📄 medicine

Associations of nine C-reactive protein-triglyceride-glucose index- derived adiposity phenotypes with incident cardiovascular-liver- metabolic multimorbidity: a prospective cohort study

This prospective cohort study of 6,534 Chinese adults demonstrates that nine C-reactive protein-triglyceride-glucose index (CTI)-derived adiposity phenotypes are significantly associated with the risk of incident cardiovascular-liver-metabolic multimorbidity, with the CTI-BMI index showing the strongest and most robust predictive performance.

Original authors: Jing Chen, Zhonghua Sun, Jiajun Qiu, Zhongqing Zhou, Kaiqi Ma, Yongxin Liang, Chunqun Li, Shicong Liang

Published 2026-07-03

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Jing Chen, Zhonghua Sun, Jiajun Qiu, Zhongqing Zhou, Kaiqi Ma, Yongxin Liang, Chunqun Li, Shicong Liang

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The Big Picture: A "Triple Threat" Warning System

Imagine your body as a complex city. Usually, problems happen in one district at a time: maybe the Heart District (Cardiovascular disease), the Sugar Factory (Type 2 Diabetes), or the Liver Warehouse (Fatty Liver).

This study looked at people who started having problems in at least two of these districts at the same time. The researchers call this "CLMM" (Cardiovascular-Liver-Metabolic Multimorbidity). It's like a city-wide power grid failure rather than just a flickering light in one room. It's a more serious, complicated state that is harder to fix.

The researchers wanted to know: Is there a single "smoke alarm" that can predict when this city-wide failure is about to happen?

The Tools: Building a Better Smoke Alarm

For a long time, doctors have used two separate tools to check for trouble:

The Inflammation Meter: Measuring C-reactive protein (CRP), which is like checking for smoke or fire in the city.
The Sugar-Fat Meter: Measuring the Triglyceride-Glucose index (TyG), which checks if the sugar and fat delivery trucks are stuck in traffic (Insulin Resistance).

The researchers created a new, super-charged tool called CTI (C-reactive protein-triglyceride-glucose index). Think of CTI as a hybrid smoke-and-traffic sensor that combines both readings into one number.

But they didn't stop there. They knew that obesity (extra weight) acts like a heavy layer of insulation that traps the heat and makes the fire worse. So, they took their hybrid sensor (CTI) and wrapped it around nine different ways of measuring body fat (like BMI, waist size, and body shape).

This created nine new "Super-Sensors" (like CTI-BMI, CTI-Waist, etc.). The goal was to see which of these nine sensors was the best at predicting the "Triple Threat" (CLMM) before it happened.

The Experiment: A 4-Year Watch

The researchers used data from 6,534 Chinese adults (mostly middle-aged and older) from a large national study called CHARLS.

The Start (2011): They measured everyone's blood and body shape.
The Wait: They waited about 4 years.
The Result: During that time, 406 people (about 6%) developed the "Triple Threat" (CLMM).

They then ran a massive comparison to see which of the nine "Super-Sensors" was the best predictor.

The Findings: Who Won the Race?

Here is what the data revealed, translated into everyday terms:

1. The "All-Rounder" Champion: CTI-BMI
The winner was CTI-BMI.

The Analogy: Imagine you are trying to predict a storm. Some sensors only look at the wind (inflammation), and some only look at the clouds (fat). CTI-BMI is like a sensor that looks at the wind and the clouds and how heavy the clouds are.
The Result: People with the highest CTI-BMI scores were 3.5 times more likely to develop the Triple Threat compared to those with low scores. It was the most accurate predictor of all nine tools.

2. The Runner-Up: CTI-CVAI
This was the second-best sensor. It's a bit more complex (it calculates "visceral" or belly fat specifically), but it was still very good at spotting the risk.

3. The "One-Size-Fits-All" Failure
Interestingly, the original sensor (just CTI, without the body fat added) was okay, but not great. It was like a smoke alarm that works, but misses the bigger picture of how much fuel is available to burn.

4. The "Shape" Sensors
Some sensors that focused purely on body shape (like how round you are or your waist-to-height ratio) didn't add much extra value once the inflammation and sugar levels were already known. They were like checking the shape of the roof when you already know the house is on fire.

The "Sweet Spot" and The "Tipping Point"

The researchers found that the relationship isn't always a straight line.

For CTI-BMI: The risk goes up steadily as the score goes up, until it hits a certain "tipping point." After that point, the risk stays high but doesn't necessarily shoot up even faster. It's like a car accelerating: once you hit a certain speed, you are already in the danger zone, and going a bit faster doesn't change the fact that you are speeding dangerously.
For other sensors: Some showed weird "U-shapes" or "J-shapes," meaning the risk went down, then up, or vice versa. This suggests those sensors might be confusing or less reliable.

The "Double Trouble" Effect

The study also looked at what happens if you have High Inflammation AND High Body Fat at the same time.

The Analogy: If you have a small fire (inflammation) and you throw a bucket of gasoline on it (obesity), the fire explodes.
The Result: People who had both high CTI and high BMI had the highest risk of all. However, the study found that they didn't necessarily "explode" in a mathematically unique way (no special interaction); rather, the obesity just added to the existing risk, making the total danger very high.

Who is Most at Risk?

The "Super-Sensor" (CTI-BMI) worked well for everyone, but it was extra sensitive for specific groups:

Men
People who smoke
People who drink alcohol

For these groups, the link between the sensor reading and the disease was even stronger.

The Bottom Line

This study suggests that to predict who is likely to get a mix of heart, liver, and diabetes problems, we shouldn't just look at blood sugar or weight alone.

The best approach is to use a combined score that looks at Inflammation + Sugar + Overall Body Weight (BMI). The study found that CTI-BMI is the most reliable "early warning system" for spotting people who are on the path to this complex, multi-disease state.

In short: If you want to know if the city is about to have a grid failure, don't just check the smoke or just check the traffic. Check the Smoke-Traffic-Weight combo. That tells you the whole story.

Technical Summary: Associations of Nine C-reactive protein-triglyceride-glucose index-derived adiposity phenotypes with incident cardiovascular-liver-metabolic multimorbidity

Problem Statement
Cardiovascular-liver-metabolic multimorbidity (CLMM), defined as the coexistence of at least two conditions among cardiovascular disease (CVD), type 2 diabetes mellitus (T2DM), and metabolic dysfunction–associated steatotic liver disease (MASLD), represents a severe state of systemic metabolic dysregulation. While insulin resistance (IR) and chronic low-grade inflammation are known pathophysiological drivers, and obesity is a major contributor, the specific utility of composite biomarkers that integrate these domains for predicting CLMM remains unclear. Specifically, the C-reactive protein-triglyceride-glucose index (CTI), which combines inflammatory (CRP) and metabolic (TyG) markers, has been proposed as a proxy for "inflammation–metabolic burden," but its association with CLMM and the added value of integrating CTI with various obesity phenotypes (e.g., BMI, waist circumference, visceral adiposity indices) have not been systematically evaluated.

Methodology
This prospective cohort study utilized data from the China Health and Retirement Longitudinal Study (CHARLS).

Study Population: The analysis included 6,534 participants (median age 58, 50.6% female) with complete baseline data from 2011–2012. Participants with prevalent CLMM or missing exposure/outcome data were excluded. The follow-up endpoint was set at 2015 (median follow-up 4.0 years).
Exposure Variables: Nine indices were calculated:
1. CTI: $0.412 \times \ln(CRP) + \ln(TG) \times FBG / 2$
2. CTI-Derived Adiposity Indices: CTI multiplied by various anthropometric measures, including CTI-BMI, CTI-WC (waist circumference), CTI-WHtR (waist-to-height ratio), CTI-BRI (body roundness index), CTI-WWI (weight-adjusted waist index), CTI-CVAI (Chinese visceral adiposity index), CTI-ABSI (a body shape index), and CTI-CI (conicity index).
Outcome: Incident CLMM, defined as the development of at least two of the three conditions (CVD, T2DM, MASLD) during follow-up. MASLD was inferred using the Lipid Accumulation Product (LAP) combined with cardiometabolic risk factors due to the lack of imaging in CHARLS.
Statistical Analysis:
- Multivariable logistic regression (crude, Model 1, and fully adjusted Model 2) was used to calculate Odds Ratios (ORs) across quartiles.
- Restricted Cubic Spline (RCS) modeling and two-piecewise logistic regression were employed to detect non-linear exposure-response relationships and inflection points.
- Predictive performance was assessed via Receiver Operating Characteristic (ROC) curves (AUC), Net Reclassification Improvement (NRI), and Integrated Discrimination Improvement (IDI).
- Joint effect and interaction analyses (multiplicative and additive) were conducted for CTI and BMI.
- Subgroup and sensitivity analyses were performed to assess stability.

Key Results

Incidence: During the follow-up, 406 participants (6.2%) developed CLMM.
Associations: In fully adjusted models, higher quartiles of all CTI-derived adiposity indices were generally associated with increased CLMM risk.
- Strongest Predictors: CTI-BMI and CTI-CVAI demonstrated the strongest associations. For CTI-BMI, the highest quartile (Q4) had an OR of 3.46 (95% CI: 2.42–4.97) compared to Q1. For CTI-CVAI, the Q4 OR was 2.37 (95% CI: 1.66–3.36).
- Weakest Predictors: CTI-ABSI and CTI-WWI showed null or non-significant associations after full adjustment.
Non-linearity: RCS analyses revealed significant non-linear exposure-response relationships for all indices except the base CTI.
- CTI-BMI showed a positive association below an inflection point of 208.698 (OR = 1.022), with the association becoming null above this threshold.
- Other indices (e.g., CTI-WC, CTI-WHtR) exhibited complex patterns, including inverse associations below inflection points followed by weak positive trends.
Predictive Performance:
- CTI-BMI achieved the highest Area Under the Curve (AUC = 0.638), significantly outperforming all other indices (all $P < 0.001$ ).
- Adding CTI-BMI to a baseline model significantly improved reclassification (continuous NRI = 28.63%) and discrimination (IDI = 0.03%).
Joint Effects: Participants with both high CTI and high BMI had the highest risk of incident CLMM (OR = 2.53). However, no statistically significant multiplicative or additive interactions were observed between CTI and BMI, suggesting their effects may be largely additive rather than synergistic.
Effect Modification: The association between CTI-BMI and CLMM was significantly stronger in men, ever-smokers, and ever-drinkers ( $P$ for interaction $< 0.001$ ).

Significance and Claims
The authors claim that this study establishes a multidimensional framework integrating inflammation, metabolism, and adiposity to assess CLMM risk.

Primary Contribution: The study identifies CTI-BMI as the most stable and robust predictor among CTI-derived obesity indices for incident CLMM in middle-aged and older Chinese adults.
Clinical Implication: The findings suggest that for risk stratification of CLMM, clinical assessments should consider both the metabolic-inflammatory burden (captured by CTI) and overall obesity phenotypes (captured by BMI), rather than relying on single-domain markers.
Mechanistic Insight: The results support the hypothesis that the "inflammation–metabolism–obesity" axis plays a critical role in the development of multimorbidity, with CTI-BMI serving as a concise surrogate for this coupled state.
Limitations: The authors acknowledge limitations including the observational nature of the study (precluding causal inference), potential residual confounding, reliance on self-reported outcomes and surrogate markers for MASLD, and the restriction of the cohort to middle-aged and older Chinese adults, which may limit generalizability to younger or other ethnic populations.

The paper concludes that while all CTI-derived indices show associations with CLMM, CTI-BMI offers the most reliable representation of the combined metabolic and obesity burden linked to this multimorbidity spectrum.