YING CHIN KO, TSU NAI WANG, LI YU TSAI, FOWN TZU CHANG, and SHUN JEN CHANG

ABSTRACT.

Objective. To explore the prevalence and related factors of hyperuricemia among adolescent Taiwan aborigines in tribes with a high prevalence of adult gout, compared with adolescents of low prevalence aboriginal and non-aboriginal tribes.

Methods. The participants were aborigines and non-aborigines in Taiwan, age 12 to 15 years and free of gout. Each participant provided information on sex, age, and parents' tribal background as well as body weight and height. Serum samples were analyzed for biochemical markers. A logistic regression model was used to study factors related to hyperuricemia.

Results. In total 940 adolescents participated. The hyperuricemia rate in tribes with high gout prevalence (57.7%) was higher than in non-aborigines (48.2%) and in aboriginal tribes with low gout prevalence (34.0%). Factors statistically significantly related to hyperuricemia were tribe, sex, obesity, creatinine, and cholesterol levels in preliminary analysis. After adjustment by the logistic regression model, obese boys with higher creatinine were most likely to have hyperuricemia. Adolescents whose parents originated from tribes with high gout prevalence had a tendency to have hyperuricemia, and those aborigines from tribes with low gout prevalence had a low prevalence of hyperuricemia compared to non-aborigines.

Conclusion. The prevalence of hyperuricemia in aboriginal adolescents mirrors the incidence of adult gout, implying a predisposition for adult gout in childhood, with genetic and/or environmental components presumably contributing to the differences between tribes; this may be of potential benefit to preventive efforts. (J Rheumatol 2002;29:837-42)

Key Indexing Terms:

HYPERURICEMIA
ADOLESCENCE
ABORIGINES
COMMUNITY

That there is a high prevalence of gout disease among adult Taiwan aborigines is well known, with incidences ranging from 15 to 44% documented for Atayal men aged over 40 years, followed by Bununs (28.1%) and Paiwans and Tsou tribes (> 5%)^1,2. In addition to high prevalence of the disease itself among aboriginal tribes, hyperuricemia is also known to be an asymptomatic feature among the adults² and children³, but not all the subtribes of Taiwan aborigines suffer from gout. According to Bellwood⁴, the ancestral homeland of Austronesians was the agricultural heartland of Southeast Asia, from where they first radiated to Formosa (Taiwan; 4000 BC), then western Polynesia (1200 BC), central Polynesia (200 BC), and New Zealand (800 AD). Studies have shown male Micronesians to have prevalence of hyperuricemia ranging from 21 to 64%^5-7, and in male Polynesians the figure ranges from 24.3 to 49%^8,9. This is higher than in male Caucasians (6 to 23%)^10-12, suggesting a direct association with gout.

To our knowledge, data on the prevalence of hyperuricemia among adolescent children are limited, especially for tribes with a high prevalence of gout. Asymptomatic hyperuricemia may be present more than 20 years before disease onset¹³, with familial aggregation^14-16. We compared the prevalence of hyperuricemia among adolescent children in tribes of Taiwan aborigines with high and low gout prevalence.

MATERIALS AND METHODS

With parental consent 940 adolescent children were enrolled in this study from 1994 to 1998. The aborigines were from the Atayal, Paiwan, Bunun, Tao, Tsou, Puyumar, and Pancah tribes, while the non-aborigines were Fukien-Taiwanese and Hakka-Taiwanese from 2 villages located near the aboriginal areas who served as controls. The adolescent children were selected from the community junior high school in their areas of residence; roughly one-third of the students were included. All the participants were aged between 12 and 15 years and free of gout. Each completed a questionnaire to give data on sex, age, parents' tribal background, and weight and height measurements. Sera were extracted from blood samples and stored at -70°C before analysis for cholesterol, triglycerides, creatinine, and uric acid with an autoanalyzer (Biotechnica; Hitachi, Tokyo).

Definition of variables. The tribal origin of both parents was assessed to classify the child's tribe; for example, an Atayal child was defined as one with both parents of Atayal lineage; 176 children were excluded as both parents did not originate from the same tribe. The children were further divided into 2 groups according to the adult gout prevalence in their tribes determined in previous studies, one from tribes with high gout prevalence (HG) and the other from tribes of low gout prevalence (LG). The Atayal, Bunun, Paiwan, and Tsou tribes are classified as HG, with incidences > 5%^1,2, whereas the Pancah, Puyumar, and Tao tribes had LG incidences < 0.3%, and the non-aboriginal adult gout prevalence was estimated to be 0.3%².

Hyperuricemia was defined as serum uric acid (SUA) concentration > 7 mg/dl for boys and > 6 mg/dl for girls¹⁷. Subgroups were generated for serum cholesterol, creatinine, and triglycerides, with cutoff concentrations of 200, 1.0, and 170 mg/dl, respectively. Since the body mass index (BMI, weight in kg/height in m²) was not distributed homogeneously between the sexes, girls having a higher mean and greater variance than the boys (girls 20.2 ± 3.5 kg/m²; boys 19.4 ± 2.9 kg/m²; p < 0.01), we defined obesity as a value greater than the 90th percentile of the distribution for the sex, i.e., if a boy's BMI was greater than 23.44 kg/m² or a girl's BMI was greater than 24.54 kg/m², they were considered obese.

Statistical analysis. The chi-square test was used to assess differences between 2 category variables, such as sex and hyperuricemia status. Analysis of variance (ANOVA) was employed for means, such as SUA for different tribal groups. The simple regression method was employed to predict the uric acid concentration from age. Odds ratios (OR) with 95% confidence intervals (95% CI) were calculated to estimate the risks of having hyperuricemia in relation to the associated factors identified by questionnaire or biochemical markers. Those showing significant OR were subsequently examined in a multiple logistic regression model to obtain adjusted OR and the adjusted 95% CI were calculated by the Mantel-Haenszel method. Where the probability was less than 0.05 or the 95% CI did not include unity, the difference was considered statistically significant.

RESULTS

Of the 940 children who participated, 778 were aborigines and 162 were non-aborigines aged between 12 and 15 years. Of the 476 boys the mean age was 13.47 years (± 1.04) and showed no significant difference from 464 girls (13.43 ± 0.97 yrs; p = 0.464). The 778 aborigines were divided into 2 groups according to their tribal adult gout prevalence: 305 children from Atayal, Tsou, Bunun, and Paiwan tribes, and 473 children from the Pancah, Puyumar, and Tao tribes. There was no significant difference in sex frequency distribution among aborigines and non-aborigines (Table 1; p = 0.58).

Table 1. Subject distribution according to tribal background and sex. There was no significant variation in the sex ratio (p = 0.58).

The boys' mean uric acid concentration was 7.21 mg/dl (± 1.92), 1.53 mg/dl higher than that for the girls (5.68 ± 1.61 mg/dl; p < 0.01) for all the participants, and in the different tribes, boys' values were always higher than girls' (Table 2; p < 0.01). The HG boys had significantly higher mean uric acid concentration than non-aborigines or LG groups (p < 0.01). Among the girls, the HG group and non-aborigines also had higher mean uric acid concentrations than the LG group (Table 2; p < 0.05).

Table 2. Uric acid concentrations by tribal background and sex.

The highest prevalence of hyperuricemia was in the Tsou tribe (12/17, 70.6%; Table 3) for both sexes, followed by the Atayal (74/107, 69.2%) and Bunun (63/105, 60.0%). The Tsou boys and Atayal girls also had the highest hyperuricemia rates among male and female groups, respectively (Table 3). For all participants, the HG hyperuricemia rate was significantly greater than for the LG aborigines (Table 3; 176/305 vs 161/473; p < 0.01) or for non-aborigines (78/162; p < 0.05), and there was a significant difference between the non-aborigines and LG groups (p < 0.01). However, for each sex, the HG group always had a higher hyperuricemia rate than the LG (p < 0.01).

Table 3. Hyperuricemia distribution by tribal background and sex. Percentages are given in parentheses.

The possible related factors for hyperuricemia were tribe, sex, age, serum cholesterol, triglycerides and creatinine, and obesity. Using the Mantel-Haenszel method to compute OR and 95% CI of factors related to hyperuricemia in the preliminary analysis, we found that all of the factors except age and triglycerides significantly affected hyperuricemia (Table 4). After adjustment by a logistic regression method, the cholesterol factor was excluded. This shows that tribe, sex, obesity, and creatinine significantly contributed to hyperuricemia; e.g., those obese boys with parents from HG tribes and with higher concentrations of serum creatinine were significantly related to hyperuricemia.

Table 4. Risk factors for hyperuricemia among Taiwan aboriginal adolescents.

The means of SUA concentrations at different ages for each sex (Figure 1) show a gradual increase with age in SUA for LG aboriginal boys, from the mean of 5.70 mg/dl at age 12 to 7.67 mg/dl at age 15 (p < 0.01). However, no age association was found in HG aborigine and non-aborigine boys (Figure 1A). HG aboriginal girls were found to have a gradual decline in SUA with age, from 6.84 mg/dl at age 12 to 5.20 mg/dl at age 15 (p < 0.01; Figure 1B). However, girls of the other tribes showed no association with age.

DISCUSSION

A number of studies have shown the prevalence of gout in Taiwan aborigines was very high in adults of the Atayal, Bunun, Tsou, and Paiwan tribes aged over 40 years^1,2, but low in the Pancah, Puyumar, and Tao tribes. Only serum uric acid concentrations of children whose parents originated from the same tribe were studied. In this study, we did not classify the children according to their parents' gout status but by the tribal adults' gout prevalence. Since encroachment by non-aborigines on aboriginal villages had been prohibited by law for many years, inbreeding had presumably been proceeding until the last 10 years. Thus, gout prevalence of the participant's parents can be considered essentially the same as that of the tribe.

Like the incidence of gout in adult aboriginals in Taiwan, the prevalence of hyperuricemia in aboriginal children was not uniformly distributed in tribes; the prevalence of adult gout and adolescent hyperuricemia were correlated. This implies that the adult gout disease appears after a long preclinical period commencing in childhood, and that genetic and/or environmental components contribute to the differences between LG and HG tribes. However, the tribes we classified with a higher prevalence of adult gout (HG) live in the highland areas that always had a low temperature, while the LG tribes live in seaside areas. Temperature differences caused by altitude may therefore affect the prevalence of hyperuricemia and gout. The other inland and coastal difference between HG and LG tribes was dietary, i.e., high fish diet on the coast.

Hyperuricemia is the major direct risk factor causing gout^18,19; other associated factors that have been documented are age, sex, diuretic usage, lead exposure, alcohol abuse, and obesity^20,21. Seegmiller²² concluded that the age of puberty in males could influence hyperuricemia and gout. However, the boys showed a higher hyperuricemia rate than girls. Age did not show a sufficient association with the hyperuricemia rate in this study as the age period studied was limited. All the participants aged between 12 and 15 years were in puberty, and a non-puberty group was not included for comparison. In addition, obesity and serum creatinine were also found to have an influence in this study, consistent with the results of Jackson, et al²³ for Western Samoans. In our study serum creatinine contributed the highest risk to hyperuricemia. The serum creatinine level is the most widely used measure not only of renal function, but also of the production, intake, and metabolism of creatinine. Reimold showed that serum creatinine level increased directly with the progression of renal failure²⁴. However, the renal function decline among these participants may be doubtful, and a further followup study of measurement on renal function should be performed.

Two enzymes involved in uric acid metabolism are documented to be important, hypoxanthine-guanine phosphoribosyltransferase (HPRT) and phosphoribosyl pyrophosphate synthetase (PRPS). HPRT^25,26 and PRPS^27-29 are constitutionally expressed enzymes that function in the purine salvage pathway. Partial deficiency of HPRT activity or upregulation of PRPS causes accumulation of uric acid, so that the subject becomes hyperuricemic. Associations between HPRT mutants and gout have been well documented^30-32, and the authors also found a new HPRT mutation in a family predisposed to gout in the Tsou tribe of Taiwan aborigines (G152A)³³. This and previous findings² indicate that genetic effects are likely to play a role in hyperuricemia or gout in Taiwan aborigines.

Although Bellwood⁴ considered that Taiwanese aborigines originated from the proto-Polynesian expansion by linguistic evidence, Melton, et al³⁴ also showed that their results from studying mitochondrial DNA polymorphisms (Polynesian motif) were consistent with Bellwood's result. Healey, et al³⁵ showed hyperuricemia was caused by the interaction of heredity and environment in Filipinos. Studies from Neel, et al³⁶ and Wyngaarden and Kelley³⁷ also considered gout to be a disease caused by multiple genetic disorders. We show the causality of hyperuricemia in adolescent Taiwan aborigines was favored by tribal effects that combined both environment and heredity. Bellwood⁴ and Melton, et al³⁴ reported that Taiwan aborigines and Filipinos are Austronesians, suggesting that they may share the same genetic risk of hyperuricemia.

The triglyceride level did not significantly influence hyperuricemia in this study. This contrasts with our previous findings² and the studies of Chou, et al³⁸. Both these studies were of adult Taiwan aborigines. Al-Arfaj³⁹ showed that uric acid levels did not correlate with triglycerides and cholesterol in female adults, but correlation was found for hyperuricemic males and non-hyperuricemic males. Jiao, et al⁴⁰ reported that cholesterol and triglycerides increased in primary gout, and suggested that alcohol intake and uric acid were the major predictors of serum triglycerides. In our previous study⁴¹, the prevalence of alcohol abuse and alcohol dependence in Taiwan aborigines was roughly 10 times the rate in Fukien Taiwanese. The correlation between hypertriglyceridemia and hyperuricemia in Taiwan adult aborigines probably relates with alcohol consumption. However, most of the adolescent Taiwan aborigines in this study were considered free of alcohol consumption. The influence of triglycerides on hyperuricemia may be due to alcohol consumption following adolescence.

The age range covered in this study was so narrow there was no overall age effect. Analysis with reference to sex and tribe (Figure 1) revealed LG aboriginal boys had an increase in uric acid concentration with age (p < 0.01). In contrast, aboriginal HG girls demonstrated a decline as they became older (p < 0.01). Since the prevalence of hyperuricemia in adults is influenced by age as an independent factor⁴², further work clearly needs to be performed to explain the apparent anomaly.

In conclusion, those at high risk of developing gout demonstrate a higher rate of hyperuricemia in adolescence; this finding may be of potential benefit to preventive efforts. This observation of high prevalence of uric acid level in Taiwan aboriginal adolescents suggests that an age adjustment in puberty is needed for a segregation analysis in family studies⁴³.

From the Graduate Institute of Medicine and School of Medicine, Kaohsiung Medical University, Kaohsiung, Taiwan.

Supported by The Department of Health and National Health Research Institute, ROC (Grant DOH88-HR-803).

Y.C. Ko, MD, PhD; S.J. Chang, PhD, Graduate Institute of Medicine, School of Medicine; T.N. Wang, PhD, School of Public Health; L.Y. Tsai, PhD, School of Technology for Medical Sciences; F.T. Chang, MSc, School of Nursing, Kaohsiung Medical University.

Address reprint requests to Dr. S.J. Chang, School of Medicine, Kaohsiung Medical University. No 100, Shih-Chuan 1st Road, Kaohsiung, Taiwan. E-mail: changsj@kmu.edu.tw

Submitted June 26, 2000; revision accepted October 29, 2001.

REFERENCES

Search PubMed for:

1. Lai SC, Chu YJ, Kuo CF, Che CL. Hyperuricemia in Ilan County. Epidemiol Bull ROC 1991;7:101-8.

2. Chang SJ, Ko YC, Wang TN, Chang FT, Cinkotai FF, Chen CJ. High prevalence of gout and related risk factors in Taiwan's Aborigines. J Rheumatol 1997;24:1364-9. [MEDLINE]

3. Chang FT, Chang SJ, Wu YY, Wang TN, Ko YC. Body mass index and hyperuricemia differences between aboriginal and non-aboriginal children in Taiwan. Kaohsiung J Med Sci 1995; 11:315-21. [MEDLINE]

4. Bellwood P. The Austronesian dispersal and the origin of languages. Sci Am 1991;265:70-5.

5. Burch TA, O'Brien WM, Need R, Kurland LT. Hyperuricemia and gout in the Marianas Islands. Ann Rheum Dis 1966;25:114-6.

6. Reed D, Labarthe D, Stallones R. Epidemiologic studies of serum uric acid levels among Micronesians. Arthritis Rheum 1972; 15:381-90.

7. Zimmet PZ, Whitehouse S, Jackson L, Thomas K. High prevalence of hyperuricemia and gout in an urbanized Micronesian population. BMJ 1978;1:1237-9.

8. Prior IA, Rose BS, Harvey HP, Davidson F. Hyperuricemia, gout and diabetic abnormality in Polynesian people. Lancet 1966; 1:333-8.

9. Darmawan J, Valkenburg HA, Muirden KD, Wigley RD. The epidemiology of gout and hyperuricemia in rural population of Java. J Rheumatol 1992;19:1595-9. [MEDLINE]

10. Jeremy R, Towson J. Serum urate levels and gout in Australian males. Med J Aust 1971;1:1116-8.

11. Garrick R, Ewan CE, Bauer GE, Neale FC. Serum uric acid in normal and hypertensive Australian subjects. Aust N Z J Med 1972;4;351-6.

12. Prior IAM, Rose BS. Uric acid, gout and public health in South Pacific. N Z Med J 1966;65:295-300.

13. Kelley WN, Wortmann RL. Gout and hyperuricemia. Textbook of rheumatology. Philadelphia: WB Saunders; 1997:1314.

14. Rice T, Vogler GP, Perry TS, Laskarzewski PM, Province MA, Rao DC. Heterogeneity in the familial aggregation of fasting serum uric acid level in five North American populations: the lipid research clinics family study. Am J Med Genet 1990;36:219-25. [MEDLINE]

15. French JG, Dodge HJ, Kjelsberg AO, Mikkelsen WM, Schull WJ. A study of familial aggregation of serum uric acid levels in the population of Tecumseh, Michigan, 1959-1960. Am J Epidemiol 1967;86:214-24.

16. Friedlander Y, Kark JD, Stein Y. Family resemblance for serum uric acid in a Jerusalem sample of families. Hum Genet 1988;79:58-63. [MEDLINE]

17. Nicholson JF, Pesce MA. Laboratory testing and reference values in infants and children. Nelson textbook of pediatrics. 15th ed. Philadelphia: WB Saunders Co., 1996:2056.

18. Hall AP, Barry PE, Dawber TR, McNamara PM. Epidemiology of gout and hyperuricemia. Am J Med 1967;42:27-37.

19. Roubenoff R, Klag MJ, Mead LA, Liang KY, Seidler AJ, Hochberg MC. Incidence and risk factors for gout in white men. JAMA 1991;266:3004-7. [MEDLINE]

20. Campion EW, Glynn RJ, Delabry LO. Asymptomatic hyperuricemia: risks and consequences in the Normative Aging Study. Am J Med 1987;82:421-6. [MEDLINE]

21. Roubenoff R. Gout and hyperuricemia. Rheum Dis Clin North Am 1990;16:539-50. [MEDLINE]

22. Seegmiller JE. Metabolism of purines and genetic defects leading to hyperuricemia and gout. In: Smyth CJ, Holers VM, editors. Gout, hyperuricemia, and other crystal-associated arthropathies. New York: Marcel Dekker Inc.; 1999:130.

23. Jackson L, Taylor R, Faaiuso S, Ainuu SP, Whitehouse S, Zimmet P. Hyperuricaemia and gout in western Samoans. J Chron Dis 1981;34:65-75. [MEDLINE]

24. Reimold EW. Chronic progressive renal failure. Rate of progression monitored by change of serum creatinine concentration. Am J Dis Child 1981;135:1039-43. [MEDLINE]

25. Gordon RB, Sculley DG, Dawson PA, Beacham IR, Emmerson BT. Identification of a single nucleotide substitution in the coding sequence of in vitro amplified cDNA from a patient with partial HPRT deficiency (HPRT_Brisbane). J Inher Metab Dis 1990; 13:692-700. [MEDLINE]

26. Tohyama J, Nanba E, Ohno K. Hypoxanthine-guanine phosphoribosyl-transferase deficiency: identification of point mutations in Japanese patients with Lesch-Nyhan syndrome and hereditary gout and their permanent expression in an HPRT- deficient mouse cell line. Hum Genet 1994;93:175-81. [MEDLINE]

27. Becker MA, Meyer LJ, Seemiller JE. Gout with purine overproduction due to increased phosphoribosylpyrophosphate synthetase activity. Am J Med 1973;55:232-42.

28. Zoref E, Vries AD, Sperling O. Mutant feedback-resistant phosphoribosyl-pyrophosphate synthetase associated with purine overproduction and gout. J Clin Invest 1975;56:1093-9.

29. Becker MA, Kostel PJ, Meyer LJ, Seegmiller JE. Human phosphoribosylpyrophosphate synthetase: increased enzyme specific activity in a family with gout and excessive purine synthesis. Proc Natl Acad Sci USA 1973;70:2749-52.

30. Cariello NF, Scott JK, Kat AG, Thilly WG, Keohavong P. Resolution of a missense mutant in human genomic DNA by denaturing gradient gel electrophoresis and direct sequencing using in vitro DNA amplification: HPRT_Munich. Am J Hum Genet 1988;42:726-34. [MEDLINE]

31. Davidson BL, Pashmforoush M, Kelley WN, Palella TD. Human hypoxanthine-guanine phosphoribosyltransferase deficiency. The molecular defect in a patient with gout (HPRT_Ashville). J Biol Chem 1989;264:520-5. [MEDLINE]

32. Fujimori S, Hidaka Y, Davidson BL, Palella TD, Kelley WN. Identification of a single nucleotide change in a mutant gene for hypoxanthine-guanine phosphoribosyltransferase (HPRT_{Ann Arbor}). Hum Genet 1988;79:39-43. [MEDLINE]

33. Chang SJ, Chang JG, Chen CJ, et al. Identification of a new single nucleotide substitution on the hypoxanthine-guanine phosphoribosyltransferase gene (HPRT_Tsou) from a Taiwanese aboriginal family with severe gout. J Rheumatol 1999;26:1802-7. [MEDLINE]

34. Melton T, Peterson R, Redd AJ, et al. Polynesian genetic affinities with southeast Asian populations as identified by mtDNA analysis. Am J Hum Genet 1995;57:403-14. [MEDLINE]

35. Healey LA, Skeith MD, Decker JL, Bayani-Sioson PS. Hyperuricemia in Filipinos: interaction of heredity and environment. Am J Hum Genet 1967;19:81-5.

36. Neel JV, Rakic MT, Davidson RT, Valkenburg HA, Mikkelson WM. Studies on hyperuricemia. II. A reconsideration of the distribution of serum uric acid values in the families of Smyth, Cotterman, and Freyburg. Am J Hum Genet 1965;17:14-22.

37. Wyngaarden JB, Kelley WN. Gout. In: Stanbury, JB, Wyngaarden JB, Fredrickson DS, editors. The metabolic basis of inherited disease. 3rd ed. New York: McGraw-Hill; 1972:889-968.

38. Chou CT, Chao PM. Lipid abnormalities in Taiwan aborigines with gout. Metabolism 1999;48:131-3. [MEDLINE]

39. Al-Arfaj AS. Hyperuricemia in Saudi Arabia. Rheumatol Int 2001;20:61-4. [MEDLINE]

40. Jiao S, Kameda K, Matsuzawa Y, Tarui S. Hyperlipoproteinaemia in primary gout: hyperlipoproteinaemic phenotype and influence of alcohol intake and obesity in Japan. Ann Rheum Dis 1986; 45:308-13. [MEDLINE]

41. Liu BH, Hsieh SF, Chang SJ, Ko YC. Prevalence of smoking, drinking and betel quid chewing and related factors among aborigines in Wufeng District. Kaohsiung J Med Sci 1994; 10:405-11. [MEDLINE]

42. Brennan PJ, Simpson, Blacket RB, McGilchrist CA. The effects of body weight on serum cholesterol, serum triglycerides, serum urate and systolic blood pressure. Aust NZ J Med 1980;10:15-20. [MEDLINE]

43. Wilk JB, Djousse L, Borecki I, et al. Segregation analysis of serum uric acid in the NHLBI Family Heart Study. Hum Genet 2000;106:355-9. [MEDLINE]