| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
CLINICAL CASE SEMINAR |
Marshfield Clinic (M.T.S.), Wausau, Wisconsin 54401; and Division of Endocrinology (A.B., F.J.S.) and Department of Surgery (G.B.T.), Mayo Clinic, Rochester, Minnesota 55905
Address all correspondence and requests for reprints to: Ananda Basu, M.D., 5-194 Joseph, Saint Mary’s Hospital, Mayo Clinic, Rochester, Minnesota 55902. E-mail: basu.ananda{at}mayo.edu.
Abstract
The diagnosis of insulinoma depends on the fulfillment of well-established criteria during the 72-h fast. However, these criteria rely on normal renal function. Spontaneous hypoglycemia that is not attributable to insulinoma may occur in persons with renal failure. We describe herein a patient with renal impairment who had undergone renal transplant and had a 20-yr history of hypoglycemic symptoms and successful resection of insulinoma. Although the results of a 72-h fast were consistent with endogenous hyperinsulinemia, their interpretation was complicated in the presence of renal impairment. Fortunately, the identification of the tumor, by endoscopic ultrasonograph, led to a correct diagnosis. This case seems to be the second report of a patient with insulinoma with concomitant renal failure.
FOLLOWING THE FIRST worldwide documentation of insulinoma at the Mayo Clinic in 1927 (1), the subsequent 60 yr of experience, at that institution, with 224 patients with initial successful operation for insulinoma contained no patients with concomitant renal insufficiency (2). We report herein the first Mayo Clinic patient with insulinoma and concomitant renal impairment.
Case Report
A 47-yr-old Caucasian man was referred for evaluation of hypoglycemia. He had experienced hunger, sweats, shakiness, palpitations, perioral numbness, and visual blurring without diplopia, 2–4 h after meals, for the previous 20 yr. The symptoms occurred more frequently after a high carbohydrate meal and were relieved with candies or cookies. With an increase in frequency and intensity in the past 3 yr, he had an episode of witnessed unconsciousness, without seizures, 2 h post cibal, with reflectance meter glucose of 22 mg/dl (1.2 mM); he was revived by the paramedical staff after two injections of glucagon, with correction of reflectance meter glucose to 56 mg/dl (3.1 mM). In addition, he had an episode of confusion, 2–3 h post cibal, while operating a motor vehicle. After he ingested several glucose tablets, the reflectance meter reading was 80 mg/dl (4.4 mM).
He was diagnosed with biopsy-proven IgA nephropathy at the age of 20 yr. He underwent a living-related-donor renal transplant in 1998. His renal function, however, did not return to normal after the transplant, with creatinine values ranging between 2.5 and 3.0 mg/dl. A recent iothalamate clearance was 27 ml/min·m2 (normal, 80–200 ml/min·m2). Medications included cyclosporin (200 mg, twice a day), prednisone (7.5 mg, every day), mycophenolate (1000 mg, twice a day), acebutolol (400 mg, twice a day), prilosec (20 mg, every day), prinivil (10 mg, every day), multivitamin (one tablet, every day), allopurinol (150 mg, every day), and caltrate (one tablet, three times a day).
Concomitant with a 70-min postmeal serum glucose of 49 mg/dl (2.7 mM) were c-peptide of 9100 pM [immunochemiluminometric assay (ICMA); sensitivity, 33 pM] and insulin of 450 pM (Access Chemiluminescent Enzyme Immunoassay; sensitivity, 10 pM; Beckman, Chaska, MN). Insulin antibodies were negative, as was screen for sulfonylureas and repaglinide (liquid chromatography mass spectrometry/mass spectrometry). A 72-h fast, performed elsewhere, and a transabdominal ultrasound of the pancreas were reportedly negative. He had been prescribed acarbose in the past without amelioration of symptoms.
There was no history of diabetes in first-degree relatives. He was an electrician by profession. He did not smoke or drink alcohol. Physical examination was normal, apart from a scar from prior renal transplantation in the right lower abdominal quadrant. He was well nourished.
Laboratory tests showed a creatinine of 2.7 mg/dl, with a fasting glucose of 100 mg/dl (5.5 mM) and glycosylated hemoglobin of 5.0% (normal, 4–7%). Liver function tests were normal. At 38 h into a repeat 72-h inpatient fast, he became profusely diaphoretic, semiconscious, and poorly responsive, with serum glucose of 22 mg/dl (1.2 mM), insulin of 366 pM (ICMA; sensitivity, 0.6 pM), c-peptide of 7800 pM (ICMA; sensitivity, 33 pM), and ß-hydroxybutyrate of 0.3 mM. The serum glucose levels at 10, 20, and 30 min after 1 mg glucagon iv, given to end the fast, were 50 (2.8 mM), 65 (3.6 mM), and 79 mg/dl (4.4 mM), with rapid relief of symptoms (Fig. 1
). Drug screens for sulfonylurea and repaglinide were negative.
|
Discussion
Hypoglycemia and renal disease
Spontaneous hypoglycemia is uncommonly (3.6% in a large case series) (3) observed in renal impairment. The most common cause of hypoglycemia in a patient with renal impairment is probably the effect of one of several medications. Although hyperinsulinemia is observed in renal insufficiency, it may not be the cause of hypoglycemia. At physiological concentrations, insulin degradation is mediated by the insulin receptor with a small contribution from nonspecific processes (4). Although the liver is the primary site of insulin clearance, accounting for 50% of portal insulin removal (5), the kidney is the major site of systemic insulin clearance, removing approximately 50% of peripheral insulin (6). Additionally, the kidney removes about 70% of c-peptide by glomerular filtration (6). Glomerular clearance of insulin occurs both by diffusion and receptor-mediated transport. Although most of this filtered insulin is reabsorbed by the tubular cells, the major route of renal insulin clearance is by receptor-mediated degradation by the renal tubular epithelial cells (7). Hence, it comes as no surprise that circulating insulin concentrations are increased in individuals with renal impairment. Furthermore, insulin resistance could also be a modifying factor contributing to hyperinsulinemia in these individuals (8).
Other purported mechanisms of spontaneous hypoglycemia in renal impairment include diminished caloric intake, leading to chronic malnutrition, reduced renal gluconeogenesis, impaired counter-regulatory hormone response (9), and use of high-glucose-containing dialysate during hemodialysis or peritoneal dialysis (10). Substrate limitation for gluconeogenesis may also be an important factor toward development of spontaneous hypoglycemia in these individuals (11). Other causes include the effect of drugs, e.g. propoxyphene (12), salicylates (13), and Cotrimoxazole (14). However, defining features of spontaneous hypoglycemia associated with uremia are that: 1) it occurs in patients who are ill-appearing; and 2) the prognosis is ominous, as over 50% of these patients succumb to their illness within months (15). In contrast, our patient was well nourished and had no signs of inanition or malnutrition, and his symptoms were clearly mediated by endogenous hyperinsulinemia.
Utility of ß-cell polypeptides in the presence of renal failure
The diagnosis of endogenous hyperinsulinemia in the absence of renal insufficiency can be confirmed by well-established criteria (16). These criteria include symptoms and signs of neuroglycopenia with concomitant serum glucose of no more than 45 mg/dl (2.5 mM), insulin of at least 36 pM (RIA) or at least 18 pM (ICMA), c-peptide of at least 200 pM (ICMA), ß-hydroxybutyrate of no more than 2.7 mM, plasma glucose response to iv glucagon (1 mg) of at least 25 mg/dl (1.4 mM) with negative plasma sulfonylurea screen (including repaglinide) and negative insulin antibodies. Our patient fulfilled all of the above criteria. However, the presence of renal impairment made it difficult for us to interpret the values of the ß-cell polypeptides. The above diagnostic guidelines are laid down after careful study of individuals, with or without insulinoma, in the presence of normal renal function. We are not aware of any established criteria that would guide the clinician in interpretation of ß-cell polypeptides when renal function is abnormal. We have previously investigated a patient with hypoglycemia and renal failure who had a positive 72-h fast (serum glucose, 2.4 mM; insulin, 132 pM; and c-peptide, 3667 pM) at 28 h into the fast. Intraoperative ultrasound and careful mobilization and palpation of the pancreas failed to detect an insulinoma.
This case and the case featured in this report are biochemically similar. The crucial difference was the identification of the insulinoma by a localization procedure (in this instance, endoscopic ultrasonography). It is our impression that, lacking positive localization, such cases of insulinoma and renal failure should have a positive selective arterial calcium stimulation test. This procedure, although contemplated in this case, was made unnecessary because of the positive ultrasound. Also, it was deemed risky to administer a dye load in the face of renal impairment.
Utility of prolonged fast, despite ß-cell peptide clearance issues
Our patient has indeed helped support the necessity for a properly conducted and supervised prolonged fast in patients with long and complicated history with confounding factors. Because we went into the fast to prove the presence or absence of Whipple’s triad, it was important that we did not end the fast at 26 h, when his plasma glucose had dropped to an asymptomatic 38 mg/dl (2.1 mM). He had no signs of neuroglycopenia and was alert, oriented, without headaches or visual symptoms, and could perform mathematical problems, including serial sevens. This is important because young healthy women, and (to a lesser extent) some healthy men, may have plasma glucose values in the range of 40 mg/dl (2.2 mM) or even lower during fasting (17).
Another notable point is the significance of the elevated ß-cell peptides despite the presence of renal impairment during steady state conditions, as in a prolonged fast. It would have been tempting to do a mixed meal test, given that most of our patient’s symptoms were indeed post prandial. However, during the extreme nonsteady state conditions of the meal, when there are rapid perturbations of glucose, insulin, and c-peptide levels, the interpretations of these values would have been impossible in the presence of renal impairment. In contrast, during steady state of the prolonged fast, the plasma concentrations of the peptides would be a logical function of the rate of secretion and the rate of clearance. Because the clearance rate during steady state is unlikely to change, the interpretation of the ß-cell peptide concentrations would be much less error-prone. In our case, the polypeptide levels were clearly and significantly elevated for the prevailing glucose concentration at the end of fast, so as to make a diagnosis of endogenous hyperinsulinemia reasonably tangible.
Utility of ß-hydroxy and glucagon tests in interpretation
Our patient also highlights the vital role played by estimation of ketone bodies and the end of fast glucagon test in the assessment and understanding of the pathophysiology of the abnormal 72-h fast. Because of the antiketogenic effect of insulin, the levels of ß-hydroxybutyrate would be low at the end of fast if hypoglycemia were mediated by insulin. Similarly, because of the glycogenic effect of insulin, glucagon administration would elevate plasma glucose through glycogenolysis at the end of fast if hypoglycemia were insulin-mediated. In contrast, noninsulin-mediated hypoglycemia, as that which could occur in renal failure, would result in increased levels of ß-hydroxybutyrate, with lack of glycemic response to glucagon at the end of fast (because hepatic glycogen stores would have been exhausted because of the relative lack of insulin) (18).
Rarity of combination of insulinoma and renal disease
Insulinoma and renal impairment are a rare combination. This case seems to be the first observed at the Mayo Clinic. Review of the literature has revealed only one prior report of insulinoma in a patient with renal failure (19). What makes this patient’s presentation more intriguing was the fact that, despite being on potentially diabetogenic drugs like prednisone (20) and cyclosporine (21), his symptoms of spontaneous hypoglycemia were progressively intensifying in severity and frequency. Another notable feature of this patient was his long history of symptoms of approximately 20 yr. The longest recorded history of symptoms in a patient with insulinoma has been 23 yr (22).
This case emphasizes that even with a properly conducted 72-h fast, the interpretation of ß-cell polypeptides in the presence of renal failure is difficult, and confirmation of suspected insulinoma may depend on identification by a localization procedure.
Acknowledgments
Footnotes
Abbreviation: ICMA, Immunochemiluminometric assay.
Received May 31, 2002.
Accepted August 13, 2002.
References
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
