Relationship between triglyceride-glucose index and new-onset hypertension in general population–a systemic review and meta-analysis of cohort studies

Figures & data

Figure 1. Flowchart of the study selection process and results.

Table 1. Baseline characteristics of the included studies.

Study ID	Participant ethnicity	Sample size	follow-up duration (years)	age (years)	Male (%)	BMI (kg/m^2)		Blood samples	TyG analysis	Adjusted variables	NOS
Gao (2023)	China	4600	6	48.1	44.7%	22.9		Fasting for 8-12 h	Q4:Q1 Q4 > 9.0 versus Q1 ≤ 8.1	Sex, age, BMI, WHR, SBP, DBP, smoking, drinking, region, urban resistance, marital status, education, occupation, dietary intake of fat, protein, carbohydrate, diabetes mellitus, eGFR, BUN, uric acid, hsCRP, hemoglobin, total protein, LDL-C, TC, HbA1c, and FBI.	9
Laura- men (2016)	Spain	2189	8.38	51.4	100%	27.2		Fasting for 8 h	Q5:Q1 8.89≤Q5 < 10.64 versus 6.40≤Q1 < 7.97	Age, sex, BMI, cigarette smoking, daily alcohol intake, lifestyle pattern, type 2 diabetes, anti-aggregation therapy, LDL-C, SBP, DBP.	8
Laura- women (2016)	Spain	1448	8.66	52.8	0%	24.9		Fasting for 8 h	Q5:Q1 8.48≤Q5 < 9.99 versus 6.43≤Q1 < 7.68	Age, sex, BMI, cigarette smoking, daily alcohol intake, lifestyle pattern, type 2 diabetes, anti-aggregation therapy, LDL-C, SBP, DBP.	8
Zhang (2023)	China	4678	3	35.8	25.3%	20.9		Fasting for over 12 h	Q5:Q1 Q5 ≥ 8.54 versus Q1 < 7.77	Age, sex, BMI, ALT, AST, TC, LDL-C, HDL-C	9
Zhao (2023)	China	10309	6	48	39.4%	23.5		Fasting for overnight	Q4:Q1 Q5 = 9.33(9.15–9.66) versus Q1 = 8.01(7.83–8.14)	Gender, age, marital status, educational level, monthly income, smoking, alcohol drinking, physical activity, family history of hypertension、total cholesterol, high-density lipoprotein cholesterol, body mass index, waist circumference.	9
Zheng (2017)	China	4686	9	40.5	67.8%	22.45		In the morning	Q4:Q1 Q4 = 9.04(8.88–9.30) versus Q1 = 7.81(7.65–7.97)	Age, gender, BMI, WC, BUN, Cr, FPG, UA, AST, ALT, γ-GGT, TC, TG, HDL-C, LDL-C, Apo-A1, Apo-B, and eGFR.	9
Liu (2023)	China	4755	5.22	57.2	46%	22.9		Fasting for overnight	Q4:Q1 Q4 = 9.29 ± 0.25 versus Q1 = 7.90 ± 0.21	Smoke, drink, heart rate, marital status, gender, height, platelets, total cholesterol, LDL-C, HbA1c, and hemoglobin.	9
Khoo (2023)	Singapore	3183	1.7	55.9	32.5%		23	Fasting for overnight	Q4:Q1 Q4 = 9.0(8.9–9.3) versus Q1 = 7.8(7.7–8.0)	Age, gender, ethnicity, BMI, LDL-C, and HDL-C.	9

Figure 2. Forest plots for the meta-analysis of the association between TyG index and new-onset hypertension in general population.

Figure 3. Funnel plots for the meta-analysis of the TyG index and risk of new-onset hypertension.

Figure 4. New funnel plot after adding three simulated missing studies.

Figure 5. Subgroup analysis for the association between TyG index and new-onset hypertension (a. Subgroup analysis according to the age of the participants b. Subgroup analysis according to the BMI of the participants c. Subgroup analysis according to the follow-up duration).

Figure 6. Subgroup analysis for the association between TyG index and new-onset hypertension (D. Subgroup analysis according to the ethnicity of the participants E. Subgroup analysis according to the sex of the participants).

Table 2. Sensitivity analysis of the association between TyG index and new-onset hypertension.

Dataset excluded	HR	95%CI	I^2	Fixed or random	P for effect
Gao (2023)	1.52	1.22–1.89	72%	Random	0.0002
Khoo (2023)	1.40	1.21–1.63	54%	Random	<0.0001
Liu (2023)	1.56	1.26–1.94	72%	Random	<0.0001
Zheng (2017)	1.51	1.23–1.84	72%	Random	<0.0001
Zhao (2023)	1.53	1.37–1.71	47%	Fixed	<0.0001
Zhang (2023)	1.49	1.23–1.80	71%	Random	<0.0001
Laura-women (2016)	1.53	1.26–1.86	73%	Random	<0.0001
Laura-men (2016)	1.46	1.21–1.76	68%	Random	<0.0001