A case of necrolytic migratory erythema managed for 24 months with intravenous amino acid and lipid infusions

Abstract

A 9-year-old castrated male Shetland sheepdog was diagnosed with necrolytic migratory erythema and hepatocutaneous syndrome. Necrolytic migratory erythema was treated with intermittent intravenous amino acids as needed to control cutaneous lesions. The addition of lipid infusions extended the treatment interval. The patient had a favorable response for 24 months.

Necrolytic migratory erythema (NME) is a rare disorder first described in 1942 that is most often associated with glucagonoma in humans (1). Other conditions associated with NME in humans include chronic liver disease or inflammatory bowel disease, malabsorption disorders, and malignancies. In veterinary medicine the condition has been referred to by several names including superficial necrolytic dermatitis, diabetic dermatopathy, pseudoglucagonoma syndrome, metabolic epidermal necrosis, and hepatocutaneous syndrome, in addition to NME (2). Although these terms all refer to the same disease syndrome, in veterinary medicine the term “hepatocutaneous syndrome (HCS)” is used when NME is associated with an underlying hepatopathy (3–5). The hepatopathy has an ultrasonographic honeycomb appearance of variable hypoechoic regions surrounded by hyperechoic borders (4). Patients diagnosed with NME exhibit erythematous skin lesions, crusting, and alopecia around the footpads, pressure points (e.g., elbows) and mucocutaneous junctions. Histological findings compatible with this disease include what is typically referred to as a “red, white, and blue pattern” or parakeratosis in the superficial epidermis, edema caused by vacuolated pale keratinocytes in the middle layer and deeply basophilic hyperplastic basal cells (6). Other conditions that have been associated with NME in veterinary medicine include phenobarbital administration (7) and glucagonoma (8–11). Most patients with NME or HCS are hypoaminoacidemic (3,7,11), and treatment with IV amino acids has been beneficial (3). Treatment for NME involves focusing on treatment of the predisposing disease and palliative therapy with intravenous (IV) amino acids infusions, supplemental dietary protein, omega-3 fatty acids, zinc, and other therapies (2).

The purpose of this report is to describe the unique management of NME that included intravenous lipids administered together with amino acids in a diabetic patient for 24 months.

Case description

A 17.5-kg 9-year-old castrated male Shetland sheepdog was referred for treatment of diabetic ketoacidosis. Five days prior to presentation the dog began vomiting, had a poor appetite, and was polyuric/polydipsic. Approximately 1 mo prior to presentation the dog was diagnosed with pododermatitis. Initial treatment included amoxicillin/clavulanic acid (Pfizer Animal Health, Deerfield, Ilinois, USA), 14 mg/kg body weight (BW), PO, q12h and hydroxyzine (Camber Pharmaceutical, Piscataway, New Jersey, USA), 2.5 mg/kg BW, PO, q12h. Pododermatitis progressed and therapy was empirically altered to cephalexin (Sun Pharmaceutical, Andheri, Indiana, USA), 25 mg/kg BO, PO, q12h, and a tapering course of prednisone (Qualitest Pharmaceutical, Huntsville, Alabama, USA), 0.25 mg/kg BW, PO, q12h initially, to which there was a favorable response.

At the time of referral examination, abnormal findings included tacky mucous membranes, moderate crusting and fissuring of the foot pads, interdigital erythema, mild periocular and perimuzzle crusting, cranial organomegaly, and a II/VI systolic heart murmur. The dog was admitted to the hospital, a jugular catheter was placed and blood was collected for a complete blood (cell) count (CBC) and serum chemistry profile. Urine was obtained via cystocentesis for urinalysis and culture. Findings included marked glucosuria and ketonuria and negative urine culture. The CBC results revealed a neutrophilic leukocytosis [total WBC count: 18.2 × 10³/μl, reference interval (RI): 6.0 to 16.9 × 10³/μL; neutrophil count: 15.8 × 10³/μL, RI: 3.3 to 12.0 × 10³/μL]. The dog had hyperglycemia (21.5 mmol/L, RI: 3.7 to 7.5 mmol/L), decreased creatinine (44.2 μmol/L, RI: 53.0 to 176.8), hypophosphatemia (0.68 nmol/L, RI: 0.84 to 2.33 mmol/L), hyperbilirubinemia (8.55 μmol/L, RI: 1.71 to 5.13); elevated alkaline phosphatase (ALP) 2112 U/L (RI: 12 to 121 U/L), gamma-glutamyl transferase (GGT) 26 U/L (RI: 2 to 10 U/L), alanine aminotransferase (ALT) 426 U/L (RI: 18 to 86 U/L), aspartate aminotransferase (AST) 314 U/L (RI: 16 to 54 U/L); hypercholesterolemia 15.1 mmol/L (RI: 2.13 to 9.23), and hypertriglyceridemia 3.81 mmol/L (RI: 0.34 to 3.63).

Diabetic ketoacidosis was treated with intravenous (IV) lactated ringer’s solution and electrolyte replacement, regular insulin constant rate infusion (CRI), and frequent blood glucose monitoring. Over the following 24 h, blood glucose values ranged between 14.6 and 16.8 mmol/L.

Thoracic radiographs were within normal limits. Abdominal ultrasound was unremarkable except for a diffusely hyperechoic enlarged liver with numerous hypoechoic nodules throughout and moderately distended gallbladder.

On the second day of treatment, three 18-gauge ultrasound-guided hepatic biopsies and two 6-mm skin biopsy samples from the footpads were obtained, routinely fixed in neutral buffered formalin and submitted for histopathological examination. Aerobic hepatic culture was negative. The hepatic biopsy revealed moderate, diffuse, chronic bridging fibrosis with bile duct hyperplasia, hepatocellular regenerative hyperplasia and multifocal hepatocellular vacuolar change; suggestive of longstanding toxic/metabolic hepatopathy. Footpad skin biopsy findings revealed moderate, diffuse hyperkeratosis (primarily orthokeratotic with foci of parakeratosis), mild to moderate hyperplasia of the superficial and follicular epidermis with acanthosis, and infrequent keratinocyte vaculation with no evidence of inflammation. The cutaneous biopsy results were consistent with NME; the combined hepatic and cutaneous results indicated a diagnosis of HCS. Evaluation of the amino acid profile in the fasted animal revealed depressed values (Table 1). Glucagon concentration was 178 ng/L [RI: 55 to 156 ng/L (4)], and blood zinc concentration was 0.46 ppm (RI: 0.7 to 2.0 ppm).

Table 1.

Plasma amino acid profiles

Amino acid	Week 1	Week 12	Week 32	Reference range ± SEM (μmol/L)
Alanine	57	170	208	389 ± 9
Arginine	6	17	20	102 ± 3
Asparagine	11	12	9	41 ± 1
Aspartic acid	3	3	2	7 ± 0.2
Citrulline	6	47	63	41 ± 2
Glutamic acid	33	17	19	24 ± 1
Glutamine	66	257	251	495 ± 9
Glycine	75	73	56	266 ± 8
Histidine	46	52	57	71 ± 2
Isoleucine	69	71	158	51 ±1
Leucine	91	137	314	120 ± 3
Lysine	54	74	84	131 ± 5
Methionine	15	29	56	57 ± 2
Ornithine	11	6	5	35 ± 2
Phenylalanine	82	57	57	45 ± 1
Proline	23	37	66	249 ± 8
Serine	38	37	48	107 ±3
Taurine	27	159	184	77 ± 2
Threonine	23	60	79	128 ± 5
Tryptophan	45	106	151	60 ± 2
Tyrosine	21	13	18	39 ± 1
Valine	174	195	396	158 ±4

SEM — standard error of the mean.

On day 3, insulin CRI was discontinued, and subcutaneous intermediate acting insulin (Humulin N; Eli Lilly, Indianapolis, Indiana, USA) was initiated. Palliative therapy for the NME skin lesions included an IV amino acid infusion at 28.5 mL/kg body weight (BW), 8.5% solution, or Travasol (Baxter Healthcare, Deerfield, Illinois, USA), 2.4 g/kg BW (2). The patient was discharged from the hospital with the following instructions: insulin (Eli Lilly), 0.5 U/kg BW, SQ, q12h, amoxicillin (Pfizer Animal Health, New York, New York, USA), 15 mg/kg BW, PO, q24h, famotidine (Teva Pharmaceutical, Sellersville, Pennsylvania, USA), 0.5 mg/kg BW, PO, q24h, 3 to 4 daily omega-3 fatty acid capsules (1000 mg, CVS Pharmacy, Marlborough, Massachusetts, USA), and an oral protein supplement (ProMod; Abbott Laboratories, Columbus, Ohio, USA), 10 g, PO, once daily mixed with food. With the exception of amoxicillin, all medications were continued chronically.

Over the next 24 mo intravenous amino acid infusions were administered 24 times as treatment of the skin lesions. Amino acid infusions were administered weekly for the first 4 wk and then every 1 to 3 wk depending upon waxing and waning of footpad and facial crusting. The client reported improvement in the patient’s skin lesions and level of activity presumably because of decreased foot pad discomfort. Crusts around the muzzle and perioccular regions nearly resolved following infusions; interdigitial hyperkeratosis improved significantly but did not completely resolve. Over the course of treatment, infusions were administered every 1.5 wk for the first 16 wk. Following the 10th infusion, additional supplementation with lipid (Intralipid; Baxter Healthcare), 7.0 mL/kg BW, 20% solution; or 1.4 g/kg was initiated; thereafter, the infusion interval averaged 6.5 wk between the subsequent 13 intravenous treatments. The solutions were premixed via a closed system into a sterile intravenous fluid bag. Infusions (500 mL 8.5% amino acid solution and 122 mL 20% lipid solution) were administered over 8 h through a peripheral catheter which has been described (12); infusions were well-tolerated without evidence of phlebitis from the hyperosmolar solution. Supplementation with glutamine (6 g PO, q12h) and taurine (500 mg PO, q12h) began 6 wk after HCS diagnosis as the patient’s glutamine and taurine levels were reduced and these amino acids were not included in the amino acid solution. Follow-up amino acid profiles were evaluated on 2 occasions, 12 and 32 wk after diagnosis (Table 1). Essential fatty acids were not assayed. The patient’s glucose curves and liver enzymes were monitored periodically over the ensuing months. Liver enzymes improved, and there were mild increases in insulin in response to hyperglycemia.

Approximately 24 mo after the initiation of treatment, the patient returned to the hospital for lethargy and decreased appetite. Physical examination revealed severe respiratory distress and tachycardia with muffled thoracic auscultation. Diagnostics revealed anemia [hematocrit 23% (RI: 40 to 59%)] and a large volume of pleural effusion which thoracocentesis confirmed as hemothorax. The owners declined further diagnostics and the patient was humanely euthanized.

Necropsy revealed a mass on the left scapula attached to the surrounding musculature with multifocal pulmonary involvement. The histological diagnosis was compatible with sarcoma with pulmonary metastasis. Chronic cirrhosis was found on examination of the liver and mild islet cell atrophy was found in the pancreas. Skin biopsy findings were compatible with the previous diagnosis of superficial necrolytic dermatitis (i.e., NME).

Discussion

This case report documents long-term management of a diabetic dog with NME, using oral supplements, intravenous amino acid infusions, and intravenous lipid infusion for 24 mo. To the authors’ knowledge, use of concurrent lipid infusion has not been previously reported in the veterinary literature. It was the authors’ observation that the addition of lipid infusion to the patient’s treatment regimen was clinically beneficial as exhibited by the increased interval between treatments (1.5 wk versus 6.5 wk, on average), and the prolonged survival. The prognosis for dogs with NME tends to be poor; reported average survivals range from 1.7 to 6.4 mo (3,5,7). The increased interval between treatments could be ascribed to other therapies (e.g., oral protein supplementation, omega-3 fatty acids, taurine, or glutamine); however, the timing strongly supports a positive effect from lipid infusions. Our dosage (1.4 g/kg BW) of lipid was extrapolated from human reports. Alexander et al (12) reported daily doses of 0.6 g/kg BW for multiple days. Bewley et al (13) described administering 1 L intralipid daily for 3 consecutive days to a 50-year-old man with octreotide-resistant NME. Linoleic and arachidonic essential fatty acid levels were low prior to infusion and improved following infusions in 1 report (13), but unchanged in another (12). Plasma fatty acid concentrations were not measured in our case and would have been useful information. Measurement of plasma fatty acid concentrations should be considered in future cases of NME. The precise role of fatty acids in the treatment of NME is unclear, but case reports in the human literature discuss its beneficial use. Two reports describe patients with glucagonoma who became refractory to therapy with octreotride, but whose NME signs were subsequently controlled with IV infusions of amino acids and lipid (12,13). Another report describes successful use of essential fatty acids in an individual with NME (14). The lipid infusion is rich in linoleic acid which is integral in ceramide production in the skin. Ceramide is important for intra-keratinocyte adhesion and may aid the prevention of keratinocyte sloughing (15).

The patient’s serum zinc concentration was mildly decreased. Initial reports of hepatocutaneous syndrome were suspected to be related to a zinc deficiency (zinc-responsive dermatosis). Unfortunately, alterations in zinc concentrations have been inconsistent and treatment with zinc supplementation shows variable responses.

In conclusion, this case report describes successful treatment of a diabetic patient with NME and hepatocutaneous syndrome for 24 mo. A novel aspect of this case was the addition of lipid to the intravenous amino acid infusions which appears to have been highly beneficial. In patients with NME, which are failing to respond to amino acid infusions, implementation of concurrent lipid infusion affords an additional treatment option. Measurement of plasma fatty acid concentrations should be considered in such cases. CVJ