- Email: [email protected]
Goitrous endemic in Guinea
Goitrous endemic in Guinea
Summary
Introduction
We identified a major goitrous area in the Republic of Guinea, characterised by an overall goitre prevalence of 70% in adults. Thyroid swelling was sometimes present at birth and affected 55% of schoolchildren. A difference between sexes appeared at puberty. Endemic cretinism, mainly in its myxoedematous form, was found in about 2% of goitrous patients. In this region, iodine deficiency is the primary causative factor (median urinary concentrations of 16 µg/L, and in 69% of inhabitants below the critical threshold of 20 µg/L). The diet contained substantial amounts of thiocyanate anions (median 6 mg/L in urine and in 27%, more than 10 mg/L) likely to further depress iodine bioavailability. Other dietary compounds, notably flavonoids were suspected to contribute. Overall nutritional and general health appeared satisfactory. The affected population is borderline euthyroid with a trend towards hypothyroidism in protracted disease. This area of Guinea may be regarded as the epicentre of the west African endemic and as one of the most severely goitrous regions ever described, requiring urgent public health measures.
The Republic of Guinea is in the goitrous belt along the Atlantic coast from west to central Africa.1 We undertook two successive missions to Fouta Djallon in the central mountainous part of the country to assess the prevalence of
goitre.
Subjects and methods Study area The Republic of Guinea (figure 1) has an area of 246 000 km2 and a population estimated at 69 million (28 per km2) in four administrative regions. Fouta Djallon is in the mid-Guinean region, a countryside of undulating foothills crossed by numerous small streams. In geological terms, it is precambian granito-gneiss in which intense weathering has occurred. Its main town, Labe, is 750 m above sea level. The commonly eaten foods are vegetable cereals such as sorghum (Sorghum vulgare), pearl-millet (Pennisetum glaucum/typhoides), and fonio (Dactylaria exilis). Maize (Zea mays), rice (Oriza sativa), groundnuts (Arachis hypogaea), cassava (Manihot esculenta/utilissima), and cruciferous plants of the Brassicae family are eaten when available. Meat, fowl, and curdled cows’ milk are occasional animal protein sources. Sea-fish, either smoked or salted, is eaten only by higher income families. Two field surveys were undertaken, a prefectorial study and a triangular study.
Prefectorial
study
The Prefecture of Labe consists of 13 sub-prefectures (figure 1), subdivided into 67 districts. 30 districts were selected by the
Direction of Public Health, Conakry, Guinea (M Konde MD, M Daffe MD, B Sylla MD, 0 Barry MD, S Diallo MD); and Department of Food Sciences, University Louis-Pasteur Strasbourg, 67400 llikirch, France (Prof Y Ingenbleek MD)
Correspondence to:
Prof Y
Ingenbleek
Figure 1: Prefecture of Labe The numbers indicate the locations of 28 randomly selected districts and star symbols the villages in the triangular study.
1675
Table 1: Prevalence of goitre in Fouta
Djallon
WHO-recommended
drawing of lots methods.2 All the subone (Garambe) were represented by at least one prefectures except district. For local reasons (a religious ceremony and an overflowing river), investigators were unable to reach 2 of the 30 districts, so the samples from the two nearest chosen districts were doubled. 7 districts were in the town of Labe and 21 in the country. 30 non-pregnant adult women, aged 18-40 years, were randomly selected from each district; hence, 909 women from 28 districts were clinically examined, and every tenth had fasting blood and urine samples taken before iodised oil (Lipiodol) was given.
Triangular study study was carried out in three rural villages in a triangular position around Labe: Karse (40 km north of Labe, 128 inhabitants), Dyohe (30 km south west of Labé, 158), and Hinde (22 km south east of Labe, 104) (figure 1: star-shaped symbols). All 390 inhabitants (180 males and 210 females) were clinically examined, and 66 adults (30 men and 36 women) had samples taken. At the end of this study, Lipiodol was also given. The second
*All values are given as mean (SD) excepot for TSH (median and range). Table 3: Indices of thyroid function in 106 rural women
(CRP), and a-acid glycoprotein (orosomucoid, AGP) were measured by immunonephelometry with the Cobas-Fara apparatus (Hoffmann-Laroche, Switzerland). Their aggregation within a prognostic inflammatory and nutritional index (PINI)6 is an indicator of health.7 The control group (n = 37) consisted of adult nonpregnant women living in areas where there is no endemic
goitre.
Analysis All values are given as means (SD), with the exception of serum TSH, urine 1271, and SCN, which are expressed as medians. Statistical interpretation of the data was carried out according to Snedecor and Cochran.8 Mann-Whitney U test9 was used to compare unpaired results.
Results
Laboratory measurements Weight and height were recorded for calculation of the body mass index (BMI). Thyroid swelling was assessed according to WHO classification.3Measurement of serum total triiodothyronine (TT3), thyroxine (TT4), free triiodothyronine (FT3), and thyroxine (FT4)
was by radioimmunoassay (Amersham, UK). thyroxine-binding globulin (TBG) and thyroidstimulating hormone (TSH) were obtained by radial immunodiffusion with LC-Partigen immunoplates, and by the Berilux chemiluminescent 290 system (Behringwerke AG, Germany), respectively. Stable iodine (127I) and thiocyanate (SCN) were estimated in the urine samples by the methods of Sandell-Kolthoff4 and Aldridge,s respectively. 1271 in 10 different samples of drinking waters was below the detection threshold of colorimetric methods and required the use of neutronic activation (CNRS, Centre of Nuclear Research, University Louis-Pasteur, France). Serum albumin, transthyretin (TTR), C-reactive protein
Serum
Table 2: Mean (SD) anthropometric, nutritional, and Inflammatory Indicators In 106 rural women
1676
Measurement of 12I in 10 well-water and stream samples in the Prefecture of Labe showed a mean iodine content of 0-78
(0’11) µg/L (range: 0,4-1’6). The triangular survey shows that the overall prevalence of goitre in 390 rural inhabitants was 57-7% in males and 80 3% in females. Stage I goitre predominated (34-1%) and stage IIIhad the lowest prevalence (17%) in males, whereas stage II predominated (38-9%) and stage III affected 9-6% of females. 5 typical cretins (1 male, 4 females) with myxoedematous symptoms were identified (1-87% of people with goitre) (table 1). The prefectorial survey shows the differences in prevalence of goitre between 190 urban women (63-1%) and 719 rural women (79-5%). This reduced overall goitre prevalence in the urban female population was mainly due to the shrinking of stages II and III, explaining that stage I culminated at 41 6%. In contrast, the rural population had the highest prevalence of stage II (34-5%) and stage III was five-fold more frequent (table 1). The prevalence of stages of goitre in triangular and prefectorial studies of rural females are similar in women free of goitre (19-7% 20-4%), in all goitres (80-3% vs 79-5%) and in stage I goitre (29-8% 29-7%). Random selection of women in the prefectorial study slightly underestimate the stage II forms (34-5% 389%) and lead to significant over-representation of stage III forms (153% v 9-6%). This discrepancy is likely to be due to iodine treatment attracting recruits aware of thyroid swelling. Table 2 shows the main anthropometric, nutritional, and inflammatory findings in 106 rural women according to
(data
not
shown),
consistent with the
universally higher
prevalence found in adult women. Moreover, 5 cases of myxoedematous cretinism with minor neurological signs (1 87% of the goitrous population) were recognised. This prevalence is surprisingly low since 7-9% in a 70 % goitrous population has been reported.13 A likely explanation is that myxoedematous children, especially those affected by the most severe neurological symptoms, suffer early and high mortality rates. Iodine deficiency is the cause of Guinean endemic and its shortage appears evenly distributed in this area. The severity of this endemic is exemplified by the very low 1271 content in drinking water and in urine (69% of the
Figure 2: Distribution of Individual values of TSH against 106 non-pregnant rural women The rectangular outline is the normal reference range.
FT,, in
BMI was at the lower limit of normal throughout. Serum albumin and TTR showed a slight and regular decline progressing with the development of goitre. CRP and AGP were within normal limits or slightly raised. PINI scores remained below the critical threshold of 10.6 Table 3 shows serum and urinary indices of thyroid function in 106 women. Median urinary output of 1271 was 16 µg/L whatever the goitrous stage, and 73 of 106 (69%) subjects had a 1271 excretion below 20 µg/L. Median urine SCN was 6 mg/L with 29 women of 106 (27%) excreting more than 10 mg/L. Serum TBG decreased in stage III, and TT4 and FT4 were significantly decreased in stage 0 compared with controls. Even in the absence of goitre, most FT4 values were at the lower limit of normal (figure 2). Serum TSH concentrations were at the upper limit of normal in subjects without goitre and showed in stages I, II, and III a progressive rise proportionate to the thyroid
goitre.
swelling. Discussion The conforms to WHO prefectorial study recommendations for evaluating the prevalence of goitre in a well-defined geographical area. This epidemiological approach has the advantage of reflecting the overall prevalence of goitre, but fails to identify sex-related and age-related aspects or discrepancies due to local genetic, dietary, and immune peculiarities. 10-12 The triangular study examined all inhabitants of three rural villages and showed that thyroid swelling sometimes exists at birth, already affects 55% of preschool children, and manifests a sexual difference at the onset of adolescence
inhabitants below the critical threshold of 20 µg/L,14) Urine SCN showed larger ranges than 1271, presumably reflecting diet before the sample. Urine SCN around 6 mg/L and exceeding 10 mg/L in 27% of individuals may increase the prevalence of goitre. 15 Because of dietary variation, a urine I/SCN index below 2 is regarded as defining the most severe goitrous area.16 This ratio is useful in stratifying the dietary risks arising from regions where cassava and other SCN-yielding vegetables are consumed daily,l’ although it fails to identify other goitrogenic substances such as glycosylflavones which have antithyroid activity through the peroxidase system.18 Since flavonoids are found in most millet varieties,19,20 which constitute the staple food in the Fouta Djallon area, the I/SCN index may underscore the goitrogenic potential of the local diet. Indeed, the dietary pattern characterising the mid-Guinean endemic is closer to the one described in subdesert savannah20,21 rather than that typical of tropical regions. 17 Most TT4 and FT4 values, even those in subjects without goitre, were at the lower limit of normal, and most TSH results at the upper limit. FT4 and TSH concentrations were negatively correlated indicating that the free-thyroxine fraction remains the primary factor responsible for feedback regulation of pituitary response, consistent with established views.22 Conversely, the enhanced production of TT3 and FT3 appears as the direct consequence of TSH hypersecretion,23 aiming at sparing iodine and maintaining euthyroidism. Most of these hormonal alterations are already detectable in apparently healthy individuals (stage 0), emphasising that, in this severely iodine-restricted area, compensatory mechanisms operate before disease becomes obvious. The progressive decline of TT3 and FT3 in spite of persisting TSH overstimulation, clearly points to hypothyroid status.23,24 One feature of the prefectorial study was the finding that the epidemiological and hormonal differences in goitrous patterns between 719 rural and 190 urban women (table 1) developed in spite of even distribution of 1271 and SCN in the whole area: the measurement of 1271 urinary output in the town of Labe was slightly but not significantly higher than that found under rural conditions (table 3), and urinary SCN similar to that in rural areas. We consider that our epidemiological and hormonal results are not due to differences in 1271 or SCN intakes, nor inflammatory burden as measured by CRP and AGP. TTR appeared to discriminate differences in nutritional status between urban (193 4 [20] mg/L, n 25) and rural (169 2 [19] mg/L, n=106) subjects, indicating better protein nutritional status in the urban setting of Labe. Investigators in the Sudan described differences in goitre epidemiology and thyroid function between equally iodinedeficient rural and urban areas and speculated that higher intake of goitrogenic substances could be responsible .21 =
1677
hypothesis seems unlikely in Guinea. Our studies suggest that improving protein status may help prevent the deleterious effects of low iodine and/or high SCN consumption: TTR determines the availability of retinol to peripheral tissues;7.26 and glycosylation and maturation of thyroglobulin could be retinoid-dependent processes,27.28 implying that any alteration in protein status, as assessed by TTR, would alter synthesis of thyroglobulin. From the epidemiological and clinical data already collected from surrounding countries (Guinea-Bissau,29 Senegambia,27 Mali 30 Liberia,31 Sierra Leone,32 and Ivory Coast33) there is no doubt that the Republic of Guinea must be regarded as the epicentre of this goitrous area of west Africa, and one of the most severe endemic areas ever described. Iodine shortage is the main causative factor, requiring urgent public health measures.
This
Our studies were supported by WHO and the International Council for Control of Iodine Deficiency Disorders (ICCIDD). We thank Mr M Fofana, Mr 0 Bangoura, and Mr M Kader (Ministry of Public Health and Social Affairs, Conakry); Mr K Bailey, Mr B deBenoist, and Mr B Lemaire (WHO-ICCIDD-AFRO, Brazzaville); Mr G Imboua-Bogui (WHO Representative), Mr R Grancourt (UNICEF Representative); and from the Louis-Pasteur University, Prof A Stahl, Prof G Ferard, Dr A Bourguignat, Dr R Sapin, Dr M Eber, and Dr A Stampfler.
References
Kelly FC, Snedden WW. Prevalence and geographical distribution of endemic goitre. In: Endemic goitre, WHO monograph series No 44,
1
2
3
4
5 6
1960: 117-53. Geneva. Bennett S, Woods T, Liyange W, Smith L. A simplified general method for cluster-sample surveys of health in developing countries. World Health Stat Quart 1991; 44: 98-106. Pérez C, Scrimshaw NS, Muñoz JA. Technique of endemic goitre surveys. In: Endemic goitre, WHO monograph series No 44, 1960: 369-83. Geneva. Sandell EB, Kolthoff IM. Micro determination of iodine by a catalytic method. Mikrochim Acta 1937; 1: 9-25. Aldridge WN. The estimation of microquantities of cyanide and thiocyanate. Analyst (London) 1945; 70: 474-75. Ingenbleek Y, Carpentier YA. A prognostic inflammatory and nutritional index scoring critically ill patients. Int J Vitamin Nutr Res
1985; 55: 91-101. 7
8 9
Ingenbleek Y, Young V. Transthyretin (prealbumin) in health and disease: nutritional implications. Annu Rev Nutr 1994; 14: 495-533. Snedecor GW, Cochran WG. Statistical methods, 6 th ed. Ames: Iowa State University Press, 1967: 91-119. Mann HB, Whitney DR. On a test of whether one or two random variables is statistically larger than the other. Ann Auth Stat 1947; 18:
50-60. 10 Koutras DA. Trace Elements, genetic and other factors. In: Stanbury JB, Hetzel BS, eds. Endemic goiter and endemic cretinism. New York: John Wiley & Sons, 1980: 255-68. 11 Benmiloud M, Bachtarzi H, Chaouki ML. Public health and nutritional aspects of endemic goitre and cretinism in Africa. In: Delange F, Ahluwalia R, eds. Cassava toxicity and thyroid research and public health issues. International Development Research Centre, Ottawa, Canada. 1983: 49-54. 12 Gaitan E, Cooksey RC, Lindsay RH. Factors other than iodine deficiency in endemic goiter: goitrigens and protein-calorie malnutrition (PCM). In: Dunn JT, Pretell EA, Daza CH, Viteri FE,
1678
eds. Towards the eradication of endemic goiter, cretinism, and iodine deficiency. PAHO-WHO Sci Publ No 502, 1986: 28-45. Pan American Sanitary Bureau, Washington DC. 13 Hershman JM, Melnick GA, Kastner R. Economic consequences of endemic goiter. In Dunn JT, Pretell EA, Daza CH, Viteri FE, eds. Towards the eradication of endemic goiter, cretinism, and iodine deficiency. PAHO-WHO Sci Publ No 502, 1986: 96-114. Pan American Sanitary Bureau, Washington DC. 14 Dunn JT, van der Haar F. A practical guide to the correction of iodine deficiency. Wageningen, Netherlands: Technical manual No 3, ICCIDD-UNICEF-WHO, 1990: 1-62. 15 Wollmann SH. Inhibition by thiocyanate of accumulation of radioiodine by thyroid gland. Am J Physiol 1962; 203: 517-24. 16 Bourdoux P, Thilly C, Delange F, Ermans AM. A new look at old concepts in laboratory evaluation of endemic goiter. In: Dunn JT, Pretell EA, Daza CH, Viteri FE, eds. Towards the eradication of endemic goiter, cretinism, and iodine deficiency. PAHO-WHO Sci Publ No 502, 1986: 115-29. Pan American Sanitary Bureau, Washington DC. 17 Courtois P, Bourdoux P, Lagasse R, Ermans AM, Delange F. Role of balance between the dietary supplies of iodine and thiocyanate in the etiology of endemic goitre in the Ubangi area. In: Delange F, Iteke FB, Ermans AM, eds. Nutritional factors involved in the goitrogenic action of cassava. International Development Research Centre, Ottawa, Canada. 1982: 65-69. 18 Gaitan E, Lindsay RH, Reichert RD, Ingbar SH, et al. Antithyroid and goitrogenic effects of millet: role of C-glycosylflavones. J Clin Endocrinol Metab 1989; 68: 707-14. 19 Klopfenstein F, Hoseney RC, Leipold HW. Goitrogenic effects of pearl millet diets. Nutr Rep Internat 1983; 27: 1039-49. 20 Osman AK. Bulrush millet (Pennisetum typhoides) a contributory factor to the endemicity of goitre in Western Sudan. Ecol Food Nutr 1981; 11: 121-28. 21 Eltom M, Hofvander Y, Torelm I, Fellstrom B. Endemic goitre in the Darfur region (Sudan). Epidemiology and aetiology. Acta Med Scand 1984; 215: 467-75. 22 Ingenbleek Y. Thyroid function in nonthyroid diseases. In: De Visscher M, ed. The thyroid gland. New York: Raven Press, 1980: 499-527. 23 Stanbury JB, Hetzel BS. Endemic goiter and endemic cretinism. Iodine nutrition in health and disease. New York: John Wiley & Sons, 1980: 1-606. 24 De Visscher M, Ingenbleek Y. Hypothyroidism. In: De Visscher M, ed. The thyroid gland. New York: Raven Press, 1980: 377-412. 25 Eltom M, Salih MA, Bostrom H, Dahlberg PA. Differences in aetiology and thyroid function in endemic goitre between rural and urban areas of the Darfur region of the Sudan. Acta Endocrinol (Kbn) 1985; 108: 356-60. 26 Ingenbleek Y, Van Den Schrieck HG, De Nayer PH, De Visscher M. The role of retinol-binding protein in protein-calorie malnutrition. Metabolism 1975; 24: 633-41. 27 Ingenbleek Y, De Visscher M. Hormonal and nutritional status: critical conditions for endemic goiter epidemiology?. Metabolism 1979; 28: 9-19. 28 Ingenbleek Y, Luypaert B, De Nayer PH. Nutritional status and endemic goitre. Lancet 1980; i: 388-92. 29 Oliveira AL, Batista JL, Silva AP, Sobrinho LG, Rocha LC. Endemic goitre in Guinea-Bissau. Bull World Health Organ 1991; 63: 347-48. 30 Roux F, Rhaly A, Bisset JP, Dumoulin B, Fichet B, Chaventre A. Le goitre endémique au Mali. Méd Nutr 1983; 19: 339-51. 31 Bediako BM. Studies on the prevalence of endemic goitre in Nimba County. Monrovia: Project report, University of Liberia College of Medicine, 1985, 1-97. 32 Wilson DC, Grundy HM, Steel RW, et al. Goitre in Sierra Leone. Trans R Soc Trop Med Hyg 1954; 48: 481-89. 33 Latapie JL, Clerc M, Beda B, et al. Aspects cliniques et biologiques du goitre endémique dans la région de Man (Côte d’Ivoire). Ann Endocrinol (Paris) 1981; 42: 517-30.
Summary
Introduction
We identified a major goitrous area in the Republic of Guinea, characterised by an overall goitre prevalence of 70% in adults. Thyroid swelling was sometimes present at birth and affected 55% of schoolchildren. A difference between sexes appeared at puberty. Endemic cretinism, mainly in its myxoedematous form, was found in about 2% of goitrous patients. In this region, iodine deficiency is the primary causative factor (median urinary concentrations of 16 µg/L, and in 69% of inhabitants below the critical threshold of 20 µg/L). The diet contained substantial amounts of thiocyanate anions (median 6 mg/L in urine and in 27%, more than 10 mg/L) likely to further depress iodine bioavailability. Other dietary compounds, notably flavonoids were suspected to contribute. Overall nutritional and general health appeared satisfactory. The affected population is borderline euthyroid with a trend towards hypothyroidism in protracted disease. This area of Guinea may be regarded as the epicentre of the west African endemic and as one of the most severely goitrous regions ever described, requiring urgent public health measures.
The Republic of Guinea is in the goitrous belt along the Atlantic coast from west to central Africa.1 We undertook two successive missions to Fouta Djallon in the central mountainous part of the country to assess the prevalence of
goitre.
Subjects and methods Study area The Republic of Guinea (figure 1) has an area of 246 000 km2 and a population estimated at 69 million (28 per km2) in four administrative regions. Fouta Djallon is in the mid-Guinean region, a countryside of undulating foothills crossed by numerous small streams. In geological terms, it is precambian granito-gneiss in which intense weathering has occurred. Its main town, Labe, is 750 m above sea level. The commonly eaten foods are vegetable cereals such as sorghum (Sorghum vulgare), pearl-millet (Pennisetum glaucum/typhoides), and fonio (Dactylaria exilis). Maize (Zea mays), rice (Oriza sativa), groundnuts (Arachis hypogaea), cassava (Manihot esculenta/utilissima), and cruciferous plants of the Brassicae family are eaten when available. Meat, fowl, and curdled cows’ milk are occasional animal protein sources. Sea-fish, either smoked or salted, is eaten only by higher income families. Two field surveys were undertaken, a prefectorial study and a triangular study.
Prefectorial
study
The Prefecture of Labe consists of 13 sub-prefectures (figure 1), subdivided into 67 districts. 30 districts were selected by the
Direction of Public Health, Conakry, Guinea (M Konde MD, M Daffe MD, B Sylla MD, 0 Barry MD, S Diallo MD); and Department of Food Sciences, University Louis-Pasteur Strasbourg, 67400 llikirch, France (Prof Y Ingenbleek MD)
Correspondence to:
Prof Y
Ingenbleek
Figure 1: Prefecture of Labe The numbers indicate the locations of 28 randomly selected districts and star symbols the villages in the triangular study.
1675
Table 1: Prevalence of goitre in Fouta
Djallon
WHO-recommended
drawing of lots methods.2 All the subone (Garambe) were represented by at least one prefectures except district. For local reasons (a religious ceremony and an overflowing river), investigators were unable to reach 2 of the 30 districts, so the samples from the two nearest chosen districts were doubled. 7 districts were in the town of Labe and 21 in the country. 30 non-pregnant adult women, aged 18-40 years, were randomly selected from each district; hence, 909 women from 28 districts were clinically examined, and every tenth had fasting blood and urine samples taken before iodised oil (Lipiodol) was given.
Triangular study study was carried out in three rural villages in a triangular position around Labe: Karse (40 km north of Labe, 128 inhabitants), Dyohe (30 km south west of Labé, 158), and Hinde (22 km south east of Labe, 104) (figure 1: star-shaped symbols). All 390 inhabitants (180 males and 210 females) were clinically examined, and 66 adults (30 men and 36 women) had samples taken. At the end of this study, Lipiodol was also given. The second
*All values are given as mean (SD) excepot for TSH (median and range). Table 3: Indices of thyroid function in 106 rural women
(CRP), and a-acid glycoprotein (orosomucoid, AGP) were measured by immunonephelometry with the Cobas-Fara apparatus (Hoffmann-Laroche, Switzerland). Their aggregation within a prognostic inflammatory and nutritional index (PINI)6 is an indicator of health.7 The control group (n = 37) consisted of adult nonpregnant women living in areas where there is no endemic
goitre.
Analysis All values are given as means (SD), with the exception of serum TSH, urine 1271, and SCN, which are expressed as medians. Statistical interpretation of the data was carried out according to Snedecor and Cochran.8 Mann-Whitney U test9 was used to compare unpaired results.
Results
Laboratory measurements Weight and height were recorded for calculation of the body mass index (BMI). Thyroid swelling was assessed according to WHO classification.3Measurement of serum total triiodothyronine (TT3), thyroxine (TT4), free triiodothyronine (FT3), and thyroxine (FT4)
was by radioimmunoassay (Amersham, UK). thyroxine-binding globulin (TBG) and thyroidstimulating hormone (TSH) were obtained by radial immunodiffusion with LC-Partigen immunoplates, and by the Berilux chemiluminescent 290 system (Behringwerke AG, Germany), respectively. Stable iodine (127I) and thiocyanate (SCN) were estimated in the urine samples by the methods of Sandell-Kolthoff4 and Aldridge,s respectively. 1271 in 10 different samples of drinking waters was below the detection threshold of colorimetric methods and required the use of neutronic activation (CNRS, Centre of Nuclear Research, University Louis-Pasteur, France). Serum albumin, transthyretin (TTR), C-reactive protein
Serum
Table 2: Mean (SD) anthropometric, nutritional, and Inflammatory Indicators In 106 rural women
1676
Measurement of 12I in 10 well-water and stream samples in the Prefecture of Labe showed a mean iodine content of 0-78
(0’11) µg/L (range: 0,4-1’6). The triangular survey shows that the overall prevalence of goitre in 390 rural inhabitants was 57-7% in males and 80 3% in females. Stage I goitre predominated (34-1%) and stage IIIhad the lowest prevalence (17%) in males, whereas stage II predominated (38-9%) and stage III affected 9-6% of females. 5 typical cretins (1 male, 4 females) with myxoedematous symptoms were identified (1-87% of people with goitre) (table 1). The prefectorial survey shows the differences in prevalence of goitre between 190 urban women (63-1%) and 719 rural women (79-5%). This reduced overall goitre prevalence in the urban female population was mainly due to the shrinking of stages II and III, explaining that stage I culminated at 41 6%. In contrast, the rural population had the highest prevalence of stage II (34-5%) and stage III was five-fold more frequent (table 1). The prevalence of stages of goitre in triangular and prefectorial studies of rural females are similar in women free of goitre (19-7% 20-4%), in all goitres (80-3% vs 79-5%) and in stage I goitre (29-8% 29-7%). Random selection of women in the prefectorial study slightly underestimate the stage II forms (34-5% 389%) and lead to significant over-representation of stage III forms (153% v 9-6%). This discrepancy is likely to be due to iodine treatment attracting recruits aware of thyroid swelling. Table 2 shows the main anthropometric, nutritional, and inflammatory findings in 106 rural women according to
(data
not
shown),
consistent with the
universally higher
prevalence found in adult women. Moreover, 5 cases of myxoedematous cretinism with minor neurological signs (1 87% of the goitrous population) were recognised. This prevalence is surprisingly low since 7-9% in a 70 % goitrous population has been reported.13 A likely explanation is that myxoedematous children, especially those affected by the most severe neurological symptoms, suffer early and high mortality rates. Iodine deficiency is the cause of Guinean endemic and its shortage appears evenly distributed in this area. The severity of this endemic is exemplified by the very low 1271 content in drinking water and in urine (69% of the
Figure 2: Distribution of Individual values of TSH against 106 non-pregnant rural women The rectangular outline is the normal reference range.
FT,, in
BMI was at the lower limit of normal throughout. Serum albumin and TTR showed a slight and regular decline progressing with the development of goitre. CRP and AGP were within normal limits or slightly raised. PINI scores remained below the critical threshold of 10.6 Table 3 shows serum and urinary indices of thyroid function in 106 women. Median urinary output of 1271 was 16 µg/L whatever the goitrous stage, and 73 of 106 (69%) subjects had a 1271 excretion below 20 µg/L. Median urine SCN was 6 mg/L with 29 women of 106 (27%) excreting more than 10 mg/L. Serum TBG decreased in stage III, and TT4 and FT4 were significantly decreased in stage 0 compared with controls. Even in the absence of goitre, most FT4 values were at the lower limit of normal (figure 2). Serum TSH concentrations were at the upper limit of normal in subjects without goitre and showed in stages I, II, and III a progressive rise proportionate to the thyroid
goitre.
swelling. Discussion The conforms to WHO prefectorial study recommendations for evaluating the prevalence of goitre in a well-defined geographical area. This epidemiological approach has the advantage of reflecting the overall prevalence of goitre, but fails to identify sex-related and age-related aspects or discrepancies due to local genetic, dietary, and immune peculiarities. 10-12 The triangular study examined all inhabitants of three rural villages and showed that thyroid swelling sometimes exists at birth, already affects 55% of preschool children, and manifests a sexual difference at the onset of adolescence
inhabitants below the critical threshold of 20 µg/L,14) Urine SCN showed larger ranges than 1271, presumably reflecting diet before the sample. Urine SCN around 6 mg/L and exceeding 10 mg/L in 27% of individuals may increase the prevalence of goitre. 15 Because of dietary variation, a urine I/SCN index below 2 is regarded as defining the most severe goitrous area.16 This ratio is useful in stratifying the dietary risks arising from regions where cassava and other SCN-yielding vegetables are consumed daily,l’ although it fails to identify other goitrogenic substances such as glycosylflavones which have antithyroid activity through the peroxidase system.18 Since flavonoids are found in most millet varieties,19,20 which constitute the staple food in the Fouta Djallon area, the I/SCN index may underscore the goitrogenic potential of the local diet. Indeed, the dietary pattern characterising the mid-Guinean endemic is closer to the one described in subdesert savannah20,21 rather than that typical of tropical regions. 17 Most TT4 and FT4 values, even those in subjects without goitre, were at the lower limit of normal, and most TSH results at the upper limit. FT4 and TSH concentrations were negatively correlated indicating that the free-thyroxine fraction remains the primary factor responsible for feedback regulation of pituitary response, consistent with established views.22 Conversely, the enhanced production of TT3 and FT3 appears as the direct consequence of TSH hypersecretion,23 aiming at sparing iodine and maintaining euthyroidism. Most of these hormonal alterations are already detectable in apparently healthy individuals (stage 0), emphasising that, in this severely iodine-restricted area, compensatory mechanisms operate before disease becomes obvious. The progressive decline of TT3 and FT3 in spite of persisting TSH overstimulation, clearly points to hypothyroid status.23,24 One feature of the prefectorial study was the finding that the epidemiological and hormonal differences in goitrous patterns between 719 rural and 190 urban women (table 1) developed in spite of even distribution of 1271 and SCN in the whole area: the measurement of 1271 urinary output in the town of Labe was slightly but not significantly higher than that found under rural conditions (table 3), and urinary SCN similar to that in rural areas. We consider that our epidemiological and hormonal results are not due to differences in 1271 or SCN intakes, nor inflammatory burden as measured by CRP and AGP. TTR appeared to discriminate differences in nutritional status between urban (193 4 [20] mg/L, n 25) and rural (169 2 [19] mg/L, n=106) subjects, indicating better protein nutritional status in the urban setting of Labe. Investigators in the Sudan described differences in goitre epidemiology and thyroid function between equally iodinedeficient rural and urban areas and speculated that higher intake of goitrogenic substances could be responsible .21 =
1677
hypothesis seems unlikely in Guinea. Our studies suggest that improving protein status may help prevent the deleterious effects of low iodine and/or high SCN consumption: TTR determines the availability of retinol to peripheral tissues;7.26 and glycosylation and maturation of thyroglobulin could be retinoid-dependent processes,27.28 implying that any alteration in protein status, as assessed by TTR, would alter synthesis of thyroglobulin. From the epidemiological and clinical data already collected from surrounding countries (Guinea-Bissau,29 Senegambia,27 Mali 30 Liberia,31 Sierra Leone,32 and Ivory Coast33) there is no doubt that the Republic of Guinea must be regarded as the epicentre of this goitrous area of west Africa, and one of the most severe endemic areas ever described. Iodine shortage is the main causative factor, requiring urgent public health measures.
This
Our studies were supported by WHO and the International Council for Control of Iodine Deficiency Disorders (ICCIDD). We thank Mr M Fofana, Mr 0 Bangoura, and Mr M Kader (Ministry of Public Health and Social Affairs, Conakry); Mr K Bailey, Mr B deBenoist, and Mr B Lemaire (WHO-ICCIDD-AFRO, Brazzaville); Mr G Imboua-Bogui (WHO Representative), Mr R Grancourt (UNICEF Representative); and from the Louis-Pasteur University, Prof A Stahl, Prof G Ferard, Dr A Bourguignat, Dr R Sapin, Dr M Eber, and Dr A Stampfler.
References
Kelly FC, Snedden WW. Prevalence and geographical distribution of endemic goitre. In: Endemic goitre, WHO monograph series No 44,
1
2
3
4
5 6
1960: 117-53. Geneva. Bennett S, Woods T, Liyange W, Smith L. A simplified general method for cluster-sample surveys of health in developing countries. World Health Stat Quart 1991; 44: 98-106. Pérez C, Scrimshaw NS, Muñoz JA. Technique of endemic goitre surveys. In: Endemic goitre, WHO monograph series No 44, 1960: 369-83. Geneva. Sandell EB, Kolthoff IM. Micro determination of iodine by a catalytic method. Mikrochim Acta 1937; 1: 9-25. Aldridge WN. The estimation of microquantities of cyanide and thiocyanate. Analyst (London) 1945; 70: 474-75. Ingenbleek Y, Carpentier YA. A prognostic inflammatory and nutritional index scoring critically ill patients. Int J Vitamin Nutr Res
1985; 55: 91-101. 7
8 9
Ingenbleek Y, Young V. Transthyretin (prealbumin) in health and disease: nutritional implications. Annu Rev Nutr 1994; 14: 495-533. Snedecor GW, Cochran WG. Statistical methods, 6 th ed. Ames: Iowa State University Press, 1967: 91-119. Mann HB, Whitney DR. On a test of whether one or two random variables is statistically larger than the other. Ann Auth Stat 1947; 18:
50-60. 10 Koutras DA. Trace Elements, genetic and other factors. In: Stanbury JB, Hetzel BS, eds. Endemic goiter and endemic cretinism. New York: John Wiley & Sons, 1980: 255-68. 11 Benmiloud M, Bachtarzi H, Chaouki ML. Public health and nutritional aspects of endemic goitre and cretinism in Africa. In: Delange F, Ahluwalia R, eds. Cassava toxicity and thyroid research and public health issues. International Development Research Centre, Ottawa, Canada. 1983: 49-54. 12 Gaitan E, Cooksey RC, Lindsay RH. Factors other than iodine deficiency in endemic goiter: goitrigens and protein-calorie malnutrition (PCM). In: Dunn JT, Pretell EA, Daza CH, Viteri FE,
1678
eds. Towards the eradication of endemic goiter, cretinism, and iodine deficiency. PAHO-WHO Sci Publ No 502, 1986: 28-45. Pan American Sanitary Bureau, Washington DC. 13 Hershman JM, Melnick GA, Kastner R. Economic consequences of endemic goiter. In Dunn JT, Pretell EA, Daza CH, Viteri FE, eds. Towards the eradication of endemic goiter, cretinism, and iodine deficiency. PAHO-WHO Sci Publ No 502, 1986: 96-114. Pan American Sanitary Bureau, Washington DC. 14 Dunn JT, van der Haar F. A practical guide to the correction of iodine deficiency. Wageningen, Netherlands: Technical manual No 3, ICCIDD-UNICEF-WHO, 1990: 1-62. 15 Wollmann SH. Inhibition by thiocyanate of accumulation of radioiodine by thyroid gland. Am J Physiol 1962; 203: 517-24. 16 Bourdoux P, Thilly C, Delange F, Ermans AM. A new look at old concepts in laboratory evaluation of endemic goiter. In: Dunn JT, Pretell EA, Daza CH, Viteri FE, eds. Towards the eradication of endemic goiter, cretinism, and iodine deficiency. PAHO-WHO Sci Publ No 502, 1986: 115-29. Pan American Sanitary Bureau, Washington DC. 17 Courtois P, Bourdoux P, Lagasse R, Ermans AM, Delange F. Role of balance between the dietary supplies of iodine and thiocyanate in the etiology of endemic goitre in the Ubangi area. In: Delange F, Iteke FB, Ermans AM, eds. Nutritional factors involved in the goitrogenic action of cassava. International Development Research Centre, Ottawa, Canada. 1982: 65-69. 18 Gaitan E, Lindsay RH, Reichert RD, Ingbar SH, et al. Antithyroid and goitrogenic effects of millet: role of C-glycosylflavones. J Clin Endocrinol Metab 1989; 68: 707-14. 19 Klopfenstein F, Hoseney RC, Leipold HW. Goitrogenic effects of pearl millet diets. Nutr Rep Internat 1983; 27: 1039-49. 20 Osman AK. Bulrush millet (Pennisetum typhoides) a contributory factor to the endemicity of goitre in Western Sudan. Ecol Food Nutr 1981; 11: 121-28. 21 Eltom M, Hofvander Y, Torelm I, Fellstrom B. Endemic goitre in the Darfur region (Sudan). Epidemiology and aetiology. Acta Med Scand 1984; 215: 467-75. 22 Ingenbleek Y. Thyroid function in nonthyroid diseases. In: De Visscher M, ed. The thyroid gland. New York: Raven Press, 1980: 499-527. 23 Stanbury JB, Hetzel BS. Endemic goiter and endemic cretinism. Iodine nutrition in health and disease. New York: John Wiley & Sons, 1980: 1-606. 24 De Visscher M, Ingenbleek Y. Hypothyroidism. In: De Visscher M, ed. The thyroid gland. New York: Raven Press, 1980: 377-412. 25 Eltom M, Salih MA, Bostrom H, Dahlberg PA. Differences in aetiology and thyroid function in endemic goitre between rural and urban areas of the Darfur region of the Sudan. Acta Endocrinol (Kbn) 1985; 108: 356-60. 26 Ingenbleek Y, Van Den Schrieck HG, De Nayer PH, De Visscher M. The role of retinol-binding protein in protein-calorie malnutrition. Metabolism 1975; 24: 633-41. 27 Ingenbleek Y, De Visscher M. Hormonal and nutritional status: critical conditions for endemic goiter epidemiology?. Metabolism 1979; 28: 9-19. 28 Ingenbleek Y, Luypaert B, De Nayer PH. Nutritional status and endemic goitre. Lancet 1980; i: 388-92. 29 Oliveira AL, Batista JL, Silva AP, Sobrinho LG, Rocha LC. Endemic goitre in Guinea-Bissau. Bull World Health Organ 1991; 63: 347-48. 30 Roux F, Rhaly A, Bisset JP, Dumoulin B, Fichet B, Chaventre A. Le goitre endémique au Mali. Méd Nutr 1983; 19: 339-51. 31 Bediako BM. Studies on the prevalence of endemic goitre in Nimba County. Monrovia: Project report, University of Liberia College of Medicine, 1985, 1-97. 32 Wilson DC, Grundy HM, Steel RW, et al. Goitre in Sierra Leone. Trans R Soc Trop Med Hyg 1954; 48: 481-89. 33 Latapie JL, Clerc M, Beda B, et al. Aspects cliniques et biologiques du goitre endémique dans la région de Man (Côte d’Ivoire). Ann Endocrinol (Paris) 1981; 42: 517-30.