Digital histology quantification of intra-hepatic fat in patients undergoing liver resection. Issue 8 (August 2015)

• Record Type: Journal Article
• Authors: Parkin, E.; O'Reilly, D.A.; Plumb, A.A.; Manoharan, P.; Rao, M.; Coe, P.; Frystyk, J.; Ammori, B.; de Liguori Carino, N.; Deshpande, R.; Sherlock, D.J.; Renehan, A.G.
• Is Part Of: European journal of surgical oncology. Volume 41:Issue 8(2015:Aug.)
• Page Start: 1020
• Page End: 1027
• Publication Date: 2015-08
• Subjects: Oncology -- Periodicals; Cancer -- Surgery -- Periodicals; Medical Oncology -- Periodicals; Neoplasms -- surgery -- Periodicals; Cancer -- Chirurgie -- Périodiques; Cancérologie -- Périodiques; Oncologie; Chirurgie (geneeskunde); Electronic journals; Electronic journals -- Sciences; Electronic journals -- Medicine; Electronic journals; 616.994059005
• DOI: 10.1016/j.ejso.2015.05.003
• Languages: English
• ISSNs: 0748-7983

Abstract

<abstract xml:lang="en" abstract-type="author" id="abs0010"> <title id="sectitle0010">Abstract</title> <sec> <title id="sectitle0015">Background</title> <p id="abspara0010">High intra-hepatic fat (IHF) content is associated with insulin resistance, visceral adiposity, and increased morbidity and mortality following liver resection. However, in clinical practice, IHF is assessed indirectly by pre-operative imaging [for example, chemical-shift magnetic resonance (CS-MR)]. We used the opportunity in patients undergoing liver resection to quantify IHF by digital histology (D-IHF) and relate this to CT-derived anthropometrics, insulin-related serum biomarkers, and IHF estimated by CS-MR.</p> </sec> <sec> <title id="sectitle0020">Methods</title> <p id="abspara0015">A reproducible method for quantification of D-IHF using 7 histology slides (inter- and intra-rater concordance: 0.97 and 0.98) was developed. In 35 patients undergoing resection for colorectal cancer metastases, we measured: CT-derived subcutaneous and visceral adipose tissue volumes, Homeostasis Model Assessment of Insulin Resistance (HOMA-IR), fasting serum adiponectin, leptin and fetuin-A. We estimated relative IHF using CS-MR and developed prediction models for IHF using a factor-clustered approach.</p> </sec> <sec> <title id="sectitle0025">Results</title> <p id="abspara0020">The multivariate linear regression models showed that D-IHF was best predicted by HOMA-IR (Beta coefficient<sub>per doubling</sub>: 2.410, 95% CI: 1.093, 5.313) and adiponectin (β<sub>per doubling</sub>: 0.197, 95% CI: 0.058, 0.667), but not by anthropometrics. MR-derived IHF correlated with D-IHF (rho: 0.626; p = 0.0001), but levels of agreement deviated in upper range values (CS-MR over-estimated IHF: regression versus zero, p = 0.009); this could be adjusted for by a correction factor (CF: 0.7816).</p> </sec> <sec> <title id="sectitle0030">Conclusions</title> <p id="abspara0025">Our findings show IHF is associated with measures of insulin resistance, but not measures of visceral adiposity. CS-MR over-estimated IHF in the upper range. Larger studies are indicated to test whether a correction of imaging-derived IHF estimates is valid.</p> </sec> </abstract>