Longitudinal and Cross-Sectional Assessment of Proton Density Fat Fraction and Metabolic Syndrome in Obese Patients undergoing Weight Loss Surgery
Curtis N. Wiens1, Alan B. McMillan1, Nathan S. Artz1,2, William Haufe3, Camilo A. Campo1, Alexandria Schlein3, Luke Funk4, Jacob Greenberg4, Guilherme M. Campos5, Claude Sirlin3, and Scott B. Reeder1,6,7,8,9

1Radiology, University of Wisconsin, Madison, WI, United States, 2Diagnostic Imaging, St. Jude Children's Research Hospital, Memphis, TN, United States, 3Radiology, University of California, San Diego, CA, United States, 4Surgery, University of Wisconsin, Madison, WI, United States, 5Surgery, Virginia Commonwealth University, Richmond, VA, United States, 6Medical Physics, University of Wisconsin, Madison, WI, United States, 7Biomedical Engineering, University of Wisconsin, Madison, WI, United States, 8Medicine, University of Wisconsin, Madison, WI, United States, 9Emergency Medicine, University of Wisconsin, Madison, WI, United States

Synopsis

The purpose of this work was to determine the relationship between proton density fat fraction (PDFF) and metabolic syndrome (MetS) in obese patients. Patients were recruited for an MRI study 1-2 days prior to weight loss surgery (WLS). A subset of these patients with biopsy confirmed hepatic steatosis were recruited for a second MRI 6 months post WLS. A cut-off PDFF of 7.5% had a sensitivity of 77% and specificity of 80% for predicting MetS prior to undergoing WLS. At 6 months follow-up, patients with confirmed hepatic steatosis had significantly decreased prevalence of MetS (91% to 52%). Additionally, other metabolic, biometric, and imaging (PDFF) markers related to MetS were significantly reduced.

Introduction

Metabolic syndrome (MetS) is a set of conditions that when present together increase the risk of cardiovascular disease and type-2 diabetes. The International Diabetes Federation defines Metabolic Syndrome as central obesity with elevated ethnic- and sex-specific waist circumference (WC), elevated triglyercerides (TG), low high density lipoproteins (HDL), elevated blood pressure (BP), and elevated fasting plasma glucose (1). MetS is strongly associated with hepatic steatosis (2,3) for which proton density fat fraction (PDFF), quantified using chemical shift encoded MRI (CSE-MRI), has been validated as a biomarker of hepatic steatosis. The main objective of this work was to determine PDFF thresholds that are predictive of the presence or absence of MetS in obese patients prior to undergoing bariatric weight loss surgery (WLS). A secondary purpose was to assess longitudinal changes in both PDFF and MetS as a result of WLS.

Methods

Patients undergoing clinical WLS ( sleeve gastrectomy or Roux-en-Y gastric bypass) were recruited at one institution for an IRB-approved MRI study 1-2 days prior to surgery (Visit 1). The subset of these patients with biopsy confirmed hepatic steatosis was recruited for a second MRI 6 months after WLS (Visit 2). Metabolic Syndrome was diagnosed using the International Diabetes Federation definition of MetS (1).

MRI studies were performed on either a 1.5T or 3T system (Signa HDxt or MR750, GE Healthcare, Waukesha. WI). At both time points, CSE-MRI fat quantification was performed using the following imaging parameters (1.5T/3T): 6 echoes, TR=13.4/8.6ms, ΔTE=2.0/1.0ms, flip angle=5°/3°, resolution=1.7x2.8x8 /1.7x3.4x8mm.

At both visits, physical measurements (blood pressure, waist circumference) and laboratory tests (fasting glucose, high density lipoproteins, triglycerides) were collected. 2 patients with Type-1 Diabetes or excess alcohol consumption (>1.5 drinks per day) were excluded from analysis.

Wilcoxon rank sum tests were used to test for statistical differences between patients with and without MetS at Visit 1 and statistical differences between measurements made at Visit 1 and 2. At Visit 1, receiver operating characteristics (ROC) were analyzed to determine PDFF’s ability to distinguish patients with or without MetS.

Results

57 patients were successfully recruited for Visit 1 (47 females, 9 males, age=50.7±12.3 years, MetS prevalence=70%) while 33 patients (with biopsy confirmed hepatic steatosis) were recruited for Visit 2 (29 females, 4 males, age= 50.4±12.1 years).

While BMI and body weight were similar for patients with and without MetS at Visit 1, PDFF of patients at Visit 1 with MetS was significantly higher (Figure 1) than of those without MetS (PDFFMetS=13.9±8.8%, PDFFNoMetS=5.6±3.2%, p-value=0.0003). Aside from PDFF, MetS risk factors of TG, HDL, systolic BP, and glucose all showed significant differences in patients with MetS present (Table 1). ROC analysis demonstrated that a PDFF threshold of 7.5% could distinguish patients with and without MetS at Visit 1 with a sensitivity=77%, specificity=80%, and an area under the curve =0.83 (Figure 2).

Substantial reductions in the prevalence of MetS were observed. Specifically, the prevalence of MetS was 90% at visit 1 and 52% at visit 2. Further, the average PDFF decreased significantly, from 15.0±4.7% at visit 1 to 4.7±3.5% at visit 2 (p-value=7.63e-9).

Other measures that changed significantly from Visit 1 to Visit 2 include weight, BMI, WC, TG, and HDL (Table 2). At Visit 2, patients with MetS had a non-statistically higher PDFF than patients without (PDFFMetS=5.1±3.3, PDFFNoMetS=4.3±3.7, p-value=0.31).

Discussion and Conclusion

Elevated PDFF (greater than 7.5%) is predictive of MetS in obese subjects prior to WLS with high AUC (0.83). Following WLS significant decreases in PDFF and the prevalence of MetS and many of its related factors (WC, TG, HDL, Glucose) were observed. Further, significant changes in BMI, Weight, and PDFF were also detected. However, PDFF was not found to be predictive of MetS at 6 months follow up after weight loss surgery. Further study is necessary to understand the time course of PDFF changes relative to metabolic changes following WLS.

Acknowledgements

The authors acknowledge the support of NIH (R01 DK083380, R01 DK088925), NSERC, and GE Healthcare.

References

1. The IDF consensus worldwide definition of the metabolic syndrome. http://www.idf.org/webdata/docs/IDF_Meta_def_final.pdf.

2. Kotronen A, Westerbacka J, Bergholm R, et al. Liver Fat in the Metabolic Syndrome. The Journal of Clinical Endocrinology & Metabolism. 2007;92:3490–3497.

3. Rehm JL, Wolfgram PM, Hernando D, et al. Proton density fat-fraction is an accurate biomarker of hepatic steatosis in adolescent girls and young women. Eur Radiol. 2015;25:2921–2930.

Figures

Figure 1: At Visit 1, the PDFF of patients with MetS is significantly higher than of those without MetS (PDFFMetS=13.9±8.8%, PDFFNoMetS=5.6±3.2%, p-value=0.0003). Sample PDFF maps and distributions for patients with and without MetS are shown (blue X’s recruited for Visit 2, red O’s were not).

Table 1: PDFF and MetS risk factors TG, HDL, Systolic BP, and Glucose were statistically difference in patients with MetS present and absent.

Figure 2: Receiver operating characteristic show a moderate ability to distinguish bariatric patients with and without metabolic syndrome at surgery (AUC=0.83). A cut-off PDFF of 7.5% has a specificity = 80% and a sensitivity = 77%.

Figure 3: At visit 2, just 6 months after surgery, substantial reductions in PDFF are observed (15.0±4.7%, 4.7±3.5%, p-value=8e-9). Sample PDFF maps and PDFF trajectories are shown for 33 patients at visit 1 and 2.

Table 2: At Visit 2, in subjects with hepatic steatosis, the prevalence of MetS had reduced from 91% to 52%. Furthermore, a statistically significant change in PDFF, weight, BMI, WC, TG, and HDL were observed.



Proc. Intl. Soc. Mag. Reson. Med. 24 (2016)
3923