Associations between anthropometric measures and cardiovascular risk biomarkers in women with short stature

Abstract

Objective

The study aimed to investigate which anthropometric indicator are most associated with cardiovascular risk (CVR) in low-income women with short stature and to determine the cut-off point that best determined CVR.

Methods

This cross-sectional study was conducted with meticulous attention to detail. We evaluated a comprehensive range of sociodemographic and anthropometric data, including body mass index (BMI), waist circumference, waist-to-hip ratio, waist-to-height ratio, and body adiposity index. Additionally, dietary and biochemical data such as glucose and lipid profile, ultrasensitive C-reactive protein (CRP), and blood pressure were assessed in 83 women. Body composition was measured using dual-energy X-ray absorptiometry.

Results

Approximately 85% had excess total body fat, 70% had low levels of high-density lipoprotein and increased levels of ultrasensitive CRP, 29.3% were hypercholesterolemic, and 13.4% were hypertensive. Multivariate logistic regression analyses indicated that BMI was the most associated factor of CVR (ultrasensitive CRP) (OR: 1.31, 95% CI: 1.10–1.56). The area under the receiver operating characteristic curve was 0.64 for cholesterol (95% CI 0.517–0.766; p = 0.04) and 0.72 for ultrasensitive CRP (95% CI 0.601–0.840; p = 0.002). BMI cut-off points with the best balance of sensitivity and specificity were 24.0 kg/m² for identifying hypercholesterolemia and 23.8 kg/m² for identifying elevated high-sensitivity CRP levels.

Conclusion

BMI was the most associated factor of CVR in low-income women with short stature. The results indicate that more sensitive BMI cut-off points should be considered to detect CVR in this population group.

Graphical abstract

This is a visual representation of the abstract.

Keywords

Body composition body mass index overweight risk factors cardiovascular disease

Introduction

The most prevalent form of malnutrition worldwide is short stature, and it is identified to a greater extent in areas of poverty and populations living in urban areas.^1,2 Short stature has been associated with adaptations in tissues and organs, which can cause an increase in the frequency and intensity of diseases throughout life.^3,4 In women, short stature has been linked to unfavorable maternal and child health conditions, including an increased likelihood of premature births⁵ and giving birth to children with low birth weight.^5,6

It is also possible to identify that short stature in adulthood has been associated with obesity,^7–9 dyslipidemia,¹⁰ hypertension,¹¹ inflammation,¹² and increased morbidity and mortality from cardiovascular diseases (CVDs).¹³ It is, therefore, important to determine how best to assess cardiovascular risk (CVR) factors in adults with short stature.

Epidemiological studies in populations without short stature have identified several anthropometric indices that are sensitive, specific, non-invasive, reproducible, and economical for predicting total and regional body fat, CVR, and CVD mortality.^14–17 In this respect, it is clear that body mass index (BMI), waist circumference (WC), waist-to-hip ratio (WHR), waist-to-height ratio (WHTR), and body adiposity index (BAI) are all associated with a higher prevalence of CVR, such as hypertension, dyslipidemia, and an increase in inflammatory markers such as C-reactive protein (CRP).^14–18

However, there is little evidence to suggest which anthropometric index are the most associated factors of CVR in women with short stature. Therefore, this study aimed to investigate which anthropometric indicator (BMI, WC, WHR, WHTR, and BAI) was most associated with elevated high-sensitivity CRP (hs-CRP) levels, an inflammatory biomarker associated with increased CVR, in low-income women with short stature.

Materials and methods

Ethical aspects

The procedures were conducted in accordance with the guidelines of the Helsinki Declaration, and all procedures involving human subjects were approved by the Research Ethics Committee of the Federal University of São Paulo (protocol number 1731/11). Written informed consent was obtained from all subjects.

Study design, location, and sample

A cross-sectional study was carried out in socially vulnerable areas of a capital city in northeastern Brazil. The study included low-income women with short stature, aged between 18 and 45 years. The women recruited were mothers of children attending the day-boarding school center at the Center for Recovery and Nutritional Education (CREN-Alagoas). Low-income was defined as a per capita household income below US$1.00 per day, consistent with extreme poverty thresholds. Socially vulnerable areas were defined as urban regions characterized by low household income, limited access to health services, reduced educational attainment, and precarious housing conditions.

Of the 100 mothers invited, 83 were included in the study. The inclusion process is described in Figure 1. This was a non-probabilistic convenience sample. Short stature was defined as height below the 25th percentile (<158 cm) of the growth curve for women at 19 years of age.¹⁹ This cut-off has been used in epidemiological studies to identify adult women exposed to chronic undernutrition and adverse socioeconomic conditions during critical periods of growth.^20–22 Women with known pathological causes of short stature, including genetic syndromes (e.g. Turner syndrome, Noonan syndrome), endocrine disorders, or congenital conditions associated with increased CVR, were not included.

Figure 1.

Flowchart showing the process of including women in the study.

Sociodemographic and lifestyle habits

Information was collected on age, per capita income, number of years of schooling, number of rooms per household, number of residents per household, hours spent in front of the television, hours of physical activity, and whether the women consumed alcohol or smoked.

Information on current medication use was collected, including lipid-lowering agents, corticosteroids, and hormonal contraceptives. Women using medications known to significantly interfere with lipid or inflammatory markers, including lipid-lowering drugs, systemic corticosteroids, chronic anti-inflammatory or immunosuppressive agents, were not included in the analysis.

Anthropometric parameters

Weight and height were collected to determine the BMI of the women. BMI was then classified as underweight (< 18.5 kg/m²), adequate weight (≥ 18.5 < 25 kg/m²), overweight (≥ 25 < 30 kg/m²), and obesity (≥ 30 kg/m²).²³ WC and hip circumference (HC) were also measured using a flexible tape measure, and the cut-off point used for WHR and WC was 0.80 and 80 cm, respectively. WHTR was calculated as the ratio between WC (cm) and height (cm), with a cut-off point for abdominal obesity associated with increased CVR as a ratio ≥ 0.5. BAI was obtained based on HC and height, according to Bergman et al.,¹⁶ using the following equation: BAI = [HC/(height)1.5]-18. Individuals with values between 33 and 38 were considered overweight, and those with values above 38 were considered obese.

All anthropometric measurements were obtained by trained health professionals following standardized protocols. Measurements were performed by two previously trained examiners, using calibrated equipment. Weight was measured with the participants barefoot and wearing light clothing. Height was measured using a stadiometer, with participants standing upright, barefoot, and with the head positioned in the Frankfurt plane. WC was measured at the midpoint between the lowest rib and the iliac crest, and HC at the widest portion of the gluteal region. Measurements were taken in duplicate, and the mean value was used for analysis.

Body composition

Body composition (trunk fat, android fat, gynoid fat, and total fat) was assessed using dual-energy X-ray absorptiometry (DXA, GE healthcare^®, Bedford, UK). The DXA was calibrated using a phantom, and quality control was assessed according to the manufacturer's specifications.

Dietary parameters

A 24-h food recall was carried out on 3 non-consecutive days, one of which was a weekend. The individual's food intake was assessed according to its nutrient content using the Nut Win program, Version 1.5.2.51 (Nutrition Support Program, developed by the Department of Health Informatics at the Federal University of São Paulo, São Paulo, Brazil). An average amount of food consumption was calculated from the three food recalls carried out.

Blood pressure

Blood pressure was measured in triplicate with an automated blood pressure monitor (OMRON HEM-705CP, Vernon Hills, IL). Women who were using antihypertensive medication or had systolic blood pressure ≥ 140 mmHg and/or diastolic blood pressure ≥ 90 mmHg were classified as hypertensive.

Biochemical parameters

For this procedure, blood samples were collected after a 12-h fast in a specialized laboratory. The values for total cholesterol (TC), high-density lipoprotein (HDL), triglycerides (TG), and glucose were determined using an enzymatic assay. The concentration of low-density lipoprotein (LDL) was determined according to Friedewald's formula: LDL = TC-(HDL + TG/5). The following ratios were calculated: the Castelli I index: TC/HDL, the Castelli II index: LDL/HDL, and TG/HDL.

Insulin values were determined using the electrochemiluminescence method. The homeostatic model assessment index for insulin resistance (HOMA-IR) was calculated using laboratory glucose and insulin concentrations using the formula proposed by Matthews et al.²⁴: HOMA-IR = insulin (mU/mL) x glucose (mg/dl)/22.5.

hs-CRP was used as an inflammatory marker and measured using an ultrasensitive immunonephelometric method. According to the literature, hs-CRP concentrations between 1 and 3 mg/L are indicative of intermediate CVR.^25,26 In the present study, elevated inflammatory status was defined as hs-CRP >1 mg/L, considering that hs-CRP acts as an early marker of CVR even at moderate concentrations²⁷ and that CVR increases progressively across the CRP distribution rather than only at markedly elevated levels.²⁸ Furthermore, this cut-off represents a more sensitive threshold for detecting low-grade systemic inflammation in apparently healthy adult women, in whom inflammatory marker levels are typically lower.^25,27

TC and hs-CRP were selected due to their well-established association with cardiovascular morbidity and mortality, and their frequent use as intermediate CVR markers in population-based studies.

In this study, hs-CRP was used as an inflammatory biomarker associated with cardiometabolic and CVR, rather than as a direct measure of CVD.

Statistical analysis

As previously described, the sample size was determined using a non-probabilistic convenience sampling approach. However, the final analytical sample (n = 83) is consistent with methodological recommendations indicating that samples ranging from approximately 50 to 100 participants allow reliable estimation of the area under the receiver operating characteristic curve (AUC) in receiver operating characteristic (ROC) analyses, particularly in exploratory diagnostic studies.^29,30 Furthermore, this sample size is comparable to that adopted in previous studies investigating cardiovascular outcomes using ROC curve analysis,^31,32 supporting the adequacy of the present sample for evaluating discriminatory performance and estimating cut-off values. The Lilliefors test was used to test normal distribution. The chi-square test was used to compare the frequency of overweight among different methods (DXA, BMI, IAC).

We conducted a series of multivariate logistic regressions, with the dependent variable being the category of hs-CRP (normal or high) and the independent variables being those that could influence CRP (anthropometrics and DXA components). All variables were adjusted for sociodemographic, behavioral, and dietary covariates (age, per capita income, education, smoking, drinking, physical activity, and saturated fat intake). Subsequently, several multivariate logistic regression models with the “forward” technique using the best likelihood ratio (including anthropometric variables and the components of DXA, separately) were built to see which parameter within each group (anthropometric measures and DXA) could be associated with the risk of presenting elevated CRP levels.

ROC curves were used to establish BMI cut-off points for SS women that could be used as a discriminatory factor for the risk of hypercholesterolemia (>200 mg/dl) and increased CVR, according to the CRP levels (CRP>1 mg/dl). The area under the ROC curve was used to determine the overall accuracy of BMI as an associated factor of high cholesterol and CRP levels. ROC curve analyses were performed using hs-CRP and TC due to their established role as early biomarkers of cardiometabolic and CVR in population-based studies.

Only participants with complete data for the variables included in each specific analysis were considered. Missing data were handled by complete-case analysis, and no imputation procedures were applied.

Statistical analyses were carried out using the R statistical package (R Foundation for Statistical Computing, Vienna, Austria). For the levels of statistical significance, alpha values of 5% were adopted.

Results

Our study sample consisted of a total of 83 women with short stature, whose characteristics are summarized in Table 1. Per capita income was <$1.0 US/day, and the average schooling was <5 years of study, with 45.8% of the subjects considered as functionally illiterate. The prevalence of overweight found by the DXA was significantly higher than the prevalence obtained by the BAI and BMI (p < 0.001) (Table 2).

Table 1.

Sociodemographic, nutritional, and metabolic characteristics of the studied sample (n = 83).

Variables	Mean	SD	Minimum	Maximum
Age, years	28.30	6.50	18.0	44.0
Sociodemographic
Number of rooms	3.63	1.02	1	6
People per household	4.92	2.02	2	17
Income per capita/day (US)	0.825	0.85	0.125	5.83
Years of study	4.94	2.97	0	11
Anthropometry
Weight, kg	58.50	10.3	39.4	83.8
Height, m	1.51	0.05	1.38	1.58
Body mass index, kg/m²	25.62	4.60	18.7	37.8
Waist circumference, cm	81.46	10.9	63.4	110.0
Waist-to-hip ratio	0.83	0.07	0.68	1.00
Waist-to-height ratio	0.54	0.07	0.42	0.71
Body adiposity index	34.60	5.23	21.8	50.15
Body composition (DXA)
Trunk fat (%)	41.10	9.23	14.3	56.2
Android fat (%)	44.51	9.90	10.3	58.3
Gynoid fat (%)	46.39	6.92	26.7	61.0
Total fat (%)	39.07	8.25	15.4	53.7
Dietary parameters
Energy intake (kcal/day)	1840.5	709.4	665.0	4864.0
Carbohydrate (%)	58.2	9.8	29.1	78.6
Protein (gr/kg/day)	1.25	0.6	0.21	3.2
Lipid (%)	26.4	8.0	12.6	46.9
Saturated fat (%)	12.5	7.4	2.4	32.8
Cholesterol (mg)	278.0	172.6	47.4	733.7
Fiber (mg)	16.3	7.30	3.0	41.0
Blood pressure
SBP, mmHg	114.88	14.82	80.0	155.0
DBP, mmHg	75.81	13.35	53.0	141.0
Biochemical parameters
Glucose, mg/dL	69.98	7.62	58.0	115.0
Insulin, mUI/ml	7.50	5.14	1.30	28.4
HOMA-IR	0.91	0.62	0.20	3.50
Cholesterol, mg/dL	177.42	34.9	103.0	272.0
HDL, mg/dL	47.60	8.54	32.0	69.0
LDL, mg/dL	111.70	31.1	39.2	206.6
Triglyceride, mg/dL	92.75	48.0	36.0	275.0
Castelli index I	3.81	0.9	1.8	6.6
Castelli index II	2.42	0.8	0.7	5.0
Triglyceride/HDLc ratio	2.05	1.2	0.72	5.5
Ultrasensitive C-reactive Protein, mg/L	3.82	4.3	0.20	21.8

DXA: dual-energy X-ray absorptiometry; SBP: systolic blood pressure; DBP: diastolic blood pressure; HDL: high-density lipoprotein; LDL: low-density lipoprotein; SD: standard deviation; HOMA-IR: homeostatic model assessment index for insulin resistance.

Table 2.

Frequency of cardiovascular risk factors in the studied sample (n = 83).

Parameters	n (%)
Anthropometry
Eutrophia (BMI > 18.5 < 25 kg/m²)	45 (54.2)
Overweight (BMI > 25 < 30 kg/m²)	23 (27.7)
Obesity (BMI > 30 kg/m²)	15 (18.1)
Overweight (BAI 33–38)	21 (25.5)
Obesity (BAI > 38)	14 (17.7)
Obesity (%fat–DXA > 30%)	70 (84.1)^a
Waist circumference (≥ 80 cm)	37 (44.4)
Waist-to-hip ratio (≥ 0.80)	48 (58)
Waist-to-height ratio (≥ 0.5)	59 (70.9)
Behavioral
Smoking	10 (12)
Alcoholism	19 (23.2)
Hours in front of television (≥ 3)	50 (61)
Sedentary life style	74 (89.2)
Metabolic
Hypertension (SBP ≥ 140mmHg and/or DBP ≥ 90 mmHg or antihypertensive use)	11 (13.4)
Glucose (> 100 mg/dL)	1 (1.2)
Insulin (< 1.9mUI/mL)	2 (2.5)
Insulin (> 23 mUI/mL)	2 (2.5)
HOMA-IR (> 2,4)	2 (2.5)
Cholesterol (≥ 200 mg/dL)	24 (29.3)
LDL cholesterol (≥ 130 mg/dL)	20 (23.8)
HDL cholesterol (< 50 mg/dL)	56 (67.9)
Triglyceride (≥ 150 mg/dL)	11 (13.8)
Castelli index I (≥ 3.5)	49 (59.3)
Castelli index II (≥ 2.5)	34 (41)
Triglyceride/HDL ratio (> 2)	32 (38.3)
Ultrasensitive C-reactive protein (> 1 mg/L)	60 (72.8)

BMI: body mass index; BAI: body adiposity index; DXA: dual-energy X-ray absorptiometry; SBP: systolic blood pressure; DBP: diastolic blood pressure; LDL: low-density lipoprotein; HDL: high-density lipoprotein; HOMA-IR: homeostatic model assessment index for insulin resistance.

Chi-square test < 0.05.

Table 3 shows the values for the multivariate logistic regression adjusted for sociodemographic, dietary, and behavioral variables with the use of CRP levels as the dependent variable. In regard to the anthropometric variables, only WHR was not significantly associated with CRP levels. The logistic regression analysis with the “forward” technique using all DXA variables (android fat, gynoid fat, trunk fat, and total fat) and based on the likelihood ratio showed that android fat was the most associated with CRP levels (OR: 1.15, 95% CI: 1.04, 1.20). The same model was performed for all anthropometric variables (BMI, WC, WHR, WHTR, and BAI) and showed that BMI was the most associated with CRP levels (OR: 1.31, 95% CI: 1.10, 1.56).

Table 3.

Multivariate logistic regression models according to ultrasensitive C-reactive protein category using the variables of interest (n = 83).

Variables	Odds-ratio ^a	CI95%	P-multivariate
DXA
Android fat (DXA), %	1.15	1.04–1.20	0.003
Gynoid fat (DXA), %	1.15	1.03–1.28	0.011
Trunk fat (DXA), %	1.13	1.04–1.23	0.003
Total fat (DXA), %	1.13	1.05–1.26	0.003
Anthropometric
Body mass index, kg/m²	1.31	1.10–1.56	0.003
Waist circumference, cm	1.12	1.04–1.21	0.003
Waist-to-hip ratio	1.08	0.98–1.18	0.119
Waist-to-height ratio	1.18	1.06–1.33	0.004
Body adiposity index	1.22	1.06–1.43	0.007

Adjusted for age, per capita income, education, smoking, alcohol consumption, physical activity and saturated fat.

CI: confidence interval; DXA: dual-energy X-ray absorptiometry.

Results from the ROC curve analysis (Figure 2(a)) was 0.64 for cholesterol (95% CI 0.517–0.766; p = 0.04) with a BMI cut-off point of 24.2 kg/m², presenting a sensitivity and specificity of 0.586 and 0.750, respectively. The area under the ROC curve was 0.72 (95% CI 0.601–0.840; p = 0.002) for CRP. The cut-off point of 23.8 kg/m² had the most suitable balance between a sensitivity of 0.661 and a specificity of 0.727 (Figure 2(b)).

Figure 2.

Area under the curve, confidence interval and cut-off point of body mass index as a predictor.

At baseline, a substantial proportion of the participants presented metabolic alterations, including elevated TC, increased LDL cholesterol, reduced HDL cholesterol, and elevated hs-CRP levels, despite the absence of clinically diagnosed CVD.

Discussion

Taking into account the scarcity of evidence proposing to establish anthropometric measures, as well as their cut-off points, that are more sensitive for identifying inflammatory and cardiometabolic alterations associated with increased CVR in adults with short stature, we set out to minimize this gap in knowledge. Our findings show that BMI was the most sensitive anthropometric indices assessed for identifying elevated hs-CRP levels, which are associated with increased CVR, compared to WC, WHR, WHTR, and BAI. In addition, the results of a ROC curve analysis showed greater sensitivity for a BMI of 23.8 kg/m². Although the observed AUC values indicate only modest discriminatory performance, they suggest that lower BMI cut-off points may be relevant for identifying early cardiometabolic alterations in women with short stature.

Establishing this type of information is relevant, given that chronic non-communicable diseases (NCDs) are responsible for almost 17 million deaths every year, 86% of which occur in low- and middle-income countries, and that CVDs are among those that most affect the population.³³ Several risk factors are related to the development of CVD, particularly age, dyslipidemia, diabetes, high BMI, smoking, and hypertension.^34–36 Among these factors, hypertension deserves attention, given its relationship with height, and it is possible to identify that the prevalence of hypertension is inversely proportional to height.¹¹

It is also important to note the relationship between CVD and anthropometric parameters such as BMI and WHR, as well as body fat percentage. In this respect, a study carried out in Mexico that used the bioimpedance technique to assess the percentage of body fat identified that individuals with short stature had a higher percentage of fat compared to taller individuals, even among those with the same BMI value, suggesting that a BMI of 25 kg/m² should be used as the cut-off point for obesity in this population.³⁷ However, the authors pointed out the limitations of their study, including the use of bioimpedance and the absence of a comparison of their results with biochemical markers,³⁷ which suggests caution in interpreting the results.

Taking this limitation into account, we evaluated the serum lipid profile of the women included in this study, given its relationship with CVD,³⁸ and it was possible to observe various alterations such as hypercholesterolemia, increased LDL, and low HDL levels. This information is relevant given the need for more information on this type of assessment in women with short stature. A study carried out in the USA between 1984 and 2005 showed that an increase of 10 cm in height was associated with a reduction of between 8 and 18% in TC in women.³⁹

An increase in the TC/HDL ratio has also been observed in low-income women living in Brazilian favelas, as well as a decrease in HDL levels,¹⁰ changes related to the lipid profile that can be observed from childhood in the poorest population.⁴⁰ Given this scenario, especially in relation to long-term changes in the lipid profile, the chance of CVD increases sharply, leading to a greater risk of coronary heart disease and myocardial infarction. This situation shows the need for permanent and continuous assessment of these biochemical parameters,⁴¹ especially in women with short stature since they are more likely to have lipid changes.⁴²

Another point that deserves attention when it comes to CVD is dietary intake. This assessment was also carried out in this study, and it was possible to see that the proportion of macronutrients, as well as saturated fat and cholesterol, was within the recommendations.⁴³ However, it has been observed that the eating habits of low-income women have changed over time. It is possible to identify the greater inclusion of ultra-processed foods in their diet.⁴⁴ This situation contributes to a higher consumption of free sugars, total fats, and saturated fats, as well as a decrease in fiber, protein, vitamins, and minerals.⁴⁵

This scenario shows that poorer women have a greater chance of compromised nutritional status over time, especially in relation to higher BMI and WC.⁴⁴ This scenario tends to get worse, given the projections related to the price of ultra-processed foods, showing that by 2030, they will be substantially cheaper than fresh/minimally processed foods, favoring a progressive increase in their consumption, especially among the poorest.⁴⁶

These findings, coupled with the physiological disparities identified in women with short stature, such as a decrease in serum levels of the thyroid hormone T3, suggest a greater physiological program to save energy,¹⁰ indicating a reduced basal metabolic rate, which can be inferred to impair their thyroid function, resulting in increased TC levels.⁴⁷ Therefore, it is possible to infer that the women in this study underwent similar physiological adaptations, leading to compromised energy expenditure and contributing to a higher proportion of body fat, increasing CVR.

In addition to the traditional CVR factors,⁴⁸ high levels of CRP, a biochemical parameter that indicates inflammation, have also been shown to be an important risk factor for CVD.^49,50 In this study, more than 70% of the women investigated had moderate or high levels of CRP. This situation is noteworthy because, for every one standard deviation increase in CRP levels, there is a 20% increase in CVD risk.⁵¹

A longitudinal study carried out in Turkey showed that in women of short stature, CRP levels are higher, and this parameter can be used as an independent risk marker for coronary heart disease in women of short stature. A height below 160 cm is a strong marker of death.⁹ The authors concluded that influences in early life may be gender-specific in determining pro-inflammatory status and that short stature was an independent risk factor for coronary heart disease.¹² Even in light of the above, evidence like this, relating to women with short stature, still needs to be discovered, highlighting the need to assess this population group better.

Based on these findings, we can suggest a possible relationship between the type of environment to which these women were exposed early in life and the presence of inflammatory mediators in adulthood,¹² a situation which, when we look at the poorest population, shows that the environment has a direct relationship with their health situation as early as childhood, directly influencing their nutritional status.^52,53 It can also be seen that the environment influences the development of NCDs, which are a risk factor for CVD,⁵⁴ as well as their eating habits.⁵⁵

Thus, when evaluating anthropometric parameters while accounting for potential confounders, BMI was the anthropometric measure most strongly associated with elevated hs-CRP levels, an inflammatory biomarker associated with increased CVR, compared to WC, WHR, WHTR, and BAI. This remained significant after adjusting for age, education, per capita income, saturated fat intake, smoking, alcohol consumption, and physical activity. This finding is important, especially given that BMI, among the anthropometric parameters assessed, is the easiest to measure,⁵⁶ especially when looking at the material resources needed to measure it, given its greater availability in health services.⁵⁷

From these findings, ROC curves were constructed to establish which BMI cut-off points could be used to identify elevated hs-CRP levels (hs-CRP >1 mg/L) and hypercholesterolemia (>200 mg/dL). The findings from the ROC curves showed that these metabolic changes are already taking place in women with a BMI of 23.8 kg/m². Similar values were found in the Asian population.³⁹ However, depending on the sample investigated, different BMI cut-off points from those usually recommended might be more reliable.⁴² When combined with the use of methods that have yet to be validated for specific populations, this may result in erroneous diagnoses and interventions. Another study demonstrated that the suitability of BMI for detecting individuals with comorbidities associated with obesity was significantly lower in Mexicans with short stature compared to those without short stature. The findings were similar even when a more sensitive cut-off point for defining individuals as overweight was used.⁵⁸

Among the strengths of this study are that we highlight the population studied, adult women with short stature, and a population that has a lack of data and/or comparisons similar to those of this study available in the literature. In addition, the technique used to evaluate CRP was ultrasensitive immunonephelometric, which increases the sensitivity and reliability of the method. In order to guarantee the sensitivity of the measurements to the differential diagnosis of an infectious process, we excluded CRP levels >10 mg/L from the study.²⁶ These considerations reinforce the interpretation that hs-CRP levels in this study most likely reflect low-grade systemic inflammation associated with increased CVR, rather than acute infection. In addition to the number of biochemical and anthropometric assessments, which better characterized the study sample, it was also possible to investigate more factors related to the outcome assessed.

This study also has some limitations, such as its cross-sectional design, which does not allow us to determine a temporal relationship between BMI and CVR. As well as the sample size. However, taking into account the number of assessments carried out, the sample size is similar to that observed in other studies. Another point concerns the fact that the presence of acute infectious disease was not assessed, a situation which may be reflected in the CRP levels found. However, it is highly unlikely that more than 70% of the individuals had an acute infection at the time of the study. In addition, the lack of controls occurred because there were no individuals in this very low-income population who reached the 50th percentile of height or were close to it. The relatively small sample size may have limited statistical power, particularly for multivariable analyses. However, the sample size is comparable to that of similar studies involving detailed biochemical and body composition assessments. The generalizability of these findings is limited, as the study population consisted of low-income adult women with short stature living in socially vulnerable urban areas. Therefore, the results may not be directly applicable to populations with different socioeconomic or demographic profiles.

Conclusion

BMI showed the strongest association with CVR biomarkers in women with short stature compared to the other anthropometric parameters investigated (WC, WHR, WHTR, and BAI). The most sensitive and specific cut-off points for identifying hypercholesterolemia risk and elevated ultrasensitive CRP levels were 24.2 kg/m² and 23.8 kg/m², respectively. These findings suggest that women with short stature may be more susceptible to early alterations in biomarkers associated with CVR.

Footnotes

Acknowledgment

To the Nutritional Recovery and Education Center (Centro de Recuperação e Educação Nutritional—CREN) of Alagoas for all logistical support.

ORCID iDs

Luiz Gonzaga Ribeiro Silva-Neto

Edward Araujo Júnior

Ethical considerations

The study was approved by the Research Ethics Committee of the Federal University of São Paulo (protocol number 1731/11).

Consent to participate

Written informed consent was obtained from all subjects.

Author contributions

Conceptualization was done by RPAB and LGRSN. Investigation was done by TMMTF and NBB. Methodology was done by APGC and ALS. Writing–original draft was done by LAC and GYC. Writing–review and editing was done by EAJ and GYC. Supervision was done by LGRSN.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by the Foundation of the State Research of Alagoas [grant number 20090132001-6]. The funding sources had no involvement in the design or the study; collection, analysis, or interpretation of the data; or decision to submit the article for publication.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

References

World Health Organization. Global nutrition targets 2025: stunting policy brief. World Health Organization, 2014.

de Onis

Branca

. Childhood stunting: a global perspective. Matern Child Nutr 2016; 12: 12–26.

Grantham-McGregor

Cheung

Cueto

, et al. Developmental potential in the first 5 years for children in developing countries. Lancet 2007; 369: 229–242.

Victora

Christian

Vidaletti

, et al. Revisiting maternal and child undernutrition in low-income and middle-income countries: variable progress towards an unfinished agenda. Lancet 2021; 397: 1388–1399.

Kozuki

Katz

Lee

ACC

, et al. Short maternal stature increases risk of small-for-gestational-age and preterm births in low- and middle-income countries: individual participant data meta-analysis and population attributable fraction. J Nutr 2015; 145: 2542–2550.

Britto

RPA

Florêncio

Benedito Silva

, et al. Influence of maternal height and weight on low birth weight: a cross-sectional study in poor communities of Northeastern Brazil. PLoS One 2013; 8: e80159.

Rangel-Baltazar

Rodríguez-Ramírez

Cuevas-Nasu

, et al. Short stature modifies the waist-to-height ratio cut-off points as an indicator of cardiovascular risk in Mexican adult women and men. Arch Med Res 2023; 54: 102839.

Black

Victora

Walker

, et al. Maternal and Child Nutrition Study Group. Maternal and child undernutrition and overweight in low-income and middle-income countries. Lancet 2013; 382: 427–451.

Bosy-Westphal

Plachta-Danielzik

Dörhöfer

, et al. Short stature and obesity: positive association in adults but inverse association in children and adolescents. Br J Nutr 2009; 102: 453–461.

10.

Florêncio

Ferreira

Cavalcante

, et al. Short stature, abdominal obesity, insulin resistance and alterations in lipid profile in very low-income women living in Maceió, North-eastern Brazil. Eur J Cardiovasc Prev Rehabil 2007; 14: 346–348.

11.

Ferreira

Florêncio

TMT

Fragoso

MAC

, et al. Hypertension, abdominal obesity and short stature: aspects of nutritional transition within a shantytown in the city of Maceió (Northeastern Brazil). Rev Nutr 2005; 18: 209–218.

12.

Onat

Ornek

Can

, et al.

Short stature is an independent risk marker for mortality and incident coronary heart disease only in women: a structural relationship?

Anadolu Kardiyol Derq 2012; 12: 289–297.

13.

Paajanen

Oksala

Kuukasjärvi

, et al. Short stature is associated with coronary heart disease: a systematic review of the literature and a meta-analysis. Eur Heart J 2010; 31: 1802–1809.

14.

Song

Jousilahti

Stehouwer

, et al. Cardiovascular and all-cause mortality in relation to various anthropometric measures of obesity in Europeans. Nutr Metab Cardiovasc Dis 2015; 25: 295–304.

15.

Aune

Sen

Norat

, et al. Body mass index, abdominal fatness and heart failure incidence and mortality: a systematic review and dose–response meta-analysis of prospective studies. Circulation 2016; 133: 639–649.

16.

Bergman

Stefanovski

Buchanan

, et al. A better index of body adiposity. Obesity 2011; 19: 1083–1089.

17.

Global Burden of Metabolic Risk Factors for Chronic Diseases Collaboration (BMI Mediated Effects). Lu

Hajifathalian

Ezzati

, et al. Metabolic mediators of the effects of body-mass index, overweight, and obesity on coronary heart disease and stroke: a pooled analysis of 97 prospective cohorts with 1·8 million participants. Lancet 2014; 383: 970–983.

18.

Choi

Joseph

Pilote

. Obesity and C-reactive protein in various populations: a systematic review and meta-analysis. Obes Rev 2013; 14: 232–244.

19.

WHO. Multicentre Growth Reference Study Group. WHO Growth reference data for 5-19 years. World Health Organization, 2007.

20.

Ferreira

Xavier Junior

AFS

Assunção

, et al. Developmental origins of health and disease: a new approach for the identification of adults who suffered undernutrition in early life. Diabetes Metab Syndr Obes 2018; 11: 543–551.

21.

Ferreira

Moura

Cabral

Jr , et al. Short stature of mothers from an area endemic for undernutrition is associated with obesity, hypertension and stunted children: a population-based study in the semi-arid region of Alagoas, Northeast Brazil. Br J Nutr 2009; 101: 1239–1245.

22.

Florêncio

Ferreira

Cavalcante

, et al. Short stature, obesity and arterial hypertension in a very low income population in North-Eastern Brazil. Nutr Metab Cardiovasc Dis 2004; 14: 26–33.

23.

World Health Organization. Physical status: the use of and interpretation of anthropometry, report of a WHO expert committee. World Health Organization, 1995.

24.

Matthews

Hosker

Rudenski

, et al. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985; 28: 412–419.

25.

Pearson

Mensah

Alexander

, et al. Markers of inflammation and cardiovascular disease: application to clinical and public health practice: a statement for healthcare professionals from the Centers for Disease Control and Prevention and the American Heart Association. Circulation 2003; 107: 499–511.

26.

Gimeno

Ferrie

Elovainio

, et al. When do social inequalities in C-reactive protein start? A life course perspective from conception to adulthood in the cardiovascular risk in young Finns study. Int J Epidemiol 2008; 37: 290–298.

27.

Ridker

. High-sensitivity C-reactive protein and cardiovascular risk: rationale for screening and primary prevention. Am J Cardiol 2003; 92: 17K–22K.

28.

Emerging Risk Factors Collaboration, Kaptoge S, Di Angelantonio E, Lowe G, et al. C-reactive protein concentration and risk of coronary heart disease, stroke, and mortality: an individual participant meta-analysis. Lancet 2010; 375: 132–140.

29.

Hajian-Tilaki

. Sample size estimation in diagnostic test studies of biomedical informatics. J Biomed Inform 2014; 48: 193–204.

30.

Hanley

McNeil

. The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology 1982; 143: 29–36.

31.

Inuzuka

Vitorino

PVO

Barroso

, et al. Pulse wave velocity of 8.2 m/s as a threshold associated with cardiovascular target organ damage presence. Arq Bras Cardiol 2023; 120: e20220934.

32.

Wang

Yan

Meng

, et al. Reliability and validity of sit-to-stand test protocols in patients with coronary artery disease. Front Cardiovasc Med 2022; 9: 841453.

33.

World Health Organization

Invisible numbers: the true extent of noncommunicable diseases and what to do about them. Geneva: World Health Organization, 2022.

34.

Flora

Nayak

MKA

. A brief review ofcardiovascular diseases, associated risk factors and current treatment regimes. Curr Pharm Des 2019; 25: 4063–4084.

35.

Francula-Zaninovic

Nola

. Management of measurable variable cardiovascular disease’ risk factors. Curr Cardiol Rev 2018; 14: 153–163.

36.

Vaduganathan

Mensah

Turco

, et al. The global burden of cardiovascular diseases and risk: a compass for future health. J Am Coll Cardiol 2022; 80: 2361–2371.

37.

López-Alvarenga

Montesinos-Cabrera

Velázquez-Alva

, et al. Short stature is related to high body fat composition despite body mass index in a Mexican population. Arch Med Res 2003; 34: 137–140.

38.

Kaneko

Itoh

Kiriyama

, et al. Lipid profile and subsequent cardiovascular disease among young adults aged < 50 years. Am J Cardiol 2021; 142: 59–65.

39.

Rashad

. Height, health, and income in the US, 1984-2005. Econ Hum Biol 2008; 6: 108–126.

40.

Alves

JFR

Britto

RPA

Ferreira

, et al.

Evolution of the biochemical profile of children treated or undergoing treatment for moderate or severe stunting: consequences of metabolic programming?

J Pediatr 2014; 90: 356–362.

41.

Hou

Chen

, et al. Exploring the causal roles of circulating remnant lipid profile on cardiovascular and cerebrovascular diseases: Mendelian randomization study. J Epidemiol 2022; 32: 205–214.

42.

Institute of Medicine Panel on Macronutrients, Institute of Medicine Standing Committee on the Scientific Evaluation of Dietary Reference Intakes. Dietary reference intakes for energy, carbohydrate, fiber, fat, fatty acids, cholesterol, protein, and amino acids / Panel on Macronutrients, Panel on the Definition of Dietary Fiber, Subcommittee on Upper Reference Levels of Nutrients, Subcommittee on Interpretation and Uses of Dietary Reference Intakes, and the Standing Committee on the Scientific Evaluation of Dietary Reference Intakes. Food and Nutrition Board, Institute of Medicine of the National Academies. National Academies Press, 2005.

43.

Zheng

Zhuang

, et al. Dietary macronutrient intake and cardiovascular disease risk and mortality: a systematic review and dose-response meta-analysis of prospective cohort studies. Nutrients 2024; 16: 152.

44.

de Medeiros

Silva-Neto

LGR

dos Santos

TLF

, et al. Higher consumption of ultra-processed foods is associated with obesity and abdominal obesity in socially vulnerable Brazilian women. Nutr Bull 2024; 49: 199–208.

45.

Martini

Godos

Bonaccio

, et al. Ultra-processed foods and nutritional dietary profile: a meta-analysis of nationally representative samples. Nutrients 2021; 13: 3390.

46.

Maia

Passos

Levy

, et al. What to expect from the price of healthy and unhealthy foods over time? The case from Brazil. Public Health Nutr 2020; 23: 579–588.

47.

VanTienhoven-Wind

Dullaart

. Low-normal thyroid function and novel cardiometabolic biomarkers. Nutrients 2015; 16: 1352–1377.

48.

Teo

Rafiq

. Cardiovascular risk factors and prevention: a perspective from developing countries. Can J Cardiol 2021; 37: 733–743.

49.

Kandelouei

Abbasifard

Imani

, et al. Effect of statins on Serum level of hs-CRP and CRP in patients with cardiovascular diseases: a systematic review and meta-analysis of randomized controlled trials. Mediators Inflamm 2022; 2022: 8732360.

50.

Kuppa

Tripathi

Al-Darraji

, et al. C-reactive protein levels and risk of cardiovascular diseases: a two-sample bidirectional Mendelian randomization study. Int J Mol Sci 2023; 24: 9129.

51.

Kaptoge

Angelantonio

Pennells

, et al. C-reactive protein, fibrinogen, and cardiovascular disease prediction. N Engl J Med 2012; 367: 1310–1320.

52.

Gomes

JMR

Silva-Neto

LGR

dos Santos

TLF

, et al. Early introduction of ultra-processed foods is associated with overweight and anaemia in socially vulnerable Brazilian children. Br J Nutr 2024; 131: 1095–1104.

53.

Silva-Neto

LGR

Clemente

APG

Bueno

, et al. The double burden of malnutrition is associated with continued breastfeeding and early consumption of ultra-processed drinks in socially vulnerable Brazilian children. Ecol Food Nutr 2024; 63: 160–173.

54.

Silva-Neto

LGR

da Silva Júnior

Bueno

, et al. Type 2 diabetes mellitus is associated with food addiction in Brazilian women living in poverty. Br J Nutr 2024; 131: 1421–1424.

55.

Silva-Neto

LGR

Bueno

dos Santos

TLF

, et al. Assessment of food insecurity in the context of COVID-19: association with emergency aid and collecting food donations among the socially vulnerable population of a capital in Northeastern Brazil. Cienc Saude Colet 2023; 28: 721–730.

56.

Gomes

dos Anjos

de Vasconcellos

MTL

. Anthropometry as a tool for assessing the nutritionalstatus of adolescents. Rev Nutr 2010; 23: 591–605.

57.

da Silva

GAS

da Rocha

CMM

de Almeida

MFL

, et al. Procedures of measurement of the body mass in children by community health agents in Macaé, Rio de Janeiro, Brazil, 2010-2011. Epidemiol Serv Saude 2017; 26: 579–588.

58.

Lara-Esqueda

Aguilar-Salinas

Velazquez-Monroy

, et al. The body mass index is a less-sensitive tool for detecting cases with obesity-associated co-morbidities in short stature subjects. Int J Obes Relat Metab Disord 2004; 28: 1443–1450.