Cardiovascular disease and health anxiety in patients with differentiated thyroid cancer

Abstract

Background

Differentiated thyroid cancer (DTC) accounts for over 90% of all thyroid malignancies, and is typically treated with a total thyroidectomy and radioactive iodine (RAI) to eliminate both macro- and microscopic cancer tissue. After these procedures, patients require levothyroxine to substitute for loss of thyroid hormones. Thyroid stimulating hormone (TSH) stimulates proliferation of both healthy and malignant thyroid cells and is suppressed by high doses of levothyroxine. Therefore, levothyroxine is administered in suppressive doses, which by definition induces a state of subclinical hyperthyroidism. This condition, when present in thyroid diseases other than DTC, is associated with negative health outcomes including cardiovascular disease. Furthermore, cancer patients are in general more susceptible to mental health challenges, such as anxiety disorders. Additionally, the TSH suppression treatment can cause symptoms, such as heart palpitations among others, which are also common in anxiety disorders. One specific form of anxiety, called health anxiety, involves an excessive preoccupation with interpreting bodily signals and connecting them to somatic illness. Therefore, it is plausible that DTC patients may be more vulnerable to this type of anxiety, however no previous research has investigated this question.

This thesis investigated cardiovascular disease and health anxiety in patients with DTC, placing a particular emphasis on providing explanations on how the TSH suppression treatment can affect these outcomes.

Study 1 and Study 2

In Study 1 and Study 2, we analysed cardiovascular mortality and incidence by investigating individuals diagnosed with DTC in Sweden between 1987 and 2013. Patients were followed from one year after DTC diagnosis until death, migration or December 31, 2014 (last study date). The data were obtained from nationwide healthcare registries that included information on hospitalization, cancer disease, causes of death, as well as data on migration. DTC patients were identified from the Swedish Cancer Registry and were assumed to be on lifelong TSH suppression treatment as according to prevalent guidelines during the study period. Six cardiovascular endpoints were considered separately (atrial fibrillation, cerebral infarction, cerebrovascular disease, ischemic heart attack, ischemic heart disease, and heart failure) as well as all the six endpoints combined. Only individuals who remained in the cohort for at least one year after DTC diagnosis were included in the final study cohort. The mortality (incidence) rates in the DTC cohort were compared to the ones of the general population by calculating standardized mortality ratios (SMRs) and standardized incidence ratios (SIRs), respectively.

We identified 6900 patients that were diagnosed with DTC, survived, and did not emigrate for at least one year after diagnosis. Study 1 did not find differences in mortality rates in the DTC cohort as compared to the rates found in the general population, when considering all cardiovascular endpoints combined. However, a gender difference was observed, with DTC males having a slightly elevated mortality rate compared to males in the general population (SMR 1.16 CI, 95% 1.02-1.31). When investigating specific cardiovascular endpoints, the whole DTC cohort displayed an increased mortality rate of atrial fibrillation (AF) in comparison to the general population (SMR 1.36, CI 95% 1.12-1.64). In study 2, analyses of SIRs revealed a higher incidence rate of AF for the entire DTC cohort as compared to the rate of the general population (SIR 1.66, CI 95% 1.41–1.94). The incidence rate ratio of cerebrovascular disease was borderline significant (SIR 1.14 CI 95% 1.01-1.29) and the result was driven by females where it was slightly elevated (SIR 1.20, CI 95% 1.04–1.38). Most importantly, when evaluating cardiovascular mortality and incidence rates for the entire DTC cohort in comparison to the corresponding rates in the general population, it's essential to note that SMRs and SIRs for these four endpoints were statistically insignificant: cerebral infarction, ischemic heart disease, ischemic heart attack, and heart failure.

Study 3

This study had two distinct yet interrelated aims. The first aim was to analyse the patterns and variability of TSH values in DTC patients during long-term follow-up. The second aim was to examine the association between the degree of TSH suppression and the risk of developing AF during follow-up for DTC. These objectives were examined in a study population comprising DTC patients diagnosed between 1995 and 2015 in the Stockholm region, and these individuals were identified from the Swedish Cancer Registry. Each patient’s medical record was thoroughly reviewed, after which cases of microcarcinoma, misclassified diagnoses and cases not relevant for the purpose of this study, were excluded. All available TSH values from the date of surgery until end of August 2022 were retrieved, and each TSH value was assigned to one of the three TSH categories: "unsuppressed" (TSH > 0.5 mE/L), "mildly suppressed" (TSH: 0.1–0.5 mE/L), and "suppressed" (TSH < 0.1 mE/L). For all the analyses described below, the follow-up period began 9 months after the date of surgery, despite the collection of certain information prior to the start of the follow-up. For the purpose of the first aim, TSH values and TSH categories were graphically analysed for up to 10 years of follow-up. For the purpose of the second aim, a nested-case control study was performed. AF status was obtained from medical records, with the baseline status extracted from the preoperative anaesthesia assessment conducted prior to the thyroidectomy for the primary tumor. Monitoring of AF status continued until the first recorded instance of AF, censoring, or August 31, 2022. AF cases were matched with controls based on age at DTC diagnosis and follow-up time since DTC diagnosis (regardless of calendar year of DTC diagnosis). AF cases and controls were categorized based on the TSH category in which they spent the majority of their follow-up time, counted from inclusion up to the point of the event or matching. Finally, a conditional logistic regression analysis was performed to compare the TSH categories between the cases and controls, calculating hazard ratios of AF. Established risk factors for AF, that were prevalent prior to DTC diagnosis, were also collected among new cases of AF. However, it is important to clarify that this additional information was not used in the nested case-control study, but was collected with the purpose to provide complementary information.

From the Swedish Cancer Registry, 1031 individuals classified as DTC were identified, and after reviewing their medical records, 608 patients remained in the final cohort. A total of 11283 TSH values were registered, most of the them (69%) were classified as “suppressed” (TSH < 0.1 mE/L), and 78% of patients were in a “suppressed” condition for more than half of their follow-up period. There were 39 new cases of AF. Among both AF cases and controls, there was a limited number of subjects categorized to the “mildly suppressed” and “unsuppressed” TSH groups. This distribution had implications for the nested case-control study, since hazard ratios of AF showed wide confidence intervals and were overlapping. Complementary data revealed that individuals with AF were older in comparison to the remaining cohort. Additionally, nearly half of these AF patients had pre-existing risk factors for AF before their DTC diagnosis, indicating a potential predisposition to AF. To conclude, most of the studied DTC patients were “suppressed” during follow-up, and the findings in this study did not yield definitive conclusions regarding the influence of TSH levels on the risk of developing AF in patients with DTC.

Study 4

This was a cross-sectional study with two aims. The first aim was to assess the prevalence of clinically significant health anxiety in DTC patients. The second aim was to estimate the association between the degree of TSH suppression and the severity of health anxiety in patients with DTC. The population in which these aims were studied was derived from a previously established nationwide population-based DTC cohort (in which health-related quality of life was researched), where the sub-population selected for this study was restricted to patients with active oncological follow-up at the Karolinska University Hospital. These patients received DTC diagnosis 2012-2017, and were treated with surgery and radioactive iodine for the primary tumor. In 2020, 201 patients were eligible and they were all invited to participate in a survey consisting of the 14-item Short Health Anxiety Inventory (SHAI-14), and a study-specific questionnaire. Each of the 14 items in the SHAI-14 was scored on a scale from 0 to 3, and a score of 18 or above was considered to be indicative of clinically significant health anxiety. The study-specific questionnaire provided information on self-reported medical conditions diagnosed by a physician, along with details about socioeconomic background. Medical records were carefully reviewed to identify the following information: the risk stage of the primary tumor, information on cancer recurrence and biomarkers including the single closest TSH value prior to completion of surveys. TSH values were categorized in three levels: “unsuppressed” (TSH > 0.5 mE/L), "mildly suppressed" (TSH: 0.1–0.5 mE/L), and "suppressed" (TSH < 0.1 mE/L). The relationship between TSH levels and SHAI-14 scores (range 0-42) was analysed by a linear regression model, where the SHAI-14 score was the dependent variable. TSH levels served as an independent variable controlling for cancer recurrence (“yes” or “no”), risk stage (“low or high”), and comorbidity (“none to one” / “at least two”).

In total, 146 patients (73% of the invited 201 patients) were included in the analysis. Among them, the mean score of SHAI-14 was 11.3 and 24 patients (16%) scored 18 or higher. The proportion of examined DTC patients experiencing clinically significant levels of health anxiety was slightly higher than that observed in the general population, but lower in comparison to other cancer patients with similar characteristics. Patients with a "mildly suppressed" TSH level scored on average, 3.28 points higher on the SHAI-14 questionnaire (pvalue 0.01), compared to patients with an "unsuppressed" TSH level. No statistically significant difference was found between patients with a “suppressed” TSH level and those with an “unsuppressed” TSH level. Therefore, contrary to expectations, we found no consistent linear association between the TSH categories and the degree of health anxiety.