Abstract
Objectives
To investigate the predictive value of serum calcium and 25-hydroxyvitamin D (25-OH-VD) on recurrence after canalith repositioning procedure (CRP) in elderly patients with benign paroxysmal positional vertigo (BPPV), and to establish a quantitative prediction model.
Methods
A single-center follow-up cohort study was conducted involving 286 elderly BPPV patients (age≥60 years) treated with CRP from January 2020 to June 2025. Patients were followed for 1 year and divided into a non-recurrence group (
n
= 213) and a recurrence group (
n
= 73) based on recurrence status. Additionally, 60 healthy elderly individuals served as controls for comparing baseline biochemical levels. Fasting serum 25-OH-VD and calcium levels were measured. Pearson correlation analysis evaluated the relationship between variables. Receiver operating characteristic (ROC) curves assessed predictive performance, and multivariate logistic regression identified independent risk factors to establish a predictive model.
Results
Serum calcium (1.94 ± 0.47 mmol/L) and 25-OH-VD (15.34 ± 3.45 ng/mL) levels were significantly lower in the recurrence group compared to the non-recurrence group (2.35 ± 0.56 mmol/L, 18.44 ± 3.74 ng/mL) and controls (2.68 ± 0.48 mmol/L, 23.61 ± 3.92 ng/mL; all
p
0.05). Levels were positively correlated (
r
= 0.637,
p
OR
= 0.301, 95%
CI
: 0.152–0.593), reduced 25-OH-VD (
OR
= 0.808, 95%CI: 0.728–0.898), multiple repositioning maneuvers (
OR
= 4.945), hypertension (
OR
= 2.237), sleep disorders (
OR
= 2.113), cervical spondylosis (
OR
= 2.167), and anxiety (OR = 2.726) were independent risk factors for recurrence. The combined predictive model yielded an AUC of 0.762 (95%
CI
: 0.697–0.827), superior to single indicators (calcium AUC = 0.707; 25-OH-VD AUC = 0.717). The optimal cut-off value was 0.339 (sensitivity 56.2%, specificity 81.2%).
Conclusion
Serum calcium and reduced 25-OH-VD are independent predictive factors for post-CRP recurrence in elderly BPPV patients, and their combination improves predictive efficacy. The predictive model provides a quantitative basis for early identification of high-risk patients. However, due to its relatively low sensitivity, the model should be used as an adjunctive tool rather than a standalone test in clinical practice. These findings require further external validation in multi-center studies, and interventional effects of calcium and vitamin D supplementation should be confirmed by randomized controlled trials.