β-Cell Transplantation Restores Metabolic Control and Quality of Life in a Patient With Subcutaneous Insulin Resistance

The patient developed type 1 diabetes at age 24 years. Subcutaneous insulin therapy initially achieved acceptable glycemic control (average insulin dose 30 IU/day, HbA_1c 7% [normal range 4–6%]; HPLC; Pharmacia Biotech, Uppsala, Sweden). However, 1 year after diagnosis, a state of insulin resistance progressively developed with difficulty controlling hyperglycemia despite the use of excessively high doses of insulin (>400 IU/day). During hospitalization, this resistance was correlated with subcutaneous insulin administration; it disappeared within 2–3 h after starting intravenous insulin therapy and reappeared within 1 day after returning to the subcutaneous route. Measurements of insulin levels confirmed low insulin resorption from subcutaneous injection sites, whereas adding aprotinin to the insulin preparation did not improve insulin absorption. There was no evidence for a condition of pseudoresistance that would be caused by a cheating patient with psychologic dysequilibrium. This clinical condition caused frequent and long hospitalizations (>500 days in 1989 and 1990) and necessitated the implantation of an insulin pump with intraperitoneal delivery in 1991 (Infusaid; Infusaid Inc., Norwood, MA). This resulted in intermediate but acceptable blood glucose control (average intraperitoneal insulin dose 60 IU/day, HbA_1c 7.0–9.9%) and markedly reduced total hospitalization time. In 1994, however, perforation of the insulin pump through the abdominal skin forced us to explant the pump. Ambulatory intravenous insulin administration was installed using a H-TRON pump (Disetronic Medical Systems, St. Paul, MN) and a subclavian catheter. The patient was followed on a semiambulatory basis with several hospitalizations for sepsis. A new attempt at subcutaneous administration with Humalog (Eli Lilly, Indianapolis, IN) was successful for several weeks but again resulted in persistent hyperglycemia despite the use of doses up to 600 IU/day. Since 1995, the patient was treated by an intraperitoneal pump (Minimed; Medtronic, Northridge, CA) at 65 IU/day, but glycemic control remained intermediate (HbA_1c 8–10%). Severe hypoglycemia under this treatment was not a major problem.

At the time of islet transplantation, the subject was 41 years of age (body weight 55 kg; BMI 18.4 kg/m²), C-peptide negative (<0.09 ng/ml, measured by time-resolved fluorescence immunoassay), and autoantibody positive (insulinoma-associated protein-2 6.1% binding vs. <0.44% in normal control subjects and insulin autoantibody 1.6 vs. <0.6% in normal control subjects). Both legs exhibited severe signs of necrobiosis lipoidica (Fig. 1).

A β-cell graft was prepared from cultured β-cell preparations as previously described (1). It contained 4.5 × 10⁶ β-cells/kg recipient body wt, which is comparable to the size used in the Edmonton study (2). Graft volume corresponded to 11,000 islet equivalents/kg body wt, but this parameter is not an adequate index for cultured preparations in view of their variable loss in dithizone staining. The graft was injected intraportally through a percutaneous transhepatic approach under ultrasound and fluoroscopic guidance. The immune suppressive regimen consisted of induction with ATG-Fresenius and maintenance therapy with tacrolimus and mycophenolate mofetil.

Daily insulin was progressively decreased over an 8-week period following transplantation, while glycemia became and remained well controlled. Before transplantation, >50% of fasting glycemia values were above 200 mg/dl, while all measurements after transplantation were well below 200 mg/dl. Six months after stopping insulin treatment, >90% of measurements fell within the normal range (60–140 mg/dl), whereas this was only the case for 15% of measurements before transplantation. HbA_1c remained below 6.5% until the present (>1 year of follow-up). Basal plasma C-peptide levels progressively increased to levels above 1.5 ng/ml (glucagon-stimulated C-peptide 2.93–3.00 ng/ml; 200–232% of basal C-peptide). Moreover, no hypoglycemic attacks were noticed after transplantation. The implantation and immediate posttransplantation periods were uneventful. Since transplantation, the patient has not been hospitalized and all necrobiosis lipoidica lesions have healed with disappearance of ulcerations (Fig. 1).

Subcutaneous in-sulin resistance is a rare but severe condition of unknown etiology (5). Enzymatic defects and antibody formation have been implicated but have not been sufficiently documented (6,7). Our patient was diagnosed as a case of idiopathic subcutaneous insulin resistance after the exclusion of other diagnostic possibilities. Under treatment with intraperitoneal and intravenous insulin administration, we succeeded in achieving a marginally acceptable glycemic control but could not avoid frequent hospitalizations. Although the follow-up period is still limited (12 months), the present case suggests that β-cell transplantation can be a better alternative for patients with subcutaneous insulin resistance.

Other than good metabolic control, the healing of the necrobiosis lipoidica lesions is of particular interest. This is an uncommon inflammatory skin disorder of unknown cause that occurs quite frequently in diabetic patients (8). The lesions are characterized by irregularly formed, indurated plaques with central atrophy and yellow pigmentation. Telangiectasias and ulcerations occur in 30% of cases. Both tighter glucose control (8) and treatment with cyclosporine (9) have been associated with a clinical improvement of the lesions, although reports are mainly casuistic. In our patient, both better metabolic control and immunosuppressive treatment may have contributed to the positive evolution.

On the basis of present observations, we propose that patients with subcutaneous insulin resistance should be considered as potential candidates for β-cell transplantation. As with other indications, the decision to undertake this intervention should depend on the balance among medical failure to adequately control disease, the clinical evolution, and the risks of immune therapy.

This work was supported by grants from the Fund Scientific Research, Flanders, Belgium (G.0375.00, G.0084.02, and clinical fellowships of C.M. and B.K.), and the Juvenile Diabetes Research Foundation (4-2001-434).