Absence of 18 F-fluorodeoxyglucose uptake using Positron Emission Tomography/Computed Tomography in Madelung's disease: A case report.

Madelung's disease is characterized by the manifestation of multiple ectopic lipomas, usually found in the cervical-thoracic region, however, clinical manifestation may vary among patients. It has been postulated that lipomas associated with Madelung's disease are linked to brown adipose tissue (BAT) due to the presence of uncoupling protein 1 (UCP1). Therefore, we here investigated whether BAT activity is present in a patient with Madelung's disease. 18 F-fluorodeoxyglucose (18 F-FDG) uptake using PET/CT after a cooling procedure was measured together with body temperature and energy expenditure. Finally, adipose tissue biopsies were taken from the lipomas for gene expression analysis and histology. 18 F-FDG uptake was not detected after the cooling procedure in the lipomas. Furthermore, adipose tissue biopsies derived from the lipomas did not express UCP1. We thus conclude that cold-stimulated BAT activity was not detected in lipomas associated with Madelung's disease. Additional research in other patients is needed to unravel the role of dysfunctional BAT in Madelung's disease.

Multiple symmetric lipomatosis or Madelung's disease is characterized by symmetric development of lipomas in the cervical‐thoracic region.1 Madelung's disease mainly affects males and is associated with alcohol abuse early in life.2, 3, 4 Clinically, patients can be divided in two sub‐groups: Type 1, with symmetrical gross masses in the upper part of the body and Type 2, with more general obesity.5

Uncoupling protein 1 (UCP1) has been detected in biopsies taken from lipomas in patients suffering from Madelung's disease.6 Normally, mitochondria build up a proton gradient across the membrane by oxidation of substrates which is used to produce ATP. Adrenergic activation of UCP1 uncouples this process resulting in heat production instead of ATP production.7 UCP1 is normally specific for brown adipose tissue (BAT), which could mean that BAT is present in the lipomas. However, cells derived from these lipomas do not respond to adrenergic receptor stimulated lipolysis,6 and this may indicate reduced fatty acid availability and related disturbed BAT metabolism. This would impair mitochondrial uncoupling and BAT thermogenesis, potentially reducing the body's protection against cold together with an elevated sympathetic nervous system activity. This could in turn lead to proliferation of adipocytes.6 In other samples taken from lipomas there was also evidence for mitochondrial dysfunction due to mutations in the mitochondrial DNA.8

Expression of UCP1 in lipomas associated with Madelung's disease suggests the presence of BAT. However, since this BAT could be dysfunctional, we hypothesize no BAT activity will be measured via uptake of 18F‐fluorodeoxyglucose (FDG) following cold exposure. Furthermore, we expect to detect UCP1 mRNA in adipose tissue biopsies derived from the lipomas.

A 68‐year‐old male patient (body mass index, BMI: 36.2 kg/m2), with clinical diagnosis of Madelung's disease (Type 1) was referred to Maastricht University Medical Centre for detection of active BAT. The patient extensively used alcohol from the age of 16. He suffered from symmetric lipomatosis mainly around the upper arms and abdomen, with relative sparing of the cervical region (Figure 1A). Recent laboratory results showed a near‐to‐normal thyroid function (TSH: 4.8 mIU/L [ref 0.3‐4.6 mIU/L]; fT4: 15.3 pmol/L [ref 10‐23 pmol/L]). The patient had not used beta‐blockers in the last year.

Figure 1

(A) Clinical view of the patient. Noticeable is the subcutaneous lipomatosis in both upper arms and abdomen, with relative sparing of the cervical region. (B) PET image using 18F‐FDG following the cooling procedure. The 18F‐fluorodeoxyglucose (FDG) uptake in the cervical region is caused by active muscle tissue due to shivering (filled arrows). Note the absence of 18F‐FDG uptake in the deep cervical adipose tissue and lipomas in both upper arms (clear arrows)

BAT presence and activity was determined in the patient using an individualized cooling procedure and 18F‐FDG‐Positron Emission Tomography/Computed Tomography (PET/CT) (Gemini TF PET/CT; Philips, Amsterdam, Netherlands) as described before.9 The patient arrived at the research institute at 8:30 am, after consuming a light breakfast at 6:00 am. At 8:45 am, the patient ingested a telemetric pill (CorTemp HT150002; HQ Inc., Palmetto, Florida). At the same time wireless temperature sensors (iButton; Maxim Integrated Products, San Jose, California) were attached at the 15 International Organization for Standardization (ISO) defined sites to measure skin temperature. An intravenous cannula was placed in the right antecubital vein. The patient was wrapped in a water‐perfused suit (ThermoWrap Universal 3166; MTRE Advanced Technologies Ltd., Yavne, Israel), which was connected to two cooling installations (Blanketrol III; Cincinnati Sub‐Zero, Sharonville, Ohio).

The water‐perfused suit was heated to 36°C to create thermoneutral conditions. Indirect calorimetry was measured using a ventilated hood system (EZCAL; Maastricht Instruments, Maastricht, Netherlands). After this thermoneutral period, the cooling procedure was started. The water temperature was lowered to 4°C each 15 minutes, until shivering. Following this, the temperature was raised to 34°C for 5 minutes to stop the shivering. Then temperature was maintained at 27°C, which was followed by a second indirect calorimetry measurement. At 12:45 pm, a bolus of 77 MBq of 18F‐FDG was injected intravenously after which the patient was instructed to lie down still for 1 hour.

After 1 hour, at 13:45 pm, the PET/CT protocol was performed with a low‐dose CT scan (120 kV, 30 mAs), immediately followed by a static PET scan (6 bed positions, 4 minutes per position) covering the area from the skull to the illiac crest. After the 18F‐FDG‐PET/CT scan, two adipose tissue biopsies were taken under local anaesthesia. One biopsy was taken at the lateral side of the right upper arm from the lipoma. The second biopsy was taken from the abdominal subcutaneous fat deposit, just right of the umbilical. The adipose tissue samples were either paraffin‐embedded or frozen in melting isopentane prior to further analysis. RNA was extracted using Trizol reagent followed by protocol described in the RNeasy kit from Qiagen (Hildenberg, Germany). UCP1 mRNA expression was determined using a CFX384 Touch Real‐Time PCR Detection System from BioRad Laboratories (Hercules, California) with a Taqman gene expression assay (Hs00222453_m1) as described before.10

Cold exposure increased resting energy expenditure from 5.2 to 9.7 kJ/min. Although the protocol was designed to induce non‐shivering thermogenesis, the patient did shiver occasionally in the final stages of the cooling procedures. Non‐shivering thermogenesis plus some shivering thermogenesis (as the per cent increase in energy expenditure during cold exposure above the resting energy expenditure) was 53.6%. The respiratory quotient remained 0.83. Core temperature was 37.3°C during the thermoneutral period and dropped to 36.5°C during cold exposure, while the mean skin temperature dropped from 34.0°C to 32.1°C.

The 18F‐FDG‐PET/CT scan showed no uptake of 18F‐FDG in the adipose tissue depots in the cervical‐thoracic region or in both upper arms (Figure 1B). Adipose tissue biopsies from the arm (Figure 2) and the abdomen (Figure 3) were stained with haematoxylin‐eosin (HE), and both showed the microscopic structure of white adipose tissue (WAT). Next, we examined UCP1 mRNA expression in biopsies taken from the arm and the abdomen. Cq values for the samples derived from the lipoma and subcutaneous WAT were 32.9 and 33.7, respectively. As a reference, we included a human hibernoma, which had a Cq value of 19.4.

Figure 2

Adipose tissue from lipoma of right upper arm (magnification: ×200)

Figure 3

Subcutaneous white adipose tissue from abdomen (magnification: ×200)

Our patient showed normal physical responses to cooling, as seen in the sustained core temperature and the temperature gradient in the lower arm. Energy expenditure increased upon cold exposure as expected for normal thermoregulation including both non‐shivering and mild shivering.

As hypothesized, the 18F‐FDG‐PET/CT scan showed no uptake of the 18F‐FDG tracer in neither the cervical‐thoracic region nor the upper arms, where the lipomas were situated in our patient. Based on the hypothesis that a dysfunctional adrenergic pathway leads to less UCP1 expression and reduced or absent thermogenesis, this could explain the lack of tracer uptake in the normal BAT regions in our patient.

In line with the 18F‐FDG‐PET/CT results, both adipose tissue biopsies showed a typical morphology of WAT (Figures 2 and 3), which was negative for UCP1 mRNA. However, it is possible that our patient has another causal mechanism, explaining the lack of UCP1.5, 6, 8 This could potentially explain why we were unable to find any sign of BAT presence in the biopsy material.

The use of an 18F‐FDG‐PET/CT scan combined with the cooling procedure is an established method to detect the presence of active BAT.9 Generally, the prevalence is highest in young healthy (lean) adults, although BAT is not always detected in all lean subjects.10 The prevalence of BAT decreases with obesity and age.9 In view of the age and BMI of our patient, this may also explain the absence of active BAT.

The defective adrenergic pathway in Madelung's disease could also explain the absence of tracer uptake in both upper arms. In the adipose tissue biopsies, we only found cells resembling white adipocytes without any expression of UCP1 mRNA. An in vitro model with primary cultured adipocytes derived from patient adipose tissue could be used to examine UCP1 expression after adrenergic stimulation. However, these experiments would require freshly isolated adipose tissue biopsies.

Concluding, in a patient with clinical Madelung's disease we found no active BAT, neither in the lipomas nor in the cervical‐thoracic region. However, since this was a single case, the results in patients may differ, and this may not exclude BAT activity in other patients. Further research could focus on the causal mechanism for Madelung's disease and the possible relationship with BAT.

CONFLICTS OF INTEREST

No conflict of interest was declared.

Author contributions

M.P.B.M., M.J.P.G.K. and E.B.M.N. performed experiments. M.P.B.M., E.B.M.N., M.J.P.G.K., D.B. and W.D.V.M.L. designed the study and wrote the manuscript.