:
CASE REPORTS
Reactive Hypoglycemic Coma Due to Insulin Autoimmune Syndrome: Case Report and Literature Review HENRYB.BURCH,M.D.,STEPHENCLEMENT,M.D.,MICHAELS.SOKOL,M.D., FRANKLANDRY,M.D., Washington,D.C.
Recurrent episodes of postprandial hypoglycemic symptoms culminated in hypoglycemic coma in a hypertensive but otherwise healthy man while he was taking hydralaxine. The patient was found to have an extreme elevation in the immunoreactive insulin level, leading to the discovery of insulin antibodies in the absence of prior exposure to exogenous insulin. Negative results of an anatomic study of the pancreas and an inability to reproduce hypoglycemia during a prolonged fast helped to exclude insuhnoma. In contrast, symptomatic hypoglycemia developed in response to oral glucose loading and was associated with an elevation in total and free insulin as well as C-peptide levels. The patient was diagnosed with insulin autoimmune syndrome, which, although a common source of hypoglycemia in Japan, has been well documented in only 15 cases from other countries. HLA typing revealed the patient to be positive for groups Cw4 and DR4, a combination that has been preliminarily associated with insulin autoimmune syndrome in Japan. Unlike the majority of cases previously reported, this patient had no clinical or serologic evidence of an underlying autoimmune disorder and had not been exposed to drugs containing sulfhydryl groups. This case adds to the world literature on insulin autoimmune syndrome, lends support to a postulated HLA association, and documents the presence of insulin autoantibodies in the absence of another underlying autoimmune disorder.
From the Department of Endocrine-Metabolic and Internal Medicine Services, Department of Medicine, Walter Reed Army Medical Center, Washington, D.C. The opinions or assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting the views of the Department of the Army or the Department of Defense: Requests for reprints should be addressed to Henry 6. Burch. M.D., Endocrine-Metabolic Service, Walter Reed Army Medical Center, Washington, D.C. 20307-5001. Manuscript submitted February 14. 1991, and accepted in revised form April 22, 1991.
T
he concurrent findings of severe hypoglycemia, an elevated immunoreactive insulin level, and insulin antibodies in the absence of prior exposure to exogenous insulin were first recognized and named the insulin autoimmune syndrome (IAS) by Hirata and Ishizu [l] in 1972. Although 124 cases of this syndrome have been described to date [2], the majority (89%) of these have occurred in Japan. Most of the 15 cases previously described outside of Japan [3-131 occurred in patients with clinical or biochemical evidence of underlying autoimmune disorders or during the course of therapy with sulfhydryl-containing drugs, which were presumed to be responsible for the development of insulin autoantibodies. We report here a new case of hypoglycemic coma due to insulin autoantibodies in a patient taking hydralazine yet with no clinical or serologic evidence of another autoimmune disease or exposure to drugs containing sulfhydryl groups.
CASEREPORT The patient is a 74-year-old man referred for evaluation of an episode of hypoglycemic coma occurring in January 1990. The patient was attending a convention on the evening of presentation and at approximately 7:30 PM had eaten a large meal, consisting primarily of seafood. He noted feeling a bit “dizzy” while bathing before retiring for the evening, but dismissed this as inconsequential. Five hours after his evening meal, the patient’s wife noted him to be nonresponsive and phoned paramedics, who on arrival at the patient’s hotel room found his fingerstick glucose level to be 1.67 mmol/L (30 mg/dL). Administration of intravenous dextrose resulted in a rapid return of consciousness. The immunoreactive insulin level was extremely high at 7.1 nmol/L (986 pIU/mL; normal range 5 to 25 &J/mL) with a concurrent C-peptide level of 2.7 nmol/L (8.1 ng/mL; normal range 0.5 to 3.0 ng/mL). The blood alcohol level was undetectable. On further questioning, the patient admitted to a severalmonth history of vague feelings of weakness and lightheadedness generally occurring while he was bowling in the evening. These episodes typically occurred 3 to 4 hours after he ate his evening meal, with amelioration of symptoms noted approximately 10 to 15 minutes after ingestion of a soft drink and a candy bar. In contrast, overnight fasting folJune
1992
The American
Journal
of Medicine
Volume
92
681
INSULIN
AUTOIMMUNE
SYNDROME
/ BURCH
ET AL
TABLE I Insulin Antibodies Reference Laboratory Smith-Kline Laboratories Total Insulin ~nmol/L)
4 ..
Eli-Lilly (to SciCor Lab., IN) 07:::::::
0
::::::::
I
9 12 13 17 21 25 33 37 41 45 49 57 61 65 69 72 Time
(hrr)
Figure 1. Seventy-two-hour fast in a patient with IAS. Total insulin reflects measurement of immunoreactive insulin by a double antibody method. The patient remained asymptomatic throughout the testing, and the serum glucose level failed to be reduced below 3.89 mmol/L (70 mg/dL).
lowed by brisk morning walks failed to produce similar symptoms. The patient’s past medical history was notable for no prior diagnosis of diabetes mellitus, and no personal or family history of insulin usage or occupation in an allied health profession. He had a history of hypertension for which he had continuously received hydralazine 50 mg twice daily, and longacting propranolol 120 mg once daily for greater than 3 years. He also had a history of podagra for which he received allopurinol 300 mg daily. The patient had no history of bowel resection, hepatic dysfunction, alcohol abuse, autoimmune disease, or psychiatric illness. Physical examination revealed an alert and cooperative man. Blood pressure was 140/84 mm Hg, and pulse was 68/min. Skin examination revealed no acanthosis nigricans. The result of thyroid examination was normal. The patient had no alopecia areata and no evidence of articular inflammation. A supervised 72-hour fast failed to produce hypoglycemia on two occasions, one of which is presented in Figure 1. Due to the extreme elevation in insulin levels, serum for insulin antibodies was obtained and sent to three different reference laboratories with confirmatory results as well as demonstration of cross-reactivity to porcine and bovine insulin (Table I). Insulin receptor antibodies by radioimmunoassay (kindly performed by Dr. Guenter Boden, University of Pennsylvania) were negative. The proinsulin level was elevated at 166 pmol/L (normal less than 53.2 pmol/L). Complete cell count, serum chemistry profile, and thyroid function test results were within normal limits. Rheumatoid factor, antinuclear antibody, anti682
June
1992
The American
Journal
of Medicine
Volume
92
Joslin Research Laboratory
Method % B/T Patient Sample 1 Sample 2 Control
Result by Species Human Porcine Bovine
NP NP NP
15 14
CASE REPORTS
Reactive Hypoglycemic Coma Due to Insulin Autoimmune Syndrome: Case Report and Literature Review HENRYB.BURCH,M.D.,STEPHENCLEMENT,M.D.,MICHAELS.SOKOL,M.D., FRANKLANDRY,M.D., Washington,D.C.
Recurrent episodes of postprandial hypoglycemic symptoms culminated in hypoglycemic coma in a hypertensive but otherwise healthy man while he was taking hydralaxine. The patient was found to have an extreme elevation in the immunoreactive insulin level, leading to the discovery of insulin antibodies in the absence of prior exposure to exogenous insulin. Negative results of an anatomic study of the pancreas and an inability to reproduce hypoglycemia during a prolonged fast helped to exclude insuhnoma. In contrast, symptomatic hypoglycemia developed in response to oral glucose loading and was associated with an elevation in total and free insulin as well as C-peptide levels. The patient was diagnosed with insulin autoimmune syndrome, which, although a common source of hypoglycemia in Japan, has been well documented in only 15 cases from other countries. HLA typing revealed the patient to be positive for groups Cw4 and DR4, a combination that has been preliminarily associated with insulin autoimmune syndrome in Japan. Unlike the majority of cases previously reported, this patient had no clinical or serologic evidence of an underlying autoimmune disorder and had not been exposed to drugs containing sulfhydryl groups. This case adds to the world literature on insulin autoimmune syndrome, lends support to a postulated HLA association, and documents the presence of insulin autoantibodies in the absence of another underlying autoimmune disorder.
From the Department of Endocrine-Metabolic and Internal Medicine Services, Department of Medicine, Walter Reed Army Medical Center, Washington, D.C. The opinions or assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting the views of the Department of the Army or the Department of Defense: Requests for reprints should be addressed to Henry 6. Burch. M.D., Endocrine-Metabolic Service, Walter Reed Army Medical Center, Washington, D.C. 20307-5001. Manuscript submitted February 14. 1991, and accepted in revised form April 22, 1991.
T
he concurrent findings of severe hypoglycemia, an elevated immunoreactive insulin level, and insulin antibodies in the absence of prior exposure to exogenous insulin were first recognized and named the insulin autoimmune syndrome (IAS) by Hirata and Ishizu [l] in 1972. Although 124 cases of this syndrome have been described to date [2], the majority (89%) of these have occurred in Japan. Most of the 15 cases previously described outside of Japan [3-131 occurred in patients with clinical or biochemical evidence of underlying autoimmune disorders or during the course of therapy with sulfhydryl-containing drugs, which were presumed to be responsible for the development of insulin autoantibodies. We report here a new case of hypoglycemic coma due to insulin autoantibodies in a patient taking hydralazine yet with no clinical or serologic evidence of another autoimmune disease or exposure to drugs containing sulfhydryl groups.
CASEREPORT The patient is a 74-year-old man referred for evaluation of an episode of hypoglycemic coma occurring in January 1990. The patient was attending a convention on the evening of presentation and at approximately 7:30 PM had eaten a large meal, consisting primarily of seafood. He noted feeling a bit “dizzy” while bathing before retiring for the evening, but dismissed this as inconsequential. Five hours after his evening meal, the patient’s wife noted him to be nonresponsive and phoned paramedics, who on arrival at the patient’s hotel room found his fingerstick glucose level to be 1.67 mmol/L (30 mg/dL). Administration of intravenous dextrose resulted in a rapid return of consciousness. The immunoreactive insulin level was extremely high at 7.1 nmol/L (986 pIU/mL; normal range 5 to 25 &J/mL) with a concurrent C-peptide level of 2.7 nmol/L (8.1 ng/mL; normal range 0.5 to 3.0 ng/mL). The blood alcohol level was undetectable. On further questioning, the patient admitted to a severalmonth history of vague feelings of weakness and lightheadedness generally occurring while he was bowling in the evening. These episodes typically occurred 3 to 4 hours after he ate his evening meal, with amelioration of symptoms noted approximately 10 to 15 minutes after ingestion of a soft drink and a candy bar. In contrast, overnight fasting folJune
1992
The American
Journal
of Medicine
Volume
92
681
INSULIN
AUTOIMMUNE
SYNDROME
/ BURCH
ET AL
TABLE I Insulin Antibodies Reference Laboratory Smith-Kline Laboratories Total Insulin ~nmol/L)
4 ..
Eli-Lilly (to SciCor Lab., IN) 07:::::::
0
::::::::
I
9 12 13 17 21 25 33 37 41 45 49 57 61 65 69 72 Time
(hrr)
Figure 1. Seventy-two-hour fast in a patient with IAS. Total insulin reflects measurement of immunoreactive insulin by a double antibody method. The patient remained asymptomatic throughout the testing, and the serum glucose level failed to be reduced below 3.89 mmol/L (70 mg/dL).
lowed by brisk morning walks failed to produce similar symptoms. The patient’s past medical history was notable for no prior diagnosis of diabetes mellitus, and no personal or family history of insulin usage or occupation in an allied health profession. He had a history of hypertension for which he had continuously received hydralazine 50 mg twice daily, and longacting propranolol 120 mg once daily for greater than 3 years. He also had a history of podagra for which he received allopurinol 300 mg daily. The patient had no history of bowel resection, hepatic dysfunction, alcohol abuse, autoimmune disease, or psychiatric illness. Physical examination revealed an alert and cooperative man. Blood pressure was 140/84 mm Hg, and pulse was 68/min. Skin examination revealed no acanthosis nigricans. The result of thyroid examination was normal. The patient had no alopecia areata and no evidence of articular inflammation. A supervised 72-hour fast failed to produce hypoglycemia on two occasions, one of which is presented in Figure 1. Due to the extreme elevation in insulin levels, serum for insulin antibodies was obtained and sent to three different reference laboratories with confirmatory results as well as demonstration of cross-reactivity to porcine and bovine insulin (Table I). Insulin receptor antibodies by radioimmunoassay (kindly performed by Dr. Guenter Boden, University of Pennsylvania) were negative. The proinsulin level was elevated at 166 pmol/L (normal less than 53.2 pmol/L). Complete cell count, serum chemistry profile, and thyroid function test results were within normal limits. Rheumatoid factor, antinuclear antibody, anti682
June
1992
The American
Journal
of Medicine
Volume
92
Joslin Research Laboratory
Method % B/T Patient Sample 1 Sample 2 Control
Result by Species Human Porcine Bovine
NP NP NP
15 14
