The neck-liver axis. Madelung disease as further evidence for an impact of body fat distribution on hepatic histology

Cheung1 reported dorsocervical lipohypertrophy to correlate with the histological severity — except for steatosis — and substantially contribute to insulin resistance (IR) in nonalcoholic fatty liver disease (NAFLD). The mechanisms underlying adipose tissue accumulation in the dorsocervical area (DCA) and the hepatic consequences of its removal are unknown. Therefore, a human disease featuring adipose tissue expansion in the neck area would represent a suitable model to explore the “neck-liver axis”. Madelung's disease (MD) might be one such model. Described by Brodie in 1846,2 MD is a rare alcohol-related3 multiple symmetric benign lipomatosis specifically involving the DCA. It primarily affects adult Mediterranean males and strongly resembles the “buffalo hump syndrome” observed in HIV infection.4 Liver disease is expected to affect patients with MD given that they display the obesity-and-alcohol abuse risk concurrence.5 However, little is known about the interaction of cytokine abnormalities, obesity, and liver disease after removal of the neck adipose tissue. A 49-year-old alcoholic man with MD (Fig. 1) fulfilling ATP III criteria6 for arterial hypertension, waist girth and HDL cholesterol, and showing bull-like neck due to subcutaneuos lipoma in DCA, quit drinking alcohol when first observed in 2001. Subsequently, the patient was subjected to anthropometric laboratory evaluation and paired liver biopsy. Owing to neurological and respiratory complications, surgical removal of 5kg adipose tissue from the neck area was performed at 2 times (2005-2007). Follow-up of this patient (June 2001– June 2007) disclosed increasing body weight (90-148 kg) coupled with decreasing triglyceride serum levels (90-38mg/dL), as if the expanded adipose tissue were avidly absorbing circulating triglycerides, thus removing them from the bloodstream.7 At difference, in the same follow-up period an increased total (91-131 mg/dL), HDL (9-30 mg/dL) and LDL (32-87 mg/dL) cholesterol was observed. The figure highlights the dorsocervical lipohypertrophy, supporting the “buffalo hump” appearance in our patient with MD. Compared to presurgery evaluation (July 2005), post-surgery (June 2007) HOMA (homeostasis model assessment) (6.42 to 8.13) and adiponectin (2.79 to 5.49 μg/mL) increased and leptin decreased (70.54 to 45.38 ng/mL). Paired liver biopsy showed alcoholic steatohepatitis in advanced fibrotic stage (2001) and precirrhotic changes without steatosis (2006), as seen in NAFLD.8 The role for genetics in fat accumulation in MD is shown by ongoing lipoma formation after alcohol abstinence and in the absence of any detectable changes among the hormones implicated in the pathogenesis of NAFLD.9 In conclusion, Dr. Cheung's study1 and the findings in our patient strongly suggest the existence of a neck-liver axis. Irrespective of the geographic area (United States or Mediterranean) and of the clinical context in which it is observed (NAFLD or MD), subcutaneous adipose tissue accumulation electively distributed in the neck-DCA appears to be associated with hepatic inflammatory fibrosis in the absence of fatty changes, such as seen in animals in which steatosis has been inhibited.10 Taken collectively, clinical and animal studies support, therefore, hepatic steatosis to be a protective mechanism against free fatty acid toxicity in the liver. Conversely, selective accumulation of fateb in the neck-DCA could be a risk factor for inflammatory-fibrotic hepatic changes. Stefano Ballestri*, Amedeo Lonardo*, Lucia Carulli*, Matteo Ricchi*, Lorena Bertozzi*, Giorgio De Santis , Mario Bondi*, Paola Loria*, * Università di Modena e Reggio Emilia, Department of Internal Medicine, Endocrinology, Metabolism and Geriatrics, Nuovo Ospedale Civile S. Agostino Estense di Baggiovara, Modena, Italy, Department of Surgery, Azienda Ospedaliero-Universitaria Policlinico di Modena, Modena, Italy.