- Email: [email protected]
Glucose tolerance and insulin response in parents of patients with insulin-dependent and juvenile-onset non-insulin-dependent diabetes mellitus
Diabetes Research and Clinical Practice, 16 (1992) 31-46 ID 1992 Elsevier Science Publishers B.V. All rights reserved
37
0168-8227/92/$05.00
DIABET 00614
Glucose tolerance and insulin response in parents of patients with insulin-dependent and juvenile-onset non-insulin-dependent diabetes mellitus Ayako Matsuda
and Takeshi Kuzuya
Division of Endocrinology and Metabolism. Jichi Medical School, Tochigi-ken, Japan
(Received 31 August 1991) (Revision accepted 11 November 1991)
Summary
Glucose tolerance and insulin response were examined using a 100 g oral glucose tolerance test (OGTT) in 108 parents of 23 patients with insulin-dependent (IDDM) and 3 1 patients with non-insulin-dependent diabetes mellitus (NIDDM), whose age of onset of diabetes was less than 35 years. Thirty-two agematched healthy volunteers without a family history of diabetes were also examined as a control group. Diabetes and impaired glucose tolerance (IGT) were significantly more frequent in parents of NIDDM (diabetes 34”/, IGT 27%) than in parents of IDDM (diabetes 7 %, IGT 13 %) (P < 0.001). At least one parent had diabetes or IGT in 307; of IDDM and 84% of NIDDM patients (P < O.OOl), and both parents had diabetes or IGT in 9% of IDDM and 39% of NIDDM patients (P -C 0.02). Even in cases with ‘normal’ glucose tolerance, the mean plasma glucose was higher in parents of NIDDM than in control subjects, suggesting a high prevalence of abnormal glucose tolerance including the marginal degree of abnormality in the families of NIDDM. The early phase insulin response was decreased more among parents of NIDDM with the greater impairment of glucose tolerance. However, among those with ‘normal’ glucose tolerance, early phase insulin response did not differ between parents of IDDM and NIDDM. and control subjects. The results confirmed a stronger familial background in NIDDM patients of younger onset than in IDDM. The different patterns of glucose tolerance among two parents of young-onset NIDDM patients suggest heterogeneity of the mode of inheritance of NIDDM among families.
Key
words:
Correspondence
Insulin-dependent (type 1) diabetes mellitus; Non-insulin-dependent (type 2) diabetes mellitus; Insulin response; Family history; Parents of juvenile-onset diabetic patients; Maturity onset type diabetes in the young; Early onset type 2 diabetes
to: A. Matsuda,
Division of Endocrinology
and Metabolism,
Jichi Medical School, Tochigi-ken,
Japan.
38 Introduction Heredity plays an important role in the etiology of both insulin-dependent (type 1) and non-insulindependent (type 2) diabetes mellitus. Of two major types of diabetes, non-insulin-dependent diabetes (NIDDM) seems to have a stronger genetic basis than insulin-dependent diabetes (IDDM) as suggested by twin studies and family history studies [l-3], but the mode of inheritance of NIDDM has not been elucidated. Previously, we found that the frequency of positive family history for diabetes in NIDDM was particularly high in patients with early onset of the disease and without previous obesity [4]. As our previous results were obtained only by questioning the patients, we attempted to study the prevalence of diabetes and impaired glucose tolerance (IGT) among parents of IDDM and early-onset NIDDM by an oral glucose tolerance test (OGTT). Insulin response during the OGTT was also studied in view of previous studies which showed that low insulin response is one of the genetic markers for NIDDM [5-71. There are reasons why we focused our attention on the parents. Firstly, if diabetes has a genetic background, at least one parent is expected to carry the diabetogenic genotype. Secondly, among NIDDM patients, the effect of the age of onset and obesity on the frequency of a positive family history for diabetes is more clearly observed in parents than in siblings [8]. Thirdly, as NIDDM is usually a disease of maturity-onset, there would be a better chance of detecting the abnormality ofglucose tolerance by examining the parents than by examining the siblings.
Subjects and Methods Subjects of this study were the 108 parents of 54 juvenile-onset diabetic patients seen at Jichi Medical School Hospital between 1977 to 1985. The diabetic patients whose age at the onset was less than 35 years and whose both parents were alive were requested to join this study. About a
half of the families consented to the proposed examination after explanation of the purpose of the study. Patients who had been treated for more than 5 years by diet alone or by sulfonylureas were classified as NIDDM. Serum C-peptide was measured in those who had been treated by insulin. When fasting serum C-peptide was below 0.5 ng/ml and maximal serum C-peptide 6-10 minutes after intravenous injection of 1 mg glucagon was less than 1.0 ng/ml, the probands were diagnosed as IDDM [9]. Patients with duration of diabetes longer than 5 years and who had a serum C-peptide higher than 1.0 ng/ml at fasting and/or higher than 1.5 ng/ml after glucagon stimulation were classified as NIDDM. Six patients who retained the C-peptide response but with a duration of diabetes shorter than 5 years, were classified as unknown type and excluded from the present analysis. The 54 probands were classified into 23 IDDM and 31 NIDDM. As a control group, 32 volunteers were recruited from hospital employees and neighbors of the diabetic patients and conformed to the following three criteria: (1) their ages were similar to the parents of diabetic patients, (2) known diabetes or other diseases were absent and (3) there was no family history of diabetes as far as was known. An oral glucose tolerance test (OGTT) was performed after an overnight fast using 300 ml Trelan G (Simizu Seiyaku Co. Japan), which was a solution of partial hydrolysate of starch corresponding to 100 g glucose and known to give a plasma glucose curve nearly the same as pure glucose solution [lo]. Blood was withdrawn from the antecubital vein into heparinized glass tubes before and 30, 60. 90, 120 and 180 min after ingestion of the solution. Glucose was measured immediately after separation of the plasma. Insulin was assayed after storage at - 20°C. Glucose was determined by a glucose oxidase method using Glucoloder S, AIC (Noda, Japan) and insulin was assayed by a double antibody method using a commercial kit (Dainabot, Japan). C-peptide was assayed by a kit from Daiichi Radioiso-
39 tope Laboratory (Tokyo). During the OGTT, the parents were interviewed, and physical examination was carried out. HLA DR typing was done in the Special Reference Laboratory by serological methods (Tokyo). Some parents with known diabetes and who were being treated with insulin were not tested by OGTT. Subjects in whom the fasting plasma glucose was 140 mg/dl or higher and/or plasma glucose at 120 min after oral glucose load was 200 mg/dl or higher were defined as diabetic. Impaired glucose tolerance (IGT) was defined when the fasting plasma glucose was less than 140 mg/dl and the 120 min plasma glucose was between 140 and 200 mg/dl. Others were defined as having normal glucose tolerance. The statistical significance was evaluated by the chi square test and Student’s t-test, either directly or after the logarithmic conversion.
some IDDM patients to become definitely insulin-dependent [ 111. The result of this classification seemed reasonable in view of the distribution of HLA DR antigens [ 121 and the frequency of ketosis in each group. Features of the parents The mean age, mean body mass index and prevalence of obese subjects (BMI 225.0) were not significantly different in parents of patients with IDDM and NIDDM (parents of IDDM and NIDDM), and in control subjects (Table 2). The frequency of known diabetes before OGTT was significantly higher in parents of NIDDM than in those of IDDM (26% vs 4%, P < 0.05). From the clinical features and plasma insulin and/or serum C-peptide levels, the types of known diabetes in parents were all judged as NIDDM. Two parents of IDDM and 8 parents of NIDDM told us that they had diabetic parents (i.e., grandparents of the probands).
Results Features of the probands Table 1 describes the basic clinical features of probands. The mean age of onset and the mean duration of diabetes were similar in IDDM and NIDDM probands. The mean maximal body mass index before the onset of diabetes was significantly higher in NIDDM (25.7) than in IDDM (21.0) (P < 0.01). Among 21 IDDM probands examined for HLA DR antigens, 19 had either HLA DR4 or DR9 or both, while 10 cases among 19 NIDDM probands had neither DR4 nor DR9. In IDDM, basal serum C-peptide was undetectable ( < 0.1 ng/ml) in 12 cases. In NIDDM, 14 cases were treated with insulin and the preservation of their serum C-peptide level was confirmed. In this study, the classification of the type of diabetes in each proband was made primarily by the C-peptide assay because of the ambiguity of other criteria for differentiating types of diabetes. Patients retaining the C-peptide response but were within 5 years of onset were excluded from the analysis, because it takes several years for
Glucose tolerance ofparents of diabetic patients and control subjects (Table 3) Plasma glucose during 100 g OGTT in parents of diabetic patients and in control subjects is shown in Table 3. Plasma glucose levels at fasting and after glucose load were significantly higher in parents of NIDDM than in the control group, while those in parents of IDDM did not differ from the control group. The distribution of fasting plasma glucose (FPG) and 2-h plasma glucose of father and mother is shown in Fig. 1. Among 46 parents of IDDM, 4 cases had slightly elevated FPG (1 lo-140 mg/dl), and one had a definitely elevated FPG of 200 mg/dl. Among 62 parents of NIDDM, FPG was slightly elevated in 12, and definitely elevated (a 140 mg/dl) in 14 including those with known diabetes. FPG exceeded 110 mg/dl in none of the control subjects. Elevated 2-h plasma glucose was more frequent in parents of NIDDM than in parents of IDDM. Even within ‘normal’ glucose tolerance, the distribution of 2-h plasma glucose appeared to be nearer the upper limit of the normal range (i.e.,
40 TABLE 1 Clinical features of the probands Type of diabetes
Number Male/female Age of onset of diabetes (yrs) Duration of diabetes (yrs) Maximal body mass index (kg/m’) Max BMI a25.0 Treatment insulin sulfonylureas diet Ketosis or ketoacidosis Serum CPR basal state CO.5 ng;ml > 0.5 nglml M + SD (ng;ml) stimulated i 1.5 ng;ml > 1.5 ng/ml M + SD (ng/ml) HLA DR DR DR DR
4:Y 4/X 9:X’ X:X’
IDDM
NIDDM
23 8115 21.2 f 8.2 (5-35)” 5.5 f 4.9 (0.5-19)’ 21.0 k 7.2 (17.7-35.1y 3:‘23
31 13/18 23.5 + 6.2 (14-34); 7.3 f 3.4 (5-19)” 25.7 + 4.1 (19.7-33.9y 16/31
‘3 0 0 13;21
14 9 8 l/31
23 0 0.28 2 0.1 Ih (
3 26 1.41 f 0.40b (0.47-2.82)u
23 0 0.44 _+0.25b (
2 27 4.00 + 1.98 (1.00-7.90)~ 0 7 1 10
I’ Ranges; ” the values are calculated in 11 cases with detectable C-peptide levels. Stimulated C-peptide is the value 6 min after intravenous injection of 1 mg glucagon.
140 mg/dl) in parents of NIDDM of IDDM.
than in parents
Glucose tolerance among parents of’ each proband
When the types of glucose tolerance in each group were classified, normal, IGT and diabetes were 80, 13 and 79: in parents of IDDM, and 39, 27 and 34”,b in parents of NIDDM, respectively (Table 2). Diabetes and IGT were significantly more prevalent in the parents of NIDDM than in parents of IDDM (P < 0.001). The percentage of
diabetes and IGT in the parents of IDDM was slightly higher than that of the control group, but the difference was not significant. Prevalence of abnormal glucose tolerance (diabetes and IGT) was similar between fathers and mothers of NIDDM. The age of onset of known diabetes was 54 and 57 years in 2 parents of IDDM and 44.6 + 18.3 (range: 32-63) years in 16 parents of NIDDM. In only 3 parents of NIDDM, had diabetes occurred before the age of 45 years. Table 4 shows the distribution of the types of
41
glucose tolerance in two parents (father and mother) of each proband. The frequency of probands who have at least one parent with ab-
normal glucose tolerance (D + IGT) was 7/23 (30%) in IDDM and 26/31 (84%) in NIDDM, respectively (P < 0.001).
TABLE 2 Features
of parents Controls
Parents of diabetic patients IDDM
NIDDM
Number (M/F) Age (yrs)
46 (23/23) 56.6 + 8.0 (43-76)J Body mass index (kg/m’) 22.6 + 3.6 (15.0-30.8) lo/46 (22”~) Obese cases (BMI >25) Known diabetes 2 (4”“) Pattern of OGTT (father : mother or male : female) normal 37 (20 : 17) IGT 6 (1 :5) diabetes 3 (2: l)b
62 (31/3l) 58.2 + 8.5 (40-77)n 23.6 f 3.1 (19.1-34.2)” 14/60 (23%) 16 (26%)
32 (l7/15) 54.6 f 7.9 (43-68) 23.2 + 2.6 (18.4-27.9)” 9/32 (28%) 0
24 (13: 11) 17 (8:9) 21 (10: 1l)b
28 (14 : 14) 4 (3: 1) 0
,’ Ranges; h including ‘known diabetes’.
TABLE 3 Changes in plasma glucose and serum insulin during 100 g OGTT in parents of diabetic patients ~Time (min)
Parents of patients with IDDM (II = 46)
Parents of patients with NIDDM (n = 62)
Controls (n = 32)
0 30 60 90 120 180 0 30 60 90 120 180 0 30 60 90 120 180
*P < 0.05. **P < 0.01, ***P < 0.001 vs corresponding
Mean + SD
and in control subjects Geometric
mean
Glucose
Insulin
Glucose
Insulin
(mgidf)
(pLJ/ml)
(mg/dl)
(nLJ/mL)
95 f 20 154 f 42 158 + 65 141 i 65 128 + 68 106 f 55 115 + 37** 198 f 67*** 222 f 97*** 211 f 97*** 191 f 101*** 158 i 96** 93& 11 141 & 26 138 + 39 120 t 42 107 + 32 100 f 21
6&4 52 + 49 48 f 24* 53 & 39 45 * 31 30 k 27 7*4 38 f 30* 58 + 52 57 _+39 59 f 39 37 + 24 6+3 55 f 35 67 f 48 51*34 46 i 37 34 + 25
94 153 147 132 117 98 111** 188*** 205*** 191*** 170*** 139** 92 139 132 113 103 97
5 41 42* 42 36 21 6 27** 42 45 47 30 6 45 55 40 34 25
values of control group.
42
Parents of IDDM patients Fasting
Parents of NIDDM patients
plasma
glucose
,I
110
140
m%dl
Mother
2-hr
Plasma
mgld 300
-
glucose
mgldl
4
300
-
$3
200
f c 2 140 ----LL
. t I
140
200
1
4
300
mg?dI
Mother
1 3
200 Mother
!
300
rnz
Fig. I. The distribution of plasma glucose values at fasting and 2 h after a 100 g glucose load in fathers and mothers of diabetic patients. One dot represents each proband. Two-h plasma glucose in known diabetics who were not examined by OGTT are shown by triangles arbitrarily placed above 300 mg/dl.
The percentage of probands whose both father and mother had an abnormal glucose tolerance was 39”, (12/31) in NIDDM and 9”,, (2/23) in IDDM (P< 0.02). Among these 12 NIDDM probands, 3 cases had two diabetic parents and 5 cases had one diabetic and one IGT parent. In 2 IDDM patients, one had one diabetic and one IGT parent, the other had two IGT parents.
The percentage of probands with one diabetic or IGT parent and the other with normal glucose tolerance was 4.5”” (14/3 1) in NIDDM (diabetes 10 and IGT 4) and 229, (5/23) in IDDM (diabetes 2 and IGT 3) (0.05 < P < 0.1). The frequencies of patients with diabetes in 3 vertical successive generations (grandparents, parents and probands) were l/23 (59;) in IDDM
43
Insulin response in parents of diabetic patients The plasma insulin responses during OGTT of each group are shown in Table 3. Geometric mean values are presented as well as arithmetic means because the distribution of raw insulin values at each time point is skewed and it becomes more symmetrical after the logarithmic transformation. Fasting insulin values did not differ significantly in parents of IDDM and NIDDM from control group. Insulin values at 30 min in parents of NIDDM and at 60 min in parents of IDDM were significantly lower than corresponding values in the control group. Fig. 2 illustrates the distribution of 30-min insulin values and the ratio of increment of insulin to that of plasma glucose at 30 min (AIRI/APG, insulinogenic index) in relation to the types of
TABLE 4 The glucose tolerance OGTT patterns two parents
of two parents of each proband
of
Normal x Normal Normal x Abnormal Normal x IGT Normal x Diabetes Abnormal x Abnormal IGT x IGT IGT x Diabetes Diabetes x Diabetes
Probands IDDM
NIDDM
16 (70”“) 5 (22”, ) 3 2 2 (99,) 1 1 0
5 (16”0) 14 (45”$) 4 10 12 (39”;) 4 5 3
and 5/29 (17%) in NIDDM (0.1 < P < 0.2). In these cases, the estimated age of onset of diabetes in parents ranged from 45 to 57 years.
Parents of patients with IDDM
NIDDM **
Parents of patients with
Control group
IDDM &
*** I g?
NlDDhn
Control group 2
”
9
2o
f
3
3’ c 9
3.0
.
0 0
0
E” 2.0
0
P 3
0
.
G3 1.0.
8 “0
3 5
2 .F3 2 0 (*> 1
g
0.5
2
0.3.
T 0.2 .E & 0.1. CI)
0
i
:
.
:
0.05 r 0.05
Types
of Gl-r
N IGT D
N IGT D
N IGT
. :
L T
5-
N tGTD
;;s:
N KjT D
N KjT
Fig. 2. Insulin response and insulinogenic index at 30 min after a 100 g glucose load in parents of diabetic patients and control group are shown according to the types of OGTT. Values are illustrated by logarithmic scale. Short horizontal lines indicate geometric means. Statistical differences are shown by * (P -C 0.05), ** (P < 0.01) and *** (P < 0.001).
44
insulin value at 120 min than in the control group. There was no difference in the insulinogenic index between parents of diabetic patients and the control group with ‘normal’ glucose tolerance (Fig. 2).
glucose tolerance. In parents of NIDDM, the 30min insulin and insulinogenic index decreased progressively with the impairment of glucose tolerance. In parents with diabetes, the insulinogenie index was less than 0.5 (~U/ml/mg/dl) except for one parent of NIDDM. In parents with IGT, a low insulinogenic index of less than 0.5 was more prevalent in parents of NIDDM (13/18) than in the control group (l/4) (0.05 < P < 0.1 vs control).
Discussion This is an extension of our previous study on the family history of diabetic patients [4,8]. In this study, glucose tolerance in parents of diabetic patients was evaluated by OGTT instead of mere inquiry. The results confirm that the frequency of diabetes and IGT was much higher in parents of early-onset NIDDM than in parents of IDDM and the control group. This strong familial tendency of abnormal glucose tolerance in NIDDM is in accordance with previous twin studies and family studies [l-4]. It is questionable whether early-onset NIDDM in this study represents NIDDM in general. Our NIDDM patients were selected only on the basis of an early age of onset. The distribution of each type of diabetes in young people differs in ethnic groups and countries. In Japan, IDDM is less
Compurison of subjects with ‘rtormal’ glucose tolerunce in parents of diabetic patients and the control group
Fig. 3 shows the mean plasma glucose and geometric mean insulin values during OGTT in cases with ‘normal’ glucose tolerance. Even though diabetes and IGT were excluded, the mean plasma glucose values were significantly higher in parents of NIDDM than in the control group at several time points. Parents of IDDM also had a slightly higher plasma glucose than controls at 30 min. Among these ‘normal’ cases, 60-min plasma glucose exceeded 200 mg/dl in l/30 control subjects, 2/37 parents of IDDM and 3/24 parents of NIDDM. Parents of NIDDM had a higher
oJ ,
0
30
60
QO
120
100
50
60’
QO
120
4 180
Time(minute) Fig. 3. Plasma glucose (mean and SD) and plasma insulin values (geometric mean) during a 100 g OGTT in parents and control parents of IDDM patients (n = 37) and --A-group with ‘normal’ OGTT. ----ok Control group (12= 28), --A-parents of NIDDM patients (n = 24). Statistical differences between parents of either IDDM or NIDDM and control group are shown by * (P < 0.05) and ** (P < 0.01).
45
than in Caucasians [ 131. Young-onset NIDDM is common in Japanese [ 141, Indians [ 151 and black Americans [ 161. The higher rate of a positive family history for diabetes in NIDDM than in IDDM is independent of the age of onset, and the decreasing tendency of the rate of positive family history with increasing age of onset is gradual and continuous [4]. Therefore, there is little reason to suppose that young-onset NIDDM as a whole constitutes a distinct category different from the usual maturity onset NIDDM in Japan. A special type of young-onset NIDDM with autosomal dominant inheritance was described by Tattersall [ 171, in which the age of onset of diabetes was also young in most other members of the family. In our NIDDM probands, five patients had diabetes in three successive generations, compatible with autosomal dominant inheritance. In all our cases, the estimated onset of diabetes in parents was after 45 years of age, although the possibility that diabetes in parents might have occurred earlier cannot be excluded. On the other hand, O’Rahilly et al. proposed another type of early-onset type 2 diabetes [ 18,191. In this type, nearly 90 % of parents of the probands had glucose intolerance. Frequent occurrence of abnormal glucose tolerance in both parents suggested a ‘double gene dose’ effect in this type of diabetes. Some NIDDM cases in which two parents had abnormal glucose tolerance are similar to this type of diabetes. It is conceivable that diabetes occurred at relatively early ages in some of these patients due to the inheritance of genetic factors from both parents. The diabetic genetic factors from two parents need not be of the same abnormality. In special types of diabetes due to insulin receptor abnormalities, there are cases in which both parents have different mutations of the receptor gene [ 201. Although the genetic abnormality is not elucidated in more common forms of NIDDM, the inheritance of susceptibility for diabetes from two parents may accelerate the onset of diabetes. There were five NIDDM patients whose both parents had ‘normal’ glucose tolerance in our
study. It is interesting that even among subjects with ‘normal’ glucose tolerance, parents of NIDDM still had significantly higher plasma glucose values during OGTT than parents of IDDM and the controls. It indicates that a slightly marginal degree of glucose intolerance is more common in parents of NIDDM. The rate of positive hereditary background among NIDDM would be further increased if these cases with marginal glucose intolerance were also taken into account. Early phase insulin response as represented by the 30-min value during OGTT was lower in parents of NIDDM than in other groups. This seems to be associated with the higher frequency of diabetes and IGT in parents of NIDDM. A decrease in early phase insulin response is not only a characteristic in definite NIDDM, but it is also frequently observed in subjects with IGT who later develop NIDDM [ 2 1,221. Previous studies have suggested that a low insulin response to glucose might be a genetic marker for NIDDM [6,7,23]. This study neither supports nor rejects the hypothesis because glucose intolerance and low insulin response occurred in parallel and it is difficult to decide the causal relationships from these data. Increased insulin concentration is found in relatives of NIDDM in Mexican Americans [24], suggesting a familial basis of insulin resistance in conferring NIDDM susceptibility. Our data are at variance from this because basal insulin levels in parents of IDDM and NIDDM were normal. in summary, we confirmed by OGTT that diabetes and IGT were more prevalent in parents of NIDDM than in those of IDDM. In parents of IDDM, the prevalence of abnormal glucose tolerance did not differ significantly from that in the control group. In young-onset NIDDM patients 84% had at least one and 39% had two with abnormal glucose parents tolerance (diabetes or IGT). The decreased early phase insulin response was found frequently in parents of NIDDM but it is mostly associated with the state of glucose intolerance. This study suggests that in early-onset NIDDM, the mode of inheritance of diabetes seems heterogeneous. Cases
46
with autosomal dominant inheritance are not frequent, and some of these patients resemble ‘earlyonset type 2 diabetes’ in Caucasians as described by O’Rahilly and Turner [ 18,191. Acknowledgements We appreciate the cooperation of the physicians of the Division of Endocrinology and Metabolism, Jichi Medical School. We also thank Miss T. Kawanago and Mrs. K. Awata for their laboratory assistance.
11
12
13
14
References Barnett, A.H., Eff, C., Leslie. R.D.G. and Pyke. D.A. (1981) Diabetes in identical twins. Diabetologia 20. 81-93. Committee on Diabetic Twins, Japan Diabetes Society (1988) Diabetes mellitus in twins: a cooperative study in Japan. Diabetes Res. Clin. Pratt. 15, 271-280. Tattersall, R.B. and Fajans, S.S. (1975) A difference between the inheritance of classical juvenile-onset and maturity-onset type diabetes of young people. Diabetes 2444-53. Kuzuya, T. and Matsuda. A. (1982) Family history of diabetes among Japanese patients with type 1 (insulindependent) and type 2 (non-insulin-dependent) diabetes. Diabetologia 22, 372-374. Cerasi, E. and Luft, R. (1967) Insulin response to glucose infusion in diabetic and non-diabetic monozygotic twin pairs. Genetic control of insulin response? Acta Endocrinol. (Copenhagen) 55, 330-345. Barnett, A.H., Spiliopoulos, A.J., Pyke, D.A., Stuggs, W.A., Burrin, J. and Alberti, K.G.M.M. (1981) Metabolic studies in unaffected co-twins of non-insulin dependent diabetics. Br. Med. J. 282, 1656-1658. Kosaka, K., Akanuma. Y., Hagura, R. and Kuzuya. T. (1979) Plasma insulin response during the 100 g glucose tolerance test in diabetic patients. In: S. Baba, T. Kaneko and N. Yanaihara (Eds.), Proinsulin, Insulin, C-peptide. Excerpta Medica ICS 468, Excerpta Medica, Amsterdam, pp. 283-293. Matsuda, A. and Kuzuya, T. (1984) Family history of Japanese patients with non-insulin-dependent diabetes. Differential implications of diabetes in parents and in siblings. Endocrinol. Jpn. 31. 335-341. 9 Matsuda, A., Kamata, I., Iwamoto, Y., Sakamoto, Y. and Kuzuya, T. (1985) A comparison of serum C-peptide response to intravenous glucagon. and urine C-peptide as indexes of insulin dependence. Diabetes Res. Clin. Pratt. 1. 161-167. 0 Nakano, M., Ito, M., Sakon, J. and Kondo. S. (1972) Changes of blood sugar. insulin and non-esteritied fatty
15
16
17 18
19
20
21
22
23
24
acid after Trelan G50 and glucose loading. A comparative study. J. Jpn. Diabetes Sot. 15, 314-315 (in Japanese). Kobayashi, T., Itoh, T.. Kosaka, K., Sato, K. and Tsuji, K. (1987) Time course of islet cell antibodies and B-cell function in non-insulin-dependent stage of type 1 diabetes. Diabetes 36, 10-17. Matsuda, A., Kuzuya, T., Iwamoto, Y. et al. (1988) HLA DR antigens in adult-onset and juvenile-onset Japanese insulin-dependent diabetic patients. Diabetes Res. Clin. Pratt. 5, 107-112. Tajima, N., LaPorte, R.E., Hibi, I., Kitagawa, T., Fujita. H. and Drash, A.L. (1985) Comparison of the epidemiology of youth-onset insulin-dependent diabetes mellitus between Japan and the United States (Allegheny County, Pennsylvania). Diabetes Care 8 (Suppl. 1), 17-23. Otani, T., Yokoyama, H., Higami, Y., Kasahara, T., Uchigata, Y. and Hirata, Y. (1990) Age at onset and type of diabetes diagnosed in Japanese patients under the age of 30. J. Jpn. Diabetes Sot. 33, 889-892 (in Japanese). Mohan, V., Ramachandran, A., Snehalatha, C., Mohan, R., Bharani, G. and Viswanathan, M. (1985) High prevalence of maturity-onset diabetes of the young (MODY) among Indians. Diabetes Care 8, 371-374. Winter, W.E., McLaren, N.K., Riley, W.J., Clarke, D.W., Kappy, M.S. and Spillar, R.P. (1987) Maturity-onset diabetes of youth in black Americans. N. Engl. J. Med. 316, 285-291. Tattersall, R.B. (1974) Mild familial diabetes with dominant inheritance. Quart. J. Med. NS 43, 339-357. O’Rahilly, S., Spivey, R.S., Holman, R.R., Nugent, A., Clark, A. and Turner, R.C. (1987) Type II diabetes of early onset: a distinct clinical and genetic syndrome? Lancet i, 923-928. O’Rahilly, S. and Turner, R.C. (1988) Early-onset Type 2 diabetes vs maturity-onset diabetes of youth: evidence for the existence of two discrete diabetic syndromes. Diabetic Med. 5, 224-229. Taylor, S.I., Kadowaki, T., Kadowaki, H., Accih, D., Cama, A. and McKeon, C. (1990) Mutations in insulinreceptor gene in insulin resistant patients. Diabetes Care 13, 257-279. Kosaka, K., Hagura, R. and Kuzuya, T. (1977) Insulin responses in equivocal and definite diabetes, with special reference to subjects who had mild glucose intolerance but later developed definite diabetes. Diabetes 26, 944-952. Kadowaki, T., Miyake, Y., Hagura, R. et al. (1984) Risk factors for worsening to diabetes in subjects with impaired glucose tolerance. Diabetologia 26, 44-49. Taylor, K.W., Sheldon, J., Pyke, D.A. and Oakley, W.G. ( 1967) Glucose tolerance and serum insulin in the unaffected first-degree relatives of diabetics. Br. Med. J. 4, 22-24. Haffner, S.M., Stern, M.P., Hazuda, H.P., Mitchell, B.D. and Patterson, J.K. (1988) Increased insulin concentrations in non-diabetic offspring of diabetic parents. N. Engl. J. Med. 319, 1297-1301.
37
0168-8227/92/$05.00
DIABET 00614
Glucose tolerance and insulin response in parents of patients with insulin-dependent and juvenile-onset non-insulin-dependent diabetes mellitus Ayako Matsuda
and Takeshi Kuzuya
Division of Endocrinology and Metabolism. Jichi Medical School, Tochigi-ken, Japan
(Received 31 August 1991) (Revision accepted 11 November 1991)
Summary
Glucose tolerance and insulin response were examined using a 100 g oral glucose tolerance test (OGTT) in 108 parents of 23 patients with insulin-dependent (IDDM) and 3 1 patients with non-insulin-dependent diabetes mellitus (NIDDM), whose age of onset of diabetes was less than 35 years. Thirty-two agematched healthy volunteers without a family history of diabetes were also examined as a control group. Diabetes and impaired glucose tolerance (IGT) were significantly more frequent in parents of NIDDM (diabetes 34”/, IGT 27%) than in parents of IDDM (diabetes 7 %, IGT 13 %) (P < 0.001). At least one parent had diabetes or IGT in 307; of IDDM and 84% of NIDDM patients (P < O.OOl), and both parents had diabetes or IGT in 9% of IDDM and 39% of NIDDM patients (P -C 0.02). Even in cases with ‘normal’ glucose tolerance, the mean plasma glucose was higher in parents of NIDDM than in control subjects, suggesting a high prevalence of abnormal glucose tolerance including the marginal degree of abnormality in the families of NIDDM. The early phase insulin response was decreased more among parents of NIDDM with the greater impairment of glucose tolerance. However, among those with ‘normal’ glucose tolerance, early phase insulin response did not differ between parents of IDDM and NIDDM. and control subjects. The results confirmed a stronger familial background in NIDDM patients of younger onset than in IDDM. The different patterns of glucose tolerance among two parents of young-onset NIDDM patients suggest heterogeneity of the mode of inheritance of NIDDM among families.
Key
words:
Correspondence
Insulin-dependent (type 1) diabetes mellitus; Non-insulin-dependent (type 2) diabetes mellitus; Insulin response; Family history; Parents of juvenile-onset diabetic patients; Maturity onset type diabetes in the young; Early onset type 2 diabetes
to: A. Matsuda,
Division of Endocrinology
and Metabolism,
Jichi Medical School, Tochigi-ken,
Japan.
38 Introduction Heredity plays an important role in the etiology of both insulin-dependent (type 1) and non-insulindependent (type 2) diabetes mellitus. Of two major types of diabetes, non-insulin-dependent diabetes (NIDDM) seems to have a stronger genetic basis than insulin-dependent diabetes (IDDM) as suggested by twin studies and family history studies [l-3], but the mode of inheritance of NIDDM has not been elucidated. Previously, we found that the frequency of positive family history for diabetes in NIDDM was particularly high in patients with early onset of the disease and without previous obesity [4]. As our previous results were obtained only by questioning the patients, we attempted to study the prevalence of diabetes and impaired glucose tolerance (IGT) among parents of IDDM and early-onset NIDDM by an oral glucose tolerance test (OGTT). Insulin response during the OGTT was also studied in view of previous studies which showed that low insulin response is one of the genetic markers for NIDDM [5-71. There are reasons why we focused our attention on the parents. Firstly, if diabetes has a genetic background, at least one parent is expected to carry the diabetogenic genotype. Secondly, among NIDDM patients, the effect of the age of onset and obesity on the frequency of a positive family history for diabetes is more clearly observed in parents than in siblings [8]. Thirdly, as NIDDM is usually a disease of maturity-onset, there would be a better chance of detecting the abnormality ofglucose tolerance by examining the parents than by examining the siblings.
Subjects and Methods Subjects of this study were the 108 parents of 54 juvenile-onset diabetic patients seen at Jichi Medical School Hospital between 1977 to 1985. The diabetic patients whose age at the onset was less than 35 years and whose both parents were alive were requested to join this study. About a
half of the families consented to the proposed examination after explanation of the purpose of the study. Patients who had been treated for more than 5 years by diet alone or by sulfonylureas were classified as NIDDM. Serum C-peptide was measured in those who had been treated by insulin. When fasting serum C-peptide was below 0.5 ng/ml and maximal serum C-peptide 6-10 minutes after intravenous injection of 1 mg glucagon was less than 1.0 ng/ml, the probands were diagnosed as IDDM [9]. Patients with duration of diabetes longer than 5 years and who had a serum C-peptide higher than 1.0 ng/ml at fasting and/or higher than 1.5 ng/ml after glucagon stimulation were classified as NIDDM. Six patients who retained the C-peptide response but with a duration of diabetes shorter than 5 years, were classified as unknown type and excluded from the present analysis. The 54 probands were classified into 23 IDDM and 31 NIDDM. As a control group, 32 volunteers were recruited from hospital employees and neighbors of the diabetic patients and conformed to the following three criteria: (1) their ages were similar to the parents of diabetic patients, (2) known diabetes or other diseases were absent and (3) there was no family history of diabetes as far as was known. An oral glucose tolerance test (OGTT) was performed after an overnight fast using 300 ml Trelan G (Simizu Seiyaku Co. Japan), which was a solution of partial hydrolysate of starch corresponding to 100 g glucose and known to give a plasma glucose curve nearly the same as pure glucose solution [lo]. Blood was withdrawn from the antecubital vein into heparinized glass tubes before and 30, 60. 90, 120 and 180 min after ingestion of the solution. Glucose was measured immediately after separation of the plasma. Insulin was assayed after storage at - 20°C. Glucose was determined by a glucose oxidase method using Glucoloder S, AIC (Noda, Japan) and insulin was assayed by a double antibody method using a commercial kit (Dainabot, Japan). C-peptide was assayed by a kit from Daiichi Radioiso-
39 tope Laboratory (Tokyo). During the OGTT, the parents were interviewed, and physical examination was carried out. HLA DR typing was done in the Special Reference Laboratory by serological methods (Tokyo). Some parents with known diabetes and who were being treated with insulin were not tested by OGTT. Subjects in whom the fasting plasma glucose was 140 mg/dl or higher and/or plasma glucose at 120 min after oral glucose load was 200 mg/dl or higher were defined as diabetic. Impaired glucose tolerance (IGT) was defined when the fasting plasma glucose was less than 140 mg/dl and the 120 min plasma glucose was between 140 and 200 mg/dl. Others were defined as having normal glucose tolerance. The statistical significance was evaluated by the chi square test and Student’s t-test, either directly or after the logarithmic conversion.
some IDDM patients to become definitely insulin-dependent [ 111. The result of this classification seemed reasonable in view of the distribution of HLA DR antigens [ 121 and the frequency of ketosis in each group. Features of the parents The mean age, mean body mass index and prevalence of obese subjects (BMI 225.0) were not significantly different in parents of patients with IDDM and NIDDM (parents of IDDM and NIDDM), and in control subjects (Table 2). The frequency of known diabetes before OGTT was significantly higher in parents of NIDDM than in those of IDDM (26% vs 4%, P < 0.05). From the clinical features and plasma insulin and/or serum C-peptide levels, the types of known diabetes in parents were all judged as NIDDM. Two parents of IDDM and 8 parents of NIDDM told us that they had diabetic parents (i.e., grandparents of the probands).
Results Features of the probands Table 1 describes the basic clinical features of probands. The mean age of onset and the mean duration of diabetes were similar in IDDM and NIDDM probands. The mean maximal body mass index before the onset of diabetes was significantly higher in NIDDM (25.7) than in IDDM (21.0) (P < 0.01). Among 21 IDDM probands examined for HLA DR antigens, 19 had either HLA DR4 or DR9 or both, while 10 cases among 19 NIDDM probands had neither DR4 nor DR9. In IDDM, basal serum C-peptide was undetectable ( < 0.1 ng/ml) in 12 cases. In NIDDM, 14 cases were treated with insulin and the preservation of their serum C-peptide level was confirmed. In this study, the classification of the type of diabetes in each proband was made primarily by the C-peptide assay because of the ambiguity of other criteria for differentiating types of diabetes. Patients retaining the C-peptide response but were within 5 years of onset were excluded from the analysis, because it takes several years for
Glucose tolerance ofparents of diabetic patients and control subjects (Table 3) Plasma glucose during 100 g OGTT in parents of diabetic patients and in control subjects is shown in Table 3. Plasma glucose levels at fasting and after glucose load were significantly higher in parents of NIDDM than in the control group, while those in parents of IDDM did not differ from the control group. The distribution of fasting plasma glucose (FPG) and 2-h plasma glucose of father and mother is shown in Fig. 1. Among 46 parents of IDDM, 4 cases had slightly elevated FPG (1 lo-140 mg/dl), and one had a definitely elevated FPG of 200 mg/dl. Among 62 parents of NIDDM, FPG was slightly elevated in 12, and definitely elevated (a 140 mg/dl) in 14 including those with known diabetes. FPG exceeded 110 mg/dl in none of the control subjects. Elevated 2-h plasma glucose was more frequent in parents of NIDDM than in parents of IDDM. Even within ‘normal’ glucose tolerance, the distribution of 2-h plasma glucose appeared to be nearer the upper limit of the normal range (i.e.,
40 TABLE 1 Clinical features of the probands Type of diabetes
Number Male/female Age of onset of diabetes (yrs) Duration of diabetes (yrs) Maximal body mass index (kg/m’) Max BMI a25.0 Treatment insulin sulfonylureas diet Ketosis or ketoacidosis Serum CPR basal state CO.5 ng;ml > 0.5 nglml M + SD (ng;ml) stimulated i 1.5 ng;ml > 1.5 ng/ml M + SD (ng/ml) HLA DR DR DR DR
4:Y 4/X 9:X’ X:X’
IDDM
NIDDM
23 8115 21.2 f 8.2 (5-35)” 5.5 f 4.9 (0.5-19)’ 21.0 k 7.2 (17.7-35.1y 3:‘23
31 13/18 23.5 + 6.2 (14-34); 7.3 f 3.4 (5-19)” 25.7 + 4.1 (19.7-33.9y 16/31
‘3 0 0 13;21
14 9 8 l/31
23 0 0.28 2 0.1 Ih (
3 26 1.41 f 0.40b (0.47-2.82)u
23 0 0.44 _+0.25b (
2 27 4.00 + 1.98 (1.00-7.90)~ 0 7 1 10
I’ Ranges; ” the values are calculated in 11 cases with detectable C-peptide levels. Stimulated C-peptide is the value 6 min after intravenous injection of 1 mg glucagon.
140 mg/dl) in parents of NIDDM of IDDM.
than in parents
Glucose tolerance among parents of’ each proband
When the types of glucose tolerance in each group were classified, normal, IGT and diabetes were 80, 13 and 79: in parents of IDDM, and 39, 27 and 34”,b in parents of NIDDM, respectively (Table 2). Diabetes and IGT were significantly more prevalent in the parents of NIDDM than in parents of IDDM (P < 0.001). The percentage of
diabetes and IGT in the parents of IDDM was slightly higher than that of the control group, but the difference was not significant. Prevalence of abnormal glucose tolerance (diabetes and IGT) was similar between fathers and mothers of NIDDM. The age of onset of known diabetes was 54 and 57 years in 2 parents of IDDM and 44.6 + 18.3 (range: 32-63) years in 16 parents of NIDDM. In only 3 parents of NIDDM, had diabetes occurred before the age of 45 years. Table 4 shows the distribution of the types of
41
glucose tolerance in two parents (father and mother) of each proband. The frequency of probands who have at least one parent with ab-
normal glucose tolerance (D + IGT) was 7/23 (30%) in IDDM and 26/31 (84%) in NIDDM, respectively (P < 0.001).
TABLE 2 Features
of parents Controls
Parents of diabetic patients IDDM
NIDDM
Number (M/F) Age (yrs)
46 (23/23) 56.6 + 8.0 (43-76)J Body mass index (kg/m’) 22.6 + 3.6 (15.0-30.8) lo/46 (22”~) Obese cases (BMI >25) Known diabetes 2 (4”“) Pattern of OGTT (father : mother or male : female) normal 37 (20 : 17) IGT 6 (1 :5) diabetes 3 (2: l)b
62 (31/3l) 58.2 + 8.5 (40-77)n 23.6 f 3.1 (19.1-34.2)” 14/60 (23%) 16 (26%)
32 (l7/15) 54.6 f 7.9 (43-68) 23.2 + 2.6 (18.4-27.9)” 9/32 (28%) 0
24 (13: 11) 17 (8:9) 21 (10: 1l)b
28 (14 : 14) 4 (3: 1) 0
,’ Ranges; h including ‘known diabetes’.
TABLE 3 Changes in plasma glucose and serum insulin during 100 g OGTT in parents of diabetic patients ~Time (min)
Parents of patients with IDDM (II = 46)
Parents of patients with NIDDM (n = 62)
Controls (n = 32)
0 30 60 90 120 180 0 30 60 90 120 180 0 30 60 90 120 180
*P < 0.05. **P < 0.01, ***P < 0.001 vs corresponding
Mean + SD
and in control subjects Geometric
mean
Glucose
Insulin
Glucose
Insulin
(mgidf)
(pLJ/ml)
(mg/dl)
(nLJ/mL)
95 f 20 154 f 42 158 + 65 141 i 65 128 + 68 106 f 55 115 + 37** 198 f 67*** 222 f 97*** 211 f 97*** 191 f 101*** 158 i 96** 93& 11 141 & 26 138 + 39 120 t 42 107 + 32 100 f 21
6&4 52 + 49 48 f 24* 53 & 39 45 * 31 30 k 27 7*4 38 f 30* 58 + 52 57 _+39 59 f 39 37 + 24 6+3 55 f 35 67 f 48 51*34 46 i 37 34 + 25
94 153 147 132 117 98 111** 188*** 205*** 191*** 170*** 139** 92 139 132 113 103 97
5 41 42* 42 36 21 6 27** 42 45 47 30 6 45 55 40 34 25
values of control group.
42
Parents of IDDM patients Fasting
Parents of NIDDM patients
plasma
glucose
,I
110
140
m%dl
Mother
2-hr
Plasma
mgld 300
-
glucose
mgldl
4
300
-
$3
200
f c 2 140 ----LL
. t I
140
200
1
4
300
mg?dI
Mother
1 3
200 Mother
!
300
rnz
Fig. I. The distribution of plasma glucose values at fasting and 2 h after a 100 g glucose load in fathers and mothers of diabetic patients. One dot represents each proband. Two-h plasma glucose in known diabetics who were not examined by OGTT are shown by triangles arbitrarily placed above 300 mg/dl.
The percentage of probands whose both father and mother had an abnormal glucose tolerance was 39”, (12/31) in NIDDM and 9”,, (2/23) in IDDM (P< 0.02). Among these 12 NIDDM probands, 3 cases had two diabetic parents and 5 cases had one diabetic and one IGT parent. In 2 IDDM patients, one had one diabetic and one IGT parent, the other had two IGT parents.
The percentage of probands with one diabetic or IGT parent and the other with normal glucose tolerance was 4.5”” (14/3 1) in NIDDM (diabetes 10 and IGT 4) and 229, (5/23) in IDDM (diabetes 2 and IGT 3) (0.05 < P < 0.1). The frequencies of patients with diabetes in 3 vertical successive generations (grandparents, parents and probands) were l/23 (59;) in IDDM
43
Insulin response in parents of diabetic patients The plasma insulin responses during OGTT of each group are shown in Table 3. Geometric mean values are presented as well as arithmetic means because the distribution of raw insulin values at each time point is skewed and it becomes more symmetrical after the logarithmic transformation. Fasting insulin values did not differ significantly in parents of IDDM and NIDDM from control group. Insulin values at 30 min in parents of NIDDM and at 60 min in parents of IDDM were significantly lower than corresponding values in the control group. Fig. 2 illustrates the distribution of 30-min insulin values and the ratio of increment of insulin to that of plasma glucose at 30 min (AIRI/APG, insulinogenic index) in relation to the types of
TABLE 4 The glucose tolerance OGTT patterns two parents
of two parents of each proband
of
Normal x Normal Normal x Abnormal Normal x IGT Normal x Diabetes Abnormal x Abnormal IGT x IGT IGT x Diabetes Diabetes x Diabetes
Probands IDDM
NIDDM
16 (70”“) 5 (22”, ) 3 2 2 (99,) 1 1 0
5 (16”0) 14 (45”$) 4 10 12 (39”;) 4 5 3
and 5/29 (17%) in NIDDM (0.1 < P < 0.2). In these cases, the estimated age of onset of diabetes in parents ranged from 45 to 57 years.
Parents of patients with IDDM
NIDDM **
Parents of patients with
Control group
IDDM &
*** I g?
NlDDhn
Control group 2
”
9
2o
f
3
3’ c 9
3.0
.
0 0
0
E” 2.0
0
P 3
0
.
G3 1.0.
8 “0
3 5
2 .F3 2 0 (*> 1
g
0.5
2
0.3.
T 0.2 .E & 0.1. CI)
0
i
:
.
:
0.05 r 0.05
Types
of Gl-r
N IGT D
N IGT D
N IGT
. :
L T
5-
N tGTD
;;s:
N KjT D
N KjT
Fig. 2. Insulin response and insulinogenic index at 30 min after a 100 g glucose load in parents of diabetic patients and control group are shown according to the types of OGTT. Values are illustrated by logarithmic scale. Short horizontal lines indicate geometric means. Statistical differences are shown by * (P -C 0.05), ** (P < 0.01) and *** (P < 0.001).
44
insulin value at 120 min than in the control group. There was no difference in the insulinogenic index between parents of diabetic patients and the control group with ‘normal’ glucose tolerance (Fig. 2).
glucose tolerance. In parents of NIDDM, the 30min insulin and insulinogenic index decreased progressively with the impairment of glucose tolerance. In parents with diabetes, the insulinogenie index was less than 0.5 (~U/ml/mg/dl) except for one parent of NIDDM. In parents with IGT, a low insulinogenic index of less than 0.5 was more prevalent in parents of NIDDM (13/18) than in the control group (l/4) (0.05 < P < 0.1 vs control).
Discussion This is an extension of our previous study on the family history of diabetic patients [4,8]. In this study, glucose tolerance in parents of diabetic patients was evaluated by OGTT instead of mere inquiry. The results confirm that the frequency of diabetes and IGT was much higher in parents of early-onset NIDDM than in parents of IDDM and the control group. This strong familial tendency of abnormal glucose tolerance in NIDDM is in accordance with previous twin studies and family studies [l-4]. It is questionable whether early-onset NIDDM in this study represents NIDDM in general. Our NIDDM patients were selected only on the basis of an early age of onset. The distribution of each type of diabetes in young people differs in ethnic groups and countries. In Japan, IDDM is less
Compurison of subjects with ‘rtormal’ glucose tolerunce in parents of diabetic patients and the control group
Fig. 3 shows the mean plasma glucose and geometric mean insulin values during OGTT in cases with ‘normal’ glucose tolerance. Even though diabetes and IGT were excluded, the mean plasma glucose values were significantly higher in parents of NIDDM than in the control group at several time points. Parents of IDDM also had a slightly higher plasma glucose than controls at 30 min. Among these ‘normal’ cases, 60-min plasma glucose exceeded 200 mg/dl in l/30 control subjects, 2/37 parents of IDDM and 3/24 parents of NIDDM. Parents of NIDDM had a higher
oJ ,
0
30
60
QO
120
100
50
60’
QO
120
4 180
Time(minute) Fig. 3. Plasma glucose (mean and SD) and plasma insulin values (geometric mean) during a 100 g OGTT in parents and control parents of IDDM patients (n = 37) and --A-group with ‘normal’ OGTT. ----ok Control group (12= 28), --A-parents of NIDDM patients (n = 24). Statistical differences between parents of either IDDM or NIDDM and control group are shown by * (P < 0.05) and ** (P < 0.01).
45
than in Caucasians [ 131. Young-onset NIDDM is common in Japanese [ 141, Indians [ 151 and black Americans [ 161. The higher rate of a positive family history for diabetes in NIDDM than in IDDM is independent of the age of onset, and the decreasing tendency of the rate of positive family history with increasing age of onset is gradual and continuous [4]. Therefore, there is little reason to suppose that young-onset NIDDM as a whole constitutes a distinct category different from the usual maturity onset NIDDM in Japan. A special type of young-onset NIDDM with autosomal dominant inheritance was described by Tattersall [ 171, in which the age of onset of diabetes was also young in most other members of the family. In our NIDDM probands, five patients had diabetes in three successive generations, compatible with autosomal dominant inheritance. In all our cases, the estimated onset of diabetes in parents was after 45 years of age, although the possibility that diabetes in parents might have occurred earlier cannot be excluded. On the other hand, O’Rahilly et al. proposed another type of early-onset type 2 diabetes [ 18,191. In this type, nearly 90 % of parents of the probands had glucose intolerance. Frequent occurrence of abnormal glucose tolerance in both parents suggested a ‘double gene dose’ effect in this type of diabetes. Some NIDDM cases in which two parents had abnormal glucose tolerance are similar to this type of diabetes. It is conceivable that diabetes occurred at relatively early ages in some of these patients due to the inheritance of genetic factors from both parents. The diabetic genetic factors from two parents need not be of the same abnormality. In special types of diabetes due to insulin receptor abnormalities, there are cases in which both parents have different mutations of the receptor gene [ 201. Although the genetic abnormality is not elucidated in more common forms of NIDDM, the inheritance of susceptibility for diabetes from two parents may accelerate the onset of diabetes. There were five NIDDM patients whose both parents had ‘normal’ glucose tolerance in our
study. It is interesting that even among subjects with ‘normal’ glucose tolerance, parents of NIDDM still had significantly higher plasma glucose values during OGTT than parents of IDDM and the controls. It indicates that a slightly marginal degree of glucose intolerance is more common in parents of NIDDM. The rate of positive hereditary background among NIDDM would be further increased if these cases with marginal glucose intolerance were also taken into account. Early phase insulin response as represented by the 30-min value during OGTT was lower in parents of NIDDM than in other groups. This seems to be associated with the higher frequency of diabetes and IGT in parents of NIDDM. A decrease in early phase insulin response is not only a characteristic in definite NIDDM, but it is also frequently observed in subjects with IGT who later develop NIDDM [ 2 1,221. Previous studies have suggested that a low insulin response to glucose might be a genetic marker for NIDDM [6,7,23]. This study neither supports nor rejects the hypothesis because glucose intolerance and low insulin response occurred in parallel and it is difficult to decide the causal relationships from these data. Increased insulin concentration is found in relatives of NIDDM in Mexican Americans [24], suggesting a familial basis of insulin resistance in conferring NIDDM susceptibility. Our data are at variance from this because basal insulin levels in parents of IDDM and NIDDM were normal. in summary, we confirmed by OGTT that diabetes and IGT were more prevalent in parents of NIDDM than in those of IDDM. In parents of IDDM, the prevalence of abnormal glucose tolerance did not differ significantly from that in the control group. In young-onset NIDDM patients 84% had at least one and 39% had two with abnormal glucose parents tolerance (diabetes or IGT). The decreased early phase insulin response was found frequently in parents of NIDDM but it is mostly associated with the state of glucose intolerance. This study suggests that in early-onset NIDDM, the mode of inheritance of diabetes seems heterogeneous. Cases
46
with autosomal dominant inheritance are not frequent, and some of these patients resemble ‘earlyonset type 2 diabetes’ in Caucasians as described by O’Rahilly and Turner [ 18,191. Acknowledgements We appreciate the cooperation of the physicians of the Division of Endocrinology and Metabolism, Jichi Medical School. We also thank Miss T. Kawanago and Mrs. K. Awata for their laboratory assistance.
11
12
13
14
References Barnett, A.H., Eff, C., Leslie. R.D.G. and Pyke. D.A. (1981) Diabetes in identical twins. Diabetologia 20. 81-93. Committee on Diabetic Twins, Japan Diabetes Society (1988) Diabetes mellitus in twins: a cooperative study in Japan. Diabetes Res. Clin. Pratt. 15, 271-280. Tattersall, R.B. and Fajans, S.S. (1975) A difference between the inheritance of classical juvenile-onset and maturity-onset type diabetes of young people. Diabetes 2444-53. Kuzuya, T. and Matsuda. A. (1982) Family history of diabetes among Japanese patients with type 1 (insulindependent) and type 2 (non-insulin-dependent) diabetes. Diabetologia 22, 372-374. Cerasi, E. and Luft, R. (1967) Insulin response to glucose infusion in diabetic and non-diabetic monozygotic twin pairs. Genetic control of insulin response? Acta Endocrinol. (Copenhagen) 55, 330-345. Barnett, A.H., Spiliopoulos, A.J., Pyke, D.A., Stuggs, W.A., Burrin, J. and Alberti, K.G.M.M. (1981) Metabolic studies in unaffected co-twins of non-insulin dependent diabetics. Br. Med. J. 282, 1656-1658. Kosaka, K., Akanuma. Y., Hagura, R. and Kuzuya. T. (1979) Plasma insulin response during the 100 g glucose tolerance test in diabetic patients. In: S. Baba, T. Kaneko and N. Yanaihara (Eds.), Proinsulin, Insulin, C-peptide. Excerpta Medica ICS 468, Excerpta Medica, Amsterdam, pp. 283-293. Matsuda, A. and Kuzuya, T. (1984) Family history of Japanese patients with non-insulin-dependent diabetes. Differential implications of diabetes in parents and in siblings. Endocrinol. Jpn. 31. 335-341. 9 Matsuda, A., Kamata, I., Iwamoto, Y., Sakamoto, Y. and Kuzuya, T. (1985) A comparison of serum C-peptide response to intravenous glucagon. and urine C-peptide as indexes of insulin dependence. Diabetes Res. Clin. Pratt. 1. 161-167. 0 Nakano, M., Ito, M., Sakon, J. and Kondo. S. (1972) Changes of blood sugar. insulin and non-esteritied fatty
15
16
17 18
19
20
21
22
23
24
acid after Trelan G50 and glucose loading. A comparative study. J. Jpn. Diabetes Sot. 15, 314-315 (in Japanese). Kobayashi, T., Itoh, T.. Kosaka, K., Sato, K. and Tsuji, K. (1987) Time course of islet cell antibodies and B-cell function in non-insulin-dependent stage of type 1 diabetes. Diabetes 36, 10-17. Matsuda, A., Kuzuya, T., Iwamoto, Y. et al. (1988) HLA DR antigens in adult-onset and juvenile-onset Japanese insulin-dependent diabetic patients. Diabetes Res. Clin. Pratt. 5, 107-112. Tajima, N., LaPorte, R.E., Hibi, I., Kitagawa, T., Fujita. H. and Drash, A.L. (1985) Comparison of the epidemiology of youth-onset insulin-dependent diabetes mellitus between Japan and the United States (Allegheny County, Pennsylvania). Diabetes Care 8 (Suppl. 1), 17-23. Otani, T., Yokoyama, H., Higami, Y., Kasahara, T., Uchigata, Y. and Hirata, Y. (1990) Age at onset and type of diabetes diagnosed in Japanese patients under the age of 30. J. Jpn. Diabetes Sot. 33, 889-892 (in Japanese). Mohan, V., Ramachandran, A., Snehalatha, C., Mohan, R., Bharani, G. and Viswanathan, M. (1985) High prevalence of maturity-onset diabetes of the young (MODY) among Indians. Diabetes Care 8, 371-374. Winter, W.E., McLaren, N.K., Riley, W.J., Clarke, D.W., Kappy, M.S. and Spillar, R.P. (1987) Maturity-onset diabetes of youth in black Americans. N. Engl. J. Med. 316, 285-291. Tattersall, R.B. (1974) Mild familial diabetes with dominant inheritance. Quart. J. Med. NS 43, 339-357. O’Rahilly, S., Spivey, R.S., Holman, R.R., Nugent, A., Clark, A. and Turner, R.C. (1987) Type II diabetes of early onset: a distinct clinical and genetic syndrome? Lancet i, 923-928. O’Rahilly, S. and Turner, R.C. (1988) Early-onset Type 2 diabetes vs maturity-onset diabetes of youth: evidence for the existence of two discrete diabetic syndromes. Diabetic Med. 5, 224-229. Taylor, S.I., Kadowaki, T., Kadowaki, H., Accih, D., Cama, A. and McKeon, C. (1990) Mutations in insulinreceptor gene in insulin resistant patients. Diabetes Care 13, 257-279. Kosaka, K., Hagura, R. and Kuzuya, T. (1977) Insulin responses in equivocal and definite diabetes, with special reference to subjects who had mild glucose intolerance but later developed definite diabetes. Diabetes 26, 944-952. Kadowaki, T., Miyake, Y., Hagura, R. et al. (1984) Risk factors for worsening to diabetes in subjects with impaired glucose tolerance. Diabetologia 26, 44-49. Taylor, K.W., Sheldon, J., Pyke, D.A. and Oakley, W.G. ( 1967) Glucose tolerance and serum insulin in the unaffected first-degree relatives of diabetics. Br. Med. J. 4, 22-24. Haffner, S.M., Stern, M.P., Hazuda, H.P., Mitchell, B.D. and Patterson, J.K. (1988) Increased insulin concentrations in non-diabetic offspring of diabetic parents. N. Engl. J. Med. 319, 1297-1301.