Objective  To investigate the association between the triglyceride-glucose (TyG) index and nocturnal hypertension (NH), and to evaluate its value for identifying different NH phenotypes.

Methods  This cross-sectional study included 1 492 patients who visited the Hypertension Department of Tianjin Kanghui Hospital between September 2023 and July 2025. Based on 24-hour ambulatory blood pressure monitoring and antihypertensive medication use, participants were categorized into four groups: isolated nocturnal hypertension (INH), sustained day-night hypertension, other forms of NH (including treated patients with controlled daytime but elevated nighttime blood pressure, NH extending to morning hypertension, and elderly patients with orthostatic hypotension accompanied by nocturnal supine hypertension), and normal nocturnal blood pressure. Demographic characteristics, lifestyle factors, medical history, and laboratory parameters were collected to calculate the TyG index. logistic regression was used to evaluate the independent association between the TyG index and NH as well as its subtypes, followed by interaction and stratified analyses. Restricted cubic spline models were applied to assess potential non-linear relationships. Receiver operating characteristic (ROC) curves, calibration plots, and decision curve analysis (DCA) were used to evaluate discriminatory performance, calibration, and clinical net benefit.

Results  Significant overall differences were observed among the four groups in demographic characteristics, medical history, blood pressure parameters, and metabolic indices (all P ＜ 0.05). Multivariate ordinal logistic regression showed that after adjusting for confounders, each 1-unit increase in the TyG index was independently associated with higher odds of INH (OR = 7.14, 95%CI: 4.82 to 10.59), sustained day-night hypertension (OR = 4.70, 95%CI: 3.39 to 6.53), and other NH (OR = 3.99, 95%CI: 2.88 to 5.51). Stratified analysis indicated stronger associations in females and individuals aged ≥ 65 years. Restricted cubic spline analysis revealed a linear relationship between the TyG index and INH risk (adjusted P for non-linearity = 0.755), while non-linear increasing trends were observed for other NH phenotypes. ROC analysis demonstrated that the TyG index had good discriminatory ability for all NH types, with the highest performance for INH (AUC = 0.814, 95%CI: 0.778–0.850). Calibration and decision curve analyses supported the use of the TyG index combined with traditional risk factors for NH screening and clinical risk stratification, and suggested that the TyG index alone was adequate for identifying INH (Brier score: 0.167 vs. 0.144; net benefit: 0.220 to 0.706 vs. 0.280 to 0.713 at thresholds of 0.200 to 0.700).

Conclusion The TyG index is independently and positively associated with multiple NH phenotypes, with the strongest predictive value for INH. This association is more pronounced in women and older adults and demonstrates a potential non-linear pattern across different NH types.