor patterns of hyperthyroidism in iodine-deficient areas.

Considerations for Thyroidectomy as Treatment for Graves Disease

Hyperthyroidism is a clinical state that results from abnormally elevated thyroid hormones. Thyroid gland affects many organ systems; therefore, patients usually present with multiple clinical manifestations that involve many organ systems such as the nervous, cardiovascular, muscular, and endocrine system as well as skin manifestations. Hyperthyroidism is most commonly caused by Graves disease, which is caused by autoantibodies to the thyrotropin receptor (TRAb). Other causes of hyperthyroidism include toxic multinodular goiter, toxic single adenoma, and thyroiditis. Diagnosis of hyperthyroidism can be established by measurement of thyroid-stimulating hormone (TSH), which will be suppressed with either elevated free T4 and/or T3 (overt hyperthyroidism) or normal free T3 and T4 (subclinical hyperthyroidism). Hyperthyroidism can be treated with antithyroid drugs (ATDs), radioactive iodine (RAI), or thyroidectomy. ATDs have a higher replacement rate when compared with RAI or thyroidectomy. Recent evidence has shown that thyroidectomy is a very effective, safe treatment modality for hyperthyroidism and can be performed as an outpatient procedure. This review article provides some of the most recent evidence on diagnosing and treating patients with hyperthyroidism.