Necrolytic migratory erythema (NME) is a characteristic skin rash most often associated with the glucagonoma, an alpha-cell tumor of the pancreatic islets. It is usually seen as a part of the glucagonoma syndrome, a paraneoplastic syndrome which includes the triad of diabetes mellitus, NME, and weight loss. However, more rarely, NME can also be seen as a part of other clinical entities such as liver disease and intestinal malabsorption; here it is termed the pseudoglucagonoma syndrome. The skin rash is characteristically extremely erythematous, shows superficial epidermal necrosis and often spreads in a centrifugal pattern. The condition is frequently misdiagnosed due in part to its rarity and often is the presenting symptom of the pancreatic islet neoplasm. Thus, timely recognition and treatment can help prevent morbidity from the epidermal dysfunction and mortality from the tumor burden itself, if the neoplasm is identified before metastasis.
The exact cause of necrolytic migratory erythema remains unknown. However, it is almost ubiquitous with the glucagonoma, where it can be the presenting symptom of the neoplasm in roughly 70% of cases. The glucagonoma, a pancreatic neuroendocrine tumor (pNET), represents a rare pancreatic tumor, encompassing less than 5% of all pancreatic neoplasms. The majority of glucagonomas are deemed to be sporadic. However, glucagonomas and the accompanying paraneoplastic NME rarely can be associated with multiple endocrine neoplasia type 1 (MEN1); this has been reported to be not more than 3% of all glucagonomas.
NME also occurs outside the context of a glucagonoma, and in these cases, it is termed “pseudoglucagonoma syndrome.” Pseudoglucagonoma syndrome has been seen to occur with a variety of different disease states such as intestinal malabsorption (Celiac disease), liver cirrhosis, inflammatory bowel disease (Crohn’s disease), chronic pancreatitis, and other malignancies (small cell lung cancer, duodenal neoplasms).
Necrolytic migratory erythema as a part of the glucagonoma syndrome is an exceedingly rare disease process with an estimated incidence of 1 in 20 million per year. In this context, NME has a peak presentation in the fifth decade of life, with an average age of onset of 52 years of age, coinciding with the peak onset of glucagonomas. However, NME has been reported in the literature in patients ranging from ages 15 to 88 years of age. Rarely, it has been seen in patients with MEN1, and in these cases, the onset of the disease is earlier. Gender predilection is unclear, as prior studies in the literature have suggested a female predominance with a female-to-male ratio of 3-4:1; however, a more recent literature review showed no gender specificity.
NME in the context of pseudoglucagonoma syndrome is an even rarer phenomenon, and as such data on its incidence and prevalence in this setting is not currently available.
The exact pathogenesis of necrolytic migratory erythema (NME) has yet to elucidated; however, hyperglucagonemia undoubtedly plays an important role, having been implicated in most cases. This is evident as tumor resection or medical therapy with glucagon antagonists often results in complete resolution of the skin symptoms, many times within one week of treatment. Elevated glucagon levels alone cannot explain the epidermal dysfunction seen in NME. Thus, how hyperglucagonemia contributes to the epidermal dysfunction and skin lesions seen in NME is likely the result of several interplaying factors including hypoaminoacidemia, deficiency of zinc and essential fatty acids, and the induction of inflammatory mediators in the epidermis.
In order to explain the interrelationship of the previously mentioned factors in the pathogenesis of NME, the multifactorial malnutrition model was created. This model is a prevailing theory as to how excess glucagon induces the changes seen in this disease state. The premise of this theory is that the catabolic state induced by excess serum glucagon results in derangement of a common metabolic pathway involving zinc, protein, amino acids, and essential fatty acids. This pathway is likely affected to due depletion of one or more of the above nutrients, thus one can classify NME as a deficiency dermatosis. Giving further credence to this theory is the fact that supplementation of both zinc and amino acids has been shown to resolve the dermatosis seen in NME.
Zinc is likely of major importance in the pathogenesis of NME, as its supplementation has been shown to resolve the skin lesions in those with glucagonoma syndrome. Zinc is an important component of metalloenzymes and transcription factors that regulate lipid, protein, and nucleic acid metabolism, as well as gene transcription. It is also involved in normal epidermal function, as evidenced by the dermatitis induced by its deficiency (acrodermatitis enteropathica). NME is similar histopathologically to the dermatitis seen in acrodermatitis enteropathica, and zinc deficiency is often observed in conjunction with NME. This can be either a true zinc deficiency secondary to malabsorption or a functional zinc deficiency due to hypoproteinemia as albumin is the main carrier of zinc in the serum. How, exactly, zinc deficiency predisposes a patient to NME remains unclear, but it is likely related to its metabolic functions.
Hypoproteinemia and hypoaminoacidemia are seen in NME and the glucagonoma syndrome and are likely a result of the catabolic state induced by excess glucagon. The hypoaminoacidemia is a result of the physiologic effect of glucagon on the liver to increase gluconeogenesis via amino acid catabolism. It is thought that the diminished availability of amino acids causes a decrease in peptide synthesis in tissues such as the epidermis which predisposes it to rapid cycling and destruction. The reduced pool of amino acids is also shown to lead to increased arachidonic acid production, and thus, an increased tendency toward inflammation in the epidermis. The hypoalbuminemia seen is likely secondary to the hypoaminoacidemia, a result of the diminished availability of the amino acids needed for its synthesis. Hypoproteinemia can result in a functional zinc deficiency, as well as a functional deficiency of essential fatty acids, as both are carried primarily by albumin in the blood. The multifactorial malnutrition model of NME emphasizes the interrelationship between all of these factors.
The deficiency of essential fatty acids that occurs in glucagonoma syndrome can be directly secondary to the hypoalbuminemia seen in the disease process; however this may be only one aspect of the phenomenon. Deficiency of fatty acids may also be secondary to the enhanced peripheral lipolysis resultant from the physiologic actions of glucagon on hormone-sensitive lipase in adipose tissue. It is also postulated that the zinc deficiency seen in NME can contribute, as it is a required co-factor in the delta-6 desaturation of linoleic acid, an essential fatty acid. If zinc is not present in sufficient quantities, then this metabolic pathway is blocked, resulting in a diversion of these precursors to the synthesis of arachidonic acid, and subsequently, increased production of inflammatory mediators such as prostaglandins and leukotrienes in the epidermis. This increase in inflammatory mediators is thought to be exacerbated by the hypoaminoacidemia and hypoproteinemia present in glucagonoma syndrome, as albumin normally sequesters fatty acids released by trauma and prevents their conversion to prostaglandins and leukotrienes. Thus, with decreased albumin, there is an increased tendency toward inflammation in the epidermis, and in the setting of trauma, one may see an exaggerated inflammatory response. This may explain why the distribution of the skin lesions seen in NME are primarily seen in areas of increased pressure and friction, for example, the perineum and lower extremities. Again the multifactorial malnutrition model shows the interrelationship between the nutrient deficiencies seen in glucagonoma syndrome and the predisposition to an inflammatory response in the epidermis.
The pathogenesis of NME in the context of pseudoglucagonoma syndrome has been less extensively studied due to the rarity of this condition. However, there are similarities to classical NME. One disease which NME has been seen without the presence of a glucagonoma is hepatic cirrhosis. It is well known that in the presence of cirrhosis the synthetic capacity of the liver is markedly reduced, resulting in hypoproteinemia and hypoaminoacidemia, which as noted above are likely related to the pathogenesis of NME seen in glucagonomas. In addition, the liver is involved with glucagon degradation, and in the presence of liver dysfunction glucagon levels can become elevated leading to NME. These 2 factors likely interplay and result in the clinical appearance of NME in this context.
Another spectrum of diseases that pseudoglucogonoma syndrome and NME occurs in are ones involving gastrointestinal dysfunction. Celiac disease, inflammatory bowel disease, and pancreatic insufficiency have all been reported to cause NME in the absence of glucagon-secreting tumors. These all can result in malabsorption of vital nutrients such as zinc and a state of malnutrition resembling that seen in glucagonoma syndrome (hypoaminoacidemia and hypoalbuminemia). Thus, the deficiency of these nutrients, as described above in the multifactorial malnutrition model, resultant from intestinal dysfunction can result in dermatitis indistinguishable from NME seen in the glucagonoma syndrome.
Although no histologic features are pathognomic for NME, some are very distinctive and help separate it from other dermatoses. It is important to recognize that the histopathologic changes are limited entirely to the epidermis, and in particular the superficial epidermis. In particular, the hallmark of NME is necrosis of the upper spinous layer of the epidermis. However, the finding of superficial epidermal necrolysis is relatively non-specific and may be only focally present; it is most likely to be found at the edge of an active lesion. Thus it is recommended to take multiple biopsies from the edges of multiple lesions when the diagnosis of NME is suspected.
Other prominent findings seen in NME include confluent parakeratosis and irregular acanthosis, with loss of the granular layer. At higher power, ballooning epidermal cells showing vacuolar degeneration, particularly in the upper spinous layer are seen. Lastly, it is possible to see mild perivascular lymphocytic or neutrophilic infiltrate along with intraepidermal bullae.
The glucagonoma syndrome is heralded by the triad of weight loss, diabetes mellitus, and necrolytic migratory erythema. NME is thought to be the presenting symptom of the disease in roughly 70% of patients. Diabetes mellitus, with its attendant symptoms such as polydipsia and polyuria, occurs in a majority of those with glucagonoma syndrome, with reports of up to 80% to 94% of patients affected. Weight loss is also prominent and can be severe in the later stages of the disease process.
On physical examination, patients with necrolytic migratory erythema will show erythematous scaly lesions with centrifugal growth, often with concomitant lesions in different stages of healing. The rash is prominent in areas of increased friction and pressure such as the intertriginous areas (i.e., inguinal creases, popliteal fossa), perineum, buttocks, groin, lower abdomen, and distal extremities. It is also commonly seen in a periorificial distribution. Patients will often complain of intense pruritus and/or pain associated with the rash. NME often appears first as intensely erythematous, well-demarcated plaques or patches with irregular edges. These lesions will then blister centrally forming flaccid bullae that then erode and crust, eventually healing with hyperpigmentation. As the central portion of the wound heals, the surrounding erythema may expand outward with new vesicles developing along the advancing serpiginous edges; eventually, these lesions may coelesce. The lesions of NME are cyclical in nature, typically waxing and waning for about 10 days; upon inspection, various lesions are seen in different stages of healing. In addition, trauma is thought to exacerbate or promote the formation of the lesions of NME.
The lesions of NME can be complicated by superimposed infection, most commonly by Staphylococcus aureus and Candida albicans. Superinfection may be heralded by the formation of pustules superimposed on the rash. Patients may be misdiagnosed with chronic candidiasis and can have a history of prior treatment with anti-fungals and/or antibiotics without improvement.
Commonly associated with the characteristic skin lesions are the mucocutaneous findings of angular cheilitis, glossitis, and stomatitis, which are present in roughly 30% of patients. Onychoschizia and other nail changes have also been seen in association with NME.
The other systemic findings seen in the glucagonoma syndrome include diarrhea, neuropsychiatric changes, anemia, and a tendency toward venous thrombosis. The diarrhea that occurs in glucagonoma syndrome is often a source of major distress, can be incapacitating and affects up to one third of patients. Neuropsychiatric alterations are present in 20% of patients and can include depression, anxiety, and other personality changes, as well as visual changes and ataxia. Hypercoagulability is commonly seen in association with glucagonomas and may manifest as migratory thrombophlebitis (Trousseau syndrome). Deep venous thrombosis, is particularly problematic with resultant venous thromboembolism occurring in about 24%, and pulmonary embolism in 11%.
As with many disease states, a thorough history and physical exam, with special attention to the skin, is necessary for proper diagnosis. Patients with suggestive skin lesions should have multiple biopsies performed at the edges of active lesions in order to increase diagnostic yield, as the histopathologic changes suggestive of NME may only be focally present.
Among laboratory tests, a serum glucagon level is of paramount importance and will be highly elevated often to levels greater than 1000 pg/nL (reference range is 50 to 200 pg/nL). Another necessary laboratory test is the fasting serum glucose and/or a glucose tolerance test, used to establish a diagnosis of diabetes mellitus. A complete blood count should be obtained in order to evaluate for anemia, which is usually of the normocytic and normochromic variety. It is also important to evaluate for accompanying nutrient deficiencies with serum zinc, amino acid, and essential fatty acid levels.
Other laboratory tests may be used to detect the presence of a neoplasm as the root cause of elevated glucagon. Serum chromogranin A levels may be measured, as this is a sensitive marker for glucagonoma. Additionally, hepatic transaminases, total bilirubin, and alkaline phosphatase may be measured to help detect the presence of liver metastases.
Since glucagonoma syndrome may rarely be apart of MEN1, it is also important to screen for this disease if clinically suspected (specifically in younger patients), with serum levels of fasting insulin, calcium, parathyroid hormone, vasoactive intestinal peptide, and prolactin.
Imaging of the abdomen is also indicated to help determine the presence of a tumor or metastases, as this helps guide treatment. A majority of glucagonomas, roughly 87%, are located in the pancreatic tail, as this is where the concentration of pancreatic islets is the highest. Generally, contrast enhanced computed tomography (CT) of the abdomen will suffice. Alternatively, one may use a gadolinium-enhanced T2 MRI of the abdomen to help localize the tumor and differentiate it from pancreatic adenocarcinoma.
One may also use selective angiography of the celiac axis to determine both the center of the neoplasm and the presence of hepatic metastases simultaneously. In fact, due to their relative hypervascularity, selective angiography of the celiac axis is the most reliable means of diagnosis and localization of glucagonoma in the face of unrevealing CT scans. It has been suggested that the combination of abdominal CT and selective celiac angiography will give an adequate preoperative assessment of the tumor and its metastases.
Lastly, somatostatin receptor scintigraphy, classically with indium-111 octreotide (In-D-Phe1-octreotide) can be used for detection of metastases, as nearly all glucagonomas reported in the literature have been somatostatin receptor positive. This may be especially useful in the case of metastases to lymph nodes, as arteriography cannot reliably detect these. Alternatively, F-fludeoxyglucose (F-FDG) positron emission tomography (PET)-CT is another means of detecting metastases. More recently however, PET-CT scanning utilizing somatostatin analogues has emerged as a promising method of detecting glucagonomas and their metastases. In particular, the use of Gallium (Ga)-labeled somatostatin analogues, specifically Ga-DOTA-NaI-Octreotide (NOC) has been shown to detect the presence of a glucagonoma when other imaging modalities have failed.
The treatment of NME and the glucagonoma syndrome can be broadly divided into curative treatment which is surgical and palliative treatment which is medical.
Curative treatment of glucagonomas and its accompanying paraneoplastic syndrome is centered around early surgical excision of the tumor, which as previously mentioned, is located in the tail of the pancreas in the majority of cases. Removal of the tumor in the setting of no distant metastases generally completely cures the dermatosis, as well as the accompanying symptoms, often, within a week. After surgical excision, patients are monitored for recurrence with serial imaging and glucagon levels at 3 and 6 months post-surgery, and every 6 to 12 months after that.
Although surgery can be curative in localized disease, many times the tumor has spread distantly at presentation, with reports of 50% to 100% of patients exhibiting metastases at the time of diagnosis. The most common sites of metastases are the liver followed by the peripancreatic lymph nodes. Unfortunately, the tumor is resistant to chemotherapy, and metastatic disease generally cannot be surgically resected. This highlights the importance of recognition of NME and early diagnosis of glucagonoma prior to metastatic spread. Fortunately, glucagonomas are slow-growing tumors and patients may survive for many years, even with metastatic disease. However, it has been shown that even in those with metastatic disease, surgical excision of the primary tumor confers a survival benefit and thus, should be pursued when feasible.
In those with metastatic disease upon presentation or who are poor candidates for surgery, medical therapy is the cornerstone. Patients may be treated palliatively with either long-acting somatostatin analogues (i.e., octreotide, lanreotide) or interferon-alpha. Somatostatin analogues in particular have been shown to be highly effective in case reports; these antagonize the effects of glucagon, effectively abolishing both the NME as well as the systemic symptoms associated with glucagonoma. In fact, lanreotide, a somatostatin analogue given as a monthly depot injection has been shown to improve progression-free survival in those with metastatic gastrointestinal neuroendocrine tumors. Thus, treatment with somatostatin analogues helps palliate symptoms as well as improve survival. However, treatment with somatostatin analogues must continue indefinitely as NME and other symptoms of glucagonoma promptly return when treatment is withdrawn. Chemotherapy with agents such as streptozotocin and 5-fluorouracil can be considered to palliate metastatic disease; however, this is not preferred due to the tumors poor response to current chemotherapeutic agents. The biologic agents sunitinib and everolimus have also shown some success in clinical trials for the treatment of pNETs including glucagonomas, and may also be considered in the appropriate clinical setting. Other novel treatments of liver metastases include liver resection and transplantation, percutaneous ablation of liver metastases, targeted radiotherapy, and chemoembolization.
As NME is considered to be a deficiency dermatosis, repletion of various nutrients has been utilized in the literature in an attempt to specifically target this entity. In particular, supplementation of zinc, essential fatty acids, and amino acids has been shown in some cases to resolve NME. One literature review found that supplementation with zinc sulfate, orally, at a dose of 440 mg per day, showed the most consistent response; even patients with normal zinc levels showed improvement of NME. In the case of repletion of essential fatty acids and amino acids, it has been shown that the route of administration is an important consideration, with parenteral administration being preferred over the oral route. This is postulated to be a result of the hepatic first pass effect, which is exaggerated in the setting of hyperglucagonemia.
Lastly, in the case of treating NME outside of glucagonoma (i.e., pseudoglucagonoma syndrome) therapy is mainly aimed at treating the underlying condition. As malnutrition seems to be a common theme in pseudoglucagonoma syndrome, one may also consider nutritional repletion with zinc, amino acids, and essential fatty acids.
In the management of NME, coordination between the entire medical care team, including but not limited to the dermatologist, oncologist, surgeon, and radiologist is imperative. Timely recognition of the disease is of utmost importance. This can many times be at the hands of the dermatologist who may receive a referral for the rash of NME, as it is often a presenting symptom of the underlying pancreatic malignancy. Delayed recognition, treatment or relaying of information between specialists and others involved with the care of the patient can result in worsened outcomes, including mortality, for the patient. Thus, it is of utmost importance to recognize this characteristic dermatitis and begin the appropriate testing and surgical referrals, as prompt surgery prior to tumor metastasis can cure the patient. In this sense education of healthcare providers on both the rash of NME as well as its accompanying symptoms can help prevent disability and death due to this relatively rare entity. (Level V)
|||van Beek AP,de Haas ER,van Vloten WA,Lips CJ,Roijers JF,Canninga-van Dijk MR, The glucagonoma syndrome and necrolytic migratory erythema: a clinical review. European journal of endocrinology. 2004 Nov [PubMed PMID: 15538929]|
|||Wu SL,Bai JG,Xu J,Ma QY,Wu Z, Necrolytic migratory erythema as the first manifestation of pancreatic neuroendocrine tumor. World journal of surgical oncology. 2014 Jul 17 [PubMed PMID: 25029913]|
|||Silva JA,Mesquita Kde C,Igreja AC,Lucas IC,Freitas AF,Oliveira SM,Costa IM,Campbell IT, Paraneoplastic cutaneous manifestations: concepts and updates. Anais brasileiros de dermatologia. 2013 Jan-Feb [PubMed PMID: 23538999]|
|||Al-Faouri A,Ajarma K,Alghazawi S,Al-Rawabdeh S,Zayadeen A, Glucagonoma and Glucagonoma Syndrome: A Case Report with Review of Recent Advances in Management. Case reports in surgery. 2016 [PubMed PMID: 26981306]|
|||Luber AJ,Ackerman LS,Culpepper KS,Buschmann CM,Koep LJ, Paediatric necrolytic migratory erythema as a presenting sign of glucagonoma syndrome. The British journal of dermatology. 2016 May [PubMed PMID: 26585841]|
|||Halvorson SA,Gilbert E,Hopkins RS,Liu H,Lopez C,Chu M,Martin M,Sheppard B, Putting the pieces together: necrolytic migratory erythema and the glucagonoma syndrome. Journal of general internal medicine. 2013 Nov [PubMed PMID: 23681843]|
|||Gupta M,Mahajan VK,Mehta KS,Chauhan PS, Zinc therapy in dermatology: a review. Dermatology research and practice. 2014 [PubMed PMID: 25120566]|
|||Brenner RR, Nutritional and hormonal factors influencing desaturation of essential fatty acids. Progress in lipid research. 1981 [PubMed PMID: 7342101]|
|||Jenkins DK,Mitchell JC,Manku MS,Horrobin DF, Effects of albumin on fatty acid, protein, and eicosanoid levels in rat mesenteric arterial bed perfusions. Canadian journal of physiology and pharmacology. 1988 Jun [PubMed PMID: 3167681]|
|||Kasper CS, Necrolytic migratory erythema: unresolved problems in diagnosis and pathogenesis. A case report and literature review. Cutis. 1992 Feb [PubMed PMID: 1563284]|
|||Marinkovich MP,Botella R,Datloff J,Sangueza OP, Necrolytic migratory erythema without glucagonoma in patients with liver disease. Journal of the American Academy of Dermatology. 1995 Apr [PubMed PMID: 7896950]|
|||Fang S,Li S,Cai T, Glucagonoma syndrome: a case report with focus on skin disorders. OncoTargets and therapy. 2014 [PubMed PMID: 25152626]|
|||Teixeira RC,Nico MM,Ghideti AC, Necrolytic migratory erythema associated with glucagonoma: a report of 2 cases. Clinics (Sao Paulo, Brazil). 2008 Apr [PubMed PMID: 18438582]|
|||Tierney EP,Badger J, Etiology and pathogenesis of necrolytic migratory erythema: review of the literature. MedGenMed : Medscape general medicine. 2004 Sep 10 [PubMed PMID: 15520626]|
|||Leichter SB, Clinical and metabolic aspects of glucagonoma. Medicine. 1980 Mar [PubMed PMID: 6987481]|
|||Qadan M,Visser B,Kim J,Pai R,Triadafilopoulos G, Abdominal mass, anemia, diabetes mellitus, and necrolytic migratory erythema. Digestive diseases and sciences. 2012 Jun [PubMed PMID: 22089253]|
|||Xu Q,Chen WH,Huang QJ, Spiral CT localization of pancreatic functioning islet cell tumors. Hepatobiliary [PubMed PMID: 15567758]|
|||Melen-Mucha G,Lawnicka H,Kierszniewska-Stepien D,Komorowski J,Stepien H, The place of somatostatin analogs in the diagnosis and treatment of the neuoroendocrine glands tumors. Recent patents on anti-cancer drug discovery. 2006 Jun [PubMed PMID: 18221040]|
|||Kindmark H,Sundin A,Granberg D,Dunder K,Skogseid B,Janson ET,Welin S,Oberg K,Eriksson B, Endocrine pancreatic tumors with glucagon hypersecretion: a retrospective study of 23 cases during 20 years. Medical oncology (Northwood, London, England). 2007 [PubMed PMID: 17873310]|
|||Sahoo MK,Gupta S,Singh I,Pahwa S,Durgapal P,Bal CS, Necrolytic migratory erythema associated with glucagonoma syndrome diagnosed by ⁶⁸Ga-DOTANOC PET-CT. Asia-Pacific journal of clinical oncology. 2014 Jun [PubMed PMID: 23279825]|
|||Hill JS,McPhee JT,McDade TP,Zhou Z,Sullivan ME,Whalen GF,Tseng JF, Pancreatic neuroendocrine tumors: the impact of surgical resection on survival. Cancer. 2009 Feb 15 [PubMed PMID: 19130464]|
|||Caplin ME,Pavel M,Ruszniewski P, Lanreotide in metastatic enteropancreatic neuroendocrine tumors. The New England journal of medicine. 2014 Oct 16 [PubMed PMID: 25317881]|
|||Yao JC,Shah MH,Ito T,Bohas CL,Wolin EM,Van Cutsem E,Hobday TJ,Okusaka T,Capdevila J,de Vries EG,Tomassetti P,Pavel ME,Hoosen S,Haas T,Lincy J,Lebwohl D,Öberg K, Everolimus for advanced pancreatic neuroendocrine tumors. The New England journal of medicine. 2011 Feb 10 [PubMed PMID: 21306238]|
|||Raymond E,Dahan L,Raoul JL,Bang YJ,Borbath I,Lombard-Bohas C,Valle J,Metrakos P,Smith D,Vinik A,Chen JS,Hörsch D,Hammel P,Wiedenmann B,Van Cutsem E,Patyna S,Lu DR,Blanckmeister C,Chao R,Ruszniewski P, Sunitinib malate for the treatment of pancreatic neuroendocrine tumors. The New England journal of medicine. 2011 Feb 10 [PubMed PMID: 21306237]|