jcem.82.3.3830

Treatment of Hypothyroidism with Once WeeklyThyroxine*

STEFAN K. G. GREBE, RUSSELL R. COOKE, HENRY C. FORD,JOCELYN N. FAGERSTROM, DIANE P. CORDWELL, NIGEL A. LEVER,GORDON L. PURDIE, AND COLIN M. FEEK

Endocrine Research Unit, Division of Endocrinology, Mayo Foundation and Clinic (S.K.G.G.),Rochester, Minnesota 55905; and the Departments of Chemical Pathology (R.R.C.), Endocrinology(J.N.F., D.P.C., C.M.F.) and Cardiology (N.A.L.), Wellington Hospital, and the Departments ofPathology (H.C.F.) and Public Health (G.L.P.), Wellington School of Medicine, Wellington, NewZealand

ABSTRACTWe compared daily T4 therapy with 7 times the normal daily dose

administered once weekly in 12 hypothyroid subjects in a randomizedcross-over trial. At the end of each treatment wemeasured serum freeT4 (FT4), free T3 (FT3), rT3, and TSH levels and multiple markers ofthyroid hormone effects at the tissue level repeatedly for 24 h.

Compared with daily administration, the mean serum TSH beforethe administration of weekly T4 was higher (weekly, 6.61; daily, 3.92mIU/mL; P , 0.0001), and the mean FT4 (weekly, 0.98; daily, 1.35ng/dL;P, 0.01) andFT3 (weekly, 208; daily, 242 pg/dL;P, 0.01)werelower. A minimally elevated serum total cholesterol during weeklyadministration (weekly, 246.8; daily, 232.6 mg/dL; P , 0.03) was the

only evidence of hypothyroidism at the tissue level.Compared with daily administration, the mean peak FT4 following

weekly administration of T4 was significantly higher (weekly, 2.71;daily, 1.59 ng/dL; P , 0.0001), as was the mean peak FT3 level(weekly, 285; daily, 246 pg/dL; P , 0.01). None of the tissue markersof thyroid hormone effect changed compared to daily T4, and therewasno evidence of treatment toxicity, including cardiac toxicity.

During weekly T4 administration, autoregulatory mechanismsmaintain near-euthyroidism. For complete biochemical euthyroidisma slightly larger dose than 7 times the normal daily dose may berequired. (J Clin Endocrinol Metab 82: 870875, 1997)

T4REPLACEMENT for hypothyroidism usually achievescomplete restoration of euthyroidism. However, daily,lifelong administration can lead to patient noncompliance. T4has an elimination half-life of about 7 days (1, 2), but itsbiological effect may be longer. It is a prohormone, the activehormone is T3 (3, 4), formed from T4 in peripheral tissues.There is strong evidence for autoregulation of the peripheralconversion of T4 to T3,with increasing conversion rates at lowserum T4 levels and decreasing conversion when serum T4 iselevated (58). Together, these properties suggest the pos-sibility of using a dosing interval longer than the traditional24 h. Weekly dosing may improve compliance in some pa-tients and could be advantageous to nurses or other care-givers when T4 must be administered to patients unable todose themselves.During the 1960s through early 1980s, studies demon-

strated that single doses of T4 up to 3 mg are well tolerated(913). However, peripheral thyroid hormone measure-ments were largely limited to total serum T4 and T3 (1113),although one study included radioactive iodide uptake, pro-tein-bound iodine, T3-resin uptake, and free T4 (FT4) in some

patients (9). Sensitive TSH assays were not available, but twostudies measured TRH-stimulated TSH responses (12, 13).No study measured free T3 (FT3). Although all studies com-mented to some degree on patient symptomatology, and onestudy measured total serum cholesterol (9), none assessedpatient symptoms and thyroid hormone effects at the tissuelevel in a systematic or comprehensive fashion.We, therefore, believed that the neither efficacy or safety

of once weekly T4 therapy was established and decided todetermine this using current thyroid function tests and mea-surements of the tissue effects of thyroid hormone. The aimof the study was to determine whether 7 times the daily doseof T4 administered once weekly was as safe and efficaciousas the usual daily dose for maintenance therapy in hypo-thyroid subjects.

Experimental Subjects

The study protocol and all procedures were approved andmonitoredby the Capital Coast Health Ethics Committee (Wellington, New Zea-land). All patients gave their informed consent.

Statistical power analysis

We used a randomized cross-over design for our study, thus achiev-ing at least 4 times the statistical power of a comparably sized groupcomparative trial. We estimated that we needed 12 or more patients toachieve 80% power at the 0.05 significance level for detecting differencesbetween daily and weekly treatment of 1060% in the mean values ofthe tests performed. In addition, we ensured that the sample size wassufficient to detect a 10% difference in mean corrected systolic timeintervals between the two treatments.

Received June 17, 1996. Revision received October 14, 1996. AcceptedNovember 27, 1996.

Address all correspondence and requests for reprints to: Dr. StefanGrebe, Dept. of Pathology, Wellington School of Medicine, Wellington,New Zealand. E-mail: [email protected].

* This work was supported by a grant form the Wellington MedicalResearch Foundation.

Supported by an Overseas Postdoctoral Fellowship grant from theHealth Research Council of New Zealand.

0021-972X/97/$03.00/0 Vol. 82, No. 3Journal of Clinical Endocrinology and Metabolism Printed in U.S.A.Copyright 1997 by The Endocrine Society

870

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 09 May 2014. at 06:28 For personal use only. No other uses without permission. . All rights reserved.

Patients

Fourteen patients were initially enrolled, and 12 subjects completedthe study; 1 subject did not commence the study, and another patientdeveloped chest pain of uncertain origin while receiving her normaldaily T4 therapy and was withdrawn before entry into the weeklytreatment phase.

Two subjects were male, and 10 were female. The mean age of thestudy subjects was 50.8 (sd 5 14.5) yr. All subjects suffered from con-firmed primary hypothyroidism, as evidenced by a clearly elevatedserum TSH at the time of diagnosis ($20 mIU/mL) and at least 1 un-successful trial of T4 withdrawal thereafter. At the time of enrollment allhad been receiving T4 replacement therapy at a stable dose for the least3 months.

Themean daily T4 dose was 1.6 mg (sd5 0.35 mg)/kg BW. At the timeof their original diagnoses, all patients had received standard instruc-tions regarding T4 administration, including taking T4 separate fromother medications and food.

The causes of hypothyroidism were autoimmune thyroiditis in fivepatients, radioiodine ablation in three patients, subtotal thyroidectomy

in two patients, and undetermined in two patients. No patient sufferedfrom severe medical illness, pituitary disease, untreated metabolic bonedisease or osteoporosis, liver disease, cardiac disease, or known abnor-malities of thyroid hormone metabolism and thyroid hormone proteinbinding or was taking medications known to interfere with thyroidfunction or thyroid hormone measurement.

Materials and MethodsStudy design

At trial entry, subjects were randomly assigned either to continuewith their usual dailymaintenance dose of T4 or to take 7 times the usualdaily dose once weekly, beginning on day 1 of the trial. Figure 1 sum-marizes the trial design. Three patients receiving other medications inaddition to T4 continued to take these in the usual manner and doseduring the duration of the trial, and all patients were instructed not tochange their normal daily habits (including diet and exercise) during thetrial period.

To achieve steady state serum thyroid hormone levels before testing,

FIG. 1. This figure depicts the cross-over trial design and testing schedule used in our study. The upper part of the figure schematically showsthe trial design in relation to the relevant time line (in days). At trial entry, 12 hypothyroid subjects receiving stable T4 replacement wererandomized to initially either continue with their normal daily T4 dose (n 5 7) or to start taking 7 times their regular daily dose once per week(n5 5). On days 43/44 and 85/86, all subjects underwent repeatedmeasurements of serum thyroid hormone levels and tissue markers of thyroidhormone effects. The testing schedule is summarized in the table inset at the bottom of the figure. The treatment cross-over occurred after theday 43/44 testing, as detailed in the footnotes to the main part of the figure.

ONCE WEEKLY T4 TREATMENT 871


all study subjects continuedwith their assigned T4 treatment regimen for6 weeks, about 6 times the elimination half-life of T4. On the 43rd and44th trial days, all subjects underwent clinical, biochemical, and bio-physical tests. The patients taking a daily dose then took 6 times theirnormal daily dose on day 44 and omitted daily T4 for the next 6 days.Oneweek after testing (day 50) they then began taking 7 times their dailydose once each week for 5 weeks. Those subjects previously on theweekly dose returned to their daily dose 1 week after testing (day 50)and continued on the daily dose for the next 5 weeks. Testing wasrepeated on the 85th and 86th trial days.

On the days of testing patients attended the Department of Endo-crinology,WellingtonHospital, between 08000900 h fasting. On arrivalthey underwent baseline testing (0 h), which comprised a standardizedquestionnaire concerning thyroid-related symptoms during the previ-ous week, self-assessment of well-being during the previous week usinga visual analog scale, echocardiographic measurement of systolic timeintervals, blood sampling for serum thyroid function tests, and mea-surement of a variety of serumanalytes used as tissuemarkers of thyroidhormone effects. Depending on the treatment period, the subjects thentook either their daily or weekly dose of T4. Further blood samples weretaken after 1, 2, 4, 8, and 24 h. Patients were permitted to eat after the4-h testing. At 8 h, the echocardiogram was repeated. Patients receivingweekly treatment underwent an additional echocardiogram at 24 h (the24 h echocardiogram was omitted in patients receiving daily treatment,because it was assumed to be equivalent to the baseline). The question-naire and self-rated scale of well-being were readministered to all sub-jects at 24 h. All testing at 24 hwas performed after an overnight fast and,for the patients receiving daily treatment, before T4 was taken.

Details of the testing procedures are summarized in Fig. 1. The detailsof the symptom questionnaire are given in Table 1. For self-assessmentof well-being, a previously validated visual analog scale was used (14).On this scale patients rated their well-being over the preceding week or24 h on a 100-mm line, with the left end corresponding to worst ever(0 mm) and the right end to best ever (100 mm).

Immediately after venipuncture and centrifugation, the serum wasseparated and stored at 220 C. At the conclusion of the study, allsamples for each patient were measured in the same assay to minimizeinterassay variation.

Serum assays

Thyroid function tests consisted of serumFT4, FT3, TSH (all measuredon the Corning ACS-1801 automated immunoanalyzer, Scianz Corp.,Auckland, New Zealand), rT3 (measured by RIA, Biodata, Milan, Italy),and serum T4-binding globulin [TBG; using an immunoradiometricassay (IRMA), Corning Medical, Midland, MI]. The functional sensitiv-ity of the TSH assay is 0.03 mIU/mL.

As general indicators of thyroid hormone effects at the tissue level,the following, previously validated, parameters (1517) were used: sexhormone-binding globulin (by IRMA, Orion Diagnostics, Finland); totalcholesterol, high density lipoprotein (HDL), and triglycerides (Hitachi717 multianalyzer, Boehringer Mannheim, Auckland, New Zealand);and apolipoprotein a (APOa; immunoturbidimetric method, Hitachi717). Low density lipoprotein (LDL) was calculated from HDL choles-terol and triglyceride concentrations.

To monitor the effects of thyroid hormone on the liver (1721), as-partate aminotransferase, alanine aminotransferase, and g-glutamyl-transferase were measured with a Hitachi 717 multianalyzer.

Serum osteocalcin and alkaline phosphatase were used as markers ofthe influence of thyroid hormones on bone turnover (2224) and weremeasured with an IRMA (Nichols Institute, San Juan Capistrano, CA)and a Hitachi 717 multianalyzer, respectively.

Echocardiography

The cardiac effects of thyroid hormone were estimated by systolictime intervals (STI)measurements, a sensitivemarker of these effects (15,2530). A two-dimensional echocardiogram was performed first, to ex-clude cardiac conditions known to interferewith STImeasurements. TheSTI were obtained after the two-dimensional study and 15 min of restby the method described by Tseng et al. (26), using a dual M-modesystem (HP 77020AC Rev F, Hewlett-Packard, Palo Alto, CA) with achart speed of 100 mm/s. Data were corrected for heart rate and gender(31), and a total of 10 cycles were analyzed to minimize variation dueto respiration.

Data analysis

Data for TSH, FT4, and rT3 followed a log-normal distribution andwere log transformed before analysis. Other noncategorical data did notneed to be transformed before analysis. Statistical analysis of all data,except for the results of the symptom questionnaire, was performedusing multivariate regression and ANOVA for repeated measures (32),with terms for treatment type (weekly vs. daily), treatment sequence(weekly or daily treatment first), and interactions. The Greenhouse-Geisser adjustment to degrees of freedom was used to account for therepeated measures on individuals (33). In addition, the untransformedTSH, FT4, and rT3 data were analyzed using nonparametric equivalentsof the parametric statistical tests (Friedman test). The results of thesymptom questionnaire were analyzed using McNemars test, compar-ing dailywithweekly treatment (32). For all statistical tests, P, 0.05wasconsidered significant.

ResultsSymptoms

All patients tolerated weekly T4 treatment well. Therewere no significant differences at either 0 or 24 h betweendaily and weekly treatments in the results of the question-naire and self-rated visual analog scale (Table 1).

Thyroid function tests

Serum thyroid hormone levels differed between weeklyand daily treatment at several time points (Fig. 2). At 0 h, themean values for FT3 (daily, 242;weekly, 208 pg/dL;P, 0.01),rT3 (daily, 26.75; weekly, 22 ng/dL; P, 0.01), and FT4 (daily,1.35; weekly, 0.98 ng/dL; P, 0.001) were significantly lower

TABLE 1. Results of the symptom questionnaire and visual analog scale of well-being

During the last week of treatment During the 24 h after T4 dose

Daily T4 Weekly T4 P Daily T4 Weekly T4 P

Tired: Y/N; N 8/4 6/6 0.5a 7/5 7/5 1.0a

Nervous/anxious: Y/N; N 3/9 3/9 1.0a 1/11 0/12 1.0a

Constipated: Y/N; N 4/8 5/7 1.0a 1/11 3/9 0.5a

Loose bowels: Y/N; N 3/9 5/7 0.69a 1/11 2/10 1.0a

Cold: Y/N; N 5/7 2/10 0.25a 1/11 1/11 1.0a

Hot/sweaty: Y/N; N 2/10 2/10 1.0a 1/11 1/11 1.0a

Visual analog scale: mean distancefrom left edge (SD;b mm)

63.5 (18.7) 57 (18.1) 0.2c 64.4 (16.5) 65.6 (17.4) 0.7c

a By McNemars test for paired proportions.b The leftmost edge (0 mm) corresponds to worst ever. The rightmost edge (100 mm) corresponds to best ever.c By Multivariate regression and ANOVA for repeated measures with Greenhouse-Geisser adjustment to degrees of freedom.

872 GREBE ET AL. JCE & M 1997Vol 82 No 3


for the group receiving weekly therapy. For this group, FT4was significantly higher at all other time points (all P ,0.005), peaking at 2 h with both daily and weekly T4 (daily,

1.6; weekly, 2.7 ng/dL). FT3 was significantly higher withweekly than with daily T4 at 4 h (daily, 240; weekly, 265pg/dL;P, 0.04) and 24 h (daily, 246;weekly, 285 pg/dL;P,0.01). rT3 was significantly lower with weekly compared todaily T4 at 2 h (daily, 30.42; weekly, 26.82 ng/day; P , 0.05)andhigher at 24 h (daily, 27.8;weekly, 33.16 ng/dL;P, 0.01).FT3 and rT3 peaked later than FT4. This delay was morepronounced with weekly T4. Serum TSH levels with weeklytherapywere significantly higher than thosewith daily treat-ment at all time points (all P , 0.04), except 24 h (P 5 NS).The largest difference occurred at 0 h, with a mean serumTSH value of 3.92 mIU/mL with daily T4 vs. 6.61 mIU/mLwithweekly T4. SerumTSH levelswithweekly T4 fell rapidlyin the first hour after T4 administration and then graduallydeclined to levels similar to those with daily treatment.Serum TBG results were subject to a treatment sequence

effect, with measurements during weekly treatment signif-icantly higher in patients who took daily T4 first than in thosereceiving initial weekly treatment. We, therefore, only com-pared measurements during the first treatment cycle witheither daily (n 5 7) or weekly (n 5 5) therapy. In this com-parison the two treatment regimens did not differ signifi-cantly. Mean TBG values were between 1.992.04 mg/dLwith daily and between 1.922.17 mg/dL with weeklytreatment.

Markers of tissue effect

There were no significant differences at any time pointbetween daily and weekly treatment in the levels of serumsex hormone-binding globulin, g-glutamyltransferase, as-partate aminotransferase, alanine aminotransferase, osteo-calcin, alkaline phoshatase (all in Table 2), HDL, and LDL(data not shown), but total serum cholesterol differed sig-nificantly at 0 h (P , 0.03), with the mean being 14.2 mg/dLhigher during weekly therapy (Table 2). No difference wasfound for serum total cholesterol measurements at othertimes. APOa was subject to a sequence effect. Consequently,as with serum TBG measurements, only the first treatmentcycle was used for analysis. No differences between groupswere found in serum APOa measurements.STI measurements did not significantly differ between

daily and weekly treatments (Table 2).

Discussion

Our results suggest that once weekly T4 replacement ther-apy for hypothyroidism is safe and arguably efficacious,making it a possible alternative to customary daily therapy.Once weekly T4 replacement was well tolerated, and therewas no indication of acute treatment toxicity compared withdaily therapy. Although serum FT4 levels rose significantlyafter treatment, changes in FT3 with once weekly therapywere slight, confirmingprevious studies (9, 11, 13).However,thyroid function tests demonstrated mild hypothyroidismbefore weekly treatment, with higher mean serum TSH andlower mean serum FT4 and FT3 values. For complete bio-chemical euthyroidism, a slightly larger dose than 7 times thenormal daily dose may be required.At the peripheral tissue level, the effect of weekly T4 treat-

ment did not differ from that of customary daily treatment

FIG. 2. From top to bottom, the four panels of this figure depict themedian serum rT3, FT3, FT4, and TSH values in 12 hypothyroidsubjects during daily (solid squares, solid lines) and weekly (opencircles, dashed lines) T4 treatment. The upper and lower error barscorrespond to the 75th and 25th percentiles of the data, respectively.For each test, the shaded areas correspond to the respective normalreference ranges [rT3, 9.035.0 ng/dL (Systeme International units,0.140.54 nmol/L; intraassay coefficient of variation, 6.5%); FT3,227.3415.6 pg/dL (Systeme International units, 3.56.4 pmol/L; in-traassay coefficient of variation, 3%); FT4, 0.781.78 ng/dL (SystemeInternational units, 1023 pmol/L; intraassay coefficient of variation,3%); TSH, 0.34.0 mIU/mL (Systeme International units, 0.34.0mIU/L; intraassay coefficient of variation, 2.8%)]. At the times de-noted by the asterisks, the medians, compared by Friedman test, aswell as the means (FT3) and log-transformed means (TSH, FT4, andrT3), analyzed by ANOVA for repeated measures, were significantlydifferent between daily and weekly therapies at the 0.05 significancelevel.



despite differences in serum thyroid hormone levels. Theonly tissue marker of thyroid hormone effect to suggesthypothyroidismwas total serum cholesterol at 0 h. This mayhave been due to statistical fluctuation. Furthermore, theobserved rise was small and seemed to be caused by a risein both HDL and LDL. It is generally believed that increasedserum HDL levels may partially compensate for elevationsin LDL (34). Hence, any potential adverse effect of the smallrise in total serum cholesterol may have been mitigated.One of the mechanisms maintaining near euthyroidism at

the tissue level might be a change in the conversion rate ofT4 to T3 (5, 6). Whereas serum FT4 levels almost tripled afterthe ingestion of weekly T4, FT3 rose by about 25%, and rT3rose by about 50%, suggesting preferential conversion of T4to themetabolically inactive rT3. By contrast, at the end of thedosing interval, FT4 levels during weekly treatment werealmost 30% lower than those during daily treatment,whereas FT3 levels were 15% lower, and rT3 levels were 18%lower. This indicates that at the end of weekly treatment,conversion of T4 to T3 increased, and conversion to rT3decreased.It could be argued that the tests of peripheral thyroid

hormone effect employed were too insensitive and the num-ber of subjects studied too small to detect subtle changes inthe thyroid state. However, even in subclinical thyroid statesmost of the tests used are discriminatory (1517, 20, 2326,35). STI are shortened in patients with subclinical hyperthy-roidism (25, 26) and lengthened in subclinical hypothyroid-ism (15, 26). All the changes observed are around 20% andwithin the range of the power of our study. The response ofSTI to changes in serum thyroid hormone levels is rapid,showing significant differences in less than 2 weeks after theonset of hypothyroidism (28, 29).

However, continuous 24-h electrocardiogrammonitoring,which generally parallels echocardiographic measurementsof cardiac status (36), was not performed, and therefore, wecannot exclude the possibility of asymptomatic cardiac ar-rhythmia as a result of weekly T4 treatment. In addition, theabsence of data on thyroid function tests and tissue markersduring days 26 after weekly treatment in our study couldunderestimate the toxicity of weekly treatment. According tosome studies, peak conversion of T4 to T3may not occur until24 days after the ingestion of large T4 doses (37, 38), al-though the rise after 24 h is slight if free T3 is measured (37).However, other studies have suggested that T4 doses be-tween 2.4300 mg will lead to a T3 peak before 24 h (11, 39).Our data suggest that FT3 levels plateau at around 4 h (andare in the lower third of the reference range), but we did notobserve a fall in mean FT3 at 24 h. Consequently, we cannotcompletely dismiss the possibility of toxicity between days26 after weekly treatment. However, in a pilot project in-volving two subjects (not included in this study) sampled at0, 1, 2, 4, 8, 24, and 72 h, FT3 values at 72 hwere between peak(24 h) and nadir (0 h) values.As we do not have firm safety data for the period between

2 and 6 days after T4 administration or any direct assessmentof potentially harmful arrhythmias, we would be hesitant touse weekly T4 treatment in individuals with ischemic heartdisease whomay be sensitive to T4 (18). We also do not knowwhether weekly T4 may be suitable to suppress TSH secre-tion, although our data suggest that the T4 dose will have tobe increased significantly above 7 times the normal dailydose to ensure suppression over a week. An increase in theweekly dose above 7 times the patients usual daily dosemight be possible without undue risk, but further study isneeded on its effects on the skeletal system (2224) and heart

TABLE 2. Effects of once weekly T4 treatment on tissue markers of thyroid hormone effect

Tissue markers of thyroid hormone effectMean (6SEM) at 0 ha Mean (6SEM) at 8 hb Mean (6SEM) at 24 hb

Daily T4 Weekly T4 Daily T4 Weekly T4 Daily T4c Weekly T4d

GeneralTotal cholesterol (mg/dL) 232.6 (12.3) 246.8 (13.6) 238.7 (12.8) 244.9 (12.9) 233.9 (12.1) 243.9 (14.8)SHBG (nmol/L) 51.1 (11.2) 47.6 (8.4) 51.9 (12.6) 44.8 (7.8) 53.3 (13.7) 46.6 (7.6)

LiverAST (U/L) 10.4 (1.5) 9.3 (1) 11.3 (1.5) 10.0 (1.1) 9.5 (1.2) 9.8 (1)ALT (U/L) 7.3 (1.4) 5.4 (1.3) 9.4 (1.9) 7.3 (1.5) 5.5 (0.8) 5.6 (1)gGT (U/L) 32.8 (8.3) 30.6 (6.8) 30.1 (7.9) 28.2 (6.2) 32.3 (7.7) 28.5 (6.2)

BoneOsteocalcin (ng/mL) 4.6 (0.5) 3.9 (0.4) 4.8 (0.6) 4.4 (0.5) 4.2 (0.6) 3.3 (0.4)Alkaline phosphatase (U/L) 51.5 (4.2) 51.9 (4.2) 53.9 (4.8) 51.8 (4) 47.9 (6) 48.9 (4.4)

Heart (STIs)Heart rate (beats/min) 69.1 (2.3) 66.6 (2.3) 69.8 (3.3) 66.7 (2) 65.7 (1.6)R-R interval (s) 0.88 (0.03) 0.9 (0.04) 0.88 (0.04) 0.91 (0.03) 0.92 (0.02)ICT (ms) 39.7 (2.7) 38.5 (2.4) 37.9 (1.7) 40.2 (1.7) 41.2 (2.7)PEPc (ms) 108.7 (3.6) 106.9 (5.1) 109.3 (3.9) 112.4 (2.9) 110.2 (3.7)LVETc (ms) 427.3 (10.6) 440.4 (11) 427.3 (9.5) 432.9 (8.5) 440.7 (8.6)PEPc/LVETc 0.26 (0.01) 0.25 (0.01) 0.26 (0.01) 0.26 (0.01) 0.25 (0.01)

SHBG, Sex hormone-binding globulin; AST, aspartate aminotransferase; ALT, alanine aminotransferase; gGT, g-glutamyltransferase;STI(s), systolic time interval(s); ICT, isovolemic contraction time; PEPc, preejection period corrected for gender and heart rate; LVETc, leftventricular ejection time corrected for gender and heart rate.

a Mean total serum cholesterol significantly higher (P , 0.03) on weekly T4. No statistically significant difference for any of the othermeasurements between daily and weekly T4.

b No statistically significant difference for any of the measurements between daily and weekly T4.c For STIs, 24-h measurements of STIs were not performed for daily dosing.d For STIs, at 24 h there was no statistically significant difference for any STI measurement with weekly T4 to corresponding 0 h values with

daily T4.

874 GREBE ET AL. JCE & M 1997Vol 82 No 3


(36). Finally, in some patients weekly treatment could also behazardous if several doses are missed. Clinicians and patientmust work out a schedule that minimizes such risks beforeswitching to weekly T4 therapy.

Acknowledgments

The authors thank Sarah Crichton, B.S., for statistical advise; CarlBurgess, MBBS, M.D., for helpful discussions on systolic time intervalmeasurements; Ian D. Hay, MBBS, Ph.D., for critical comments on themanuscript and helpful discussion; the staff of the Wellington Hospitalendocrine and radiochemistry departments for their help and support;and last, but not least, all the patients who volunteered for this study.

References

1. Stock JM, Surks MI, Oppenheimer JH. 1974 Replacement dosage of l-thy-roxine in hypothyroidism. A re-evaluation. N Engl J Med. 290:529533.

2. Haynes Jr RC. 1990 Thyroid and antithyroid drugs. In: Goodman-Gilman A,Rall TW, Nies AS, Taylor P, eds. Goodman and Gilmans the pharmacologicalbasis of therapeutics, 8th ed. New York: Pergamon Press; 13611383.

3. Shepard A, Eberhardt NL. 1993 Molecular mechanisms of thyroid hormoneaction. Clin Lab Med. 13:531541.

4. Schwartz HL, Strait KA, Oppenheimer JH. 1993 Molecular mechanisms ofthyroid hormone action. A physiological perspective. Clin Lab Med.13:543561.

5. Keck FS, Loos U. 1988 Peripheral autoregulation of thyromimetic activity inman. Horm Metab Res. 20:110114.

6. Nicoloff JT, Lum SM, Spencer CA, Morris R. 1984 Peripheral autoregulationof thyroxine to triiodothyronine conversion in man. Horm Metab Res.14(Suppl):7479.

7. Lum SM, Nicoloff JT, Spencer CA, Kaptein EM. 1984 Peripheral tissue mech-anism for maintenance of serum triiodothyronine values in a thyroxine-defi-cient state in man. J Clin Invest. 73:570575.

8. Hay ID, Gorman CA, Burman KD, Jiang N-S. 1985 Stereospecific determi-nation and in vivomonoiodination of thyroxine enantiomers in euthyroidman.Metabolism. 34:266271.

9. Bernstein RS, Robbins J. 1969 Intermittent therapy with l-thyroxine. N EnglJ Med. 281:14441448.

10. Wallack MS, Adelberg HM, Nicoloff JT. 1970 A thyroid suppression testusing a single dose of l-thyroxine. N Engl J Med. 283:402405.

11. Sekadde CB, Slaunwhite Jr WR, Aceto Jr T, Murray K. 1974 Administrationof thyroxine once a week. J Clin Endocrinol Metab. 39:759764.

12. Thein-Wai W, Larsen PR. 1980 Effects of weekly thyroxine administration onserum thyroxine, 3,5,39-triiodothyronine, thyrotropin, and the thyrotropin re-sponse to thyrotropin-releasing hormone. J Clin Endocrinol Metab.50:560564.

13. Goretzki P, Roeher HD, HoreyseckG. 1981 Prophylaxis of recurrent goitre byhigh dose l-thyroxine. World J Surg. 5:855857.

14. CarrD,McLeodDT, ParryG, ThornesHM. 1988 Fine adjustment of thyroxinereplacement dosage: comparison of the thyrotropin releasing hormone testusing a sensitive thyrotropin assay with measurement of free thyroid hor-mones and clinical assessment. Clin Endocrinol (Oxf). 28:325333.

15. Staub JJ, Althaus BU, Engler H, et al. 1992 Spectrum of subclinical and overthypothyroidism: effect on thyrotropin, prolactin, and thyroid reserve, andmetabolic impact on peripheral target tissues. Am J Med. 92:631642.

16. Foldes J, Tarjan G, Banos C, Nemeth J, Varga F, Buki B. 1991 Biologic bloodmarkers reflecting thyroid hormone effect at peripheral tissue level in patientsreceiving levothyroxine replacement for hypothyroidism. Acta Med Hun.48:3343.

17. GowSM, Caldwell G, Toft AD, et al. 1987 Relationship between pituitary andother target organ responsiveness in hypothyroid patients receiving thyroxinereplacement. J Clin Endocrinol Metab. 64:364370.

18. Leslie PJ, Toft AD. 1988 The replacement therapy problem in hypothyroidism.Bailliere Clin Endocrinol Metab. 2:653670.

19. Ross DS. 1988 Subclinical hyperthyroidism: possible danger of overzealousthyroxine replacement therapy. Mayo Clin Proc. 63:12231229.

20. Gow SM, Caldwell G, Toft AD, Beckett GJ. 1989 Different hepatic responsesto thyroxine replacement in spontaneous and 131I-induced primary hypothy-roidism. Clin Endocrinol (Oxf). 30:505512.

21. Azizi F. 1982 g-Glutamyl transpeptidase levels in thyroid disease. Arch InternMed. 142:7981.

22. Mosekilde L, Eriksen EF, Charles P. 1990 Effects of thyroid hormones on boneand mineral metabolism. Endocrinol Metab Clin North Am. 19:3563.

23. Giannini S, Nobile M, Sartori L, et al. 1994 Bone density and mineral me-tabolism in thyroidectomized patients treatedwith long-term l-thyroxine. ClinSci. 87:593597.

24. Marcocci C, Golia F, Bruno-Bossio G, Vignali E, Pinchera A. 1994 Carefullymonitored levothyroxine suppressive therapy is not associated with bone lossin premenopausal women. J Clin Endocrinol Metab. 78:818823.

25. Banovac K, Papic M, Bilsker MS, Zakarija M, McKenzie JM. 1989 Evidenceof hyperthyroidism in apparently euthyroid patients treated with levothyrox-ine. Arch Intern Med. 149:809812.

26. Tseng KH, Walfish PG, Persaud JA, Gilbert BW. 1989 Concurrent aortic andmitral valve echocardiography permits measurement of systolic time intervalsas an index of peripheral tissue thyroid functional status. J Clin EndocrinolMetab. 69:633638.

27. Amidi M, Leon DF, DeGroot WJ, Kroetz FW, Leonard JJ. 1968 Effect of thethyroid state on myocardial contractility and ventricular ejection rate in man.Circulation. 38:229239.

28. Grossmann G, Keck FS, Wieshammer S, Goller V, Schmidt A, Hombach V.1995 Systolic ventricular function in acute hypothyroidism: a study usingDoppler echocardiography. Exp Clin Endocrinol. 102:104110.

29. Price DE, OMalley BP, Northover B, Rosenthal FD. 1991 Changes in circu-lating thyroid hormone levels and systolic time intervals in acute hypothy-roidism. Clin Endocrinol (Oxf). 35:6769.

30. Wakonig P, Rothlauer W, Koltringer P, Lind P, Eber O. 1986 Diagnostik derperipheren Schilddrusen-hormonwirkung mittels Sex-Hormone-Binding-Globulin (SHBG) und systolischer Zeitintervalle (STI). ActaMedAustr. 13:58.

31. Weissler AM,HarrisWS, Schoenfeld CD. 1968 Systolic time intervals in heartfailure in man. Circulation. 37:149159.

32. SAS Institute. 1991 SASlanguage and procedures, version 607. Cary: SASInstitute

33. Greenhouse SW, Geisser S. 1959 On methods in the analysis of profile data.Psychometrika. 32:95112.

34. Gordon T, Kannel WB, Castelli WP, Dawber TR. 1981 Lipoproteins, cardio-vascular disease, and death. The Framingham study. Arch Intern Med.141:11281131.

35. Kung AWC, Pang RWC, Janus ED. 1995 Elevated serum lipoprotein(a) insubclinical hypothyroidism. Clin Endocrinol (Oxf). 43:445449.

36. Biondi B, Fazio S, Carella C, et al. 1993 Cardiac effects of long term thyro-tropin-suppressive therapy with levothyroxine. J Clin Endocrinol Metab.77:334338.

37. LeBoff MS, Kaplan MM, Silva JE, Larsen PR. 1982 Bioavailability of thyroidhormone from oral replacement preparations. Metabolism. 31:900905.

38. Wenzel KW, Meinhold H, Schleusener H. 1975 Different effects of oral dosesof triiodothyronine or thyroxine on the inhibition of thyrotropin releasinghormone (TRH)mediated thyrotropin (TSH) response inman.Acta Endocrinol(Copenh). 80:4248.

39. Symons RG, Murphy LJ. 1983 Acute changes in thyroid function tests fol-lowing ingestion of thyroxine. Clin Endocrinol (Oxf). 19:539546.



jcem.82.3.3830

Documents