Preoperative Thyroid-Stimulating Hormone Levels and Three-Year Mortality in Elderly Hip Fracture Patients: Insights from a Prospective Cohort Study

Introduction

Hip fracture is a major health concern for elderly individuals, leading to significant mortality rates [1]. Elderly individuals who have experienced hip fractures often face a higher risk of mortality after surgery due to their existing comorbidities [2,3]. One-year mortality rates for patients with HF have been reported to be as high as 20–24%, with mortality risks persisting beyond 5 years [4].

Hip fracture is frequently a result of bone fragility and low-energy trauma [5]. Studies have revealed that various factors can cause bone fragility, including aging, endocrine diseases, osteoporosis, and medications [6,7]. Bone mineral density (BMD), a reliable marker for osteoporosis, has been reported to be affected by both hyperthyroidism and hypothyroidism [8]. In addition, a recent study indicated that low BMD was linked to higher mortality risk after hip fracture surgery in the elderly [9]. Therefore, thyroid dysfunction might be correlated with mortality in geriatric hip fracture patients after surgery.

There have only been a few studies to investigate the relationship between thyroid-stimulating hormone (TSH) levels and postoperative mortality in patients with hip fracture [10]. A recent study from Denmark found that hip fracture patients who underwent surgery and had plasma TSH levels higher than 1.41 mIU/L upon admission had a higher 30-day mortality rate [10]. However, it is still unclear whether preoperative TSH level is associated with long-term mortality in geriatric hip fracture patients.

Identifying risk factors associated with postoperative mortality is crucial for enhancing outcomes following hip fractures [11]. The purpose of this study was to investigate whether TSH is correlated with 3-year postoperative mortality in geriatric hip fracture patients.

Material and Methods

STUDY DESIGN:

The current study was conducted at a single tertiary hospital located in Beijing, China, which has implemented an orthogeriatric co-management (OGC) pathway for elderly hip fracture patients. Ethics approval was obtained from the Institutional Review Board at Peking University Health Science Centre (IRB00001052-17021) and the Biomedical Ethics Committee at Beijing Jishuitan Hospital (201807-11), before starting the study. All participants provided written informed consent before data collection. For illiterate participants, informed consent to participate was obtained from their literate legal guardians. This post hoc analysis used baseline data gathered prospectively in our prior observational research, which investigated the effect of OGC model on elderly patients with hip fracture in China (Clinical Trials.gov Identifier: NCT03184896) between November 2018 and November 2019. The clinical approach was previously explained [12].

INCLUSION AND EXCLUSION CRITERIA:

Inclusion criteria: We included patients who underwent hip fracture surgery within 3 weeks after injury from November 2018 to November 2019. The study comprised a total of 1092 participants.

Exclusion criteria: Patients were excluded if they had incomplete baseline data, lacked admission thyroid function tests, suffered from liver or kidney failure, had a history of thyroid surgery, or had pathological fractures or terminal malignancies.

DATA COLLECTION:

Demographic information and perioperative records were gathered prospectively in the current study. Age, sex, height, weight, body mass index (BMI), alcohol consumption, and education attainment were all included in the demographic data. Medical conditions such as hypertension, diabetes, cognitive impairments, and visual impairments were also documented. Educational level was categorized into 5 levels, ranging from illiterate to university or higher. Cognitive function was evaluated with the Mini-Mental State Examination – China (MMSE) [13], and participants with scores of 23 or less were considered cognitively impaired. As a measure of overall health, the Charlson Comorbidity Index (CCI) was calculated [14]. During the perioperative assessment, various tests and information were gathered, including time to surgery (TTS), admission blood tests, fracture type, American Society of Anesthesiologists (ASA) scores, anesthesia method, length of stay (LOS), and rehabilitation data. Nutritional status was indicated by albumin level. Cardiac function was indicated by N-terminal pro-brain natriuretic peptide (NT-proBNP). The thyroid function test primarily includes TSH, triiodothyronine (T3), thyroxine (T4), free T3 (FT3), and free T4 (FT4). Given that FT3 and FT4 remain unaffected by the concentration and binding properties of thyroid-binding proteins, we used FT3 and FT4 to assess thyroid function when TSH was abnormal [15].

Patients with intertrochanteric fractures (ITFs) were treated with intramedullary nail fixation. For those with femoral neck fractures (FNFs), osteosynthesis or arthroplasty was used depending on fracture stability. In this study, we classified operation types into arthroplasty and internal fixation (including intramedullary nail fixation and osteosynthesis).

INTERPRETATION OF THYROID FUNCTION TESTS AND CATEGORIZATION OF PATIENTS:

In this study, the initial TSH measurement taken upon admission to the emergency department was used to investigate its association with postoperative mortality. To precisely elucidate the association between admission TSH levels and postoperative mortality, patients were grouped based on their median and quartile TSH levels as well as their thyroid function status at admission.

Patients were divided into 2 groups by a median TSH value of 1.64 mIU/L and 4 groups by quartiles: Q1 (TSH <0.98 mIU/L), Q2 (0.98≤ TSH <1.64 mIU/L), Q3 (1.64≤ TSH <2.86 mIU/L), and Q4 (TSH ≥2.86 mIU/L).

Additionally, patients were classified into 5 categories according to their thyroid function status at admission: euthyroid (0.27≤ TSH ≤4.20 mIU/L), hypothyroidism or subclinical hypothyroidism (TSH >4.20 mIU/L with low FT3/FT4 levels or normal FT3/FT4 levels respectively), hyperthyroidism or subclinical hyperthyroidism (TSH <0.27 mIU/L with high FT3/FT4 levels or normal FT3/FT4 levels, respectively) [15]. Patients with inconsistent results between TSH and FT3/FT4 were excluded from the analysis.

OUTCOME MEASUREMENT:

A 3-year follow-up telephone assessment was conducted by orthopedic specialists for all patients, with a median follow-up duration of 3.1 years (from 2018–2019 to 2021–2022) for the study cohort. The primary objective of this investigation was to ascertain the overall mortality rate within the 3-year period subsequent to hip fracture surgery.

STATISTICAL ANALYSIS:

The median and interquartile range (IQR) were used for non-parametric data, while means and standard deviations (SD) were used for parametric data. Categorical variables are displayed in numerical distribution and frequency. The chi-square test was used to compare categorical variables. Student’s t-tests or Mann-Whitney U tests were applied for continuous variables, depending on their distribution (parametric or non-parametric). Covariates were defined as baseline variables that were considered clinically relevant or univariate with outcomes. The association between TSH levels and mortality risk was calculated using multivariate Cox proportional hazards models, both with and without adjusting for covariates. The 3-year mortality was analyzed using Kaplan-Meier analyses, and the difference between strata was determined using log-rank analyses.

The data were analyzed using R 4.1.1 (http://www.R-project.org, The R Foundation), with a two-tailed test, and P<0.05 was considered statistically significant.

Results

Figure 1 presents the flowchart of the study. Of 799 eligible patients, 92.7% (741/799) completed the follow-up, 20.6% (153/741) deceased during the 3-year follow-up, including 7.0% (52/741) of deaths during the first year. Upon inclusion, 621 patients were euthyroid, 33 had hypothyroidism, 68 had subclinical hypothyroidism, 6 had hyperthyroidism, and 9 had subclinical hyperthyroidism. Four patients showed inconsistent TSH and FT3/FT4 results. Table 1 presents the baseline characteristics of 2 groups categorized by median admission TSH levels. The cohort had an average age of 79.7 years, consisting of 73.3% females and 52.1% ITFs. Patients with lower TSH levels were older and exhibited a higher incidence of TTS within 48 hours compared to those with higher TSH levels. Additionally, the lower TSH group had more patients suffering from cognitive impairment and received post-surgery physiotherapy more frequently than the higher TSH group.

The Kaplan-Meier analysis in Figure 2 shows no significant difference in survival probability between the lower and higher TSH groups (P=0.140). Similarly, Figure 3 indicates no significant variance in survival probability from Q1 to Q4 (P=0.440). Lastly, Figure 4 reveals no notable disparity in survival probability among euthyroid, hypothyroidism, subclinical hypothyroidism, hyperthyroidism, and subclinical hyperthyroidism patients (P=0.140).

As shown in Table 2, after adjusting for all covariates, there was no significant correlation between TSH levels and mortality risk when treated as a continuous variable (HR 0.98 95% CI 0.92–1.05, P=0.639). Similarly, dividing patients into 2 groups based on median TSH value revealed no significant difference in mortality risk between the higher and lower TSH groups (HR 0.86, 95% CI 0.62–1.20, P=0.379). This pattern continued when patients were split into 4 groups according to TSH quartiles. Q4, Q3, and Q2 showed no significant difference in mortality risk compared to Q1 (HR 1.13, 95% CI 0.73–1.74, P=0.575; HR 0.93, 95% CI 0.60–1.46, P=0.768; HR 0.90, 95% CI 0.56–1.44, P=0.664, respectively). Lastly, the thyroid function status upon admission did not demonstrate a statistically significant impact on the 3-year mortality risk in patients with hypothyroidism, subclinical hypothyroidism, hyperthyroidism, or subclinical hyperthyroidism when compared to euthyroid patients (HR 1.34, 95% CI 0.72–2.49, P=0.359; HR 0.77, 95% CI 0.38–1.60, P=0.489; HR 1.15, 95% CI 0.16–8.30, P=0.890; HR 1.07, 95% CI 0.34–3.38, P=0.913, respectively).

Discussion

Hip fracture has emerged as a prevailing global public health concern. According to data from the 2018 Asian Federation of Osteoporosis Societies, it is foreseen that by the year of 2050, the number of individuals afflicted by hip fracture in Asia will increase by 2.28 times from 2018 [16], which will lead to an increase from 1.1 million to approximately 2.5 million [17]. The data reveals that the perioperative mortality rate for elderly hip fracture patients ranges from 2.3% to 13.9%, with a mortality rate of 8.0% to 10.0% within one month, 12.0% to 23.0% within 6 months, and 20.0% to 40.0% within 1 year [18,19]. Osteoporosis, advanced age, malnutrition, congestive heart failure, and renal insufficiency are frequently acknowledged as adverse prognostic factors for hip fracture outcomes. Simultaneously, hyperthyroidism is commonly recognized as a risk factor for osteoporosis and fractures, inducing heightened bone resorption, expediting bone turnover, and ultimately culminating in diminished bone density and an elevated risk of fractures [20]. Thyroid hormone levels are intimately interconnected with endocrine factors governing bone metabolism. Evidence suggests that TSH, via mediation by the TSH receptor (TSHR) on osteoblasts, osteoclast precursors, and osteoclasts, directly influences both the bone-forming and bone-resorbing components, thus regulating bone remodeling [21]. This also substantiates the notion that TSH could independently influence bone formation and resorption processes, thus affecting bone density [22]. However, there is a relative scarcity of studies addressing whether TSH hormone levels can directly impact the short-term and long-term mortality rates following hip fracture surgery.

In this study, TSH levels were not found to be associated with 3-year mortality in elder hip fracture patients following surgeries. In a prospective observational study including 599 elderly male participants from the Netherlands, the mortality risk decreased by 23% (HR 0.77, 95% CI 0.63–0.94, P=0.009) for each standard deviation (SD) increase of thyrotropin by 2.71 mIU/L [23]. The study suggested a survival benefit associated with increased thyrotropin levels and decreased free thyroxine levels, both indicative of lower thyroid function. Gussekloo et al [23] reported that lower metabolic rate is related with increased survival in several species and it may suggest that a lower metabolic rate underlies the lower mortality in humans with decreased free thyroxine. The correlation between TSH and outcome after hip fracture has been examined by Ling et al [24] in 2013 and Rapacki et al [10] in 2019. Ling et al [24] found that thyroid dysfunctions increased the risk of hip fracture complications following surgery. However, they did not find an association between TSH and 30-day mortality, which is partially consistent with our findings. Nevertheless, Rapacki et al [10] reported that TSH level was related with 30-day mortality. Their study enrolled 914 patients aged >60 years with hip fracture from 1996 to 2012 and found that higher TSH level (>1.41 mIU/L) was linked to increased 30-day mortality in hip fracture patients after surgery (HR 2.1, 95% CI 1.38–3.26, P<0.001). A prospective study conducted by Kalra et al [25] in the United Kingdom involving 131 elderly patients (average age of 82) concluded that subclinical thyroid dysfunction (TSH <0.35 mIU/L or TSH >5.50 mIU/L without overt thyroid dysfunction) did not affect the 1-year mortality rate of elderly patients who received surgical treatment for hip fracture (P=0.477) [25]. Their study is consistent with the conclusions drawn in the present study regarding the impact of TSH on long-term mortality rates after hip fracture surgery.

This study has the advantage of comprehensive and prospective data collection, which minimizes recall bias. By thoroughly and meticulously recording patients’ baseline information, many potential confounding factors that could lead to bias could be mitigated. This contributes to a more comprehensive conclusion regarding the impact of TSH on postoperative mortality rates following hip fracture.

One limitation of this study is that the statistical data regarding patients’ thyroid function were derived from the TSH hormone levels upon admission, which could be influenced by long-term use of medications such as calcitonin, bisphosphonates, estrogens, and corticosteroids. Moreover, since the data used in this study came from baseline data collected prospectively in our previous observational study, it is difficult to track specific medication usage at that time, potentially introducing bias. On the other hand, because all patients in this study experienced hip fractures within a span of 3 weeks, acute major trauma could trigger an internal stress response, thereby affecting TSH levels in the blood [26,27]. Due to these reasons, the TSH hormone values collected by patients upon admission may lack representatives. Finally, although this study included a considerable number of patients, the categorization of patients based on thyroid function status led to smaller sample sizes for hyperthyroidism and subclinical hyperthyroidism, which may have impacted the study’s findings. Therefore, future research with a larger sample size is warranted.

Conclusions

In summary, this study found that there was no significant correlation between TSH levels at admission and 3-year postoperative mortality in geriatric patients after hip fracture surgery.