Growth in juvenile diabetes mellitus
Growth in Juvenile Diabetes Mellitus
0 Department of Pediatrics, University of British Columbia , Vancouver , Canada
Summary. Height, weight, height velocity and skeletal maturity information was collected on a group of 94 juvenile diabetics in a mixed longitudinal study, and compared to a control population. Familial growth trends were taken into consideration where possible. Although the hormonal environment in the juvenile diabetic is probably unusual and fluctuating, there is no evidence that juvenile diabetes is likely to interfere with growth provided that a reasonable degree of control is maintained.
Growth; juvenile diabetic; maturation; height; weight; bone age
T h a t chronic disease tends to interfere with growth
is a widely accepted fact, although often the evidence
for such belief is meagre. Juvenile diabetes mellitus
m a y be considered a chronic metabolic disease;
however it differs from such conditions as congenital
cardiac disease or chronic renal disease in t h a t the
abnormalities involve endocrine-metabolic
relationships which are central to growth processes.
In the literature of the early insulin era, references
on growth and juvenile diabetes are numerous, and
generally suggest subnormal growth is to be expected,
although a few suggest "good control" ~11 result in
normal growth. However in more recent years, since
current modes of management have been adopted,
references are rare, and further equally divided
between the opinions t h a t juvenile diabetes, even if well
controlled, is, [1--4] or is not, [
] associated with
growth disturbances. The object of the present mixed
longitudinal s t u d y was to investigate whether growth
disturbances were usually associated with controlled
Material and Methods
The s t u d y population consists of a total of 94
juvenile diabetics ranging from three to t w e n t y years
of age at the present time. Duration of diabetic state
varied from 1 to 15 years. All subjects have been
followed longitudinally over a period of at least one
year, and some as long as four or five years. Subjects
are normally seen three or four times a year.
The great majority of heights have been recorded
with a Harpenden stadiometer. Skeletal m a t u r i t y was
determined from posteroanterior films of the left hand
and wrist. M a t u r i t y is rated b y the method of Tanner
Growth velocity was obtained from measurements
over periods of at least one year. This is essential to
eliminate the seasonal effects on growth rate [
Unfortunately it is not usually possible to schedule visits
at exact intervals, and thus the seasonal effect has not
been entirely excluded b u t at least minimized.
Height and weight have been compared with the
standards (3rd, 50th and 97th centiles) reported b y
Tanner et al. (1966) [
]. I n a previous paper [
] we have
reported the height and weight of healthy Vancouver
school children using the same standards. Skeletal
m a t u r i t y has been plotted and compared with the same
control population. Skeletal m a t u r i t y films were
available in 74 subjects.
Information regarding growth patterns, final height,
and age of menarche was sought on close relatives
(siblings, parents and parent siblings). Useful data
were obtained in f o r t y subjects.
Discrepancies (of a n y sign and magnitude)
between apparent skeletal m a t u r i t y and chronological
age were examined in the light of family data. I n view
of its greater precision and reliability, only menarchial
age in female relatives was utilized.
"Good control" is herein defined as absent or
minimal giueosuria at most times; absence of ketosis or
prolonged h e a v y giucosuria, except when infection is
present; infrequent hypoglycemic episodes; and
evidence of satisfactory adherence to a measured (but not
weighed) dietary regimen, and urine testing. Insulin
t h e r a p y should have been adjusted appropriate to
urine testing findings.
Achieved stature. Figs. 1 and 2 show achieved
stature of all subjects at their most recent
measurement. This cross-sectional method has been used b y
almost all previous workers. I t will be seen from Table
a t e s
1 that compared to the standards used, the skewing to
the higher centfles seen in our control population is
fully reflected in our female subjects, but not in our
male subjects. This apparent excess of shorter males
will be discussed below.
Family data. The reliability of family stature d a t a
is open to question, as in few cases was it measured b y
us. Stature of close relatives is described in Table 4.
All male relatives and females, except for i m m a t u r e
compared to the standards. H o w e v e r our control
subjects of b o t h sexes showed a similar skewing, and
we conclude t h a t the diabetic children show no
significant abnormalities of skeletal maturation.
Height velocity. We compared the height velocity
in 76 subjects with the appropriate standard of Tanner
et al. [
], and recorded the average centile status of the
individual over the period of observation in Table 3.
Some subjects showed considerable variation in their
actual centile position from y e a r to year. This m a y be
due to natural variations or measurement errors; it
might also be due to fluctuations in diabetic control,
although we have been unable to show a n y such
correlation from our clinical records.
siblings, show distributions similar to the control
As will be seen f r o m Table 5, in a b o u t half the
instances the bone age discrepancy of the diabetic is
compatible with the family trend. I n those where the
difference was not compatible, in the m a j o r i t y it was
better t h a n (i. e. less than) the family trend.
Serum growth hormone levels, which are high in
u n t r e a t e d or inadequately treated juvenile diabetics,
fall when control is established [
]. However in
longstanding diabetes, good control is often associated
with higher growth hormone levels . Recent reports
demonstrate t h a t sleep-related plasma growth hormone
elevation is normal in juvenile diabetics even in the
presence of hyperglycemia [
Insulin also fulfills the criteria of a growth hormone;
it is anabolie, promotes cell growth, and an
insufficiency leads to growth stunting. Like somatotropin itself,
we face the p a r a d o x t h a t the secretion and circulating
levels of a m a j o r growth-promoting hormone are so
variable, depending on the nutritional status of the
individual at the m o m e n t . I t is clear t h a t insulin levels
achieved in the t h e r a p y of juvenile diabetes are
"unphysiological", in the sense t h a t t h e y cannot reflect
rapidly changing metabolic needs. I t is likely t h a t the
mean level of plasma insulin in a juvenile diabetic is
usually less t h a n in a non-diabetic, since to avoid
periods of incapacitating hypoglycemia during the
post-absorptive phase, we accept a degree of
hyperglycemia, and thus of hypoinsulinemia. Growth failure
is certainly a feature of uncontrolled juvenile diabetes,
this being p a r t of the "Mauriac s y n d r o m e " [
A t present the possible interplay of these, at times,
conflicting hormonal p a t t e r n s on growth cannot be
perceived. The problem is whether " g o o d " control will
p r e v e n t the growth retardation characteristic of
undertreated juvenile diabetes.
Nutritional factors are prominent among the
requirements for optimal growth. The diabetic diet, with
emphasis on variety, adequate caloric intake, and with
its relatively high protein content (typically 20% of
calories compared to the usual 12--15%) should also
The data on achieved stature are not in favour of
poor growth in our subjects. Examination of
individuals shorter than expected reveals familial factors in
most instances. Height velocity data, although more
difficult to interpret, are much more valuable than
single determinations. Again no obviously abnormal
trend is evident. We were not able to correlate
fluctuations in height velocity with variation in diabetic
Skeletal maturity patterns, allowing for familial
trends, were again comparable to our control
Included in this series were 11 subjects (7 male,
4 female), whose diabetic control status, even b y our
moderate standards, was for prolonged periods judged
unacceptable b y the criteria noted above. Of these,
8 were unusually short, and 7 markedly underweight.
0 n l y 4 had radiography, and while one showed much
delay in bone maturation, two were normal and one
was accelerated for age. None of these subjects had
The apparent excess of shorter diabetic males
compared to the control population must be contrasted
with the growth rates of these males, which are normal
when compared to the controls. Thus the shorter
stature was not due to diminished growth rates during
the period of observation. Familial factors certainly
were responsible in some instances. Contrary to the
observation of Bergqvist [
], there was no
relationship between shorter stature and early onset of
diabetes in our series. The poorly controlled short subjects
noted above were also included in the series and
affected the distribution of stature.
I t is concluded t h a t diabetes mellitus, with
adequate therapy, is not usually associated with growth
delay or short stature.
1. Bergqvist , N. : The growth of juvenile diabetics . Acta endoer . 15 , 133 -- 165 ( 1964 }.
2. Larsson , Y. , Sterky , G. : Long-term prog~aosis in juvenile diabetes mellitus . Acta Pedlar . 51 , Suppl . 130 ( i962 ).
3. Knowles , H.C. , Guest , G.M. , Lampe , J. , Kessler , M. , Skillman , T. G. : The course of juvenile diabetes treated with an unmeasured diet . Diabetes 14 , 239 -- 273 ( 1965 ).
4. Sterky , G. : Growth patterns in juvenile diabetes . Acta pediat. Scandinav. Suppl . 177 , 80 -- 82 ( 1967 ).
5. Jackson , R.L. , Kelly , H.G. : Growth of children with diabetes mellitus in relationship to level of control of the disease . J. Pediat . 29 , 316 -- 328 ( 1946 ).
6. Joos , T.H. , Johnston , J.A.: A long-term evaluation of the juvenile diabetic . J. Pedlar . 50 , 133 -- 137 ( 1957 ).
7. Tanner , J.M. , Whitehouse, l%.H., Healey , M.J.]~.: A new system for estimating skeletal maturity from the hand and wrist . Paris: Int. Child. Centre ( 1962 ).
8. -- -- Takaishi , M. : Standards from birth to maturity for height, weight, height velocity and weight velocity: British children , 1965 . Arch. Dis. Childh. 41 , 454 -- 471 and 613 -- 635 ( 1966 )
9. Birkbeek , J.A. : Growth standards for British Columbia children . Submitted for publication.
10. t-Iansen , A.P. , Johansen , K. : Diurnal patterns of blood glucose, serum free fatty acids, insulin, glucagon, and growth hormone in normals and juvenile diabetics . Diabetologia 6 , 27 -- 33 ( 1970 ).
11. Johansen , K. , Hansen , A.P. : Diurnal serum growth hormone levels in poorly and well-controlled diabetics . Diabetes 20 , 239 -- 245 ( 1971 ).
12. Parker , D.C. , l%ossman, L.G. : Sleep release of human growth hormone in treated juvenile diabetics . Diabetes 29 , 691 -- 695 ( 1971 ).
13. Mauriac , P. : Hepatomegalies de l' enfance avec troubles de la croissanee et du metabolisme des glucides . Paris Med . 2 , 525 -- 528 ( 1934 ).
14. Traisman , H.F. : Mauriae's syndrome. A complication of poorly managed diabetes . Clin. Ped . 8 , 520 -- 522 ( 1964 ).