Hyperuricemia and Gout in Thyroid Endocrine Disorders

Hyperuricemia and gout in thyroid endocrine disorders

N. Giordano, C. Santacroce, G. Mattii, S. Geraci, A. Amendola, C. Gennari

Institute of Internal Medicine, University of Siena, Siena, Italy

Abstract Objective A significant correlation between thyroid function and purine nucleotide metabolism has been established in hypothyroidism. On the contrary, the relationship between hyperthyroidism and purine metabolism is more controversial. The present study evaluates the prevalence of hyperuricemia and gout in patients affected by primary hypothyroidism and hyperthyroidism. Methods We studied 28 patients with primary hypothyroidism and 18 patients with primary hyperthyroidism, all hospitalized because of endocrine dysfunction. All underwent a series of clinical, biochemical and instrumental evaluations; in particular, thyroid-stimulatin hormone (TSH), free thyroxine (fT4), blood urea, serum creatinine, creatinine clearance, serum and urinary uric acid levels were measured. Results In comparison to the prevalence reported in the general population, a significant increase of both hyperuricemia and gout was found in the hypothyroid patients, and of hyperuricemia in the hyperthyroid patients. In hyperthyroidism the hyperuricemia is due to the increased urate production, while in hypothyroidism the hyperuricemia is secondary to a decreased renal plasma flow and impaired glomerular filtration. Conclusions Our findings confirm the data in the literature concerning the high prevalence of hyperuricemia and gout in hypothyroidism. It shows that hyperthyroidism can cause a significant increase in serum uric acid, as well, although lower than the hyperuricemia due to thyroid hormone deficiency. Key words Hyperuricemia, gout, hypothyroidism, hyperthyroidism, pathogenesis.

Clinical and Experimental Rheumatology 2001; 19: 661-665. Hyperuricemia and gout in thyroid endocrine disorders / N. Giordano et al.

Nicola Giordano, MD, Assistant Introduction ed secretion of thy roid horm o n e, i n Professor; Clorinda Santacroce, MD, The association between hypothyroid- particular the manifestations of cardio- Assistant Professor; Giancarlo Mattii, ism and hyperuricemia was first sug- vascular, central nervous and cutane- MD, Assistant Professor; Simone Geraci, gested in 1955 by Kuzell and colleag- ous system invo l vement. More ove r, MD, Assistant Professor; Alessandra ues (1), who examined 520 pat i e n t s they satisfied the biochemical criteria Amendola, MD, Assistant Professor; suffering from gout and found hypo- for the diagnosis of primary hypothy- Carlo Gennari, MD, Full Professor. thyroidism in 20% of the males and in roidism (21), in particular, low serum Please address correspondence and 30% of the females. Subsequent stud- l evels of fT4 and increase titres of reprint requests to: Dr. Nicola Giordano, ies (2-10) confirmed this association, TSH. Patients with secondary (pitui- MD, Institute of Internal Medicine, University of Siena, Viale Bracci no. 1, furthermore suggesting that hypothy- t a ry and hypothalamic) hy p o t hy ro i d- 53100 Siena, Italy. roid hyperuricemia could be due to a ism, transient hypothyroidism,or a his- E-mail: [email protected] decrease in both the renal plasma flow t o ry of renal disease we re ex cl u d e d Received on December 12, 2000; and urate excretion (11-14). from the case study. In the 12 months accepted in revised form on May 22, 2001. On the contra ry, the association be- preceding their admission, 7 (25.5%) t ween hy p e rt hy roidism and hy p e r- of the 28 patients (5 women and 2 © Copyright CLINICAL AND EXPERIMENTAL RHEUMATOLOGY 2001. uricemia has always been more contro- men) manifested relapsing monoarthri- versial. In 1989 Ford et al. (15),in con- t i s , wh i ch was diagnosed as acute t rast with previous rep o rts (16, 1 7 ) , gouty attacks by their doctors. These 7 demonstrated that hyperthyroidism can p atients we re tre ated with a non- cause hy p e ru ricemia through the in- s t e roidal antiinfl a m m at o ry, but they c rease of purine nu cleotide turn ove r had stopped any drug that might have and the decrease of renal urate excre- interfered with uric acid metabolism at tion. Other studies (18,19) confirmed least 2 months before their admission their findings and also suggested that to hospital. They were all on a normal hyperthyroid hyperuricemia might be purine diet. c o rrected by antithy roid drugs such The fo l l owing lab o rat o ry tests we re methimazole. However, in 1999 Raber c a rried out: cy t o m e t ric full bl o o d et al. (20) did not find any significant count, total cholesterol, triglycerides, s t atistical diffe rence in serum urat e p l a s m atic electro ly t e s , s e rum cre at i- c o n c e n t ration between hy p e rt hy ro i d nine and blood urea, creatinine clear- patients and euthyroid ones. ance, serum uric acid, 24-hour urinary Here we report our experience regard- uric acid, thyroid-stimulation hormone ing the possible relationship between (TSH), and free thyroxine (fT4) (TSH uric acid metabolism and thyroid dis- and fT3 were dosed by an automated orders, presenting data obtained from chemiluminescence immu n o a s s ay ) . patients suffering from either primary The fo l l owing eva l u ations we re also hypothyroidism or hyperthyroidism. performed: blood pressure, electrocar- di o gram (ECG), ech o c a rd i o gram , che s t Patients and methods X - ray, renal and uri n a ry tract sono- Th i rty-eight patients affected by pri- grams and thyroid sonogram. The bio- mary hypothyroidism, 20 (71.4%) wo- chemical para m e t e rs (with the ex- men and 8 (28.5%) men, aged between c eption of thy roid hormones) we re 34 and 81 years (average age, 65 ± 9.5 m e a s u red seve ral times during each years), were examined. Eighteen pri- patient’s hospitalization and 2 months mary hyperthyroid patients were stud- after the beginning of the substitute ied as well, 14 (77.7%) women and 4 hormone therapy with L-thyroxin (L- (22.2%) men, aged between 28 and 79 T4) (doses ra n ging between 75 and years (average age, 55 ± 8.3 years). All 100 g/day), while fT4 and TSH were the patients were admitted to the Insti- measured at 0, 30 and 60 days from the tute of Internal Medicine of the Uni- beginning of L-thyroxin therapy. Thy- versity of Siena during the period 1998 roid sonogram, chest X-ray, echocar- Ð 2000 due to signs and symptoms of diogram and renal sonogram were car- thyroid dysfunction. ried out only once, at the moment of admission. ECG and blood pre s s u re Hypothyroid patients (group A) eva l u ations we re ve ri fied fre q u e n t ly, The hypothyroid patients presented the and in particular at 30 and 60 day s classic signs and symptoms of decreas- after beginning the hormone substitu-

662 Hyperuricemia and gout in thyroid endocrine disorders / N. Giordano et al. tion therapy, in order to monitor the ef- of hospitalization. Both patients pre- normalization of TSH, fT4 and serum ficacy and tolerability of the treatment. sented increased values of uric acid in and urinary uric acid levels, as well as the serum and of urate crystals in the of creatinine and creatinine clearance Hyperthyroid patients (group B) joint fluid. In these 2 patients a diagno- values. A statistically significant corre- The 18 hyperthyroid patients presented sis of gout was made and they were lation was found between serum TSH the classic signs and symptoms of treated with colchicine (average dose: and fT4 on the one hand, and creati- increased secretion of thyroid hormone 3 mg/day, for 4 days). They had pre- nine clearance on the other (p < 0.05, (Basedow’s disease), in particular dif- sented the same articular disease at Pearson’s correlation coefficient) in the fuse goiter and manifestations of car- home before their admission. Table I group of hypothyroid patients. d i ovascular and nervous system in- shows the demographic and laboratory Table III shows the demographic and vo l vement. More ove r, t h ey sat i s fi e d data of the hypothyroid patients. laboratory data on the 20 hyperthyroid the biochemical criteria for the diagno- Table II describes the biochemical par- patients. The results first of all demon- sis of primitive hyperthyroidism (21): ameters in the 7 hyperuricemic hypo- strate that 5 (27.7%) of the hyperthy- elevated serum levels of fT4 and low thyroid patients under basal conditions roid patients presented stable hy p e r- l evels of TSH. Patients with other and 2 months after L-thyroxin treat- uricemia (demonstrated by at least 3 causes of thyrotoxicosis and patients ment. Among the hypothyroid patients, co n s e c u t i ve tests); these 5 patients wer e with a history of renal insuffi c i e n cy 9 (32.1%) presented stable hyperuri- a ffected by asymptomatic hy p e ru ri- we re ex cluded from this case study. cemia, a decrease in uric acid excretion cemia, and 3 (60%) of them were fe- None of the patients enrolled in the (in 3 consecutive tests), increased crea- males. Of the 13 (72.2%) normourice- study presented signs or symptoms of tinine, and decreased creatinine clear- mic hyperthyroid patients, 11 (84.6%) gout, either previously or at the mo- ance (Table I). Six (66.6%) of these 9 were females. Between the two groups ment of hospitalization. All were on a patients were female, and 3 (33.3%) of hyperthyroid patients, no significant normal purine diet. we re male. A stat i s t i c a l ly signifi c a n t difference was found as regards serum The same battery of tests and examina- d i ffe rence was found between the 9 creatinine or creatinine clearance val- tions described above were carried out hy p e ru ricemic hy p o t hy roid pat i e n t s ues. Table IV shows the values for the in this group of patients. Thyroid hor- and the 19 normouricemic hypothyroid biochemical parameters in the 5 hyper- mones were measured on day 0 and patients as regards TSH, fT4, serum uricemic hyperthyroid patients before then two months after beginning the and urinary uric acid, serum creatinine, and at the end of metronidazole treat- h o rmone suppression therapy with and cre atinine cl e a rance (Table I). ment. The drug resulted in the normal- methimazole (attack dose: 15-20 mg/ Table II shows that the administration ization of fT4 and TSH values, and of day; maintenance dose: 5-10 mg/day), of L-thy roxin in the 9 hy p o t hy ro i d - the serum and urinary uric acid levels using an automated ch e m i l u m i n e s- hyperuricemic patients resulted in the (Table IV). Lastly, in the hyperthyroid cence immunoassay.

Statistical analysis Table I. Personal and biochemical data in patients (n = 28) with hypothyroidism, compris- The statistical evaluation of the results ing 9 patients with increased serum uric acid levels and 19 patients with normal serum uric obtained was carried out using Stu- acid levels (expressed as means ± SD). dent’s t-test to analyse the significance Hyperuricemic Normouricemic P of the difference between the averages (n = 9) (n = 19) of data paired under basal conditions, and befo re and after substitute hor- Age (yrs.) 60.0 ± 3.4 61 ± 3.6 NS mone therapy with L-thyroxin (hypo- Sex (no. of females) 6 (65.5%) 13 (68.4 %) NS thyroid patients, group A) or suppres- TSH sive therapy with metronidazole (hy- (nv 0.40-5.50 mUI/ml) 10.0 ± 1.2 7.5 ± 1.5 < 0.05 p e rt hy roid pat i e n t s , group B), a n d FT4 Pe a rs o n ’s corre l ation coefficient to (nv 7.5-15 pg/ml) 3.9 ± 1.2 6.4 ± 0.9 < 0.05 determine the possible correlations be- Blood urea tween the numerous variables on one (nv 20-45 mg/dl) 53 ± 1.9 51 ± 2.3 NS side and TSH and fT4 on the other.The Serum creatinine results were expressed as averages ± (nv 0.5-1.2 mg/dl) 1.9 ± 0.2 0.7 ± 0.2 < 0.05 standard deviation (SD). Statistical sig- Creatinine clearance nificance was assigned as p < 0.05. (nv 70-120 cc/min) 59 ± 4 94 ± 4.9 < 0.05 Serum uric acid Results (nv 3-7 mg/dl) 8.5 ± 1.2 4.8 ± 1.8 < 0.05 Among the 28 hypothyroid patients, 2 Urinary uric acid (7.1%) (one male and one female) had (nv 300-800 mg/ 24h) 280 ± 17.4 616 ± 21.1 < 0.001 monolateral gonarthritis at the moment

663 Hyperuricemia and gout in thyroid endocrine disorders / N. Giordano et al. patients no statistically significant cor- Table II. Laboratory data in patients (n = 9) with hypothyroidism and hyperuricemia, relation was found between TSH and before and after treatment (2 months) with L-thyroxin (expressed as means ± SD). fT4 on the one hand, and serum creati- Parameter Pre-treatment Post-treatment p nine and creatinine clearance on the other. TSH (nv 0.40 - 5.50 mUI/ml) 10 ± 1.2 3.9 ± 1.4 < 0.001 Discussion FT4 Our study evaluated the possible inter- (nv 7.5 - 15 pg/ml) 3.9 ± 1.2 10.6 ± 2.2 < 0.001 relationship between purine nucleotide Blood urea metabolism and thyroid endocrine dis- (nv 20 - 45 mg/dl) 53 ± 1.9 32 ± 2 < 0.05 orders, in particular primary hypo- and Serum creatinine hy p e rt hy ro i d i s m , by examining the (nv 0.5 - 1.2 mg/dl) 1.9 ± 0.2 0.8 ± 0.2 < 0.05 d ata in the literat u re and compari n g Creatinine clearance this with the data observed in a case (nv 70 - 120 cc/min) 59 ± 4 101 ± 3 < 0.001 study. In the literature, while the corre- Serum uric acid l ation between hy p o t hy roidism and (nv 3 - 7 mg/dl) 8.5 ± 1.2 5.1 ± 1 < 0.05 hyperuricemia is well established (1- Urinary uric acid 14), the connection between hyperthy- (nv 300-800\mg/ 24h) 280 ± 17.4 560 ± 15.1 < 0.001 roidism and hy p e ru ricemia still ap- pears to be under debate (15-20). Table III. Personal and biochemical data in patients (n = 18) with hyperthyroidism, com- The present study, even if based on a prising 5 patients with increased serum uric acid levels and 13 patients with normal serum limited number of cases, s h owed a uric acid levels (expressed as means ± SD). high prevalence of hy p e ru ricemia in hypothyroid patients, 33.3%, which is Hyperuricemic Normouricemic s u b s t a n t i a l ly in accordance with the patients (n = 5) patients (n =13) p values rep o rted by others (1, 3 ,7 ,8 ) . Age (yrs.) 56 ± 3.3 55 ± 7.1 NS This finding is part i c u l a rly re l eva n t when compared with the prevalence, Sex F 3 (60%) 11 (84.6% ) 0.05 ranging from 2 to 10%, in the general TSH p o p u l ation (22, 23). More ove r, o u r (nv 0.40 - 5.50 mUI/ml) 0.30 ± 0.1 0.25 ± 0.1 NS s t u dy showed that hy p e ru ricemia in hypothyroidism is associated with in- FT4 creased serum creatinine and decreas- (nv 7.5-15 pg/ml) 19.4 ± 1.6 20 ± 1.7 NS ed creatinine clearance.This fact sug- Blood urea (mean ± SD) gests that hypothyroid hyperuricemia (nv 20 - 45 mg/dl) 40 ± 0.9 41 ± 1.2 NS is secondary to a reduction in re n a l plasma flow and glomerular filtration, Serum creatinine (nv 0.5 - 1.2 mg/dl) 0.9 ± 0.2 0.8 ± 0.3 NS already well demonstrated in thyroid h o rmone defi c i e n cy syndromes (11- Creatinine clearance 14). (nv 70 - 120 cc/min) 85 ± 1.3 89 ± 2.6 NS The finding of a connection between Serum uric acid gender and hypothyroid hyperuricemia (nv 3 - 7 mg/dl) 7.9 ± 1.4 4.2 ± 1.2 < 0.05 seem to be important all the same. Our data and those in literature (3,4,7,8) Urinary uric acid indicate that hypothyroid hyperurice- (v.n. 280 - 800 mg/ 24 h) 855 ± 13.3 605 ± 16.2 < 0.05 mia shows the same prevalence in both sexes, while it is known that hyperuricemia and gout are strongly associ- Examining the data on the hyperthy- ar e not able to completely explain these ated with the male sex (22,23). The roid patients, the high prevalence of divergences: the selection criteria (out- same prevalence of hypothyroid hyper- hy p e ru ricemia ap p e a rs ev i d e n t : t h e p atients or patients admitted to hos- uricemia in both sexes can be explain- prevalence was 27.7% in our case stu- pital) and different laboratory methods ed as follows: hypothyroidism is more dy, significantly higher than the preva- used to define THS and fT3 values (en- frequent in females (21); hypothyroid lence of hyperuricemia in the general zy m e - l i n k ed immun o ab s o r ption assays , hyperuricemia is due to a reduction in p o p u l ation (22). Our findings agre e ra d i o i m mu n o a s s ay s ,e t c.) could explain renal plasma flow and glomerular fil- with those reported by some authors in part the different results. tration secondary to thyroid hormone (15,18,19) but disagree with those des- The increase in serum and urinary uric deficiency. cribed by others (20). At present, we acid levels in our hyperthyroid patients

664 Hyperuricemia and gout in thyroid endocrine disorders / N. Giordano et al.

Table IV. Laboratory data in patients (n = 5) with hyperthyroidism and hyperuricemia, 7. ERICKSON A R , E N Z E NAUER RJ, N O R D- before and after treatment (2 months) with methimazole (expressed as the mean ± SD). STROM DM, MERENICH JA:The prevalence of hypothyroidism in gout. Am J Med 1994; Parameter Pre-treatment Post-treatment p 97: 231-4. 8. M O N T E N E G RO J, G O N Z A L E S O, SA R AC H O TSH R et al.: Changes in renal function in primary (nv 0.40 - 5.50 mUI/ml) 0.30 ± 0.1 4.9 ± 1.6 < 0.001 hypothyroidism. Am J Kidney Dis 1996; 27: 195-8. FT4 9. MOORAKI A, BASTANI B: Reversible renal (nv 7.5-15 pg/ml) 19.4 ± 1.6 13.5 ± 2.4 < 0.05 insufficiency, hyperuricemia and gouty arthritis in a case of hypothyroidism. Clin Neph - Blood urea rol 1998; 49: 59-61. (nv 20-45 mg/dl) 40 ± 0.1 39 ± 1.4 NS 10. MAKINO Y, FUJII T, KURODAS et al.: Exac- e r b ation of renal fa i l u re due to hy p o t hy- Creatinine roidism in a patient with ischemic nephropa- (nv 0.5-1.2 mg/dl) 0.9 ± 0.2 0.8 ± 1.1 NS thy. Nephron 2000; 84: 267-9. 11. K ATZ A L , EMMANOUEL DS, L I N D H E I M E R Creatinine clearance M D: Thy roid hormone and the kidney. (nv 70-120 cc/min) 85 ± 1.3 93 ± 2.6 NS Nephron 1975; 15: 223-9. 12. L A D E N S O N P W: R e c ognition and manage- Serum uric acid ment of cardiovascular disease related to thy- (nv 3-7 mg/dl ) 7.9 ± 1.4 4.6 ± 1.3 < 0.05 roid dysfunction. Am J Med 1990; 88: 638- 41. Urinary uric acid 13. STEIGER MJ, WATSON A R , M O R G A N AG: (nv 300-800 mg/24 h) 855 ± 13.3 623 ± 16.6 < 0.05 Hypothyroidism and renal impairment. J R Soc Med 1991; 84: 688-9. 14. MCLANGHLIN KJ, M ACTIER R A: R e n a l impairment in hypothyroidism. Nephrol Dial seems to be due to accelerated purine cemia, as evidenced in our study. Transplant 1994; 9: 1521-2. nucleotide turnover.This hypothesis is Despite the small number of the pa- 15. F O R D H C , L I M W C , C H I S NA L L W N, shared by Sato and colleagues (19), but tients ex a m i n e d, the present study PEARCE JM: Renal function and electrolyte l evels in hy p e rt hy ro i d i s m : u ri n a ry pro t e i n not by Di Shiroba and colleagues (18). shows that hyperuricemia is a common excretion and the plasma concentrations of In fact, Di Shiroba et al. (18) sustain feature not only of hypothyroidism, but urea, creatinine, uric acid, hydrogen ion and that hyperthyroid hyperuricemia is sec- also of hyperthyroidism. Therefore, we electrolytes. Clin Endocrinol 1989; 30: 293- ondary to the direct action of thyroxin would emphasize the importance of the 301. 16. YO KOGOSHI Y, S A I TOS: A b n o rmal seru m on the kidney, consisting in a decrease routine evaluation of serum and uri- uric acid level in endocrine disorders. Nippon in uric acid tubular excretion, not cor- nary uric acid levels, both in patients Rinsho 1996; 54: 3360-3. rected by an increase of cre at i n i n e a ffected by hy p o t hy roidism and in 17. SMYTH CJ: Disorders associated with hyper- clearance.The significant increase in patients with hyperthyroidism. In this uricemia. Arthritis Rheum 1975; 18: 713-9. 18. SHIROTA T, SHINODA T , YAMADA T, AIZA- urinary uric acid levels in our hyper- way we will be able to correct the pos- WA T: Alteration of renal function in hyperthyroid hyperuricemic patients seems sibly altered purine nucleotide metabo- t hy ro i d i s m : i n c reased tubular secretion of to demonstrate that hyperthyroid hype- lism and to prevent the onset of gout, creatinine and decreased distal tubule deliv- ruricemia is linked to the increase in wh i ch can wo rsen thy roid endocri n e ery of chloride. Metabolism 1992; 41: 402- 405. purine nucleotide turnover, rather than disorders. 19. S ATO A , S H I ROTA T, S H I N O DA T et al.: to a decrease in uric acid renal excre- H y p e ru ricemia in patients with hy p e rt hy- tion. As was observed in the hypothy- References roidism due to Graves’ disease. Metabolism roid patients, in the hyperthyroid sub- 1. KU Z Z E L W C , S C H A F FARZICK RW, NAU- 1995; 44: 207-11. GLER WE et al.: Some observations on 520 20. RABER W, V U KOVICH T, VIERHAPPER H: jects as well hy p e ru ricemia did not gouty patients. J Chronic Dis 1955; 2: 64-8. Serum uric acid concentration and thyroid- s h ow a significant prevalence in the 2. KUHLBACK B: Creatinine and creatine me- s t i mu l at i n g - h o rmone (TSH): results of male sex. This finding can be explain- tabolism in thyreotoxicosis and hypothyroid- screening for hyperuricaemia in 2359 con- ed by the high prevalence of hyperthy- ism. Acta Med Scand 1957; 159: 1-70. secutive patients with various degrees of thy- 3. LEEPER RD, BENUA RS, BRENER JL, RAW- roid dysfunction. Wien Klin Wo ch e n s ch r roidism in females and the increase in S O N RW: H y p e ru ricemia in my xedema. J 1999; 111: 326-8. purine nucleotide turnover caused by Clin Endocrinol Metab 1960; 20: 1457-66. 21. L A Z A RU S J H , O BU O B I E K: Thy roid disor- thyroid hyperfunction. 4. BOYLE JA , GREIG W, DUNCAN AM et al.: ders: an update. Postgrad Med J 2000; 76: With regard to the prevalence of gout Serum uric acid values in various states of 529-36. thyroid function. Acta Rheum Scand 1996; 22. M I K K E L S E N W M , D O D G E H J,VA L K E N- in the two populations, we noted that 12: 204-9. BURG H:The distribution of serum uric acid no hyperuricemic patient had signs or 5. RYC K E WA E R T A , M A S S E C , JURMAND S H values in a population unselected as to gout symptoms of this metabolic disease. et al.: Goutte et myxoedème. Sem Hop Paris or hyperuricemia. Am J Med 1965; 39: 242- This fact could be due to the lower val- 1967; 43: 3059-62. 51. 6. BLAND JH, FRYMOYER JW: Rheumatic syn- 23. MEYERS OL,MONTEAGUDOFSE: A compar- ues of hy p e rt hy roid hy p e ru ri c e m i a dromes of myxedema. N Engl J Med 1970; ison of gout in man and women. S Afr Med J than those of hy p o t hy roid hy p e ru ri- 282: 1171-4. 1986; 70: 721-2.

665