0021-972X/91/7305-1056$03.00/0 Journal of Clinical Endocrinology and Metabolism Copyright © 1991 by The Endocrine Society
Vol. 73, No. 5 Printed in U.S.A.
Insulin-Resistant Diabetes Mellitus and Hypermetabolism in Mandibuloacral Dysplasia: A Newly Recognized Form of Partial Lipodystrophy* D. L. CUTLERf, S. KAUFMANN, AND G. R. FREIDENBERG Departments of Medicine (D.L.C) and Pediatrics (S.K., G.R.F), University of California-San Diego, La Jolla, California 92093
ABSTRACT. Mandibuloacral dysplasia (MAD) is a syndrome characterized by partial lipodystrophy and a distinct phenotype, which includes progressive osteolysis of the mandible and clavicles, cutaneous atrophy, joint contractures, and diabetes mellitus. We now describe the results of hyperinsulinemic glucose clamps performed in conjunction with indirect calorimetry in two subjects with MAD. At a glucose level of 5 mmol/L and insulin concentration of over 6.5 X 104 pmol/L, glucose disposal rates were less than 20% of maximum insulin-stimulated glucose disposal in five nondiabetic controls. Basal hepatic glucose out-
M
put was elevated in the two patients and was incompletely suppressed by a 1200 mU/m2-min infusion of insulin. Glucose and lipid oxidation rates were inappropriately elevated, reflecting marked hypermetabolism. Pharmacological concentrations of insulin failed to normally suppress lipid oxidation, diminish FFA levels, or adequately suppress glucagon levels. In summary, MAD is a unique form of lipodystrophic diabetes characterized by typical somatic features, extreme insulin resistance, and marked hypermetabolism. (J Clin Endocrinol Metab 73: 10561061,1991)
sent was obtained from each subject and an adult guardian, as appropriate. Subjects were admitted at least 48 h before testing. The metabolic parameters of these two subjects and the five nondiabetic males who served as controls are displayed in Table 1.
ANDIBULOACRAL dysplasia (MAD) is a syndrome characterized by partial lipodystrophy and a phenotype which includes mandibular and clavicular hypoplasia, dental abnormalities, acroosteolysis, speckled atrophic skin, and joint contractures (1-10). Occasionally associated with this syndrome are hypogonadism, short stature, and alopecia (1-4, 7, 8,10). Recently, we reported the occurrence of insulin-resistant diabetes mellitus and hyperlipidemia in three subjects with MAD (11). We have now further characterized the insulinresistant state in two of these patients, employing the hyperinsulinemic clamp technique and indirect calorimetry. These studies represent the first indirect calorimetry measurements in subjects with any type of lipodystrophy.
Glucose clamp studies
Materials and Methods Patients All studies were approved by the Human Subjects Review Committee, University of California-San Diego. Informed conReceived January 24,1991. Address all correspondence and requests for reprints to: David L. Cutler, M.D., UCSD Medical Center, H-203,225 Dickinson Street, San Diego, California 92103. * This work was supported by General Clinical Research Center Grant M01-RR-00827. t Supported in part by the Juvenile Diabetes Foundation (Grant 388348).
The ability of insulin to stimulate peripheral glucose uptake and suppress hepatic glucose production was assessed using a modification of the hyperinsulinemic clamp technique, as previously described (12-14). After an overnight fast, a primed (30 fiCi) continuous (0.3 j^Ci/min) infusion of [3-3H]glucose (New England Nuclear, Boston, MA) was started and continued throughout the study. After 2 h, a primed continuous infusion of insulin was initiated at a rate of 1200 mU/m2 • min in the subjects with MAD; in selected studies, cyclic somatostatin (Bachem Chemicals, Torrance, CA) was infused at a rate of 0.16 /ig/kg-min in combination with insulin to suppress endogenous insulin and glucagon release. Glucose levels were maintained at 3.3, 5.0, or 10.0 mmol/L by a variable rate infusion of a 20% dextrose solution. In controls, plasma glucose was maintained at 5.0 mmol/L, and insulin was infused at rates of 20, 40, and 240 mU/m2-min, yielding steady state serum insulin levels of 445 ± 29, 789 ± 43, and 1.07 X 104 ± 1.2 X 103 pmol/L. The glucose disappearance rate (Rd), rate of appearance (Ra), and hepatic glucose output (HGO) were calculated based on glucose specific activity measured in arterialized plasma during the last 40 min of each study period, using the equations of Steele in their modified derivative form (15,16). O2 consump-
1056
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 November 2015. at 11:48 For personal use only. No other uses without permission. . All rights reserved.
HYPERMETABOLISM IN MAD tion and CO2 production were measured during the final 40 min of each study, using a Deltatrac Metabolic Monitor (Sensormedics Corp., Anaheim, CA). Net glucose oxidation (GOJ and lipid oxidation (Lox) rates were determined according to the equations of Frayn (17). Nonoxidative glucose metabolism (Gnox), which is thought to represent primarily glycogen synthesis (18, 19), was determined by subtracting Gox from Rd. Resting energy expenditure was calculated by the equation of de Weir (20). Analytical methods Plasma glucose concentrations were measured using a YSI analyzer (Yellow Springs Instruments, Yellow Springs, OH). Glucose specific activity was determined on perchloric acid extracts of plasma. Plasma FFA levels were determined by the method of Novak (21). Glucagon (22), insulin (23), and catecholamine (24) levels were measured by standard RIAs. CPeptide concentrations were determined using a commercial kit (Incstar Corp., Stillwater, MN). Data analysis All calculations and data analysis were performed using the Clinfo Data Base Management and Analysis System. Data from control subjects are expressed as the mean ± SEM.
1057
Glucose disposal
As expected, when glucose levels were maintained at 5.0 mmol/L, insulin produced a 6- to 7-fold rise in Rd in the control subjects to 62.2 ± 3.0 /imol/kg-min at an insulin infusion rate of 240 mU/m2-min (Fig. 1). In contrast, insulin-stimulated Rd was markedly depressed in the two MAD subjects. At a glucose concentration of 5.0 mmol/L and an insulin infusion rate of 1200 mU/ m2-min, Rd values were 13.0 and 12.1 /imol/kg-min in subjects 1 and 2, respectively. By comparison, insulin levels of only 445 ± 29 pmol/L, reached during the 20 mU/m2 • min clamp in the controls, were associated with an Rd of 21.4 ± 2.2 ^mol/kg • min. The ability of exogenous insulin to increase Rd under hyperglycemic conditions was examined in subject 1. When the plasma glucose level was maintained at 10.0 mmol/L, Rd rose from 13.2 /immol/kg-min in the basal state to a maximum of 24.1 /xmrnol/kg-min. This small effect of insulin on Rd was easily offset by a change in plasma glucose, as illustrated by the fall in Rd to baseline when the plasma glucose level was lowered from 10.0 to 5.0 mmol/L (Fig. 1). HGO
Results Baseline metabolic characteristics The two MAD patients had fasting hyperglycemia, marked hyperinsulinemia, and hypertriglyceridemia (Table 1). Levels of GH, cortisol, and T4 were normal. Antiinsulin and antiinsulin receptor antibodies were absent. Baseline indirect calorimetry showed that both subjects were markedly hypermetabolic compared to expected caloric requirements (25).
Basal HGO was 9.5 ± 1.1 ^mol/kg-min in the control subjects and was fully suppressed at insulin levels at and above those achieved during the 20 mU/m2 • min clamp (Table 2). Relative to the control values, basal HGO was approximately 50% higher in the MAD subjects despite endogenous insulin levels of 1055 pmol/L in subject 1 and 639 pmol/L in subject 2. Furthermore, insulin failed to completely suppress HGO even at the maximum insulin levels achieved in the MAD subjects. 70
TABLE 1. Baseline metabolic characteristics
Age(yr) Body mass index (kg/m2) Fasting glucose (mmol/L) Fasting insulin (pmol/L) Cholesterol (mmol/L) Triglyceride (mmol/L) FFA (mmol/L) Glucagon (ng/L) Cortisol (nmol/L) GH (Mg/L) T< (nmol/L) Resting energy expenditure (Cal/day) Expected resting energy expenditure (Cal/day) Values are the mean ± SEM. " Mean ± SD.
^
60 50
Subject 1
Subject 2
Controls
*§
16 24 10.0 1055 4.55 4.84 0.57 180 83 0.5 91 2373
17 17 14.7 639 5.56 3.74 0.44 358 91 5.2 115 2060
28 ± 24 ± 5.1 ± 57 ±
I 40
1700 ± 56
1690
1480
1770 ± 54
6° 1 0.2 7
o
I 30
I20
10
0.59 ± 0.04 155 ± 9
Insulin Infusion Rale (mUMftmin) Glucose (mmoVL)
0
20
40
240
0
5.0
5.0
5.0
5.0
10.0
Controls
1200
1200
1200
10.0
Subject 1
5.0 , , 14.7
5.0
Subject 2
FIG. 1. Insulin-stimulated glucose disposal in control and MAD subjects. In five nondiabetic controls, Rd was measured at euglycemia in the fasting state and at insulin infusion rates of 20, 40, and 240 mU/ m2-min. Rd was also measured in two MAD subjects (subjects 1 and 2) in the basal state and during an insulin infusion rate of 1200 mU/m2 • min, under hyperglycemic and euglycemic conditions. Rd is plotted as micromoles per kg/min and shown as the mean ± SEM for the controls.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 November 2015. at 11:48 For personal use only. No other uses without permission. . All rights reserved.
CUTLER, KAUFMANN, AND FREIDENBERG
1058
JCE & M • 1991 Vol73«No5
TABLE 2. Metabolic characteristics during insulin infusion Subject no.
Control
I
(mU/m2-min)° Basal 20 40 240
2
Glucagon (ng/L) 155 ± 9 103 ± 5 131 ± 7 114 ± 5
CT?TTT')
5.0 5.0
Basal 1200 1200 1200 1200 1200 Basal 1200
1
VJ1UCOS6
(mmol/L) 5.0 5.0 10.0 10.0 10.0
180 129 81 182 85 75 379 334
+
5.0
5.0 3.3 14.7 5.0
FFA
(mmol/L) 0.59 ± 0.04 0.19 ± 0.01 0.23 ± 0.02 0.20 ± 0.02 0.57 0.35 0.27 0.47 0.29 0.30 0.44 0.36
+ +
C-Peptide (pmol/L)c — -
1950 1520
EPI
(pmol/L)d — 136 240 60
HGO
0tmol/kg-min) 9.5 ± 1.1 0.6 ± 0.3 0.2 ± 0.3 0.2 ± 0.2 13.9
2150
1.7 1.9 8.6 2.7 4.5
1850
131
15.6
560
202
4.1
50 200
Vol. 73, No. 5 Printed in U.S.A.
Insulin-Resistant Diabetes Mellitus and Hypermetabolism in Mandibuloacral Dysplasia: A Newly Recognized Form of Partial Lipodystrophy* D. L. CUTLERf, S. KAUFMANN, AND G. R. FREIDENBERG Departments of Medicine (D.L.C) and Pediatrics (S.K., G.R.F), University of California-San Diego, La Jolla, California 92093
ABSTRACT. Mandibuloacral dysplasia (MAD) is a syndrome characterized by partial lipodystrophy and a distinct phenotype, which includes progressive osteolysis of the mandible and clavicles, cutaneous atrophy, joint contractures, and diabetes mellitus. We now describe the results of hyperinsulinemic glucose clamps performed in conjunction with indirect calorimetry in two subjects with MAD. At a glucose level of 5 mmol/L and insulin concentration of over 6.5 X 104 pmol/L, glucose disposal rates were less than 20% of maximum insulin-stimulated glucose disposal in five nondiabetic controls. Basal hepatic glucose out-
M
put was elevated in the two patients and was incompletely suppressed by a 1200 mU/m2-min infusion of insulin. Glucose and lipid oxidation rates were inappropriately elevated, reflecting marked hypermetabolism. Pharmacological concentrations of insulin failed to normally suppress lipid oxidation, diminish FFA levels, or adequately suppress glucagon levels. In summary, MAD is a unique form of lipodystrophic diabetes characterized by typical somatic features, extreme insulin resistance, and marked hypermetabolism. (J Clin Endocrinol Metab 73: 10561061,1991)
sent was obtained from each subject and an adult guardian, as appropriate. Subjects were admitted at least 48 h before testing. The metabolic parameters of these two subjects and the five nondiabetic males who served as controls are displayed in Table 1.
ANDIBULOACRAL dysplasia (MAD) is a syndrome characterized by partial lipodystrophy and a phenotype which includes mandibular and clavicular hypoplasia, dental abnormalities, acroosteolysis, speckled atrophic skin, and joint contractures (1-10). Occasionally associated with this syndrome are hypogonadism, short stature, and alopecia (1-4, 7, 8,10). Recently, we reported the occurrence of insulin-resistant diabetes mellitus and hyperlipidemia in three subjects with MAD (11). We have now further characterized the insulinresistant state in two of these patients, employing the hyperinsulinemic clamp technique and indirect calorimetry. These studies represent the first indirect calorimetry measurements in subjects with any type of lipodystrophy.
Glucose clamp studies
Materials and Methods Patients All studies were approved by the Human Subjects Review Committee, University of California-San Diego. Informed conReceived January 24,1991. Address all correspondence and requests for reprints to: David L. Cutler, M.D., UCSD Medical Center, H-203,225 Dickinson Street, San Diego, California 92103. * This work was supported by General Clinical Research Center Grant M01-RR-00827. t Supported in part by the Juvenile Diabetes Foundation (Grant 388348).
The ability of insulin to stimulate peripheral glucose uptake and suppress hepatic glucose production was assessed using a modification of the hyperinsulinemic clamp technique, as previously described (12-14). After an overnight fast, a primed (30 fiCi) continuous (0.3 j^Ci/min) infusion of [3-3H]glucose (New England Nuclear, Boston, MA) was started and continued throughout the study. After 2 h, a primed continuous infusion of insulin was initiated at a rate of 1200 mU/m2 • min in the subjects with MAD; in selected studies, cyclic somatostatin (Bachem Chemicals, Torrance, CA) was infused at a rate of 0.16 /ig/kg-min in combination with insulin to suppress endogenous insulin and glucagon release. Glucose levels were maintained at 3.3, 5.0, or 10.0 mmol/L by a variable rate infusion of a 20% dextrose solution. In controls, plasma glucose was maintained at 5.0 mmol/L, and insulin was infused at rates of 20, 40, and 240 mU/m2-min, yielding steady state serum insulin levels of 445 ± 29, 789 ± 43, and 1.07 X 104 ± 1.2 X 103 pmol/L. The glucose disappearance rate (Rd), rate of appearance (Ra), and hepatic glucose output (HGO) were calculated based on glucose specific activity measured in arterialized plasma during the last 40 min of each study period, using the equations of Steele in their modified derivative form (15,16). O2 consump-
1056
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 November 2015. at 11:48 For personal use only. No other uses without permission. . All rights reserved.
HYPERMETABOLISM IN MAD tion and CO2 production were measured during the final 40 min of each study, using a Deltatrac Metabolic Monitor (Sensormedics Corp., Anaheim, CA). Net glucose oxidation (GOJ and lipid oxidation (Lox) rates were determined according to the equations of Frayn (17). Nonoxidative glucose metabolism (Gnox), which is thought to represent primarily glycogen synthesis (18, 19), was determined by subtracting Gox from Rd. Resting energy expenditure was calculated by the equation of de Weir (20). Analytical methods Plasma glucose concentrations were measured using a YSI analyzer (Yellow Springs Instruments, Yellow Springs, OH). Glucose specific activity was determined on perchloric acid extracts of plasma. Plasma FFA levels were determined by the method of Novak (21). Glucagon (22), insulin (23), and catecholamine (24) levels were measured by standard RIAs. CPeptide concentrations were determined using a commercial kit (Incstar Corp., Stillwater, MN). Data analysis All calculations and data analysis were performed using the Clinfo Data Base Management and Analysis System. Data from control subjects are expressed as the mean ± SEM.
1057
Glucose disposal
As expected, when glucose levels were maintained at 5.0 mmol/L, insulin produced a 6- to 7-fold rise in Rd in the control subjects to 62.2 ± 3.0 /imol/kg-min at an insulin infusion rate of 240 mU/m2-min (Fig. 1). In contrast, insulin-stimulated Rd was markedly depressed in the two MAD subjects. At a glucose concentration of 5.0 mmol/L and an insulin infusion rate of 1200 mU/ m2-min, Rd values were 13.0 and 12.1 /imol/kg-min in subjects 1 and 2, respectively. By comparison, insulin levels of only 445 ± 29 pmol/L, reached during the 20 mU/m2 • min clamp in the controls, were associated with an Rd of 21.4 ± 2.2 ^mol/kg • min. The ability of exogenous insulin to increase Rd under hyperglycemic conditions was examined in subject 1. When the plasma glucose level was maintained at 10.0 mmol/L, Rd rose from 13.2 /immol/kg-min in the basal state to a maximum of 24.1 /xmrnol/kg-min. This small effect of insulin on Rd was easily offset by a change in plasma glucose, as illustrated by the fall in Rd to baseline when the plasma glucose level was lowered from 10.0 to 5.0 mmol/L (Fig. 1). HGO
Results Baseline metabolic characteristics The two MAD patients had fasting hyperglycemia, marked hyperinsulinemia, and hypertriglyceridemia (Table 1). Levels of GH, cortisol, and T4 were normal. Antiinsulin and antiinsulin receptor antibodies were absent. Baseline indirect calorimetry showed that both subjects were markedly hypermetabolic compared to expected caloric requirements (25).
Basal HGO was 9.5 ± 1.1 ^mol/kg-min in the control subjects and was fully suppressed at insulin levels at and above those achieved during the 20 mU/m2 • min clamp (Table 2). Relative to the control values, basal HGO was approximately 50% higher in the MAD subjects despite endogenous insulin levels of 1055 pmol/L in subject 1 and 639 pmol/L in subject 2. Furthermore, insulin failed to completely suppress HGO even at the maximum insulin levels achieved in the MAD subjects. 70
TABLE 1. Baseline metabolic characteristics
Age(yr) Body mass index (kg/m2) Fasting glucose (mmol/L) Fasting insulin (pmol/L) Cholesterol (mmol/L) Triglyceride (mmol/L) FFA (mmol/L) Glucagon (ng/L) Cortisol (nmol/L) GH (Mg/L) T< (nmol/L) Resting energy expenditure (Cal/day) Expected resting energy expenditure (Cal/day) Values are the mean ± SEM. " Mean ± SD.
^
60 50
Subject 1
Subject 2
Controls
*§
16 24 10.0 1055 4.55 4.84 0.57 180 83 0.5 91 2373
17 17 14.7 639 5.56 3.74 0.44 358 91 5.2 115 2060
28 ± 24 ± 5.1 ± 57 ±
I 40
1700 ± 56
1690
1480
1770 ± 54
6° 1 0.2 7
o
I 30
I20
10
0.59 ± 0.04 155 ± 9
Insulin Infusion Rale (mUMftmin) Glucose (mmoVL)
0
20
40
240
0
5.0
5.0
5.0
5.0
10.0
Controls
1200
1200
1200
10.0
Subject 1
5.0 , , 14.7
5.0
Subject 2
FIG. 1. Insulin-stimulated glucose disposal in control and MAD subjects. In five nondiabetic controls, Rd was measured at euglycemia in the fasting state and at insulin infusion rates of 20, 40, and 240 mU/ m2-min. Rd was also measured in two MAD subjects (subjects 1 and 2) in the basal state and during an insulin infusion rate of 1200 mU/m2 • min, under hyperglycemic and euglycemic conditions. Rd is plotted as micromoles per kg/min and shown as the mean ± SEM for the controls.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 November 2015. at 11:48 For personal use only. No other uses without permission. . All rights reserved.
CUTLER, KAUFMANN, AND FREIDENBERG
1058
JCE & M • 1991 Vol73«No5
TABLE 2. Metabolic characteristics during insulin infusion Subject no.
Control
I
(mU/m2-min)° Basal 20 40 240
2
Glucagon (ng/L) 155 ± 9 103 ± 5 131 ± 7 114 ± 5
CT?TTT')
5.0 5.0
Basal 1200 1200 1200 1200 1200 Basal 1200
1
VJ1UCOS6
(mmol/L) 5.0 5.0 10.0 10.0 10.0
180 129 81 182 85 75 379 334
+
5.0
5.0 3.3 14.7 5.0
FFA
(mmol/L) 0.59 ± 0.04 0.19 ± 0.01 0.23 ± 0.02 0.20 ± 0.02 0.57 0.35 0.27 0.47 0.29 0.30 0.44 0.36
+ +
C-Peptide (pmol/L)c — -
1950 1520
EPI
(pmol/L)d — 136 240 60
HGO
0tmol/kg-min) 9.5 ± 1.1 0.6 ± 0.3 0.2 ± 0.3 0.2 ± 0.2 13.9
2150
1.7 1.9 8.6 2.7 4.5
1850
131
15.6
560
202
4.1
50 200
