Body mass index and post-thyroidectomy hypocalcemia: a protective effect of overweight through non-surgical mechanisms—a propensity score-matched study

Abstract

Background

Postoperative hypocalcemia is a common complication after total thyroidectomy (TT). Recent studies suggest that body mass index (BMI) may influence its occurrence. This study aimed to investigate whether overweight reduces postoperative hypocalcemia after TT and to investigate the impact of surgical and non-surgical factors on postoperative hypocalcemia in patients undergoing total thyroidectomy.

Methods

A retrospective analysis was conducted on 228 patients who underwent TT for papillary thyroid carcinoma between January 2021 and January 2024. Patients were categorized into overweight (BMI ≥ 25 kg/m², n = 96) and non-overweight groups (BMI < 25 kg/m², n = 132). Propensity score matching (PSM) was performed to balance confounding factors. Postoperative hypocalcemia, hypoparathyroidism rates, and related biochemical markers were compared between matched groups.

Results

After PSM (51 pairs), baseline characteristics were balanced except for Total Cholesterol (TC) (1.40 ± 0.83 vs 1.99 ± 1.80 mmol/L, P = 0.036), High-density lipoprotein cholesterol (HDL-C) (1.30 ± 0.34 vs 1.10 ± 0.23 mmol/L, P < 0.001), and fatty liver presence (35.5% vs 75.0%, P = 0.001). Postoperative hypocalcemia was significantly higher in the non-overweight group (78.4% vs 54.9%, P = 0.020), while hypoparathyroidism rates showed no significant difference (54.9% vs 62.7%, P = 0.546). Postoperative calcium levels were higher in the overweight group (2.11 ± 0.12 vs 2.05 ± 0.12 mmol/L, P = 0.014).

Conclusions

Non-overweight patients are more likely to develop hypocalcemia after TT compared to overweight patients. The protective effect of overweight appears to operate through non-surgical mechanisms, as evidenced by similar hypoparathyroidism rates between groups. Further research is needed to elucidate the specific mechanisms involved.

Background

Thyroid cancer ranks ninth in global cancer incidence according to the WHO International Agency for Research on Cancer (IARC) 2020 database, with papillary thyroid cancer accounting for approximately 80% of cases [1]. Total thyroidectomy (TT) is one of the standard treatments for differentiated thyroid cancer, with surgical approach selection based on various factors including tumor characteristics, patient factors, and current guidelines [2]. However, postoperative hypocalcemia, particularly after TT, is a significant concern with reported incidence of temporary and permanent hypoparathyroidism ranging from 4.6% to 51.9% [3]. This complication typically presents as numbness in the extremities and may affect respiratory function, significantly impacting patient recovery [4].

Concurrently, obesity and overweight have emerged as global health challenges [5]. These conditions are associated with various comorbidities, including diabetes and hypertension, which traditionally are considered to increase perioperative complications [6]. According to the World Health Organization Asian Standard, overweight is defined as BMI ≥ 25 kg/m² [7]. Interestingly, recent studies have suggested a protective effect of higher BMI in certain clinical scenarios, a phenomenon termed the "obesity paradox" [8, 9].

The relationship between BMI and postoperative hypocalcemia appears complex and potentially involves both surgical and non-surgical factors. While surgical factors, such as technical difficulty in patients with larger neck circumference, have been well-documented [10], the potential protective mechanisms associated with higher BMI remain poorly understood [11]. Understanding these mechanisms could have significant implications for perioperative management strategies.

Therefore, this study aimed to investigate whether overweight reduces the occurrence of postoperative hypocalcemia after TT and to investigate the impact of surgical and non-surgical factors on postoperative hypocalcemia in patients undergoing total thyroidectomy. We hypothesized that overweight might influence postoperative calcium homeostasis through non-surgical pathways, independent of direct surgical factors affecting parathyroid function.

Methods

Study design and setting

A retrospective cohort study was conducted at the Fourth Affiliated Hospital, Zhejiang University School of Medicine between January 2021 and January 2024. The study protocol was approved by the Institutional Ethics Committee (approval number: K2024098).

The study enrolled 228 patients who underwent total thyroidectomy. Patients were eligible if they met the following criteria: preoperative serum calcium and parathyroid hormone (PTH) within normal range, no previous thyroid surgery, postoperative pathology confirming papillary thyroid cancer, and provided informed consent. We excluded patients with diseases affecting PTH and calcium metabolism, vital organ damage, parathyroid disease, or recent use of medications affecting calcium and PTH metabolism. Following WHO guidelines, patients were categorized based on BMI into overweight (≥ 25 kg/m2, n = 96) and non-overweight (< 25 kg/m2, n = 132) groups.

Data collection

We extracted clinical data from electronic medical records including demographic characteristics (age, gender, marriage status, education level), preoperative comorbidities (hypertension, diabetes, Hashimoto thyroiditis, goiter, fatty liver), surgical parameters (approach, tumor size, multifocality, lymphadenectomy extent, capsular/nerve/vascular invasion, operation duration, blood loss, hospital stay), and laboratory measurements (TC, TG, LDL-C, HDL-C, creatinine, preoperative calcium and PTH levels, postoperative calcium and PTH levels) and the incidence of postoperative hypocalcemia and hypoparathyroidism.

Outcome measures

The primary outcome was the incidence of postoperative hypocalcemia, defined as serum calcium < 2.12 mmol/L (8.5 mg/dL) within one week after surgery, or presence of symptoms including limb twitching, numbness, and sensory abnormalities despite normal calcium levels. Secondary outcome was the incidence of postoperative hypoparathyroidism, defined as PTH < 15 pg/mL within one week after surgery.

Propensity score matching

To minimize selection bias and control for potential confounding factors, propensity score matching was performed using all variables except postoperative outcomes (postoperative calcium levels, postoperative PTH levels, and the incidence of postoperative hypocalcemia and hypoparathyroidism). The matching was conducted using 1:1 nearest-neighbor matching with a caliper width of 0.02. Matching quality was evaluated by comparing baseline characteristics between groups.

Statistical analysis

Data analysis was performed using SPSS version 25.0. Continuous variables were expressed as mean ± standard deviation, and categorical variables as numbers and percentages. Between-group comparisons were conducted using χ2 test for categorical variables. Spearman correlation analysis was performed to assess the relationship between BMI and postoperative calcium homeostasis markers. Statistical significance was set at P < 0.05.

Results

Baseline characteristics

Between January 2021 and January 2024, 1135 patients were screened at our institution. After applying inclusion and exclusion criteria, 228 patients were enrolled in the study (Fig. 1), with 135 cases (59.2%) experiencing postoperative hypocalcemia. Patients were categorized into non-overweight (n = 132) and overweight (n = 96) groups based on BMI.

Fig. 1

Flow diagram of patient selection and propensity score matching process

Baseline comparison showed no significant differences between non-overweight and overweight groups in multiple parameters: age (P = 0.311), marital status (P = 0.687), education level (P = 0.502), surgical approach (P = 0.510), Hashimoto’s thyroiditis (P = 0.055), goiter (P = 0.151), maximal tumor diameter (P = 0.199), tumor multifocality (P = 0.873), extent of lymphadenectomy (P = 0.999), nerve invasion (P = 0.265), vascular invasion (P = 0.702), operative duration (P = 0.921), intraoperative blood loss (P = 0.201), length of hospital stay (P = 0.283), total cholesterol (P = 0.849), LDL-C (P = 0.546), preoperative calcium (P = 0.945), and preoperative PTH levels (P = 0.464).

However, significant differences were observed in several characteristics: gender distribution (male-to-female ratio, P = 0.003), comorbidities including hypertension (P = 0.006), diabetes (P = 0.015), and fatty liver (P < 0.001), tumor characteristics (capsular invasion, P = 0.033), and laboratory parameters including triglycerides (P < 0.001), HDL-C (P < 0.001), and serum creatinine (P = 0.002) (Table 1). To address these baseline differences, propensity score matching was subsequently performed.

Table 1 Baseline characteristics of the two groups before PSM

Baseline characteristics after propensity score matching

Propensity score matching successfully paired 51 patients from each group. After matching, baseline characteristics were well balanced between groups for 23 covariates (all P > 0.05), including demographic features (age, gender, marital status, education level), preoperative comorbidities (hypertension, diabetes, Hashimoto’s thyroiditis, goiter), surgical parameters (approach, tumor diameter, multifocality, lymphadenectomy extent, capsular/nerve/vascular invasion, operation duration, blood loss, hospital stay), and laboratory measurements (total cholesterol, LDL-C, preoperative calcium and PTH levels) (Table 2).

Table 2 Baseline characteristics in the propensity score–matched sample

However, three metabolic parameters remained significantly different between groups. The overweight group showed higher prevalence of fatty liver (75.0% vs 35.5%, P = 0.001), higher triglyceride levels (1.99 ± 1.80 vs 1.40 ± 0.83 mmol/L, P = 0.036), and lower HDL-C levels (1.10 ± 0.23 vs 1.30 ± 0.34 mmol/L, P < 0.001). These persistent differences suggest that overweight patients maintain distinct metabolic characteristics even after matching for other confounding factors.

Postoperative outcomes and calcium metabolism

In the propensity score-matched cohort, the incidence of postoperative hypocalcemia was significantly lower in the overweight group compared with the non-overweight group (54.9% vs 78.4%, P = 0.020). Postoperative serum calcium levels were also higher in the overweight group (2.11 ± 0.12 vs 2.05 ± 0.12 mmol/L, P = 0.014).

However, no significant differences were observed between groups in either the incidence of postoperative hypoparathyroidism (62.7% vs 54.9%, P = 0.546) or postoperative PTH levels (13.94 ± 13.19 vs 15.11 ± 12.69 pg/mL, P = 0.649) (Table 3).

Table 3 Postoperative complications and biochemical indicators

Additionally, correlation analysis revealed a positive correlation between BMI and postoperative calcium levels (Spearman R = 0.289, P = 0.003) (Fig. 2A), suggesting that higher BMI was associated with better maintenance of calcium homeostasis after surgery. However, no significant correlation was found between BMI and postoperative PTH levels (Spearman R = −0.005, P = 0.958) (Fig. 2B), further supporting the hypothesis that BMI’s effect on calcium levels operates through mechanisms independent of parathyroid function.

Fig. 2

BMI correlations with postoperative calcium homeostasis markers at 24 h. A Serum calcium levels (B) PTH levels

Discussion

In this propensity score-matched study of patients undergoing total thyroidectomy, we identified several key findings. First, overweight patients (BMI ≥ 25 kg/m2) demonstrated a significantly lower incidence of postoperative hypocalcemia compared with non-overweight patients (54.9% vs 78.4%, P = 0.020). Second, despite the difference in hypocalcemia rates, the incidence of postoperative hypoparathyroidism was comparable between groups (62.7% vs 54.9%, P = 0.546), suggesting that surgical impact on parathyroid glands was similar. Third, after propensity score matching, while most baseline characteristics were well-balanced, overweight patients maintained distinct metabolic features, including higher TG levels (1.99 ± 1.80 vs 1.40 ± 0.83 mmol/L, P = 0.036), lower HDL-C levels (1.10 ± 0.23 vs 1.30 ± 0.34 mmol/L, P < 0.001), and higher prevalence of fatty liver intra-hepatic fat deposits (75.0% vs 35.5%, P = 0.001).

Our findings were supported by Soelling et al. [12], who analyzed data from the American College of Surgeons National Surgical Quality Improvement Program (2016–2020) and found that patients with BMI < 25 kg/m2 had a higher risk of hypocalcemia after total thyroidectomy. Additionally, increased hypocalcemia rates have been reported in patients following bariatric surgery [13]. Similar incidence of postoperative hypoparathyroidism between groups (62.7% vs 54.9%, P = 0.546), supported by a recent case-matched study by Rossi et al. [14], challenge the traditional concept that higher BMI leads to increased incidence of postoperative hppoparathyroidism. Their study demonstrated that obesity is not associated with increased postoperative hppoparathyroidism in thyroid surgery, aligning with our observation. Our matched analysis revealed persistent metabolic differences between groups, including higher TG levels, lower HDL-C, and increased prevalence of fatty liver in overweight patient. The relationship between lipid metabolism and calcium homeostasis has been previously documented [15], though its specific role in post-thyroidectomy calcium regulation remains unclear. These metabolic differences might contribute to the observed variation in postoperative calcium homeostasis between overweight and non-overweight patients, though this hypothesis needs validation through future research.

Our results indicate a more complex relationship between BMI and postoperative outcomes. The comparable rates of postoperative hypoparathyroidism between groups in our study (62.7% vs 54.9%, P = 0.546) are consistent with previous reports of transient hypoparathyroidism rates ranging from 4.6% to 51.9% [3]. In our institution, PTH levels are routinely measured in all patients 24 h after total thyroidectomy as part of our standard protocol, regardless of calcium levels or symptoms. This systematic screening approach may explain our higher detection rate of hypoparathyroidism compared to studies that only measure PTH in symptomatic patients.

The dissociation between hypoparathyroidism and hypocalcemia incidence in our study suggests that factors beyond parathyroid function influence postoperative calcium levels. While surgical trauma to the parathyroid glands remains the primary cause of post-thyroidectomy hypocalcemia [16, 17], our observations suggest a hypothesis that metabolic factors associated with BMI might play a role in modifying post-thyroidectomy calcium homeostasis. Several potential mechanisms might explain this observation. Adipose tissue serves as an important endocrine organ [18], secreting various adipokines that could influence calcium homeostasis [19]. Additionally, overweight patients typically maintain higher metabolic reserves and body fat stores [20], which might provide better buffering capacity against acute changes in calcium metabolism following surgery.

These findings have several important clinical implications. First, they suggest that BMI should be considered in the risk assessment for post-thyroidectomy hypocalcemia, with particular attention paid to non-overweight patients who might be at higher risk. Second, understanding that the risk of hypocalcemia may be influenced by both surgical and metabolic factors could help in developing more targeted preventive strategies. Third, our findings indicate that higher BMI does not necessarily increase surgical risk in thyroid surgery, at least in terms of calcium homeostasis.

Several limitations of our study should be acknowledged. First, its observational design, although strengthened by propensity score matching, cannot completely eliminate selection bias or account for unknown confounders. Second, data were collected from a single institution, potentially limiting the generalizability of our findings. Third, the substantial reduction in sample size after propensity score matching (from 228 to 102 patients) might have affected statistical power. Fourth, the lack of long-term follow-up data prevents us from assessing the durability of the observed protective effect. Additionally, while we found a statistically significant correlation between BMI and postoperative calcium levels, the correlation coefficient was relatively weak (R = 0.289), suggesting that BMI alone explains only a small portion of the variance in postoperative calcium levels.

Future research should focus on elucidating the specific mechanisms underlying the relationship between BMI and postoperative calcium homeostasis through prospective multicenter studies with larger sample sizes and longer follow-up periods. Investigation of specific adipokines and metabolic factors could provide insights for developing targeted interventions.

Conclusions

This propensity score-matched study demonstrates that overweight patients have a significantly lower risk of postoperative hypocalcemia following total thyroidectomy compared with non-overweight patients, despite similar rates of postoperative hypoparathyroidism between groups. These findings suggest that the protective effect of higher BMI operates through non-surgical mechanisms independent of direct parathyroid trauma. While the exact pathophysiological mechanisms remain to be elucidated, our results challenge traditional perspectives on BMI-related surgical risk and suggest that metabolic factors associated with body weight may play an important role in maintaining calcium homeostasis after thyroid surgery. Future experimental studies are needed to investigate the specific mechanisms underlying this protective effect, which could lead to more targeted strategies for preventing post-thyroidectomy hypocalcemia.