Obesity is an increasing worldwide problem that has taken on epidemic proportions [
The metabolic syndrome is a common metabolic disorder that may result from the increasing prevalence of obesity. Metabolic syndrome is a cluster of cardiometabolic risk factors characterized by abdominal obesity, insulin resistance, and chronic systemic inflammation [
The mechanisms underlying the relationship between impaired lung function and the metabolic syndrome are unclear. The chronic low-grade systemic inflammation that is associated with obesity might explain this relationship. Our hypothesis is that this low-grade systemic inflammation causes inflammation in the lungs and, hence, lung function impairment.
We therefore performed a study to investigate (1) the association between lung function and the metabolic syndrome in morbidly obese subjects and (2) to determine the effect of systemic inflammation on the relationship between lung function impairment and the metabolic syndrome.
The subjects included in this study were consecutive patients who underwent preoperative screening for bariatric surgery in the Sint Franciscus Gasthuis in Rotterdam, The Netherlands, between October 2009 and May 2011. Eligibility criteria for bariatric surgery were age between 18 and 60 years and body mass index (BMI) either ≥40 kg/m2 or ≥35 kg/m2 combined with the presence of comorbidity, for example, diabetes mellitus, hypertension, or proven obstructive sleep apnea syndrome (OSAS). Subjects underwent baseline physical examinations that included routine assessment of anthropometry, blood pressure, pulmonary function, and blood samples.
Height and weight were measured wearing light clothes and no shoes. Body mass index was calculated as weight (in kg) divided by height (in m) squared. Abdominal circumference was measured directly to the body surface midway between the lower rib margin and the ileac crest.
The Epworth Sleepiness Scale questionnaire [
All subjects gave informed consent, and the local ethics committee (Toetsingscommissie Wetenschappelijk Onderzoek Rotterdam e.o, Trial no. NL25637.101.08) approved the study protocol (The Netherlands Trial Register no. NTR3204).
Metabolic syndrome was diagnosed according to the National Cholesterol Education Program’s Adult Treatment Panel III report (NCEP ATP-III) criteria when ≥3 of the following 5 risk factors were present: abdominal obesity, an elevated level of serum triglycerides, low serum level of high-density lipoprotein cholesterol (HDL-C), elevated blood pressure, and high serum glucose level or treatment for any of these disorders [
Spirometry was performed with Vmax spirometer (Vmax SensorMedics Viasys, type Encore 20/22/229/62 Encore, Cardinal Health, USA) before and after 400
Blood was taken by venipuncture during routine preoperative screening. Laboratory measurements were performed according to standard procedures by our Department of Clinical Chemistry. Plasma-cholesterol, HDL-cholesterol, glucose, triglycerides, and CRP were measured using LX 20 and DxC analyzers (Beckman Coulter, Miami, FL, USA). LDL-cholesterol was calculated using the Friedewald formula. Total IgE and specific plasma IgE were determined with a solid-phase two-step chemiluminescent immunoassay on the Immulite 2000 (Siemens, Los Angeles, CA). A positive inhalation screen was defined as at least one increased amount of specific IgE against one of the following allergens:
Unadjusted between-group comparisons were performed using Student’s
A total of 452 subjects were included in the study (97 males and 355 females). Also 293 subjects (64.8%) fulfilled the criteria for metabolic syndrome. Table
General characteristics of the population included in the study.
No metabolic syndrome ( |
Metabolic syndrome ( |
|
|
---|---|---|---|
Age (years) | 37 ± 11 | 43 ± 10 | <0.0011 |
BMI (kg/m2) | 45.4 ± 6.7 | 46.0 ± 6.5 | 0.395 |
Gender (% female) | 87.4% | 73.7% | 0.0012 |
Ethnicity (% Caucasian) | 80.5% | 85.0% | 0.222 |
Smoking | 0.152 | ||
% never smoked | 49% | 40% | |
% stopped smoking | 29% | 32% | |
% smokes | 22% | 28% | |
Pack years3 | 6.0 ± 11.2 | 9.9 ± 14.8 | 0.003 |
Abdominal circumference (cm) | 126.8 ± 14 | 134.3 ± 16 | <0.001 |
Systolic blood pressure (mm Hg) | 136.4 ± 17.9 | 145.98 ± 16.6 | <0.001 |
Diastolic blood pressure (mm Hg) | 83.6 ± 11.8 | 87.83 ± 10.1 | <0.001 |
Comorbidities | |||
Diabetes mellitus | 3.8% | 30.3% | <0.001 |
Hypertension | 18.9% | 42.8% | <0.001 |
Hypercholesterolemia | 1.9% | 21.7% | <0.001 |
Self-reported asthma | 19.6% | 21.0% | 0.723 |
Use of inhaled corticosteroids | 3.8% | 5.5% | 0.413 |
1Data presented as mean ± standard deviation,
2Data presented as percentage,
3Data were log transformed for statistical analysis.
Patients with the metabolic syndrome did not have a higher total leukocyte count or neutrophil percentage in the peripheral blood (
Comparison of blood parameters between subjects with and without metabolic syndrome.
No metabolic syndrome ( |
Metabolic syndrome ( |
|
|
---|---|---|---|
Leukocytes (109/L) | 8.6 ± 2.3 | 8.9 ± 2.1 | 0.2531 |
Neutrophils (%) | 70.1 ± 8.6 | 68.9 ± 9.1 | 0.186 |
Lymphocytes (%) | 23.7 ± 7.1 | 23.8 ± 7.0 | 0.844 |
Monocytes (%) | 4.9 ± 1.8 | 5.3 ± 1.9 | 0.044 |
Eosinophils (%) | 0.82 (0.05–1.07) | 1.00 (0.45–1.85) | 0.0022 |
CRP (mg/L)3 | 9.6 ± 7.4 | 9.5 ± 8.0 | 0.865 |
Cholesterol (mmol/L) | 5.2 ± 1.0 | 5.1 ± 1.1 | 0.198 |
HDL-cholesterol (mmol/L) | 1.3 ± 0.3 | 1.1 ± 0.2 | <0.001 |
LDL-cholesterol (mmol/L) | 3.5 ± 0.9 | 3.3 ± 1.1 | 0.108 |
Triglyceride (mmol/L) | 1.0 ± 0.4 | 1.6 ± 1.0 | <0.001 |
Glucose (mmol/L) | 6.3 ± 1.3 | 8.2 ± 3.9 | <0.001 |
HbA1C (%) | 5.52 ± 0.4 | 6.4 ± 1.5 | <0.001 |
Insulin (mE/L)3 | 56.2 ± 58.1 | 75.0 ± 79.4 | <0.001 |
Vitamin D (nmol/L) | 39.5 ± 22.9 | 38.5 ± 17.7 | 0.661 |
IgE (kU/L)3 | 213.5 ± 479.9 | 197.8 ± 391.3 | 0.141 |
Positive inhalation screen | 46.2% | 44.0% | 0.694 |
1Data presented as mean ± standard deviation,
2Data presented as median (1st–3rd quartiles),
3Data were log transformed for statistical analysis.
The subjects with the metabolic syndrome showed a significantly lower FEV1/FVC ratio—a measure for bronchial obstruction—compared to the group without the metabolic syndrome (76.7% and 78.2%, resp.,
Pulmonary function test of subjects with and without metabolic syndrome.
No metabolic syndrome ( |
Metabolic syndrome ( |
|
|
---|---|---|---|
FEV1 (% predicted) | 93.0 ± 13.9 | 91.2 ± 14.5 | 0.2063 |
FVC (% predicted) | 100.6 ± 14.0 | 98.5 ± 14.7 | 0.206 |
FEV1/FVC | 78.2 ± 6.9 | 76.7 ± 6.4 | 0.032 |
FEF25–75 (% predicted) | 77.4 ± 24.0 | 73.9 ± 23.3 | 0.152 |
Reversibility FEV1 > 12% | 6% | 7% | 0.5232 |
FeNO (ppb)1 | 19.3 ± 22.4 | 17.6 ± 12.6 | 0.541 |
1Data were log transformed before comparison.
2Data presented as percentage,
3Data presented as mean ± standard deviation,
FEV1: forced expired volume in 1 second; FVC: forced vital capacity; FEF25–75: forced expiratory flow 25%–75%; TLC: total lung capacity; FeNO: fractional expiratory nitric oxide.
There was no difference in the prevalence of self-reported asthma between subjects with or without the metabolic syndrome (21.0% versus 19.6%, resp.,
In a subgroup analysis of subjects with and without reversibility (
Since the FEV1/FVC ratio was the only variable of spirometric function tests which was different between subjects with and without the metabolic syndrome and our hypothesis was that the metabolic syndrome causes a lower FEV1/FVC ratio—not the other way around—we investigated which variables were related to the FEV1/FVC ratio.
There was an association between the peripheral blood eosinophils percentage and the FEV1/FVC ratio (univariate linear regression coefficient = −0.806,
Univariate linear regression analysis of FEV1/FVC.
Regression coefficient | Standard error |
|
|
---|---|---|---|
Monocytes (%) | −0.079 | 0.162 | 0.626 |
Eosinophils (%) | −0.806 | 0.290 | 0.006 |
CRP (mg/mL) | 0.075 | 0.045 | 0.098 |
Abdominal circumference (cm) | −0.058 | 0.021 | 0.006 |
Body mass index (kg/m2) | 0.041 | 0.047 | 0.387 |
OSAS (Epworth Sleepiness Scale)1 | 0.311 | 0.141 | 0.029 |
GERD (GERD questionnaire) | 0.068 | 0.157 | 0.667 |
1Missing values
2Data presented as linear regression coefficient,
OSAS: obstructive sleep apnea syndrome; GERD: gastro esophageal reflux disease.
Since the metabolic syndrome was associated with the FEV1/FVC ratio, we investigated which of the components of the metabolic syndrome contributed to this relationship. Only hypertension was significantly related to FEV1/FVC ratio (regression coefficient = −1.612,
Univariate regression analysis of relation of components of metabolic syndrome and FEV1/FVC ratio.
Regression coefficient | Standard error |
|
|
---|---|---|---|
Abdominal circumference | NA | ||
Hypertriglycemia | −1.119 | 0.739 | 0.130 |
Low serum HDL-cholesterol | 0.712 | 0.686 | 0.300 |
Hypertension | −1.612 | 0.774 | 0.038 |
High serum glucose | −0.010 | 0.753 | 0.989 |
1Data presented as linear regression coefficient,
Abdominal circumference is not applicable, because all patients fulfill the criteria for abdominal circumference according to the NCEP-ATP III criteria for metabolic syndrome.
After correction for the use of inhaled corticosteroids and the number of pack years, we found an association between blood eosinophils and the FEV1/FVC ratio (adj. B −0.113,
Multiple regression analysis FEV1/FVC ratio.
Whole group ( |
|||
---|---|---|---|
Regression coefficient | 95% CI |
|
|
Eosinophils (%) | −0.113 | −1.247–−0.118 | 0.018 |
Use of inhalation corticosteroids | −0.025 | −3.635–2.122 | 0.606 |
Pack years | −0.259 | −0.162–−0.076 | <0.001 |
Variables associated with FEV1/FVC ratio in univariate analysis at a
This study shows that obese patients with the metabolic syndrome have a higher proportion of blood monocytes and eosinophils and a lower FEV1/FVC ratio, indicating airway obstruction, than obese patients without the metabolic syndrome. Blood eosinophils (%) in morbidly obese subjects were related to FEV1/FVC, whereas monocytes were not. After adjustment for multiple variables, the relationship between FEV1/FVC ratio and eosinophils remained intact. Although the differences are small, it strengthens our hypothesis that the presence of the metabolic syndrome could play a role in lung function impairment, through the induction of systemic inflammation, in particular mediated by blood eosinophils. Whether this also leads to asthma on the long term still remains to be elucidated.
To our knowledge, this is the first study concerning lung function and the metabolic syndrome in a cohort of only morbidly obese subjects. Secondly, comorbid conditions associated with obesity such as OSAS and GERD were taken into account in the current study. Thirdly, adiposity was not only assessed with BMI, but also with abdominal circumference. Although one would expect that all our subjects have the metabolic syndrome, we found a 65% prevalence of the metabolic syndrome in our group, which corresponds with 60% of the morbidly obese in the NHANES III cohort [
Various studies have shown that obesity causes a modest reduction in total lung capacity (TLC) and a larger reduction in functional residual capacity (FRC) [
The prevalence of self-reported asthma was similar in the two groups. We did not perform methacholine-provocation tests; so, a definitive diagnosis of asthma was often not possible. Since misdiagnosis of asthma is a relevant issue, also in the obese [
Since the metabolic syndrome was associated with the FEV1/FVC ratio in univariate analysis, we investigated which of the components of the metabolic syndrome contributed to this relationship. Hypertension was the only component that was significantly associated with a reduced FEV1/FVC ratio after multiple backward regression analysis. Hypertension may have the strongest association with systemic inflammation, as proposed by Irace et al. [
Obesity is a state of chronic low-grade systemic inflammation. Leukocyte count is considered a marker of systemic inflammation. Several epidemiological studies have already noted a relationship between some components of metabolic syndrome and leukocytes [
The question remains why there is relative eosinophilia in the obese. Traditionally, eosinophils are related to allergic diseases, asthma, and parasitic infections. The fat tissue is a source of adipokines, which are considered to play a role in the low-grade systemic inflammation in obesity [
Another question is whether this relative eosinophilia in the peripheral blood also has local effects on the bronchial tissue and causes or enhances the bronchial inflammation as seen in asthma. Asthma-like symptoms are common in patients with the metabolic syndrome, and their pulmonary function is impaired [
Our study includes several limitations. We did not measure TLC values; so, we were not able to firmly assess a restrictive lung function pattern. Also, we did not perform methacholine-provocation tests; so, a definite diagnosis of asthma was often not possible. The FEV1/FVC ratio, however, is easy to measure and widely used in clinical practice. Although the differences in FEV1/FVC and eosinophils between subjects with and without the metabolic syndrome were small, the difference was significant despite the fact that our study groups consisted of unselected subjects. There probably is a selection bias since we only included patients who were willing to undergo bariatric surgery. It is widely known that most of the subjects who undergo bariatric surgery are females. This explains the female predominance among our subjects. Furthermore, we did not measure adipocytokines or high-sensitivityCRP. We did not include a non-obese control group as comparison for low-grade inflammation, and we cannot fully exclude that smoking might have contributed to a state of low-grade inflammation. However, we corrected for smoking in the multiple regression analysis. Finally, we realize that a cross-sectional association between metabolic syndrome and lung function cannot establish causality. However, Naveed et al. [
In summary, our study shows that there is a small, but statistically significant, difference in eosinophils and FEV1/FVC between subjects with and without the metabolic syndrome. After correction for other variables, an association between blood eosinophils and FEV1/FVC remained. Although the differences we have found were relatively small, it might support our hypothesis that the presence of the metabolic syndrome may influence lung function impairment, through the induction of systemic inflammation, in particular, mediated by blood eosinophils. Further research with a firm diagnosis of asthma and assessment of peripheral airway inflammation is necessary. Moreover, in order to establish causality between the metabolic syndrome and lung function impairment, longitudinal studies in morbidly obese patients before and after bariatric surgery are needed.
Body mass index
C-reactive protein
Epworth Sleepiness Scale
Forced expiratory flow 25%–75%
Fraction of exhaled nitric oxide
Forced expiratory volume in 1 second
Forced vital capacity
Gastroesophageal reflux disease
High-density lipoprotein
National Cholesterol Education Program’s Adult Treatment Panel III report
Obstructive sleep apnea syndrome.
The authors declare no conflict of interests.
The authors wish to thank Mrs. Sandra Reijnhart for editing the paper. The authors are grateful for the help of all the staff in the Respiratory Laboratory and members of the Bariatric Surgery Team at Sint Franciscus Gasthuis. This research was supported by grants from Foundation of Research and Development, Department of Internal Medicine, Sint Franciscus Gasthuis (Stichting Onderzoek en Ontwikkeling Interne Specialismen Sint Franciscus Gasthuis).