17

TRANSACTIONS OF THE ROYAL SOCIETY OF TROPICAL. MEDICINE AND HYGIENE, Vor. 76, No. 1, 1982

Plasma albumin, parasitic infection and pubertal development in Egyptian boys;

T. J. COLE, SOHAIR I. SALEM, A. S. HAFEZ, 0. M. GA~AL AND A. MASSOUD Dept. of Public Health and Environmental Medicine, Ein Shams Faculty of Medicine and Child Health Dept.,

National Research Centre, Cairo, Egypt and MRC Dunn Nutrition Laboratory, Cambridge, UK

Summary Pubertal stage was assessed in 453 Egyptian boys

aged from nine to 17 years. Weights and albumin concentrations were measured and parasites identified from stool samples. The mean age to reach each stage of puberty was significantly delayed in the presence of low albumin levels and parasites. The magnitude of the delay was greater for genital stage 2 than for later stages. Weight was also lower in children with reduced albumin concentration or parasites. Among children of the same age, weight and albumin were positively correlated (P<O.OOl), suggesting that albu- min status is involved in the timing of puberty.

Introduction Nutritional status during childhood is conven-

tionally assessed using anthropometry, but plasma albumin concentration is an additional index of nutritional status (WHITEHEAD et al., 1973) which is closely related to the development of oedematous malnutrition, kwashiorkor, in early childhood (Cow- ARD & FIOROTTO, 1979). Puberty is another period of life in which growth velocity is high and nutritional status is believed to have a significant influence on pubertal development (SATYANARAYANA & NAIDU, 1979; GREENWOOD & RICHARDSON, 1979; COBLE et al., 1971). To investigate these factors, weight-for-age and plasma albumin concentrations were measured in pubertal Egyptian boys and relationships between weight-for-age, albumin status, timing of puberty, social conditions and the effect of parasitic infestation were examined.

Materials and Methods The study consisted of a cross sectional sample of

453 boys, from four different schools in the Abbassia district of Cairo, Egypt. The boys ranged in age from nine years one month to 16 years 10 months. They were classified into four pubertal stages, G2 to G5,

according to the development of their genitalia (TANNER, 1962), and all but 13 were further divided into six classes according to the occupation and educational level of the father. Further details of the study and the classification system are given else- where (HAFEZ et al., 1981).

Stool and urine samples taken from each child were examined for the presence of parasites. Blood samples for the estimation of albumin were taken from each child at the school clinic between 10 a.m. and 11 a.m., and plasma was separated within three hours. Albu- min was estimated using the bromocresol green method of MISBET et al, (1973).

Statistical Analysis Statistical association between variables was meas-

ured by multiple regression. To estimate the mean age for each pubertal stage, the proportions of children at each age who had reached the stage were transformed to logits, and a linear trend on age fitted. The GENSTAT statistical computing language was used for the analysis (ALVEY et al., 1977).

Results Four types pf parasitic ova or cysts were found:

Entamoeba histolytica cysts, Hymenolepis nana ova, Ascaris ova and Giardia cysts. Table I gives their prevalences by age, with the significance of the age-trend, and the numbers at risk. E. histolytica was common throughout the age range, H. nana and Asdaris were seen mainly in the younger children, and Giardia was observed only in those aged 12 and over.

The risk of having E. histolytica of H. nana was unrelated to social class, but for both Ascaris and Giardia the parasite was more common in the lower social classes (4, 5 and 6); in particular, Ascaris was not seen in classes 1 and 2. There was no evidence that family size influenced risk of infection.

Table I-Numbers at risk, and percentage prevalence by age of parasite infection

Level of

9+ lO+ ll+ 12+ 13+ 14+ 15+ 16+ Total significance of

age trend

E. histolytica 33 32 51 45 49 35 39 44 42 DO.2 H. nana

Ascaris

:oo : 1: 6 0 8 0 P<O* 1

Giardia 0 0 0 1;

i

13 1: 1: ii

: P-CO.05

8 P<O*OOl

Numbers at risk 30 37 51 82 78 74 51 50 453 -

18 PUBERTAL DEVELOPMENT IN EGYPTIAN BOYS

Table II-Albumin levels by social class, correctea to age 13; the differences between classes are highly signilicant (P<OWll)

Significance Social Albumin relative to class Number g/l Social Class 1

:. 46 44.2 61 42.5 P>F2

: E 41.6 P<O.OS 38.5 P<O~OOl

5 1::

39.3 PcO.01 6 40.0 P<O*Ol

Table III-Estimated effects of parasites on albu- min level. Values given are the reductions in albumin (+SE) associated with the presence of each parasite, after correcting for age

Parasite

Giardia H. nana Ascaris H. histolytica

Albumin Level of g/l reduction significance

3.9+ 1.3 P<O*Ol 3.2+ 1.5 P<O*O5 3.5+1*9 P<O. 1 0.7kO.7 E-o.2

Plasma albumin concentrations remained steady from nine to 13 years at around 42 g/l, but rose steeply beyond age 14 (P<O*OOl). There were signifi- cant differences (P<O*OOl) in albumin concentration between the social classes after standardizing for age, as shown in Table II, which also gives the numbers in each social class. There was a steady fall in albumin from class 1 to class 4, while classes 4, 5 and 6 were all similar.

Table III gives the effects of parasitic infections on plasma albumin. E. histolytica had no significant effect, whereas the other three were all associated with a drop of between 3 and 4 g/l in albumin level.

Albumin was itself found to be strongly related to the mean ages of attainment of the four pubertal stages. Table IV gives the mean albumin levels by age and pubertal stage, showing a steady rise over-all with increasing maturity, and a relatively late rise with age. Within the body of the Table for boys of the same pubertal stage, there is a clear negative trend with age, which largely cancels out the opposite trend with maturity. After correction for age and stage, albumin shows no difference between social classes (E-0.2). The magnitude of the age-trend, and also the effects of parasites on the mean age of each pubertal stage, are shown in Table V. A plasma albumin level 10 g/l below the average for age was associated with a delay of nearly two years in reaching stage G2, while for stages G3, G4 and G5 the delay was about one year.

Table IV-Mean albumin values (g/l) by age and stage, with numbers in parentheses

Stage

Pre-pubertal stage

G2

G3

G4

G5

9+ 10+

38.1 37.2 (19) (21) 46.1 45.9 (11) (16) - -

- -

- -

11-t

37.9 (29) 47.9 (22) -

-

-

Age All 12+ 13+ 14+ 15+ 16+ Ages

31.4 34.9 30.5 35.8 (28) (9) (1) (107) 40.3 39.2 32.0 37.5 - 40.9 (39) (33) (22) (2) (145) 48.6 47.5 45.8 41.5 38.7 45.8 (13) (29) (25) (18) (2) (87) 46.7 47.8 48.8 45.5 46.3 47.0 (2) (7) (25) (26) (13) (73) - - 53.5 51.9 50.0 50.3

(1) (5) (35) (41)

All stages 41.0 41.0 42.2 38.7 42.6 42.6 44.4 48.6 42.5 (30) (37) (51) (82) (78) (74) (51) (50) (453)

Table V-Delay in maturation associated with parasitaemia, or albumin levels 10 g/l below the mean for age

Delay

~iZ2W

Delay (years) when each parasite present Pubertal

Stage albumin E. histolytica H. nana Ascaris Giardia

E”3 1.9*** . *

8.: 1.7*

1.1***

Et 0.9*** 0.5** 0”:;

;:: ;:; 0.7 1.1*

0.9*** 0*4** 1.3 1.0 0.4

Level of significance *P<O.O5, **P<O~Ol,***P<O~OO1

Delay

kG?i

parasites combined

;::

;:;

T. J.

Table VI-Percentage fall in weight-for-age associ- ated with an albumin level 10 g/l below the average for age, or the presence of parasites. Weight is expressed as a percentage of the median weight-for- age for Tanner et al. (1966), and is age-corrected

% Fall Level of

significance

Albumin 6.5&l+0

Ascaris 8.3f4.0 H. nana 7.3k3.5 Giardia 6.71k2.7 E. histolytica l-2+1.5

P<O.OOl

P<O-05 P<O*O5 P<O*O5 P>O.2

Furthermore there were delays associated with the presence of parasites which, for stage G2, involved a further delay of 0.6 years.

A fall in weight-for-age was also associated with reduced albumin concentration and the presence of parasites. To standardize for age the weights were expressed as percentages of the median weight for age from TANNER et al. (1966). In addition, differences in mean weight-for-age between years were allowed for. There was a strong correlation between weight-for- age and albumin (0.31, P<O*OOl), and weight-for-age was also affected by parasitic infection, as shown yn Table VI. These values indicated that weieht-for-aae in the whole population would have been 1.3% great& on average in the absence of parasites.

After standardizing for the pubertal stage reached as well as for age! the relationships shown in Table VI all became non-significant, suggesting that their effect on weight-for-age was related to their inverse associa- tion with pubertal stage.

Discussion The contribution of the presence of parasites to

delayed sexual maturation and to growth retardation is clear in this study. However the delaying effects of the parasites varied according to the pubertal stage involved; there was a suggestion that stage 2 was more sensitive than the later stages, in that its mean age was retarded by 0.6 years. This finding disagrees with that of COBLE et al. (1971) who reported from a different part of Egypt that parasitic infections did not contribute to delayed sexual maturation in the boys they studied. However, they only investigated 26 boys, of whom all but three were infected, and their normal children were appreciably retarded relative to the Tanner pubertal indices.

An upward trend in plasma albumin concentration with age, as shown in the present study, was not seen bv ROUND et al. (1979) in their loneitudinal studv of British boys. They found that albun& did not change significantly with the growth spurt, although it is not absolutely clear how they looked for it. However the range of their values (43 to 54 g/l) was comparable to that of our boys in the same age period.

Despite the over-all rise in plasma albumin concen- tration with age shown in the present study, there is a trend in the -opposite direction when each stage is examined seoaratelv (Table IV). The older the children in a- given- stage the poorer their albumin status. When the same table is viewed by columns,

COLE et al. 19

i.e. for children of the same age, those at a more advanced pubertal stage have higher albumin concen- trations. This is exactly the same pattern as seen for weight-for-age (HAFEZ et al., 1981).

An inevitable concomitant of puberty is the growth spurt. It follows that children who are advanced for their age are further into their growth spurt, and thus are taller and heavier than other children of the same age. This explains why weight-for-age is positively correlated with pubertal stage among children of the same age.

The same explanation cannot be invoked for plasma albumin concentration, since it is not reognized as being influenced by the growth spurt (ROUND et al., 1979). Thus our findings are unexpected.

The findings suggest that a low plasma albumin concentration is related to delayed puberty either by its influencing the marked physiological changes accompanying this period, or by its reflecting an inadequacy of protein intake. The latter alternative seems the more likely, given the poor nutritional status of many of the ~children in ihe sample. We observed previously (HAFEZ et aZ., 1981) deficits of 10% in mean weight-for-age associated with low social class, and it is reasonable To assume a similar range in albumin status. On the other hand, rats feeding a control diet ad libitum showed a significant rise (P<O.Ol) in nlasma albumin concentration durine puberty (Salem et al., unpublished data), suggesti@ a physiological explanation in this animal model. It is quite possible that both mechanisms contribute to our findings.

It is clear from this study that pubertal changes can be affected by a number of factors, particularly at the start of puberty. In the later stages their influence is less, which may suggest that it is the initiation of puberty that is most sensitive, and once pubertal changes have begun to occur they are less easily delayed. Alternatively, as discussed elsewhere (HAFEZ et al., 1981), the education system may select out the most underprivileged children, so that older schoolchildren are more homogeneous as regards their nutritional status. Within classes, there is no evidence of a general trend in nutritional status over the period of puberty.

Acknowledgements We thank the referees for their comments. Sohair I.

Salem from the National Research Centre in Cairo acknowledges the receipt of a fellowship from the Wellcome Trust to work at the Dunn Nutritional Laboratory.

Alvey, N. G., Banfield, C. F., Baxter,, R. I. et al. (1977). GENSTAT a general statistical program. Rothamsted, Harpenden, UK.

Cable, Y. D., Bardin, C. W., Ross, G. T. & Darby, W. T. (1971). Studies of endocrine function in boys with retarded growth, delayed sexual matura- tion and zinc deficiency. Journal of Clinical Endocrinology, 32, 361-367.

Coward, W. A. & Fiorotto, M. (1979). The pathogenesis of oedema in kwashiorkor-the role of plasma proteins. Proceedings of the Nutrition Society, 38, 51-59.

20 PUBERTAL DEVELOPMENT IN EGYPTIAN BOYS

Greenwood, C. T. & Richardson, D. P. (1979). Nutrition during adolescence. World Review. Nutrition and Dietetics, 33, 1-41.

Hafez, A. S., Salem, Sohair, I., Cole, T. J., Galal, 0. M. & Massoud, A. (1981). Sexual maturation and growth pattern in Egyptian boys. Annals of Human Biology, 8, 461-467.

Misbet, J. A., Jorvis, A. & Feta, A. (1973). Measure- ment of human serum albumin with bromocresol green. Clinical Chimica Acta, 45, 189.

Round, J. M., Butcher, S. & Steele, R. (1979). Changes in plasma inorganic phosphorus and alkaline phosphatase activity during the adolescent growth spurt. Annals of Human Biology, 6, 129- 136.

menarche in rural Hyderabad. Annals of Human Biology, 6, 163-165.

Tanner, J. M. (1962). Growth at adolescence. (2nd edit.). Oxford: Blackwells.

Tanner; J. M., Whitehouse, R. H. 8z Takaishi, M. (1966). Standards from birth to maturity for height, weight, height velocity and weight veloc- ity: British children 1965. Archives of Disease in Childhood, 41, 453-471; 613-635.

Whitehead, R. G., Coward, W. A. & Lunn, P. G. (1973). Serum albumin concentration and the onset of kwashiorkor. Lancet, i, 63-66.

Satyanarayana, K. & Naidu, N. (1979). Nutrition and Accepted for publication 7th March, 1981.