- Email: [email protected]
Endocrine pancreatic response of children with onset of insulin-requiring diabetes before age 3 and after age 5
36
Jul.v 1982 The Journal o f P E D l A T R 1 C S
Endocrine pancreatic response of children with onset of insulin-requiring diabetes before age 3 and after age 5 The increased incidence of severe hypog@cemia reported in young children with diabetes is consistent with a deject in glueagon secretion or a generalized abnormality in islet hormone secretion. To assess pancreatic hormone and gastric inhibitory po@peptide secretion in children with ear@ onset diabetes, 12 children with onset of diabetes prior to the age o f 28 months were studied and the data compared to the hormone responses observed in t ] children with LOD, diagnosed after the age o f 5 years. Plasma glucose, C-peptide, gtucagon, pancreatic po@peptide, and gastric inhibitory peptide concentralions were measured during and following an arginine infusion (500 mg/kg over 60 minutes) and a mixed meal. During arginine infusion, plasma glucose and glucagon increased similar@ in both groups and returned to basal concentrations following discontinuation of arginine infusion. In contrast, plasma C-peptide, hPP, and GIP concentrations did not change. Following the mixed meal plasma glucose, hPP, and GIP concentrations increased similar@ in the two groups of children, but no change was observed in either plasma glucagon or C-peptide concentrations in either group. These data demonstrate that EOD and LOD are associated with insulin insufficiency alone and that abnormalities in secretion o f other pancreatic islet hormone or GIP cannot be implicated in the high incidence of severe hypog@cemia observed in children with EOD.
Christine Ternand, M.D., Vay L. W. Go, M.D., John E. Gerich, M.D., and Morey W. Haymond, M.D.,* Rochester, Minn.
AMONG THE CHILDREN with permanent insulin-requiring diabetes, two subpopulations may exist, depending on the age of onset. Children with early onset of diabetes, prior to the age of 3 years, are reported to have a number of factors which distinguish them from patients with a later onset of diabetes mellitus. These include: high prevalence of diabetes in the family, ~a predominance of affected males, ~-3less frequent transient remission of disease ("honeymoon"),4 an apparent increase in susceptibility to recurrent and severe episodes of hypoglycemia, particularly with
From the Departments o f Pediatrics and Medicine, Endocrine and Gastroenterology Research Units, Mayo Medical School and Foundation, Mayo Clinic. Supported by grants from the Wasie Foundation, Mayo Foundation, and the National Institutes of Health (AM-20973, AM-2041 I). *Reprint address: Mayo Clinic, Rochester, MN 55905.
Vol. 10l, No. 1, pp. 36-39
infections,t'5 and few episodes of ketoacidosis?'~ Absence of seasonal variation of onset of symptoms in the children less than 3 years of age has been observed by some but not by all investigators.:. 3 Abbreviations used EOD: early-onsetdiabetes LOD: late-onsetdiabetes GIP: gastricinhibitory peptide hPP: pancreatic polypeptide These clinical and cpidemiologic differences between classical insulin-dependent,ketosis-prone diabetes and early-onset diabetes suggest that the etiology and/or pathophysiology of the hyperglycemia in these two groups of children may be different. The increased susceptibility to severe hypoglycemia and decreased propensity to develop diabetic ketoacidosis would be consistent with a defect in the secretion of the counterregulating hormone, glucagon, as well as decreased insulin secretion?'7
0022-3476/82/070036+04500.40/0 9 1982 The C. V. Mosby Co.
Volume 101 Number 1
Endocrine pancreatic response with insulin-requiring diabetes
We recently studied a child with onset of diabetes in the neonatal period who, at 18 months of age, had no evidence of either endocrine or exocrine pancreatic function and was presumed to have pancreatic aplasia or hypoplasia.8 On the basis of this single patient, it was postulated that other children with EOD may have a generalized defect in pancreatic function. Therefore, the present studies were designed to determine if the pancreatic endocrine response to arginine and a mixed meal in children with onset of diabetes prior to age 3 is different from that observed in children with onset of disease after age 5 years. PATIENT POPULATION Twelve patients with EOD and 11 with onset after age 5 years (LOD) were studied (Table). The EOD group was younger (P > 0.1) at the time of study (mean 9.5 _+ 2.0 years) than the LOD group (mean 13.3_+ 0.9 years) and had a preponderance (P < 0.05) of males (Table). However, the duration of diabetes was greater (P < 0.01) in the EOD group. The LOD children were slightly smaller for their age than the EOD group (Table), but the insulin dose (U/kg/day) was higher (P < 0.01) in the LOD group than in the EOD children. The number of patients with primary and secondary relatives affected with diabetes in the EOD group when compared to the LOD group (9 vs 4, respectively) was not significantly different. In addition, it could not be determined whether or not these relatives had insulin-dependent or insulin-independent diabetes. Seven of the 12 EOD children had less than two episodes of ketoacidosis, whereas 7 of 11 LOD patients had two or more episodes (P > 0.1). Only one of nine EOD subjects had keotacidosis associated with intercurrent illness, whereas four of eight LOD patients had diabetic ketoacidosis in association with an intercurrent illness. Only three of 11 LOD children had seizures or loss of consciousness associated with hypoglycemia whereas 11 of 12 EOD patients had histories of severe insulin reactions (P < 0.01). STUDY PROTOCOL Following informed consent, all subjects were admitted to the Clinical Research Unit and were studied in the postabsorptive state after an overnight fast. Two intravenous needles were placed (one for administration of arginine and the other for blood sampling). Following three baseline blood samples, a 60-minute arginine infusion (500 mg/kg) was begun at 0 minutes. At 115 minutes the subjects received their usual morning dose of insulin and consumed their previously prescribed breakfast between 120 and 135 minutes. Four milliliters of blood was drawn at 15- to 30-minute intervals throughout the study and placed immediately in iced tubes containing EDTA
37
Table. Patient population studies Early-onset Later-onset diabetes (EOD) diabetes (LOD) Number A g c at onset (too)
Sex (male/female) Duration of diabetes (yr) Insulin dose at study (U/kg/day) Height (percentile for age) Weight (percentile for age) Number of subjects with severe hypoglycemia (seizure or coma)
12 16 ~+ 2*
10/2 8.2 _+ 2.0 0.69 • 0.03 41 _+ 10 58 _+ 9 11
I1 104 • 1 i t
4/7~ 4.7 _+ 1.1 t 1.06 + 0.099 33 _+ 8 42 +_ 8 3w
*Data expressed as mean _+ SE. t P < 0.01 from EOD group by standard t test. :~P < 0.05 from EOD group by chi square. w < 0.01 from EOD group by chi square.
and 5,000 IU of trasylol (FBA Pharmaceutical Co., New York). Plasma was separated by centrifugation and frozen at - 2 0 ~ until assay. Glucose, C-peptide, glucagon, hPP, and GIP values were obtained on each blood sample. M A T E R I A L S AND M E T H O D S Plasma glucose was determined by a glucose oxidase method using a Beckman glucose analyzer (Sunnydale, Calif.). C-peptides (Cal BioChem, Lot No. 72,90,93), glucagon (Unger 30K antiserum), gastric inhibitory polypeptide (J. D. Brown antiserum), and human pancreatic polypeptide were determined by radioimmunoassay. The reproducibility and detection limits of these various peptides have been described elsewhere. 9 Arginine was obtained from Cutter Laboratories (Berkeley, Calif.). Data are expressed as mean _+ SE. Differences in hormone and glucose concentrations between study groups were analyzed by the Student t test and analysis of variance. Differences in demographic information were analyzed by chi square analysis using the Fischer exact test or Student t test. RESULTS The baseline plasma glucose concentrations were similar in the EOD and LOD groups (247 _+ 31 and 234 +_ 28 mg/dl, respectively) and increased similarly to 310 _+ 25 and 311 _+ 30 mg/dl, respectively, during arginine infusions (Fig. 1). Upon discontinuation of the arginine, plasma glucose values returned toward basal values. Plasma C-peptide concentrations were low (less than 1.8 ng/ml in both groups) and did not change significantly in either group (Fig. 1). Two individuals with EOD had circulating antibodies to glucagon and could not be included in the
38
Ternand et al.
The Journal of Pediatrics July 1982
arginine ] infusion 0.5g/Kg
mixed meal
I
400
I
I
PLASMA GLUCOSE
300
arginine infusion O.Eg/Kg
I000
Insulin/ se~+
mixed meal
PLASIMA HPP
,nsul,n~ i
500
I
(D.
B
E
200
100
I
1000
E
0
O,-,-OLOD
i J~
I
PLASMA
o~ Q.
0
A 2 _ PLASM I L
I
I 0
I I
I
I
C-PEPTIDE I
n
a
,
60
120
180
MINUTES Fig. 1. Plasma glucose, C-peptide, and glucagon concentrations in response to an arginine infusion and mixed meal challenge in 10 patients with EOD and in 11 children with LOD. Two children with EOD had antibodies to glucagon and were excluded from analysis of the glucagon data.
data analyses. Plasma glucagon concentrations were 145 _+20 and 210 _+ 41 p g / m l in the LOD and EOD, respectively, increased similarly in both groups (660 _+ 80 for E O D and 880 + 100 p g / m l for LOD in response to arginine infusion, and decreased to baseline values following discontinuation of the amino acid infusion (Fig. 1). Although there appears to be a difference in glucagon response to arginine in the two groups, the differences were not significant by Student t test or analysis of variance, whether raw data, change from baseline, or percentage of change from baseline were used. Two children with EOD a n d one with L O D had circulating antibodies to hPP. Only one subject (EOD) had circulating antibodies to GIP. Plasma hPP (311 -+ 38 vs 316 -+ 37 p g / m l in E O D and LOD, respectively) and GIP (184 + 39 vs 137 + 33 p g / m l in EOD and LOD) concentrations were similar in both groups and showed no response to arginine infusion (Fig. 2). Following ingestion of the mixed meal, plasma glucose
I I I
~
EOD
I
iI
I
I I
500
500
0
I I
o,p,
GLUCAJGON
e~
e~
1000
I
i
I
0
60
120
I
180
MINUTES Fig. 2. Plasma pancreatic polypeptide and GIP concentration in response to an arginine infusion and mixed meal challenge in children with EOD and children with LOD. Of the EOD children, one had circulating antibodies to GIP and two had antibodies to hPP. One subject with LOD had circulating antibodies to hPP, but none had antibodies to GIP. Subjects with antibodies to specific hormones were excludcd from analysis of the respective hormone data.
concentration increased to similar values (296 _+ 16 m g / d l for E O D and 281 _+ 26 m g / d l for LOD, Fig. 1). No significant change in plasma C-peptide or glucagon was observed in either group during the 60 minutes following the meal challenge. Plasma hPP and G I P concentrations increased similarly in the EOD and L O D children following mixed meal challenge, and no significant difference was observed between the two groups. DISCUSSION In the present studies, children with E O D had little, if any, differences in plasma glucose, glucagon, C-peptide, hPP, and GIP responses to arginine or mixed meal when compared to children with onset of diabetes after the age of 5 years. These data suggest that both groups of children have a primary abnormality in B-cell function. Therefore, a generalized defect in islet cell function cannot be implicated in the pathophysiology of the carbohydrate intolerance in the children with E O D studied, as was reported in a child with presumed pancreatic agenesis, s The very high incidence of severe hypoglycemia observed in the EOD group children in the present study and previously reported ~,5 most likely is not the result of an
Volume 101 Number 1
Endocrine pancreatic response with insulin-requiring diabetes
abnormality in glucagon counterregulation, since administration of arginine increased plasma glucagon radioimmunoreactivity and plasma glucose concentrations. Other factors which might increase insulin sensitivity in EOD children, such as hypopituitarism or malabsorption, would seem unlikely since all were of appropriate height and weight for age. The 35% lower dose of insulin in the EOD group is difficult to interpret since these studies were not carried out under euglycemic conditions. However, the significantly lower insulin requirements in the EOD group when compared to those of the LOD children may suggest differences in insulin sensitivity, since no difference in plasma C-peptide or glucagon values was observed between the groups following either arginine infusion or a mixed meal challenge. Therefore, the increased frequency of severe hypoglycemia observed in the EOD children, despite their relatively lower dose of insulin, may relate to our inability to make appropriate small incremental changes (< 5%) in the dose of insulin required to adequately control the glycemic excursions in these young infants and children using commercially available insulin and insulin syringes; this finding suggests that more precise methods of insulin delivery (e.g., use of diluted insulin) may be of great importance in decreasing the severity and frequency of hypoglycemic seizures in this particularly vulnerable subgroup of juvenile-onset diabetic patients. The similarity of hormonal responses to arginine and a mixed meal in the LOD and E O D children in the present study provides biochemical evidence that the two groups of diabetic children are similar and thus represent a continuum of a common process, i.e., failure of the r to secrete
39
insulin. However, the clinical and epidemiologic data suggest that the etiology of the/3-cell failure in these two groups of subjects may be different.
REFERENCES
i. Francois R, Hermier M, Jurlot B, kamboreuf JC, David L, and Pourhande JM: Occurrence of diabetes in infants less than one year old, Mod Probl Pediatr 12:60, 1972. 2. Bloom A, Hayes TM, and Gamble DR: Register of newly diagnosed diabetic children, Br Med J 3:580, 1975. 3. Fleegler FM, Rogers KD, Drash A, Rosenbloom AL, Travis LB, and Court JM: Age, sex and season of onset of juvenile diabetes in different geographic areas, Pediatrics 63:374, 1979. 4. White P, and Graham CA: The child with diabetes, in Marble A, White P, Brandley R, and Krall L, editors: Joslin's diabetes mel!itus, ed 11, Philadelphia, 1971, Lea & Febiger, Publishers, p 339. 5. Grunt JA, Banion DM, Ling L, Saigel O, and Frost A: Problems in the care of the infant diabetic patient, Clin Pediatr 17:772, i978. 6. Unger R: Diabetes and the alpha-cell, Diabetes 25:136, 1976. 7. Gerich J, Haymond M, Rizza R, Verdonk C, and Miles J: Hormonal and substrate determinants of hepatic glucose production in man, in Veneziale CM, editor: Regulation of carbohydrate formation and utilization in mammals, Baltimore, 198!, University Park Press, 419. 8. Howard CP, Go VLW, lnfante A J, Perrault J, Gerich JE, and Haymond MW: Long4erm survival in a case of functional pancreatic agenesis, J PEDIATR97:786, 1980. 9. Service FJ, Nelson RL, Rubenstein AH, and Go VLW: Direct effect of insulin, glucagon, gastric inhibitory polypeptide and gastrin secretion during maintenance of normoglycemia, J Clin Endocrinol Metab 47:488, 1978.
Jul.v 1982 The Journal o f P E D l A T R 1 C S
Endocrine pancreatic response of children with onset of insulin-requiring diabetes before age 3 and after age 5 The increased incidence of severe hypog@cemia reported in young children with diabetes is consistent with a deject in glueagon secretion or a generalized abnormality in islet hormone secretion. To assess pancreatic hormone and gastric inhibitory po@peptide secretion in children with ear@ onset diabetes, 12 children with onset of diabetes prior to the age o f 28 months were studied and the data compared to the hormone responses observed in t ] children with LOD, diagnosed after the age o f 5 years. Plasma glucose, C-peptide, gtucagon, pancreatic po@peptide, and gastric inhibitory peptide concentralions were measured during and following an arginine infusion (500 mg/kg over 60 minutes) and a mixed meal. During arginine infusion, plasma glucose and glucagon increased similar@ in both groups and returned to basal concentrations following discontinuation of arginine infusion. In contrast, plasma C-peptide, hPP, and GIP concentrations did not change. Following the mixed meal plasma glucose, hPP, and GIP concentrations increased similar@ in the two groups of children, but no change was observed in either plasma glucagon or C-peptide concentrations in either group. These data demonstrate that EOD and LOD are associated with insulin insufficiency alone and that abnormalities in secretion o f other pancreatic islet hormone or GIP cannot be implicated in the high incidence of severe hypog@cemia observed in children with EOD.
Christine Ternand, M.D., Vay L. W. Go, M.D., John E. Gerich, M.D., and Morey W. Haymond, M.D.,* Rochester, Minn.
AMONG THE CHILDREN with permanent insulin-requiring diabetes, two subpopulations may exist, depending on the age of onset. Children with early onset of diabetes, prior to the age of 3 years, are reported to have a number of factors which distinguish them from patients with a later onset of diabetes mellitus. These include: high prevalence of diabetes in the family, ~a predominance of affected males, ~-3less frequent transient remission of disease ("honeymoon"),4 an apparent increase in susceptibility to recurrent and severe episodes of hypoglycemia, particularly with
From the Departments o f Pediatrics and Medicine, Endocrine and Gastroenterology Research Units, Mayo Medical School and Foundation, Mayo Clinic. Supported by grants from the Wasie Foundation, Mayo Foundation, and the National Institutes of Health (AM-20973, AM-2041 I). *Reprint address: Mayo Clinic, Rochester, MN 55905.
Vol. 10l, No. 1, pp. 36-39
infections,t'5 and few episodes of ketoacidosis?'~ Absence of seasonal variation of onset of symptoms in the children less than 3 years of age has been observed by some but not by all investigators.:. 3 Abbreviations used EOD: early-onsetdiabetes LOD: late-onsetdiabetes GIP: gastricinhibitory peptide hPP: pancreatic polypeptide These clinical and cpidemiologic differences between classical insulin-dependent,ketosis-prone diabetes and early-onset diabetes suggest that the etiology and/or pathophysiology of the hyperglycemia in these two groups of children may be different. The increased susceptibility to severe hypoglycemia and decreased propensity to develop diabetic ketoacidosis would be consistent with a defect in the secretion of the counterregulating hormone, glucagon, as well as decreased insulin secretion?'7
0022-3476/82/070036+04500.40/0 9 1982 The C. V. Mosby Co.
Volume 101 Number 1
Endocrine pancreatic response with insulin-requiring diabetes
We recently studied a child with onset of diabetes in the neonatal period who, at 18 months of age, had no evidence of either endocrine or exocrine pancreatic function and was presumed to have pancreatic aplasia or hypoplasia.8 On the basis of this single patient, it was postulated that other children with EOD may have a generalized defect in pancreatic function. Therefore, the present studies were designed to determine if the pancreatic endocrine response to arginine and a mixed meal in children with onset of diabetes prior to age 3 is different from that observed in children with onset of disease after age 5 years. PATIENT POPULATION Twelve patients with EOD and 11 with onset after age 5 years (LOD) were studied (Table). The EOD group was younger (P > 0.1) at the time of study (mean 9.5 _+ 2.0 years) than the LOD group (mean 13.3_+ 0.9 years) and had a preponderance (P < 0.05) of males (Table). However, the duration of diabetes was greater (P < 0.01) in the EOD group. The LOD children were slightly smaller for their age than the EOD group (Table), but the insulin dose (U/kg/day) was higher (P < 0.01) in the LOD group than in the EOD children. The number of patients with primary and secondary relatives affected with diabetes in the EOD group when compared to the LOD group (9 vs 4, respectively) was not significantly different. In addition, it could not be determined whether or not these relatives had insulin-dependent or insulin-independent diabetes. Seven of the 12 EOD children had less than two episodes of ketoacidosis, whereas 7 of 11 LOD patients had two or more episodes (P > 0.1). Only one of nine EOD subjects had keotacidosis associated with intercurrent illness, whereas four of eight LOD patients had diabetic ketoacidosis in association with an intercurrent illness. Only three of 11 LOD children had seizures or loss of consciousness associated with hypoglycemia whereas 11 of 12 EOD patients had histories of severe insulin reactions (P < 0.01). STUDY PROTOCOL Following informed consent, all subjects were admitted to the Clinical Research Unit and were studied in the postabsorptive state after an overnight fast. Two intravenous needles were placed (one for administration of arginine and the other for blood sampling). Following three baseline blood samples, a 60-minute arginine infusion (500 mg/kg) was begun at 0 minutes. At 115 minutes the subjects received their usual morning dose of insulin and consumed their previously prescribed breakfast between 120 and 135 minutes. Four milliliters of blood was drawn at 15- to 30-minute intervals throughout the study and placed immediately in iced tubes containing EDTA
37
Table. Patient population studies Early-onset Later-onset diabetes (EOD) diabetes (LOD) Number A g c at onset (too)
Sex (male/female) Duration of diabetes (yr) Insulin dose at study (U/kg/day) Height (percentile for age) Weight (percentile for age) Number of subjects with severe hypoglycemia (seizure or coma)
12 16 ~+ 2*
10/2 8.2 _+ 2.0 0.69 • 0.03 41 _+ 10 58 _+ 9 11
I1 104 • 1 i t
4/7~ 4.7 _+ 1.1 t 1.06 + 0.099 33 _+ 8 42 +_ 8 3w
*Data expressed as mean _+ SE. t P < 0.01 from EOD group by standard t test. :~P < 0.05 from EOD group by chi square. w < 0.01 from EOD group by chi square.
and 5,000 IU of trasylol (FBA Pharmaceutical Co., New York). Plasma was separated by centrifugation and frozen at - 2 0 ~ until assay. Glucose, C-peptide, glucagon, hPP, and GIP values were obtained on each blood sample. M A T E R I A L S AND M E T H O D S Plasma glucose was determined by a glucose oxidase method using a Beckman glucose analyzer (Sunnydale, Calif.). C-peptides (Cal BioChem, Lot No. 72,90,93), glucagon (Unger 30K antiserum), gastric inhibitory polypeptide (J. D. Brown antiserum), and human pancreatic polypeptide were determined by radioimmunoassay. The reproducibility and detection limits of these various peptides have been described elsewhere. 9 Arginine was obtained from Cutter Laboratories (Berkeley, Calif.). Data are expressed as mean _+ SE. Differences in hormone and glucose concentrations between study groups were analyzed by the Student t test and analysis of variance. Differences in demographic information were analyzed by chi square analysis using the Fischer exact test or Student t test. RESULTS The baseline plasma glucose concentrations were similar in the EOD and LOD groups (247 _+ 31 and 234 +_ 28 mg/dl, respectively) and increased similarly to 310 _+ 25 and 311 _+ 30 mg/dl, respectively, during arginine infusions (Fig. 1). Upon discontinuation of the arginine, plasma glucose values returned toward basal values. Plasma C-peptide concentrations were low (less than 1.8 ng/ml in both groups) and did not change significantly in either group (Fig. 1). Two individuals with EOD had circulating antibodies to glucagon and could not be included in the
38
Ternand et al.
The Journal of Pediatrics July 1982
arginine ] infusion 0.5g/Kg
mixed meal
I
400
I
I
PLASMA GLUCOSE
300
arginine infusion O.Eg/Kg
I000
Insulin/ se~+
mixed meal
PLASIMA HPP
,nsul,n~ i
500
I
(D.
B
E
200
100
I
1000
E
0
O,-,-OLOD
i J~
I
PLASMA
o~ Q.
0
A 2 _ PLASM I L
I
I 0
I I
I
I
C-PEPTIDE I
n
a
,
60
120
180
MINUTES Fig. 1. Plasma glucose, C-peptide, and glucagon concentrations in response to an arginine infusion and mixed meal challenge in 10 patients with EOD and in 11 children with LOD. Two children with EOD had antibodies to glucagon and were excluded from analysis of the glucagon data.
data analyses. Plasma glucagon concentrations were 145 _+20 and 210 _+ 41 p g / m l in the LOD and EOD, respectively, increased similarly in both groups (660 _+ 80 for E O D and 880 + 100 p g / m l for LOD in response to arginine infusion, and decreased to baseline values following discontinuation of the amino acid infusion (Fig. 1). Although there appears to be a difference in glucagon response to arginine in the two groups, the differences were not significant by Student t test or analysis of variance, whether raw data, change from baseline, or percentage of change from baseline were used. Two children with EOD a n d one with L O D had circulating antibodies to hPP. Only one subject (EOD) had circulating antibodies to GIP. Plasma hPP (311 -+ 38 vs 316 -+ 37 p g / m l in E O D and LOD, respectively) and GIP (184 + 39 vs 137 + 33 p g / m l in EOD and LOD) concentrations were similar in both groups and showed no response to arginine infusion (Fig. 2). Following ingestion of the mixed meal, plasma glucose
I I I
~
EOD
I
iI
I
I I
500
500
0
I I
o,p,
GLUCAJGON
e~
e~
1000
I
i
I
0
60
120
I
180
MINUTES Fig. 2. Plasma pancreatic polypeptide and GIP concentration in response to an arginine infusion and mixed meal challenge in children with EOD and children with LOD. Of the EOD children, one had circulating antibodies to GIP and two had antibodies to hPP. One subject with LOD had circulating antibodies to hPP, but none had antibodies to GIP. Subjects with antibodies to specific hormones were excludcd from analysis of the respective hormone data.
concentration increased to similar values (296 _+ 16 m g / d l for E O D and 281 _+ 26 m g / d l for LOD, Fig. 1). No significant change in plasma C-peptide or glucagon was observed in either group during the 60 minutes following the meal challenge. Plasma hPP and G I P concentrations increased similarly in the EOD and L O D children following mixed meal challenge, and no significant difference was observed between the two groups. DISCUSSION In the present studies, children with E O D had little, if any, differences in plasma glucose, glucagon, C-peptide, hPP, and GIP responses to arginine or mixed meal when compared to children with onset of diabetes after the age of 5 years. These data suggest that both groups of children have a primary abnormality in B-cell function. Therefore, a generalized defect in islet cell function cannot be implicated in the pathophysiology of the carbohydrate intolerance in the children with E O D studied, as was reported in a child with presumed pancreatic agenesis, s The very high incidence of severe hypoglycemia observed in the EOD group children in the present study and previously reported ~,5 most likely is not the result of an
Volume 101 Number 1
Endocrine pancreatic response with insulin-requiring diabetes
abnormality in glucagon counterregulation, since administration of arginine increased plasma glucagon radioimmunoreactivity and plasma glucose concentrations. Other factors which might increase insulin sensitivity in EOD children, such as hypopituitarism or malabsorption, would seem unlikely since all were of appropriate height and weight for age. The 35% lower dose of insulin in the EOD group is difficult to interpret since these studies were not carried out under euglycemic conditions. However, the significantly lower insulin requirements in the EOD group when compared to those of the LOD children may suggest differences in insulin sensitivity, since no difference in plasma C-peptide or glucagon values was observed between the groups following either arginine infusion or a mixed meal challenge. Therefore, the increased frequency of severe hypoglycemia observed in the EOD children, despite their relatively lower dose of insulin, may relate to our inability to make appropriate small incremental changes (< 5%) in the dose of insulin required to adequately control the glycemic excursions in these young infants and children using commercially available insulin and insulin syringes; this finding suggests that more precise methods of insulin delivery (e.g., use of diluted insulin) may be of great importance in decreasing the severity and frequency of hypoglycemic seizures in this particularly vulnerable subgroup of juvenile-onset diabetic patients. The similarity of hormonal responses to arginine and a mixed meal in the LOD and E O D children in the present study provides biochemical evidence that the two groups of diabetic children are similar and thus represent a continuum of a common process, i.e., failure of the r to secrete
39
insulin. However, the clinical and epidemiologic data suggest that the etiology of the/3-cell failure in these two groups of subjects may be different.
REFERENCES
i. Francois R, Hermier M, Jurlot B, kamboreuf JC, David L, and Pourhande JM: Occurrence of diabetes in infants less than one year old, Mod Probl Pediatr 12:60, 1972. 2. Bloom A, Hayes TM, and Gamble DR: Register of newly diagnosed diabetic children, Br Med J 3:580, 1975. 3. Fleegler FM, Rogers KD, Drash A, Rosenbloom AL, Travis LB, and Court JM: Age, sex and season of onset of juvenile diabetes in different geographic areas, Pediatrics 63:374, 1979. 4. White P, and Graham CA: The child with diabetes, in Marble A, White P, Brandley R, and Krall L, editors: Joslin's diabetes mel!itus, ed 11, Philadelphia, 1971, Lea & Febiger, Publishers, p 339. 5. Grunt JA, Banion DM, Ling L, Saigel O, and Frost A: Problems in the care of the infant diabetic patient, Clin Pediatr 17:772, i978. 6. Unger R: Diabetes and the alpha-cell, Diabetes 25:136, 1976. 7. Gerich J, Haymond M, Rizza R, Verdonk C, and Miles J: Hormonal and substrate determinants of hepatic glucose production in man, in Veneziale CM, editor: Regulation of carbohydrate formation and utilization in mammals, Baltimore, 198!, University Park Press, 419. 8. Howard CP, Go VLW, lnfante A J, Perrault J, Gerich JE, and Haymond MW: Long4erm survival in a case of functional pancreatic agenesis, J PEDIATR97:786, 1980. 9. Service FJ, Nelson RL, Rubenstein AH, and Go VLW: Direct effect of insulin, glucagon, gastric inhibitory polypeptide and gastrin secretion during maintenance of normoglycemia, J Clin Endocrinol Metab 47:488, 1978.