Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2025 Jul 30:12:1612403.
doi: 10.3389/fmed.2025.1612403. eCollection 2025.

Development of a clinical nomogram for predicting sarcopenia in patients with chronic obstructive pulmonary disease based on NHANES data

Xingfu Fan #  1 , Jin Zhao #  1 , Yang Luo  1 , Xiaofang Li  1 , Wenqin Tan  1 , Shiping Liu  1
Affiliations

Development of a clinical nomogram for predicting sarcopenia in patients with chronic obstructive pulmonary disease based on NHANES data

Xingfu Fan et al. Front Med (Lausanne). .

Abstract

Background: The prevalence of sarcopenia in COPD patients is high, and the mutual influence between COPD and sarcopenia creates a vicious cycle. The goal of this research is to create a nomogram model that can forecast when sarcopenia will strike people with COPD.

Methods: 2011 to 2018 data were retrieved from four NHANES database cycles. The 7:3 proportion was applied to split the dataset randomly to separate validation and training datasets. Multivariate logistical regression and LASSO regression were applied to design nomogram design and to select predictors. In addition, multicollinearity existence among final predictor variables remaining in model were tested, among other variables. Calibration curve, decision curve analysis (DCA), and area under receiver operating characteristic curve (AUC) were applied in testing performance in prediction model.

Results: The nomogram was constructed based on four predictive factors: gender, height, BMI, and WWI. The AUC for the training set was 0.94 (95% CI 0.91-0.97), and the AUC for the validation set was 0.91 (95% CI 0.83-0.98), indicating excellent predictive performance. Furthermore, the clinical applicability of the model has been thoroughly validated.

Conclusion: We established a nomogram model to provide an easy and convenient way for early screening of sarcopenia in COPD patients, and to allow for effective guidance to perform early intervention and manage patient prognosis in an optimal way.

Keywords: COPD; NHANES; nomogram; risk factors; sarcopenia.

PubMed Disclaimer

Conflict of interest statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Figures

Flowchart outlining a study design using NHANES data from 2011-2018, starting with 39,156 participants. After exclusions based on criteria such as age under 20, absence of COPD, chronic bronchitis, or emphysema, and missing data, 555 participants were enrolled. They were divided into a training group (388) and validation group (167). LASSO and logistic regression analyses were conducted, leading to a nomogram diagram creation. This was followed by receiver operating characteristic, calibration curve, and decision curve analysis, culminating in model evaluation.
FIGURE 1
The flow chart of study population screening and statistical analysis.
A line plot depicting the relationship between coefficients and the logarithm of lambda values. Multiple colored lines represent different coefficients converging towards zero as lambda increases. Vertical lines indicate \(\lambda_{min}\) in blue and \(\lambda_{1se}\) in red.
FIGURE 2
LASSO regression model cross-validation plot.
Plot showing binomial deviance versus log lambda. Red dots represent data points, with deviance decreasing and then increasing as log lambda moves from -10 to -2. Gray error bars are present for each point.
FIGURE 3
Coeffïcient profïle plot of predictors.
Diagram displaying a nomogram with scales for Points, Gender, Height, BMI, WWI, Total Points, Linear Predictor, and Risk. Points range from 0 to 100. Gender is marked as 1 and 2. Height ranges from 140 to 190 centimeters. BMI from 15 to 65. WWI from 8.5 to 14. Total Points from 0 to 180. Linear Predictor from -16 to 4. Risk ranges from 0.1 to 0.9.
FIGURE 4
The nomogram of patients with COPD complicated by sarcopenia. Gender: 1 represents male, 2 represents female.
Correlation heatmap showing relationships among gender, height, BMI, and weight-adjusted waist index (WWI). Gender and height negatively correlate at -0.64. BMI and WWI have a positive correlation of 0.57. Color gradient legend ranges from -1 (light pink) to 1 (dark red).
FIGURE 5
Spearman correlation heatmap among gender, height, BMI, and WWI.
Two ROC curve graphs labeled A and B. Graph A, in orange, shows a high curve with an Area Under the Curve (AUC) of 0.94, with a 95% confidence interval (CI) of 0.91 to 0.97. Graph B, in blue, also depicts a high curve, with an AUC of 0.91 and a 95% CI of 0.83 to 0.98. Both graphs plot sensitivity against 1-specificity.
FIGURE 6
Receiver operating characteristic curves of nomogram based on the data of the training group (A) and validation group (B).
Two calibration plots labeled A and B compare predicted probabilities to actual probabilities. Both plots display three lines: apparent, bias-corrected, and ideal. Plot A has a Hosmer-Lemeshow p-value of 0.431, showing close alignment of apparent and bias-corrected lines with the ideal line. Plot B presents a p-value of 0.806, with lines also closely matching the ideal. Axes are labeled with predicted probabilities on the x-axis and actual probabilities on the y-axis.
FIGURE 7
Calibration curve of the nomogram based on the data of training group (A) and validation group (B).
Two graphs labeled A and B show net benefit against high-risk threshold. Both graphs include three lines for ‘Model’, ‘All’, and ‘None’. The lines generally decrease from left to right, with some fluctuations. Graph A exhibits sharper declines and more variance compared to Graph B. Legends are in the top right corners of each graph.
FIGURE 8
Decision curve analysis curve of the nomogram based on the data of training group (A) and validation group (B).
See this image and copyright information in PMC

References

    1. Rabe K, Watz H. Chronic obstructive pulmonary disease. Lancet. (2017) 389:1931–40. 10.1016/S0140-6736(17)31222-9 - DOI - PubMed
    1. GBD 2021 Causes of Death Collaborators. Global burden of 288 causes of death and life expectancy decomposition in 204 countries and territories and 811 subnational locations, 1990-2021: a systematic analysis for the Global Burden of Disease Study 2021. Lancet. (2024) 403:2100–32. 10.1016/S0140-6736(24)00367-2 - DOI - PMC - PubMed
    1. Boers E, Barrett M, Su J, Benjafield A, Sinha S, Kaye L, et al. Global burden of chronic obstructive pulmonary disease through 2050. JAMA Netw Open. (2023) 6:e2346598. 10.1001/jamanetworkopen.2023.46598 - DOI - PMC - PubMed
    1. Byun M, Cho E, Chang J, Ahn C, Kim H. Sarcopenia correlates with systemic inflammation in COPD. Int J Chron Obstruct Pulmon Dis. (2017) 12:669–75. 10.2147/COPD.S130790 - DOI - PMC - PubMed
    1. Jones S, Maddocks M, Kon S, Canavan J, Nolan C, Clark A, et al. Sarcopenia in COPD: prevalence, clinical correlates and response to pulmonary rehabilitation. Thorax. (2015) 70:213–8. 10.1136/thoraxjnl-2014-206440 - DOI - PubMed

LinkOut - more resources