Association between C-reactive protein and sarcopenia: The National Health and Nutrition Examination Survey Authors: Yang Li, Zhi-Wen Zhang Published: 2024 ABSTRACT C-reactive protein (CRP), a clinical biomarker, is frequently used to evaluate the inflammatory status of the body. However, the association between CRP levels and sarcopenia among the general adult population in the USA is unknown. This study focused on assessing whether CRP levels were associated with sarcopenia. This cross-sectional study collected adult data of adults from the 2015 to 2018 National Health and Nutrition Examination Survey. Four-extremity total muscle mass was used to evaluate sarcopenia (appendicular lean mass). In addition, dual-energy X-ray absorptiometry was adopted to measure appendicular lean mass. The CRP levels were used to assess inflammation status. Odds ratios (OR) and 95% confidence intervals (95% CI) were determined using multivariable logistic regression to analyze the association between CRP levels and sarcopenia. A multivariable-adjusted restricted cubic spline model was built to plot OR curves at 4 knots. Among the 3710 participants involved in this study (average age, 39.4 [11.54] years; 1801 [48.5%] men), 352 (9.5%) displayed characteristics of sarcopenia, while 3358 (90.5%) did not. Compared with participates in the lowest quartile (Q1) of CRP level (Q1; ≤0.08 to ≤0.7), those in the highest quartile (Q4; ≤4.3 to ≤188.5) had an adjusted OR for sarcopenia of 2.74 (95% CI, 1.65–4.57; P < .001). Based on the multivariable restricted cubic spline model, CRP levels showed a nonlinear association with sarcopenia (P < .001). The adjusted OR of sarcopenia of 1.86 (95% CI, 1.37–2.51; P < .001) was determined by 2 piecewise regression models for those having the CRP level of 1.8. Based on subgroup analysis, CRP levels were related to sarcopenia in males (OR, 1.03; 95% CI, 1–1.05) and individuals aged <50 years (OR, 1.03; 95% CI, 1.01–1.05), drinking (OR, 1.02; 95% CI, 1–1.03), and body mass index ≥ 25 kg/m2 (OR, 1.02; 95% CI, 1–1.03). Our results indicated that CRP levels showed a nonlinear correlation with sarcopenia among adults in the US. 1. Introduction Sarcopenia, a Greek phrase implying fresh poverty, was initially depicted in 1989 by Rosenberg. Sarcopenia is an aging-related loss of skeletal muscle mass and strength.[1,2] It is related to multiple pathophysiological events, including mitochondrial dysfunction, denervation, and hormonal and inflammatory alterations that negatively affect health, such as decreased function, falls, mortality, and frailty due to reduced lean body mass.[3] Approximately 10% of individuals over 65 years of age are affected by the condition, including up to 30% of men over 80 years.[4–6] This phenomenon will be complicated by the increase in the >60-year-old population, which may reach approximately 2 billion by 2050. Although the pathogenesis of sarcopenia remains unknown, inflammation may have a critical effect. Inflammation is also associated with sarcopenia.[4–6] With old age, the body may experience a phenomenon known as “inflammatory aging,” which is characterized by a persistent low-level inflammatory response. In this condition, the body produces excessive amounts of pro-inflammatory cytokines such as interleukin-6 (IL-6) and tumor necrosis factor-alpha (TNF-α). The pro-inflammatory cytokines are thought to promote muscle breakdown and inhibit muscle protein synthesis, which accelerates the loss of muscle mass. However, findings from previous studies on the relationship between C-reactive protein (CRP) levels and sarcopenia are limited. CRP has been increasingly suggested as an acute inflammatory protein with a critical effect on infection responses and inflammatory processes in the host, such as apoptosis, the complement pathway, phagocytosis, nitric oxide production, and cytokine generation, especially TNF-α and IL-6.[7–11] CRP activates many types of immune cells, including macrophages and dendritic cells, making them more active and involved in the inflammatory process, which helps the body to respond quickly to potential threats. CRP can also lead to an excessive inflammatory response. Understanding how CRP specifically affects these processes is important for the development of novel therapeutic approaches to aging-related diseases. However, the association between CRP level and sarcopenia needs to be further investigated. To bridge the existing knowledge gap, the association of CRP with sarcopenia was assessed based on the National Health and Nutrition Examination Survey (NHANES) data. Therefore, we investigated whether CRP level was independently related to sarcopenia in adults from the 2015 to 2018 NHANES. 2. Materials and methods 2.1. Data source The NHANES database is a cross-sectional, population-based survey that collects information on the health and nutrition of US household populations.[5] In the current cross-sectional study, data were extracted from adult participants of the 2015 to 2018 NHANES cycles. These cycles were chosen because sarcopenia was only obtained from the Miscellaneous Pain Questionnaire. Data were collected between October 2023 and November 2023. There were 19,225 participants in the 2015 to 2018 NHANES cycles. This study excluded participants if the status of these participants was any of the following: (1) Their diet data were unavailable (n = 4185); (2) The data about sarcopenia were missing (n = 8009); (3) The data about covariates were missing (n = 3321). Finally, 3710 participants were included, of which 3358 participants without sarcopenia and 352 participants with sarcopenia. 2.2. Standard protocol approval, registration, and patient consent In December 1998, the NHANES was approved by the NCHS Ethics Review Committee. Informed consent was obtained from all participants before enrollment. No additional Institutional Review Board approval was required to perform secondary analysis. 2.3. Study design and population Participants aged over 20 years who completed the survey interviews were included in this study. Those with insufficient information regarding sarcopenia or CRP levels, those in pregnancy, and those without data for calculating CRP levels and covariates were eliminated. The exclusion criteria included those with insufficient information concerning sarcopenia or CRP, those who were pregnant, and those with missing data needed to calculate CRP and covariates. 2.4. Sarcopenia and CRP assessment We evaluated sarcopenia using the total 4-extremity muscle mass (appendicular lean mass) with dual-energy X-ray absorptiometry using NHANES.[12] Sarcopenia was defined as follows: The sarcopenia index was employed to define sarcopenia, with values of <0.789 and 0.512 indicating sarcopenia in male and female, respectively. Baseline CRP levels were determined and used as a continuous variable. The eventual outcome variable (dichotomous variable) was obtained based on current studies and guidelines. The covariates included age, sex, race, marital status, education, drinking, smoking, hypertension, diabetes, body mass index (BMI), and CRP levels. The race was divided into Mexican American, other Hispanics, non-Hispanic White, non-Hispanic Black, and other race-including multiracial. Marital status was classified as married, widowed, divorced, separated, never-married, or living with a partner. Education level was classified as <9 years, 9 to 12 years, high school graduate, some college or AA degree, and college graduate or above. In line with the predetermined literature definitions, drinking status was classified as never drinking or drinking. Underlying diseases (diabetes and hypertension) were defined in accordance with the questionnaire inquiries regarding whether the physician knew the condition previously. BMI was determined using a standardized approach based on height and weight. Latex-enhanced nephelometry was used to determine the CRP levels. 2.5. Statistical analysis All estimates were calculated by accounting for NHANES sample weights. Categorical variables are shown as proportions (%), while continuous variables are presented as means (standard deviation) or medians (interquartile range). To compare intergroup differences, chi-square tests (categorical variables), one-way analysis of variance (normal distribution), and Kruskal–Wallis tests (skewed distribution) were conducted. To examine the correlation between sarcopenia and CRP levels, odds ratios (ORs) and 95% confidence intervals (95% CIs) were calculated. In Model 1, sociodemographic features, including age, sex, race, marital status, education, and drinking status were adjusted. Sociodemographic features and previous diseases were adjusted for in model 2. In addition, age, sex, marital status, race, education, drinking, smoking, diabetes, hypertension, and body mass index (BMI) were fully adjusted in Model 3. The smoothing 2-piecewise logistic regression model was adopted to analyze the association threshold of sarcopenia with CRP in Model 3. In addition, inflection points were examined using the bootstrap resampling approach and the likelihood-ratio test. We evaluated possible modifications for the association between sarcopenia and CRP, such as sex, age (<50 vs ≥50 years), drinking, and BMI (<25 vs =25 kg/m2). Multivariate logistic regression was used to evaluate inter-subgroup heterogeneity, while likelihood-ratio testing was performed to analyze subgroup-CRP interactions. Free Statistics software version 1.9 and statistical software package R 3.3.2 (http://www.R-project.org) were used for statistical analysis. A 2-sided test was made for this descriptive research on every subject, and P < .05 represented statistical significance. 3. Results 3.1. Study population A total of 19,225 participants completed the interview, including 15,040 participants with diet data. We excluded patients with no available information on sarcopenia (n = 8009) and covariates (n = 3321). Finally, 3710 participants were recruited from the 2015 to 2018 NHANES survey in the current cross-sectional study for analysis. Figure 1 presents more details regarding patient eligibility criteria. Inclusion and exclusion process was based on the 2015 to 2018 National Health and Nutrition Examination Survey. 3.2. Baseline features Table 1 shows the baseline features of the participants based on the corresponding CRP quartiles. A total of 352 (9.5%) participants had sarcopenia. The mean age of all participants was (39.4 ± 11.5) years, including 1909 (51.5%) women. Those with higher CRP levels were usually older, female, married, non-Hispanic White, some college or AA degree, drinkers, a decreased risk of diabetes and hypertension, and obese. Table 1Characteristics of participants by quartiles of the CRP level in the NHANES 2015 to 2018 cycles.VariablesTotal (n = 3710)Q1 (≤0.08 to ≤0.7)Q2 (≤0.71 to ≤1.79)Q3 (≤1.8 to ≤4.29)Q4 (≤4.3 to ≤188.5)P-valueGender<.001 Male1801 (48.5)514 (55.6)498 (55.3)468 (49.3)321 (34.3) Female1909 (51.5)410 (44.4)402 (44.7)482 (50.7)615 (65.7)Age39.4 ± 11.536.8 ± 11.539.6 ± 11.340.5 ± 11.340.8 ± 11.3<.001Race<.001 Mexican American648 (17.5)106 (11.5)178 (19.8)193 (20.3)171 (18.3) Other Hispanic414 (11.2)87 (9.4)97 (10.8)112 (11.8)118 (12.6) Non-Hispanic White1231 (33.2)298 (32.3)292 (32.4)309 (32.5)332 (35.5) Non-Hispanic Black737 (19.9)185 (20)153 (17)192 (20.2)207 (22.1) Other race-including multiracial680 (18.3)248 (26.8)180 (20)144 (15.2)108 (11.5)Education level<.001 <9229 (6.2)44 (4.8)61 (6.8)69 (7.3)55 (5.9) 9–12374 (10.1)74 (8)89 (9.9)101 (10.6)110 (11.8) High school graduate850 (22.9)188 (20.3)204 (22.7)227 (23.9)231 (24.7) Some college or AA degree1255 (33.8)301 (32.6)289 (32.1)328 (34.5)337 (36) College graduate or above1002 (27.0)317 (34.3)257 (28.6)225 (23.7)203 (21.7)Marital status<.001 Married1776 (47.9)416 (45)449 (49.9)449 (47.3)462 (49.4) Widowed54 (1.5)7 (0.8)12 (1.3)19 (2)16 (1.7) Divorced341 (9.2)61 (6.6)78 (8.7)106 (11.2)96 (10.3) Separated131 (3.5)24 (2.6)30 (3.3)34 (3.6)43 (4.6) Never-married925 (24.9)282 (30.5)232 (25.8)214 (22.5)197 (21) Living with partner483 (13.0)134 (14.5)99 (11)128 (13.5)122 (13)Drinker<.001 No1074 (28.9)246 (26.6)233 (25.9)273 (28.7)322 (34.4) Yes2636 (71.1)678 (73.4)667 (74.1)677 (71.3)614 (65.6)BMI29.2 ± 6.924.5 ± 4.227.4 ± 4.930.8 ± 6.234.0 ± 7.7<.001Hypertension<.001 No2960 (79.8)791 (85.6)725 (80.6)745 (78.4)699 (74.7) Yes750 (20.2)133 (14.4)175 (19.4)205 (21.6)237 (25.3)Diabetes<.001 No3294 (88.8)880 (95.2)837 (93)824 (86.7)753 (80.4) Yes416 (11.2)44 (4.8)63 (7)126 (13.3)183 (19.6)Sarcopenia<.001 No3358 (90.5)901 (97.5)842 (93.6)834 (87.8)781 (83.4) Yes352 (9.5)23 (2.5)58 (6.4)116 (12.2)155 (16.6)Mean + SD for continuous variables: P value was calculated by weighted linear regression model.Percentage for categorical variables: P value was calculated by weighted chi-square test.CRP = C-reactive protein, NHANES = National Health and Nutrition Examination Survey, SD = standard deviation. Characteristics of participants by quartiles of the CRP level in the NHANES 2015 to 2018 cycles. Mean + SD for continuous variables: P value was calculated by weighted linear regression model. Percentage for categorical variables: P value was calculated by weighted chi-square test. CRP = C-reactive protein, NHANES = National Health and Nutrition Examination Survey, SD = standard deviation. 3.3. Association between sarcopenia and CRP As shown in Table 2, high CRP levels were correlated with a higher prevalence of sarcopenia when age, sex, marital status, race, educational level, BMI, hypertension, and diabetes were adjusted. Correlation was maintained if the CRP levels were converted to categorical variables as quartiles. Compared with participates with quartile1 (Q1) of CRP (≤0.08 to ≤0.7), those with Q3 (≤1.8 to ≤4.29) and Q4 (≤4.3 to ≤188.5) had the adjusted ORs for sarcopenia of 2.43 (95% CI; 1.49–3.98; P < .001) and 2.74 (95% CI; 1.65–4.57; P < .001). Table 2Association between CRP level and sarcopenia among adult participants in the NHANES 2015 to 2018 cycles.NoCrude OR (95% CI)P-valueModel 1 OR (95% CI)P-valueModel 2 OR (95% CI)P-valueModel 3 OR (95% CI)P-valueQ19241 (Ref)1 (Ref)1 (Ref)1 (Ref)Q29002.7 (1.65–4.41)<.0012.17 (1.31–3.59).0032.12 (1.28–3.51).0031.6 (0.96–2.68).074Q39505.45 (3.45–8.61)<.0014.64 (2.9–7.42)<.0014.44 (2.77–7.11)<.0012.43 (1.49–3.98)<.001Q49367.77 (4.97–12.17)<.0017.57 (4.75–12.08)<.0016.96 (4.35–11.14)<.0012.74 (1.65–4.57)<.001Trend test37101.84 (1.65–2.06)<.0011.92 (1.69–2.17)<.0011.86 (1.64–2.11)<.0011.36 (1.18–1.57)<.001Model 1: Sociodemographic features, including age, gender, race, marital status, education, and drinking status were adjusted. Model 2: Sociodemographic features and previous disease were adjusted. Model 3: Age, gender, marital status, race, education, drinking, smoking, diabetes, hypertension, and body mass index (BMI) were fully adjusted.CI = confidence interval, NHANES = National Health and Nutrition Examination Survey, OR = odds ratio, Ref = reference. Association between CRP level and sarcopenia among adult participants in the NHANES 2015 to 2018 cycles. Model 1: Sociodemographic features, including age, gender, race, marital status, education, and drinking status were adjusted. Model 2: Sociodemographic features and previous disease were adjusted. Model 3: Age, gender, marital status, race, education, drinking, smoking, diabetes, hypertension, and body mass index (BMI) were fully adjusted. CI = confidence interval, NHANES = National Health and Nutrition Examination Survey, OR = odds ratio, Ref = reference. Therefore, sarcopenia exhibited a nonlinear relation with CRP levels (P < .001) using the restricted cubic spline model (Fig. 2). Additionally, the adjusted OR of sarcopenia of 1.86 (95% CI, 1.37–2.51; P < .001) was obtained for participates with CRP levels ≥ 1.8, using 2-piecewise regression models (Table 3). Table 3Association between CRP level and sarcopenia using 2-piecewise regression models.Crude modelAdjusted modelCRP levelOR (95%)P-valueOR (95%)P-value<1.81 (Ref)1 (Ref)≥1.83.61 (2.79–4.67)<.0011.86 (1.37–2.51)<.001Age, gender, marital status, race, education, drinking, smoking, diabetes, hypertension, and body mass index (BMI) were fully adjusted.CRP = C-reactive protein, OR = odds ratio, Ref = reference. Association between CRP level and sarcopenia using 2-piecewise regression models. Age, gender, marital status, race, education, drinking, smoking, diabetes, hypertension, and body mass index (BMI) were fully adjusted. CRP = C-reactive protein, OR = odds ratio, Ref = reference. Association between CRP level and sarcopenia. Solid and dashed lines indicate the predicted value and 95% CI. CRP = C-reactive protein, CI = confidence interval. 3.4. Subgroup analysis The results of the subgroup analysis are shown in Figure 3. CRP levels were related to sarcopenia in males (OR, 1.03; 95% CI, 1–1.05) and individuals aged <50-year-old (OR, 1.03; 95% CI, 1.01–1.05), drinking (OR, 1.02; 95% CI, 1–1.03), and BMI ≥ 25 kg/m2 (OR, 1.02; 95% CI, 1–1.03). No association was found between CRP levels and sarcopenia in female or participants aged ≥50-years, those with no drinking, or those with BMI < 25 kg/m2 (Fig. 3). Association between CRP level and sarcopenia according to the general characteristics. CRP = C-reactive protein, OR = odds ratio, CI = confidence interval, BMI = body mass index. 4. Discussion In the current nationwide cross-sectional study, CRP levels showed a nonlinear association with sarcopenia among the US adult population. The correlation was robust in the sensitivity and subgroup analyses. Sarcopenia has been suggested to be closely related to inflammatory reactions. Chronic low-grade inflammation may occur during aging among older people.[13] Skeletal muscle cells generate inflammatory cytokines during inflammation, causing sarcopenia, which is characterized by weak muscle strength, low muscle mass, and low physical performance, resulting in negative outcomes. Loss of mass, strength, and fitness of skeletal muscle within the 4 extremities in males and females are related to increased levels of pro-inflammatory cytokines, including IL-6, TNF-α, and CRP.[14–16] However, many studies have been conducted on obese populations or those with additional chronic disorders.[17–19] Therefore, their conclusions are likely to be influenced by these underlying diseases. As a result, the association between muscle mass and inflammatory cytokine levels remains unknown. In contrast to the previous results, we examined NHANES data by multivariable regression after adjusting for possible confounders. Thus, our findings could be generalized to all US adults. In this study, it was shown that the CRP levels were associated with sarcopenia in males (OR, 1.03; 95% CI, 1–1.05), individuals aged <50-year-old (OR, 1.03; 95% CI, 1.01–1.05), drinking (OR, 1.02; 95% CI, 1–1.03), and BMI ≥ 25 kg/m2 (OR, 1.02; 95% CI, 1–1.03). According to dose–response analyses, CRP levels showed a nonlinear relationship with sarcopenia. In particular, the risk of sarcopenia increased with increasing CRP levels in those with the CRP levels > 1.8. In addition, the association between CRP levels and sarcopenia was robust in the subgroup and sensitivity analyses. In addition to contraction and relaxation, skeletal muscle is also a secretory organ.[13,20,21] Recently, this view has been applied extensively. During skeletal muscle contraction, cytokines are generated and released. These cytokines are called myokines, which include IL-6, IL-8, and IL-15, and may influence the activity of skeletal muscle and metabolism of additional organs and tissues.[22–24] Similar to typical hormones, myokines exert their effects through endocrine, autocrine, and paracrine mechanisms to change target tissue physiological activity. Interactions between skeletal muscle and additional organs or tissues are dependent on myokines.[25–27] Exercises can upregulate the gene levels of certain cytokines in skeletal muscles and increase the levels of different circulating cytokines.[28] However, cytokine gene levels in skeletal muscles may not be causally associated with increased circulating cytokines.[16,29,30] These findings suggest that the physiological mechanism of resistance training is the present mainstream interventional measure to prevent sarcopenia. Physical activity and diet are key interventional measures for sarcopenia patients because they are important for modulating the systemic inflammatory state in patients with sarcopenia. Thus, investigating sarcopenia-related inflammatory cytokines is of great significance, as these cytokines are vital for diagnosing sarcopenia, developing innovative interventions, and evaluating their efficacy. Dietary patterns have been suggested to be tightly related to sarcopenia.[31,32] The association between dietary patterns and inflammation can be interpreted as a correlation with sarcopenia. In particular, the American diet results in increased CRP and IL-6 levels in the serum, while Dietary Approaches to Stop Hypertension and Mediterranean diets have decreased inflammatory factor levels. In a follow-up study, we investigated whether dietary patterns were associated with sarcopenia. However, this study had several limitations. First, this study focused on the US population; thus, more investigations are needed to confirm whether our findings are generalizable to more populations. Second, the residual confounders were not eliminated. Multivariate logistic regression models were used in this study. Sensitivity and subgroup analyses were used to control for the impact of potential confounders on the association of CRP levels with sarcopenia. Third, due to the lack of sarcopenia occurrence or severity data in the NHANES database, we were unable to evaluate the correlation of CRP levels with the clinical characteristics of sarcopenia. Owing to its cross-sectional nature, it was impossible to determine the causal relationship between CRP levels and sarcopenia. Therefore, more longitudinal studies need to be performed to determine whether our observed correlation between CRP level and sarcopenia is causal. 5. Conclusions Based on our results, CRP levels showed a nonlinear association with sarcopenia among adults in the United States. The CRP levels were related to sarcopenia in males, the individuals aged <50-year-old, drinking and BMI ≥ 25 kg/m2. In addition, the association between CRP levels and sarcopenia was robust in the subgroup and sensitivity analyses. Thus, it is necessary to consider the possible role of CRP levels in sarcopenia when treating and preventing these conditions. Therefore, the studies on the molecular mechanisms underlying the pathogenesis of sarcopenia should be performed to better prevent and treat sarcopenia. Acknowledgments We appreciate all the staff members of our institution. We gratefully thank Dr Jie Liu from the Department of Vascular and Endovascular Surgery, Chinese PLA General Hospital, for his constant efforts in statistical support, consultations of study design, and comments on this study. Author contributions Data curation: Yang Li. Funding acquisition: Zhi-Wen Zhang. Methodology: Yang Li. Supervision: Zhi-Wen Zhang. Writing – original draft: Yang Li.