and Thyroxine Content of Desiccated Tablets

Robert W. Rees-Jones and P. Reed Larsen

Triiodothyronine (Ts) and thyroxine (T,) variation in T, and T, content between the were measured by ~dioimmunoassay in four lots of each tablet strength for each Pronase hydrolysates of four lots each of product was 10% or less. These estimates l- and IT-grain tablets of desiccated thy- of T, and T, content are 15 to 2-fold roid (Thyroid, Armour) and thyroglobulin greater than those previously published. (Proloid, Warner-Chilcott). The method- This difference probably results from the ology used was verified by studies of more sophisticated methodology now tablets containing known quantities of available which does not require chro- T, and T,. One grain of desiccated thyroid matographic separation of T, and T, or contained 12 rt 1 and 44 rf: 3 rg (mean it iodometry. Using calculations based on SD) of T, and T, per tablet, respectively published estimates of T, and T, absorp- (T,/T, molar ratio, 4.3). A l-grain tablet tion and of the T,/T, potency ratio, it of thyroglobulin contained 16 f 2 and would appear that the T, content of 55 * 5 pg of T, and T,, respectively with desiccated thyroid and thyroglobulin pro- a T,/Ts ratio of 2.9. Two-grain tablets vide approximately 39% and !il%, re- generally contained twice the quantity of spectively, of the thyromimetic activity T, and T4 in the l-grain preparations. The of these two medications.

NITIAL OBSERVATIONS by Gull’ in 1873 of a disease similar to cretinism I but occurring in adults, subsequently referred to as myxedema, led to Murray’s successful attempt to ameliorate such symptoms with injections of extracts of sheep thyroid glands.2 This was closely followed by the successfui use of fresh animal thyroid glands (or extracts) orally in the treatment of myxedema. 3.4 Since that time, products derived from animal have been widely utilized in the successful treatment of . Despite the current availability of synthetic preparations of L-thyroxine (T4) and L- triiodothyronine (T,), these extracts continue to be widely used. Although pre- cise information is not available, some pharmaceutical industry estimates of the proportion of patients still treated with animal thyroid preparations as op- posed to synthetic T, and/or T, is as high as 50”;. Despite the widespread use of dessicated thyroid (Thyroid, Armour), the U.S. Pharmacopeia specifies only that this medication contain between 0.17”,, and 0.230_0 iodine by weight. No assurances of hormonal iodine content are required. While, in general, these preparations are therapeutically efficacious, occasional previous investigations have reported significant variations in potency despite an acceptable iodine content. 5 ’ In addition to these occasional

From the Thyroid Unit, Department of Medicine,Peter Bent Brigham Hospjtai and Harwrd Medi- caf School, Boston. Mass. Received for publication Februarv 2. 1977. Reprint requests should be addressed to P. Reed Larsen. M.D.. Thyroid Unit, Peter Bent Brigham Hospital, 721 Huntington Avenue, Boston, Mass. 02115. Supported b”v NIH Grant AM 18616. P. R. Larsen is an Investigator of the Howard Hughes Medical Institute. R. W. Rees-Jones was supported by the Laboratory Assistants Program fbr Medical Students of the Howard Hughes Medical Institute. ,g I977 bv Grtme & Stratton, Inc. ~026~~495.

Metabolism, Vol. 26, No. 11 (November), 1977 1213 1214 LARSEN problems, the only previous data available relative to the T4 and T, content of Thyroid gives results which appear to be low with respect to the quantities of pure required for adequate replacement. Since we have been unable to find more recent studies of this widely used medication, we have ap- plied currently available radioimmunoassay techniques to provide these data.

MATERIALS AND METHODS Preparations of Thyroid were obtained from seven Boston pharmacies. All seven carried only two preparations classified as desiccated thyroid (Thyroid, Armour) and thyroglobulin (Proloid, Warner-Chilcott). Four different lots of 1- and 2-grain tablets of each brand were evaluated. All samples from each lot were assayed for iodine content, which averaged 0.2% and was not different in the two preparations. Four tablets from each lot were dissolved and thyroglobulin was digested using methods slightly modified from that of Inoue and Taurog for thyroid tissue.’ The tablets were dissolved in 0.6 ml 0.1 I M NaCL-0.04 M Tris (pH 8.5) containing 0.05 M I-methyl,2-mercapto . Pronase, 0.13 mg/mg stated thyroid extract, was added. “‘I-T, or “‘I-T, was also present in the digestion solution to allow recovery calculations. Tubes were incubated anaerobically at 37°C for 20 hr and iodothyronines were then extracted using a 99/ 1 (volume/volume) solution of methanol/ NHdOH (2N) and counted for recovery, which was 86% * 4% for T, and 93% 5 3% for T, (mean f SD). The extracts were then diluted appropriately (several thousand-fold) so that T, and T, could be determined in duplicate at two dilutions by radioimmunoassays.‘O1” Four tablets each of a T, preparation (Synthroid, Flint Division, Travenol Laboratories) 0.1 mg and 0.2 mg, and of a combination T, + T, preparation (Thyrolar I. Armour, containing 50 rg T, + 12.5 pg T3) were also digested and assayed for iodothyronines in the same manner to quantitate unlabeled hormone recovery from tablets. Samples of purified bovine thyroglobulin (type I) and porcine thyroglobulin (type II) obtained from Sigma Chemical Corporation with stated iodine content of approximately 1% were also di- gested and assayed for T, and T, content in the same manner. Statistical analyses were per- formed by standard techniques.j2

RESU ITS The mean T3 and T, content of T, tablets and T,-Tj preparations with and without pronase areshown in Table 1. Recovery of the theoretical quantities of T4 in these tablets averaged 87% in the presence of pronase, modestly but sig-

Table 1. T, and T, Content of Thyroxine and T, + T, Combination Tablets After 20hr of Incubation With or Without Pronare

No Pranare PronasePresent lodothyroniner Recovery lodothyroniner Recovery @g/tablet) (“h) Wt/tabW (Oh)

Tablet T3 T4 T3 T4 T3 T4 T3 T4

1 thyroxine lOOKI 1.3 i 0.1* 96k 1 96 1.6 f 0.4 83+1 - 83t 200 PQ 4.2 z!z0.3 203+ 13 - 101 3.2 f 0.41 186zk 16 - 93

T3 + T4 13 l 0.5 55 f 2 107 110 12 zk 0.8 42 f 1 97 85t (50 fig T, + 12.5 w T3)

Mean f SD 102+ 7 87 + 5

*Meon f SD. tp < 0.001 for difference from recovery in the absence of pronase. $p i 0.025 for difference from recovery in the absence of pronase. TRIIODOTHYRONINE AND THYROXINE 1215

Table 2. T, and T, Content of I-Grain Desiccated Thyroid Preparations After Pronase Digestion (fig/Tablet, Mean f SD)

T3 *4 T4/T3 (Molar)

Desiccated thyroid Lot 1 13 i 1.0 65 i 4 4.3 2 12 It 1.2 61 zk 4 4.3 3 12 f 0.4 63 + 2 4.5 4 13+ 1.2 66 i 2 4.2 Mean f SD 12 f 1.1 64 + 3 4.3 f 0.1

Thyroglobulin Lot 1 18+ 1.6 52h 1 2.5 2 lb+ 2.6 58 f 9 3.0 3 17 * 0.9 54 f 4 2.7 4 15 * 0.9 57* 2 3.2 Mean f SD lb& 1.8 55 * 5 2.9 zt 0.3 nificantly (p < 0.05) reduced from 102% in the absence of pronase. In the T4-T, tablet, T, recovery was not significantly affected by Pronase and averaged 102”;. The apparent contamination of the T, preparations with T, was in all cases less than 2.17; and did not increase during incubation with pronase. Tables 2 and 3 show the results of T, and T, analyses in the hydrolyzed tablets and the T,/T, molar ratios for each of four lots of desiccated thyroid and thyroglobulin, l- and 2-grain tablets. The T, and T4 content of 2-grain tablets was generally twice that found in the 1 grain. The single exception was in the T, content of the 2-grain desiccated thyroid tablet, which was significantly less than would be present in two l-grain tablets. The differences in T3 and T, content between desiccated thyroid and thyroglobulin are highly significant (allp < 0.001) for both l- and 2-grain preparations. Mean TI/T, molar ratios for desiccated thyroid were 4.3 and 5.2 (1 and 2 grains, respectively), the dif- ference being significant (p < 0.001). T,/T, ratios for thyroglobulin prepara- tions were 2.9 (1 grain) and 2.8 (2 grains) with no significant difference. Dif- ferences in the molar ratios of the desiccated thyroid and thyroglobulin were highly significant (p < 0.01).

Table 3. Ts and T, Content of l-Grain Desiccated Thyroid Preparations After Pronase Digestion (pg/Tablet, Mean f SD)

T3 T4 T4/T3 (Molar)

Desiccated thyroid Lot 1 19 f 1.1 121 f 15 5.3 2 19 f 0.9 117* 7 5.2 3 19~ 1.6 119+ 7 5.2 4 21 f 1.8 128+ 7 5.1 Mean f SD 20 i 1.6 121 * 10 5.2 f 0.1

Thyroglobulin Lot 1 30 f 0.8 102+ 4 2.8 2 34 f 2.0 113* 10 2.8 3 30 f 4.7 101 l 10 2.9 4 32 f 2.9 104 f 13 2.7 Mean ziz SD 31 f 3.1 105+ 9 2.8 f 0.1 1216 REES-JONES AND LARSEN

Table 4. T,, T,, and Iodine Content of Bovine and Porcine Thyroglobulin Hydrolysates (pg/65 mg, Mean f SD)

I content

(“h) T3 T4 T4/T3 (Molar)

Bovine thyroglobulin 1 44 =t 2.6 481 f 36 9.1 f 0.4 Porcine thyroglobulin 1 52 f 6.9 399 f 2 6.6 * 0.9

To determine whether or not there were significant quantities of free iodo- in either preparation, tablets were assayed for T, and T, content without prior digestion. In the 2-grain desiccated thyroid tablet, 0.2 pg T, and 1.7 wg T, were present. Comparable results for thyroglobulin were 0.6 pg T, and 1.7 pg T, per pill. These results indicate that free iodothyronines constitute less than 4% of the total present in each tablet. Table 4 shows the T,, T,, and iodine content as well as T,/T, molar ratios for 65 mg (1 grain) of bovine and porcine thyroglobulin. The differences be- tween the two preparations in T, content and the T,/T, molar ratio were both significant (p < 0.025 and p < 0.01, respectively), but the T, content was the same.

DISCUSSION Previous studies of the T, and T, content of pharmacologic preparations of animal thyroid glands have used chromatographic methods for separation of the iodothyronines with quantitation by either iodine determinations or gas- liquid chromatographic techniques. 13-‘* Both of these methods may result in artifactual conversion of T, to T3, thus leading to an inappropriately low cal- culated T4 to T, ratio and an underestimation of the total T4 content.‘gm21 The only previous attempts to analyze individual tablets of thyroid have given T4 content results of 20-30 pg and 7-10 pg of T, per l-grain tablet.‘4.‘5,‘7 Our data suggest that this represents a considerable underestimation of the T4 content of both desiccated thyroid and thyroglobulin and a smaller, but still significant, underestimate of the T, content. The measured content in the present studies is not inconsistent with the long-standing clinical observations that 2 grains of Thyroid is an adequate replacement dosage for most hypothyroid subjects. The disagreement with the previous results could be due to the improved tech- nology or due to a change in the tablets themselves over the past 10 yr. The for- mer seems more likely based on the fact that clinical experience has not sug- gested a need for a lower dosage in recent years. The radioimmunoassay approach used in the present studies avoids the need for separation of T4 and T, due to the high specificity of the antibodies em- ployed. To the best of our knowledge, there are no previous reports of analyses of thyroid extract using these techniques. While there is up to a 13:/, loss of T, from T, tablets during a 20-hr incubation with Pronase, the results shown in Table 1 indicate that T3 is not a substantial by-product of this process. Because it is not certain when the T, is released during the digestion of the Thyroid tablets, no correction for this loss has been made and this may result in a slight underestimation of the T4 content. The T3 and T4 content of both the desiccated thyroid and the thyroglobulin TRIIODOTHYRONINE AND THYROXINE 1217 tablets from these manufacturers were remarkably consistent from tablet to tablet and from lot to lot. The variation of 100/o or less for both preparations is within the range of the coefficient of variation of the radioimmunoassay tech- niques. Evidence of this consistency is reassuring and has not been previously available. The differences between the two preparations were also consistent, in agreement with earlier studies,‘3,‘4,‘6 and are best illustrated in the significantly lower T4/T3 molar ratio of thyroglobulin. This ratio was 2.8, similar to the 2.5 ratio claimed by the manufacturer. Thyroglobulin preparations are stated to originate from porcine thyroid, whereas the source of desiccated thyroid is not specified and presumably contains both bovine and porcine thyroglobulin. Radioimmunoassay studies of chemically pure porcine thyroglobulin (Table 4) confirm that it has a lower T,/T, ratio than the bovine material. This could, in part, explain the lower ratio found in the thyroglobulin preparation. However, either due to changes occurring during the manufacturing process or to dif- ferences in the iodine content of the thyroglobulin itself, the T,/T, molar ratios for both thyroid extract and thyroglobulin were substantially lower than those found in the two purified thyroglobulin preparations. Since these analyses were performed under conditions of maximal hydrolysis, the quantities of T, and T, available for absorption after ingestion of these preparations by patients may be overestimated. In vivo bioavailability studies will be necessary to determine the completeness of hydrolysis under these cir- cumstances. Nevertheless, these results allow estimation of the maximum amounts of iodothyronines available to patients receiving these brands. Since the absorption of T, and T, is approximately 70”;, and loo?:,, respectively,*? Z4 and since T, is about 2.5 times more potent metabolically than T4 on a weight basis when given intravenously,25 it follows that roughly 39% and 51”, of the thyromimetic effect of desiccated thyroid and thyroglobulin, respectively, are due to their content of T,. Recent data have indicated that in patients receiv- ing L-thyroxine therapy for primary hypothyroidism, serum T, concentrations can be used as an estimate of the adequacy of thyroid hormone replacement.16 In these studies, patients who were clinically and biochemically euthyroid had normal serum T, concentrations (mean 8.1 pg/dl). Because of its relatively high T, content, this would not be the case in patients receiving metabolically equivalent quantities of Thyroid. According to our results, the generally recom- mended dosage of 2 grains of desiccated thyroid or thyroglobulin would pro- vide a maximum of 127 or 110 pg of T, (depending on the preparation used) as compared with the stated equivalent dosage of 200 vg of T4.27m29Thus it is im- portant for the physician to recognize that a low-normal or even slightly sub- normal serum T, concentration is consistent with an adequate replacement dosage of these medications.

REFERENCES

I. Gull Sir WW: On a cretinoid state super- tract of the thyroid gland of a sheep. Br Med J vening in adult life in women. Trans Clin Sot 2:796-797. 1891 7:181-185, 1873 3. Fox EL: A case of myxoedema treated by 2. Murray CR: Note on the treatment of taking extract of thyroid by the mouth. Br Med myxoedema by hypodermic injections of an ex- J 2:941, 1892 1218 REES-JONES AND LARSEN

4. MacKenzie HWG: A case of myxoedema 17. Moody JE Jr, Hohmann JR, Kaplan GB: treated with great benefit by feeding with fresh Chemical determination of the potency of thyroid glands. Br Med J 2:940, 1892 thyroid preparations. J Pharm Sci 57:634-639, 5. Catz B, Ginsburg E, Salenger S: Clinically 1968 inactive thyroid USP. A preliminary report. N 18. Bilous R, Winheuser JJ: Determination Engl J Med 266:136-137, 1962 of and thyroxine in dried thyroid 6. Williams AD, Meister L, Florsheim WH: by GLC. J Pharm Sci 62:274-277, 1973 Chemical identification of defective thyroid 19. Misuma T, Nihei N, Gershengorn MC, preparations. J Pharm Sci 52:8333839, 1963 et al: Serum triiodothyronine measurements in 7. Braverman LE, Ingbar SH: Anomalous human serum by radioimmunoassay with cor- effects of certain preparations of desiccated roboration by gas-liquid chromatography. J thyroid on serum protein bound iodine. N Engl Clin Invest 50:2679-2688, 197 I J Med 270:439-442, 1964 20. Larsen PR: Technical aspects of the esti- 8. Mangieri CN, Lund MH: Potency of mation of triiodothyronine in human serum: United States Pharmacopeia desiccated thyroid Evidence of conversion of thyroxine to tablets as determined by the anti-goitrogenic triiodothyronine during the assay. Metabolism assay in rats. J Clin Endocrinol Metab 30:102- 20:6099624, I97 1 104, 1970 21. Chopra IJ, Fisher DA, Solomon DH, 9. Inoue K, Taurog A: Digestion of 13’1- et al: Thyroxine and triiodothyronine in the labeled thyroid tissue with maximum recovery human thyroid. J Clin Endocrinol Metab 36: of ‘3’I-iodothyronines. Endocrinology 81:319- 311 316, 1973 332, 1967 22. Hays MT: Absorption of triiodothyronine 10. Larsen PR: Direct immunoassay of in man. J Clin Endocrinol Metab 30:675-677, triiodothyronine in human serum. J Clin Invest 1970 51:1939-1949, 1972 23. Read DG, Hays MT, Hershman JM: 11. Larsen PR, Dockalova J, Sipula D, Absorption of oral thyroxine in hypothyroid et al: Immunoassay of thyroxine in unextracted and normal man. J Clin Endocrinol Metab 30: human serum. J Clin Endocrinol Metab 37: 7988799, 1970 177-182, 1973 24. Wenkel KW, Kirschieper HE: Aspects of 12. Snedecor GW, Cochran WG: Statistical the absorption of oral L-thyroxine in normal Methods (ed 6). Ames, Iowa State University man. Metabolism 26:1-g, 1977 Press, 1967 25. Blackburn CM, Keating FR Jr: Com- 13. Devlin WF, Stephenson NR: The chemi- parative effectiveness of daily doses of L- cal determination of liothyronine and thyroxine triiodothyronine or L-thyroxine in the control in enzymic hydrolysates of pork thyroid. J of myxedema. J Clin Invest 33:918, 1954 Pharm Pharmacol 14:597-604. 1962 26. Stock JM, Surks MI, Oppenheimer JH: 14. Pileggi VJ, Golub OJ, Lee ND: Deter- Replacement dosage of L-thyroxine in hypo- mination of thyroxine and triiodothyronine in thyroidism. N Engl J Med 290:529-533, 1974 commercial preparations of desiccated thyroid 27. lngbar SH. Woeber KA: The thyroid and thyroid extract. J Clin Endocrinol Metab gland, in Williams RH (ed): Textbook of Endo- 25~949-956. 1965 crinology (ed 5). Philadelphia, WB Saunders, 15. Lemieux R, Talmage JM: The determina- 1974, p 208 tion of liothyronine and thyroxine in thyroid preparations. J Pharm Pharmacol 18:94-100, 28. Robbins J, Rail JE, Gorden P: The thyroid and iodine metabolism. in Bondy PK. 1966 16. Kologlu S, Schwartz HL, Carter AC: Rosenberg LE (eds): Duncan’s Diseases of Quantitative determination of the thyroxine, Metabolism (ed 7). Philadelphia, WB Saunders, triiodothyronine, monoiodotyrosine and 1974, p 1085 diiodotyrosine content of desiccated thyroid. 29. Werner SC, Ingbar SH: The Thyroid Endocrinology 78:23 l-239, 1966 (ed 3). New York, Harper & Row, 1971, p 835