Você está na página 1de 6

Diabetes Research and Clinical Practice 77 (2007) 327–332


Metabolic improvement of male prisoners with type 2

diabetes in Fukushima Prison, Japan
Masamitsu Hinata a,*, Masami Ono a, Sanae Midorikawa b, Koji Nakanishi c
Medical Affairs Section, Fukushima Prison, 1 Uehara, Minamisawamata, Fukushima 960-8254, Japan
Department of Internal Medicine III, Fukushima Medical University, School of Medicine, Fukushima 960-1295, Japan
Department of Endocrinology and Metabolism, Toranomon Hospital, Tokyo 105-0001, Japan
Received 25 May 2006; received in revised form 4 September 2006; accepted 3 October 2006
Available online 8 January 2007

Imprisonment often improves metabolic control in prisoners with type 2 diabetes; however, the reasons for this remain unclear.
Here, we investigated the metabolic control of male prisoners with type 2 diabetes in Japan. Retrospective analysis of 4385 medical
charts of male prisoners in Fukushima Prison from 1998 to 2004 revealed 109 prisoners (all Asian) with type 2 diabetes
(mean  S.D.: 51  10 years). All were followed up during their imprisonment (14  10 months). During imprisonment, mean
fasting plasma glucose and hemoglobin Alc (HbA1c) levels dramatically decreased from 184  74 to 113  38 mg/dl ( p < 0.001)
and 8.4  2.1 to 5.9  1.2% ( p < 0.001), respectively. In addition, 5 of 18 prisoners (28%) treated with insulin and 17 of 34 (50%)
treated with oral hypoglycemic agents were able to discontinue their treatment and maintain good metabolic control. Most prisoners
in Japanese prisons work 8 h a day 5 days a week, consuming a high dietary fiber diet including boiled rice with barley,
‘‘Mugimeshi’’. These findings suggest that a well-regulated lifestyle and long-term intake of high dietary fiber may have beneficial
effects on metabolic control in patients with type 2 diabetes.
# 2006 Elsevier Ireland Ltd. All rights reserved.

Keywords: Type 2 diabetes; Prison; Lifestyle; Dietary fiber; Barley

1. Introduction America where correctional institutions are provided

with a manual for diabetes management [4], In Britain,
The number of patients with type 2 diabetes is on the it was reported that good diabetic control was achieved
increase in Japan [1] and western countries [2], with in prisoners with diabetes possibly because most of
changing lifestyles and western dietary habits thought these individuals probably led chaotic life-styles before
to be the most important contributing factors [3]. These imprisonment [5]. However, the exact reasons remain
factors have also lead to an increase in prisoners with unknown because levels of fasting plasma glucose and/
type 2 diabetes [4]. The state of diabetes in Japanese or HbA1c have not been evaluated, and analysis of the
prisons is unclear because of the lack of diabetes effect of diet and exercise has not been described.
specialists in Japanese correctional institutions, unlike We recently noted that metabolic control of male
prisoners with type 2 diabetes often shows a dramatic
improvement during imprisonment [6]. In the present
study, we document the metabolic control of male
* Corresponding author. Tel.: +81 24 557 2222;
fax: +81 24 555 2039. prisoners with type 2 diabetes in Fukushima Prison,
E-mail address: mhinata@hotmail.com (M. Hinata). Japan, and evaluate the factors involved.

0168-8227/$ – see front matter # 2006 Elsevier Ireland Ltd. All rights reserved.
328 M. Hinata et al. / Diabetes Research and Clinical Practice 77 (2007) 327–332

2. Patients and methods medical officers; none of the subjects were administered
oral hypolipidemic agents during imprisonment.
With about 650 inmates, Fukushima Prison is one of
the smallest male prisons in Japan. Most prisoners of 2.1. Statistical analysis
Fukushima Prison are repeat offenders mainly charged
with theft, drug abuse, fraud, intrusion, bodily harm, Results are expressed as the mean  S.D. Statistical
robbery, blackmail, and so on. Sentences of subjects significance was determined using the student’s paired,
varied from a few days to 8 years but were usually unpaired t-test or x2 analysis for independence tests (x2-
between 6 months and 4 years. Medical records were test), and comparisons between groups were conducted
only investigated by the two prison medical officers using one-way analysis of variance (ANOVA). A p-level
after obtaining permission from Fukushima Prison of less than 0.05 was considered statistically significance;
Ethics Committee, and anonymity and confidentiality all tests were two-tailed. All statistical analyses were
were maintained throughout by assigning each prisoner performed on a personal computer using Dr. SPSS II for
a number. Retrospective analysis of 4385 medical charts Windows (SPSS Japan Inc., Tokyo, Japan). The non-
between January 1998 and December 2004 identified diabetic reference range for the FPG and HbA1c assays
157 prisoners (3.6%) with type 2 diabetes at the were 70–109 mg/dl and 4.3–5.8%, respectively.
beginning of imprisonment according to the criteria of
the National Diabetes Data Group [7]. Of these, 109 (all 3. Results
Asian) were followed up and included in this study.
Almost all of the 48 prisoners excluded could not be Among the prison inmates, 109 prisoners with type 2
included because of transfer to another prison, parole or diabetes were followed-up for determination of the
release. The following information was obtained for all clinical course (Table 1). Subjects were aged 51  10
diabetes prisoners: age, height, weight, body mass index years, and were followed-up for 14  10 months (range:
(BMI), duration of diabetes history or imprisonment, 1–44, median: 12). Most were within the normal BMI
fasting plasma glucose (FPG), hemoglobin A1c range. Diabetes treatments were usually the same as
(HbA1c), total cholesterol (TC), triglyceride (TG) those before imprisonment. All 18 prisoners treated
and high density lipoprotein cholesterol (HDL-C) with insulin had been undergoing insulin therapy for
levels, the mode of treatment of diabetes, daily calorific several years before imprisonment. Thirty-four prison-
intake, diabetic complications, and past medical history ers were treated with glibenclamide (1.25–7.5 mg/day),
and treatments. These items were measured every 1–6 and of these, 25 had been undergoing this treatment for
months. Administration of oral hypoglycemic agents several years before imprisonment. The remaining 57
and insulin injections were supervised by prison prisoners were treated with diet alone.

Table 1
Characteristics of the 109 male prisoners with type 2 diabetes in Fukushima Prison
Initial assessment Final assessment p-Value
Age (years) 51  10
Duration of diabetes (years) 77
Follow-up period (months) 14  10
Height (cm) 165  7
Body weight (kg) 65  13 62  10 <0.001
BMI (kg/m2) 23.7  3.7 22.7  2.6 <0.001
FPG (mg/dl) 184  74 113  38 <0.001
HbA1c (%) 8.4  2.1 5.9  1.2 <0.001
TC (mg/dl)a 165  40 149  39 <0.001
TG (mg/dl)a 140  100 119  78 0.099
HDL-C (mg/dl)a 40  13 38  11 0.144
Insulin 18 (17%) 13 (12%)
Glibenclamide 34 (31%) 17 (16%) 0.007
Diet 57 (52%) 79 (72%)
Values represent the mean  S.D. BMI: body mass index, FPG: fasting plasma glucose, HbA1c: hemoglobin A1c, TC: total cholesterol, TG:
triglycerides, HDL-C: high-density lipoprotein cholesterol; an = 52 samples.
M. Hinata et al. / Diabetes Research and Clinical Practice 77 (2007) 327–332 329

Table 2
Metabolic control and treatment changes during imprisonment for each therapy group
Insulin (n = 18) Glibenclamide (n = 34) Diet (n = 57)
FPG (mg/dl)
Initial assessment 161  80 190  87 182  62
Final assessment 102  26* 118  47y 115  35y
HbA1c (%)
Initial assessment 7.7  2.0 8.6  2.5 8.4  1.8
Final assessment 5.7  1.2y 6.1  1.5y 5.9  1.0y
Changes to the treatment profile
Increase 1 (6%) 2 (6%) –
No change 6 (33%) 11 (32%) –
Reduction 6 (33%) 4 (12%) –
Discontinued 5 (28%) 17 (50%) –
Values represent the mean  S.D.; *p = 0.012; yp < 0.001, initial vs. final assessment.

Mean FPG and HbA1c values were 184  74 mg/dl different between groups, and all decreased dramatically
and 8.4  2.1%, respectively, at the initial assessment. during imprisonment. Five of 18 diabetic prisoners (28%)
Repeated measurements of FPG and HbA1c were treated with insulin therapy and 17 of 34 (50%) treated
made after at least 1 month revealing a dramatic with glibenclamide therapy were able to discontinue
decrease to 113  38 mg/dl ( p < 0.001) and 5.9  treatment and maintain normal metabolic control until
1.2% ( p < 0.001), respectively. Ninety-two (84%) of release. There was no statistically significant relationship
the 109 prisoners with type 2 diabetes showed improved between the improved metabolic control and age, daily
FPG and HbA1c values. In addition, the number of calorific intake, duration of diabetes, diabetic complica-
prisoners who achieved FPG and HbA1c values of less tions, and past medical history and treatments.
than 130 mg/dl and 7.0%, respectively, satisfying the
values recommended by the American Diabetes
Association for adults with diabetes mellitus [4],
dramatically increased from 11 (10%) to 82 (75%)
during imprisonment. Furthermore, body weights and
BMI also significantly decreased from 65  13 to
62  10 kg ( p < 0.001) to 23.7  3.7 to 22.6  2.6
( p < 0.001), respectively, between the initial and final
assessments. TC levels also significantly decreased
from 165  40 to 149  39 mg/dl ( p < 0.001); how-
ever, TG levels tended to decrease but not significantly
( p = 0.078). HDL-C levels showed no difference
between initial and final assessments ( p = 0.144).
The ratio expressing the number of prisoners with type
2 diabetes in each treatment group was significantly
changed between initial and final assessments
( p = 0.007).
The time courses of FPG and HbA1c are shown in
Fig. 1. Levels of both significantly decreased after 1
month. FPG levels then stabilized after 4 months, while
levels of HbA1c stabilized after about 6 months. Both
remained close to their upper normal limit until release.
In addition, FPG and HbA1c levels were also analyzed
according to the three treatment groups, insulin, Fig. 1. Time course of fasting plasma glucose and HbA1c levels in 109
glibenclamide and diet therapy, respectively (Table 2). male prisoners with type 2 diabetes after the initial assessment. Values
Initial levels of FPG and HbA1c were not significantly represent the mean  S.D. *p < 0.01, yp < 0.05, vs. the control.
330 M. Hinata et al. / Diabetes Research and Clinical Practice 77 (2007) 327–332

4. Discussion type 2 diabetes [11], suggesting that work and exercise

improve glucose tolerance with or without weight loss,
In the present study, overall metabolic control of possibly highlighting one of the most important
prisoners with type 2 diabetes was shown to be good contributions to improved metabolic control among
during imprisonment. These findings were perhaps prisoners with type 2 diabetes.
related to the specific lifestyles and nutrition of The second factor specific to Japanese prisons that
Japanese prisoners. First, almost all prisoners were might have affected the findings is that meals are taken
shown to lead well-regulated lifestyles during their regularly and tend to be healthy. Standard nutrition
imprisonment in Fukushima Prison (Table 3). More- values of Japanese males (2003) [12] and male prisoners
over, many probably led irregular lifestyles before with type 2 diabetes in Fukushima Prison are shown in
imprisonment with, for example, excessive eating, Table 4. The prisoners with type 2 diabetes tend to
alcohol and/or sweetened beverage consumption, little consume a 2518  226 kcal/day diet including boiled
exercise, poor compliance with treatment regimens, and rice with barley, known as ‘‘Mugimeshi’’, a traditional
so on. In Japanese prisons, on the other hand, there is no Japanese food, as do most prisoners in Fukushima
access to alcohol, sweetened drinks or fast food, and Prison. The calorific intake of prisoners, however, is
smoking, eating between meals and night time snacks higher than that of the average Japanese male, as is the
are prohibited. Excessive alcohol [8], junk food [9], and energy ratio between carbohydrates and fat. The
sugar-sweetened drinks [10] are all risk factors of type 2 composition of these two nutrients is close to that of
diabetes. high-carbohydrate low fat diets in western countries,
In addition, prisoners experience limited but regular which reportedly lead to high postprandial glucose and
periods of work and exercise. Almost all prisoners work insulin levels in patients with type 2 diabetes [13].
for 8 h a day 5 days a week while in prison, although this However, in this study, good metabolic control was
work is not particularly strenuous; for example, cooking achieved despite this. This is possibly because the
meals, washing clothes, cleaning inside and outside, prisoners consume a high amount of dietary fiber, about
setting the tables at mealtimes, sewing, woodworking, twice the intake of Japanese males. Insoluble dietary
wiring, handicrafts, and so on. There is little physical fiber intake was not so different between Japanese males
labor in Japanese prisons and exercise, for example, and the male prisoners; however, the soluble dietary
walking, push-ups, and so on, is usually limited to fiber intake of the prisoners was 5.5-fold more.
30 min in a narrow area twice or three times a week. One possible reason for this is that in the 1950s most
Therefore, their work and exercise schedules do not Japanese tended to eat ‘‘Mugimeshi’’ as a staple food
consume a lot of calories, usually less than 800 kcal/ because rice was expensive, unlike barley, which was
day, and as a result, the body weights and BMI of only relatively cheap as well as filling. However, recently,
24 (22%) of 109 prisoners with type 2 diabetes people tend to eat boiled polished rice without barley.
increased during imprisonment; nevertheless, most As a result, daily intake of high dietary fiber in Japan
achieved good metabolic control. Overall, mean body decreased from 20.5 g in 1952 [14] to 14.3 g in 2003
weights and BMI values significantly decreased by 3 kg [12]. All prisoners of Fukushima Prison, as well as other
and 1 kg/m2 over 14 months, respectively. It has been
reported that exercise training improves metabolic Table 4
control regardless of body weight loss in patients with Comparison of the nutritional intake of Japanese men (2003) [12] and
male prisoners with type 2 diabetes in Fukushima Prison
Table 3
Japanese men Prisoners
Typical daily schedules of prisoners in Fukushima Prison
with diabetes
Weekday Weekend and holiday
Calorific intake (kcal/day) 2138 2518  226
Get up 6:40 7:30
Constituents (% of total energy)
Breakfast 7:10 8:00
Carbohydrate 56 66–61
On duty 7:50–9:30 –
Protein 15 13–14
9:45–12:00 –
Fat 24 21–25
Lunch 12:00 12:00
Others 5 0
On duty 12:40–14:30 –
14:45–16:30 – Dietary fiber (g/day)
Supper 17:00 16:20 Total 14.5 28.4  2.1
Free time 18:30–21:00 17:30–21:00 Soluble 3.4 18.7  1.2
Bedtime 21:00 21:00 Insoluble 11.1 9.7  0.9
M. Hinata et al. / Diabetes Research and Clinical Practice 77 (2007) 327–332 331

Japanese prisons, tend to eat ‘‘Mugimeshi’’ daily (rice: enhanced consumption of soluble fiber improves blood
barley = 7:3) as a staple food. Barley is known to glucose disposal by increasing skeletal muscle plasma
contain an abundance of soluble and insoluble fiber and membrane GLUT-4 content [28]. These mechanisms
its endosperm cell walls are rich in b-glucan, a might also explain why high dietary fiber not only
polysaccharide formed by linear chains with b-1,3 improves postprandial glucose but also lowers FPG
and b-1,4 linkages. b-Glucan is partially soluble and levels and improves glucose tolerance. Thus, high
highly viscous in a soluble condition [15]. As a result, dietary fiber foods such as ‘‘Mugimeshi’’ might be
b-glucan probably reduces glucose and insulin levels by beneficial for patients with type 2 diabetes or impaired
increasing the viscosity of the stomach and small glucose tolerance.
intestine contents, decreasing the absorption of digested In the present study, imprisonment also improved
nutrients from the small intestine [16]. Moreover, lipid metabolism in prisoners with type 2 diabetes.
glucose transport in the intestinal wall is inhibited partly Levels of TC significantly decreased, while TG and
by increased resistance of the mucosal diffusion barrier HDL-C values tended to decrease but not significantly.
as a result of increased viscosity of the intestinal bolus Several studies have shown that lowered cholesterol can
[15]. Thus, delayed carbohydrate digestion and absorp- be attributed to the intake of high soluble fiber
tion appears to be a major factor responsible for [21,29,30], increased excretion of bile acids or neutral
reductions in the plasma glucose response [17]. The sterols, increased catabolism of low-density lipoprotein
prison diet also includes vegetables and other high cholesterol, and reduced fat absorption [31]. Soluble
dietary fiber foods such as beans, mushrooms, and fibers have also been shown to be fermented in the colon
seaweeds. Carbohydrates with a high dietary fiber giving rise to short-chain fatty acids, which can be
content might delay the absorption of glucose, thereby absorbed and potentially inhibit hepatic cholesterol
permitting a better match between the timing of insulin synthesis [29]. It has also been reported that dietary
release and blood glucose peaks [18]. fiber intake has no effect on plasma TG levels in studies
Animal [19], epidemiologic [20], and human inter- using normal and hypercholesteromic subjects [28].
vention [21] studies have all shown that a high soluble Indeed, in the present study, plasma TG concentrations
fiber diet might improve glucose and insulin control and tended to decrease, but not significantly, with the high
hence reduce the risk of type 2 diabetes. The high dietary fiber diet followed in Fukushima Prison. One
viscosity of b-glucan might therefore be particularly reason for this is related to the initial TG values varied
effective at reducing postprandial glycemia, causing a widely, and the number of samples used to obtain the
significant reduction in the glycemic response [22]. This initial TG values; that is, less than the number available
finding is consistent with those of several previous for FPG and HbA1c analysis. In addition, prisoners are
studies whereby an increase in total dietary fiber intake generally detained by police for about 2–3 months
(predominantly soluble fiber) was shown to significantly before sentencing, during which time their meals are
improve glycemic control and decrease the degree of also controlled. Thus, we believe that the values of TC,
hyperinsulinemia in patients with type 2 diabetes [21,23]. TG and HDL-C also improved compared to before
Several large-scale prospective studies have shown imprisonment; improvement of FPG and HbA1c values
that diets high in dietary fiber and low in glycemic load takes longer.
are inversely related to the incidence of type 2 diabetes in Since this research was not a randomized controlled
humans [20,24,25]. However, these studies were short- study, it is unclear whether a well-regulated lifestyle, high
term and, until now, no research has investigated the dietary fiber intake or both, predominantly contribute to
effect of high dietary fiber intake for more than 1 year in good metabolic control in prisoners with type 2 diabetes,
addition to a well-regulated lifestyle including no alcohol and consequently, further studies are needed.
or eating between meals in patients with type 2 diabetes. In conclusion, this study shows that good diabetic
The findings of this research are therefore important in care could be dramatically achieved in a Japanese
suggesting appropriate management of type 2 diabetes. prison environment, indicating the importance of a
Dietary fiber can significantly alter proglucagon well-regulated lifestyle and long-term intake of high
gene expression and modulate glucagon-like peptide-1 dietary fiber in patients with type 2 diabetes.
(GLP-1) and insulin secretion [26]. GLP-1 plays an
important role in the disposition of glucose absorbed Acknowledgments
from the intestine, and circulating GLP-1 has a
regulatory effect on the basal islet output of glucagons, We thank Messrs. Kazuhiro Sugai, Yasunori Kan,
and consequently, fasting glycemia [27]. In addition, Hirotaka Honda, Manabu Fumoto, Shigeru Denbo,
332 M. Hinata et al. / Diabetes Research and Clinical Practice 77 (2007) 327–332

Michio Saito and Terufumi Takeno in Fukushima Prison [16] J. Adiotomre, M.A. Eastwood, C.A. Edwards, W.G. Brydon,
for their assistance. Dietary fiber: in vitro methods that anticipate nutrition and
metabolic activity in humans, Am. J. Clin. Nutr. 52 (1990)
References [17] J.H. Lavin, N.W. Read, The effect on hunger and satiety of
slowing the absorption of glucose: relationship with gastric
[1] N. Yoshiike, Y. Matsumura, M. Iwaya, M. Sugiyama, M. Yama- emptying and postprandial blood glucose and insulin responses,
guchi, National nutrition survey in Japan, J. Epidemiol. 6 (1996) Appetite 25 (1995) 89–96.
S189–S200. [18] M. Rendell, Dietary treatment of diabetes mellitus, N. Engl. J.
[2] S. Liu, J.E. Manson, M.J. Stampfer, F.B. Hu, E. Giovannucci, Med. 342 (2000) 1440–1441.
G.A. Colditz, et al., A prospective study of whole-grain intake [19] J. Li, T. Kaneko, L.Q. Qin, J. Wang, Y. Wang, A. Sato, Long-term
and risk of type 2 diabetes mellitus in US women, Am. J. Public effects of high dietary fiber intake on glucose tolerance and lipid
Health 90 (2000) 1409–1415. metabolism in GK rats: comparison among barley, rice, and
[3] S. Nakanishi, M. Okubo, M. Yoneda, K. Jitsuiki, K. Yamane, N. cornstarch, Metabolism 52 (2003) 1206–1210.
Kohno, A comparison between Japanese-Americans living in [20] K.A. Meyer, L.H. Kushi, D.R. Jacobs Jr., J. Slavin, T.A. Sellers,
Hawaii and Los Angeles and native Japanese: the impact of A.R. Folsom, Carbohydrates, dietary fiber, and incident
lifestyle westernization on diabetes mellitus, Biomed. Pharmac- type 2 diabetes in older women, Am. J. Clin. Nutr. 71 (2000)
other. 58 (2004) 571–577. 921–930.
[4] D.L. Lorber, S. Chavez, J. Dorman, L.K. Fisher, S. Guerken, [21] M. Chandalia, A. Garg, D. Lutjohann, K. von Bergmann, S.M.
L.A. Haas, et al., Diabetes management in correctional institu- Grundy, L.J. Brinkley, Beneficial effects of high dietary fiber
tions, Diabetes Care 27 (2004) S114–S121. intake in patients with type 2 diabetes mellitus, N. Engl. J. Med.
[5] L.A. MacFarlane, G.V. Gill, E. Masson, N.H. Tucker, Diabetes in 342 (2000) 1392–1398.
prison: can good diabetic care be achieved? BMJ 304 (1992) [22] G.F. Keogh, G.J. Cooper, T.B. Mulvey, B.H. McArdle, G.D.
152–155. Coles, J.A. Monro, et al., Randomized controlled crossover study
[6] M. Hinata, K. Sugai, M. Ono, K. Nakanishi, S. Midorikawa, of the effect of a highly beta-glucan-enriched barley on cardi-
Metabolic improvement of prisoners with type 2 diabetes in ovascular disease risk factors in mildly hypercholesterolemic
Fukushima Prison, Japan, Diabetes 54 (Suppl. 1) (2005) A425– men, Am. J. Clin. Nutr. 78 (2003) 711–718.
A426. [23] R.C. Turner, C.A. Cull, V. Frighi, R.R. Holman, Glycemic
[7] National Diabetes Data Group, Classification and diagnosis of control with diet, sulfonylurea, metformin, or insulin in patients
diabetes mellitus and other categories of glucose intolerance, with type 2 diabetes mellitus: progressive requirement for multi-
Diabetes 28 (1979) 1039–1057. ple therapies (UKPDS 49). UK Prospective Diabetes Study
[8] S. Carlsson, N. Hammar, S. Efendic, P.G. Persson, C.G. Osten- (UKPDS) Group, JAMA 281 (1999) 2005–2012.
son, V. Grill, Alcohol consumption, type 2 diabetes mellitus and [24] J. Salmeron, J.E. Manson, M.J. Stampfer, G.A. Colditz, A.L.
impaired glucose tolerance in middle-aged Swedish men, Dia- Wing, W.C. Willett, Dietary fiber, glycemic load, and risk of
betes Med. 17 (2000) 776–781. non-insulin-dependent diabetes mellitus in women, JAMA 277
[9] J. Gittelsohn, T.M. Wolever, S.B. Harris, R. Harris-Giraldo, A.J. (1997) 472–477.
Hanley, B. Zinman, Specific patterns of food consumption and [25] J. Salmeron, A. Ascherio, E.B. Rimm, G.A. Colditz, D. Spiegel-
preparation are associated with diabetes and obesity in a Native man, D.J. Jenkins, et al., Dietary fiber, glycemic load, and risk of
Canadian community, J. Nutr. 128 (1998) 541–547. NIDDM in men, Diabetes Care 20 (1997) 545–550.
[10] M.B. Schulze, J.E. Manson, D.S. Ludwig, G.A. Colditz, M.J. [26] R.A. Reimer, M.I. McBurney, Dietary fiber modulates intestinal
Stampfer, W.C. Willett, et al., Sugar-sweetened beverages, proglucagon messenger ribonucleic acid and postprandial secre-
weight gain, and incidence of type 2 diabetes in young and tion of glucagon-like peptide-1 and insulin in rats, Endocrinol-
middle-aged women, JAMA 292 (2004) 927–934. ogy 137 (1996) 3948–3956.
[11] N.G. Boule, E. Haddad, G.P. Kenny, G.A. Wells, R.J. Sigal, [27] D.A. D’Alessio, R. Vogel, R. Prigeon, E. Laschansky, D. Koer-
Effects of exercise on glycemic control and body mass in type 2 ker, J. Eng, et al., Elimination of the action of glucagon-like
diabetes mellitus: a meta-analysis of controlled clinical trials, peptide 1 causes an impairment of glucose tolerance after
JAMA 286 (2001) 1218–1227. nutrient ingestion by healthy baboons, J. Clin. Invest. 97
[12] The National Nutrition Survey in Japan (2003), Tokyo: Office for (1996) 133–138.
Life-Style Related Disease Control, General Affairs Division, [28] Y.J. Song, M.M. Sawamura, K. Ikeda, S. Igawa, Y. Yamori,
Health Service Bureau, Ministry of Health, Labour and Welfare, Soluble dietary fibre improves insulin sensitivity by increasing
(Accessed August 21, 2006, at http://www.mhlw.go.jp/houdou/ muscle GLUT-4 content in stroke-prone spontaneously hyper-
2005/04/h0421-1b.html), in Japanese. tensive rats, Clin. Exp. Pharmacol. Physiol. 27 (2000) 41–45.
[13] Y.D. Chen, A.M. Coulston, M.Y. Zhou, C.B. Hollenbeck, G.M. [29] K.M. Behall, D.J. Scholfield, J. Hallfrisch, Effect of beta-glucan
Reaven, Why do low-fat high-carbohydrate diets accentuate level in oat fiber extracts on blood lipids in men and women, J.
postprandial lipemia in patients with NIDDM? Diabetes Care Am. Coll. Nutr. 16 (1997) 46–51.
18 (1995) 10–16. [30] D.J. Jenkins, T.M. Wolever, A.V. Rao, R.A. Hegele, S.J. Mitch-
[14] S. Nakaji, K. Sugawara, D. Saito, Y. Yoshioka, D. MacAuley, T. ell, T.P. Ransom, et al., Effect on blood lipids of very high intakes
Bradley, et al., Trends in dietary fiber intake in Japan over the last of fiber in diets low in saturated fat and cholesterol, N. Engl. J.
centry, Eur. J. Nutr. 41 (2002) 222–227. Med. 329 (1993) 21–26.
[15] P. Wursch, F.X. Pi-Sunyer, The role of viscous soluble fiber in the [31] R.M. Kay, A.S. Traswell, Effect of citrus pectin on blood lipids
metabolic control of diabetes. A review with special emphasis on and fecal steroid excretion in man, Am. J. Clin. Nutr. 30 (1997)
cereals rich in beta-glucan, Diabetes Care 20 (1997) 1774–1780. 171–175.