Changes in C-reactive protein, neopterin and lactoferrin differ after conservative and surgical weight loss in individuals with morbid obesity

Design and participants

Adults referred to the obesity centre at Innlandet Hospital Trust in South-Eastern Norway from December 2012 through September 2014 were eligible to this prospective study16. In the present sub-study, individuals with blood samples analysed for biomarkers and with complete data on weight, height, physical activity, and comorbidities comprising diabetes, hypertension and chronic obstructive pulmonary disease (COPD), were included. Information from three visits was used; a first visit to the hospital, a second visit around 5 months later after the conservative weight loss intervention and before the bariatric surgery, and a third visit 6 month after surgery.

Inclusion and exclusion criteria

Inclusion criteria were age between 18 and 65 years and MO, defined as BMI > 40 kg/m2 or BMI > 35 kg/m2 with obesity-related comorbidity17. Exclusion criteria were major psychiatric disorders, drug or alcohol addiction, and serious somatic disorders not judged as obesity-related, i.e. conditions that would preclude usual weight loss treatment at the clinic. The participants thus represented the usual selection of individuals treated at the clinic and were receiving standard care for any comorbidities. They were asked to report any ongoing illnesses, and blood sampling was performed during routine visits when they reported to be well.

The first visit

At the first study visit, clinical data were registered, and blood samples were drawn. Information about comorbidity was provided by the participant on a case report form. This information was reviewed by a physician with full access to the hospital records.

The conservative weight loss program

The participants underwent a five month long conservative weight loss intervention with changes in diet and physical activity18. This intervention was performed in groups and was based on a series of outpatient visits. It was comparable to weight loss programs used in other Norwegian hospitals in this time period19 with some small local changes due to availability of personnel. All included participants started with three hour-long consultations with three health professionals; a nurse, a nutritionist and a physician, which included personalized advice on diet and physical activity. These consultations were usually scheduled in three separate weeks to give the participant time to implement suggested changes. Subsequently, all participants were enrolled in a patient group with weekly four-hour meetings during seven consecutive weeks. The meetings included group counseling together with other participants with morbid obesity, and were led by nurses specialized in obesity, nutritioni sts, surgeons and a psychologist.

The dietary advice was based on reduction of total energy intake and choice of food rich in micronutrients18,20. The participants were recommended to eat more fiber and protein, less sugar and fat, and to distribute their food intake into 4–6 meals each day with 2–4 hour intervals between meals. Advice on specific dishes was individualized by a nutritionist based on the individual's former diet and food preferences. 21 days before the second study visit the participants were advised to follow a "crisp bread diet" containing 1000 kcal/day. The recommendations for this diet was intake of six pieces of crisp bread with low-fat high-protein topping (cheese, meat or fish), 4.5 dl of low-fat milk, a small dinner plate (meat or fish with vegetables), free amounts of vegetables (not corn, olives, or avocados) and free amounts of beverages without calorie content (preferentially above 2 litres). As an alternative to the crisp bread diet, the participants could use meal re placement powder giving 800–900 kcal with vegetables. All participants were told that acceptance to the clinic's public, free-of charge bariatric surgery program partly depended on their adherence to the given advice.

The second visit, before surgery

The second study visit took place after the conservative weight loss program, on average 23 weeks after the first visit, one week before planned bariatric surgery. Clinical data were registered, and blood samples were drawn.

The surgical procedure

The participants subsequently underwent bariatric surgery, either as laparoscopic Roux-en-Y Gastric Bypass21 or laparoscopic gastric sleeve22.

The follow-up visit six months after surgery

The third visit was a postoperative follow-up at the outpatient clinic for MO about six months after bariatric surgery, on average 49 weeks after the first visit. Clinical data were registered, and blood samples were drawn.

Variables

The present study used a subset of the variables in the main study. Age, sex, body mass index (BMI, calculated as the weight in kilograms divided by the square of the height in meters), comorbidities (chronic obstructive pulmonary disease, diabetes, hypothyroidism, previous myocardial infarction, angina, previous stroke, hypertension), total cholesterol, LDL cholesterol, HDL cholesterol and HbA1c were registered in addition to serum concentrations of the inflammatory biomarkers CRP, neopterin and lactoferrin.

CRP and HbA1c were analysed at Innlandet Hospital Trust Gjøvik with a Cobas c501 instrument with the reagents CRPL3 and Tina-quant HbA1C (Roche Diagnostics GmbH, Mannheim, Germany). Reference values were <5 mg/L for CRP and 4.0–6.0% for HbA1c. Neopterin was quantified using a commercial enzyme immunoassay (Genway Biotech, San Diego, USA). Lactoferrin was analysed using an enzyme immunoassay as previously reported23. For these assays, neopterin values above 10 nmol/L are evaluated as pathological. In a group of 302 blood donors, lactoferrin values were below <250 ug/L (V Videm, unpublished data). Total cholesterol, low- and high-density lipoprotein (LDL and HDL, respectively) cholesterol were analysed using standard methods at Innlandet Hospital Trust Gjøvik.

Statistical analysis

Data are given as mean (standard deviation – SD) or number (percentage). Statistical analysis was performed using mixed models. The time variable was coded as time 0 (first visit), 23 weeks (second visit) and 49 weeks (follow-up visit after bariatric surgery). Mixed models allow for simultaneous analysis of the three repeated (i.e. correlated) measurements in each individual and direct comparison of the changes after conservative weight loss with the changes following bariatric surgery. It also allows for individual trajectories of the variables in each individual, and different numbers of participants included at each time point due to attrition during follow-up. These are realistic assumptions for biomarkers of inflammation, which show large individual variations, and weight-loss interventions, where it is unrealistic that all participants will complete all steps of treatment.

Data were first analysed in univariate models with CRP, neopterin, lactoferrin, BMI, or HbA1c as the dependent variable, in order to investigate the observed values and changes after the two weight loss interventions. Each biomarker was thereafter analysed in a multivariable model including BMI and HbA1c to investigate associations of the biomarker changes with changes in adiposity and glucose tolerance, also including adjustments for age and sex to avoid confounding due to increased low-grade inflammation with older age or to sex-related differences. To evaluate whether the associations between each biomarker and BMI or HbA1c were different after the conservative and the surgical weight loss interventions, interaction terms with time were evaluated. Model fit was assessed using residual plots.

For an analysis of bivariate associations, Pearson's correlation coefficients were calculated between changes in CRP, neopterin and lactoferrin as well as between changes in each biomarker and changes in BMI or HbA1c after the two weight loss interventions. This method does not take within-person correlation into account. Therefore, unadjusted mixed models including one biomarker as the dependent variable and another as the independent variable as well as an interaction term with time were also analysed, thus accounting for within-person correlation. Likewise, unadjusted mixed models including each biomarker as the dependent variable and either BMI or HbA1c as the independent variable and an interaction term with time were analysed to assess unadjusted associations between changes in each biomarker and changes in adiposity or glucose tolerance. Overall association in each of these models was evaluated using the between-person Snijders/Bosker R2 for mixed models. P-values <� ��0.05 were considered significant, and all tests were two-sided. Stata (v15.1, College Station, Texas, USA) was used for all analyses.

To assess the robustness of the findings in the analyses to investigate associations of the biomarker changes with changes in BMI and HbA1c, we performed sensitivity analyses with additional adjustment for smoking (coded as ever, former or present smoker) or gastric sleeve vs. gastric bypass operation. We also analysed models where body weight was included instead of BMI.

Ethics

The study was approved by the Regional Committee for Medical and Health Research Ethics South East Norway (reference 2012/966) and conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from all participants included in the study.