CLIN. CHEM. 41/2, 226-231 (1995) #{149}Lipids and Llpoprotelns
226 CLINICAL CHEMISTRY, Vol. 41, No. 2, 1995
Changes in Serum Lipoprotein(a) and Lipids During Treatment of HyperthyroidismAnnie W.C. Kung,”4 Richard W.C. Pang,2 Ian Lauder,3 Karen S.L. Lam,’ and Edward D. Janus2
Because of suggestions that thyroid hormones modulateserum Iipoprotein(a) [Lp(a)] concentration, we evaluatedprospectively the serial changes of serum Lp(a), mea-sured as apolipoprotein(a) [apo(a)], and other lipoproteinsin 40 subjects with hyperthyroidism treated with radioac-tive iodine (RAI) therapy. Hyperthyroid patients had lower(P <0.001) concentrations of apo(a), total cholesterol(IC), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), and apo B, buthigher apo A-I concentrations compared with age-matched controls [geometric mean (range)]; apo(a) 81(17-614) vs 187 (17-1 808 lU/L): TC 4.07 ± 0.8 vs 5.22 ±1.00 mmoVL (mean ± SD); LDL-C 2.47 ± 0.89 vs 3.40 ±
0.88 mmoVL; HDL-C 1.05 ± 0.33 vs 1.24 ± 0.34 mmol/L;apo B 0.66 ± 0.23 vs 1.13 ± 0.34 g/L, and apo A-I 2.07± 0.42 vs 1.46 ± 0.28 gIL, respectively. Euthyroidismwas associated with normalization of serum TC, LDL-C,and apo B within 1 month of treatment. However, apo(a)required 4 months to normalize, and HDL-C and apo A-I
were still abnormal 6 months after RAI. Serum apo(a), TC,LDL-C, and ape B were negatively correlated with serumthyroxine (14), free thyroxine index, and triiodothyronine(13) and positively correlated with thyrotropin during thetransitional period from hyperthyroidism to euthyroidism.Parallel changes of these lipoproteins and thyroid hor-mones were also observed after treatment of hyperthy-roidism. In conclusion, thyroid hormones do modulatelipoproteins, particularly Lp(a). The delay in normalizationof apo(a) but not LDL suggests an effect on apo(a)production rather than on LDL removal.
Indexing Ternis thyrotoxicosis/iudioactive iodine/apo!ipoproteins/choJest
Lipoprotein(a) [Lp(a)1 has been demonstrated tohave a thrombogemc and atherogenic influence.5 Itconsists of a carbohydrate-rich apolipoprotein(a)[apo(a)], very similar in structure to plasminogen,covalently linked to a low-density lipoprotein (LDL)-like particle that has apo B100 as its structural protein(1-3). Epidemiologic studies have confirmed that car-diovascular risk is related to increased plasma concen-trations of Lp(a) (4). Lp(a) possesses atherogenic and invitro antifibrinolytic properties and favors the deposi-
Departments of’ Medicine, 2 Clinical Biochemistry, and3 Sta-tistics, The University of Hong Kong, Queen Mary Hospital,Pokfulam Road, Hong Kong.
for correspondence. Fax 852-872-5828.5Nonstandard abbreviations: Lp(a), lipoprotein(a); RAI, radio-
active iodine; TC, total cholesterol; LDL-, HDL-, VLDL-C, low-,high-, and very-low-density lipoprotein cholesterol; ape, apoli-poprotein; T4, thyroxine; FF1, free thyroxine index; T3, triiodothy-romne; TSH, thyrotropin; and TG, triglyceride.
Received June 8, 1994; accepted October 19, 1994.
tion of cholesterol in lesions of the endothelium, thus
posing an increased atherogenic risk (3, 5).Lp(a) is highly heterogeneous in structure, density,
and composition, and its concentrations are understrong genetic influence (6). Recently other factors,including thyroid hormones, sex steroids, and growthhormone, have also been shown to be important in theregulation of circulating Lp(a) concentrations (7-10).However, conflicting data are reported regarding Lp(a)concentrations in patients with thyroid disorders. DeBruin et al. (7) reported that higher Lp(a) concentra-tions are observed in hypothyroid subjects comparedwith euthyroid subjects, whereas others have beenunable to detect significant changes in Lp(a) concen-trations in thyroid disorders (11, 12). The increasedLp(a) concentration may contribute to the increasedrisk of premature coronary artery disease in the hypo-thyroid state.
The present study aimed to evaluate prospectivelyLp(a) and other serum lipoprotein concentrations in 40Chinese subjects with hyperthyroidism and to followthe changes in these lipoproteins over 6 months afterradioactive iodine (RAI) therapy.
Subjects and Methods
Subjects
We prospectively studied 40 subjects (11 men, 29women, including 10 postmenopausal women) withnewly diagnosed hyperthyroidism due to Graves dis-
ease, who were scheduled for RAT. The diagnosis of
Graves disease was based on clinical assessment, con-firmed by a diffuse increased uptake on thyroidal scan
and increased total serum thyroxine (T4) concentration
or free thyroxine index (FTI), with or without thepresence of antithyroid antibodies. All patients hadsuppressed thyrotropin (TSH) concentrations. Theyhad had thyrotoxic symptoms for 4 ± 2 months beforepresentation. Blood samples were taken at diagnosisand at 1,2,4, and 6 months after RAT therapy. Healthylaboratory staff and police recruits of comparable age(n = 119) were used as the reference population forserum lipid and lipoprotein concentrations. Serum ali-quots were stored at -20#{176}Cuntil determination ofserum lipids in batches. All subjects gave informed
consent, and the study was approved by the EthicsCommittee of the University of Hong Kong.
Thyroid Function Tests and Lipid Profiles
Serum total T4 and T-uptake tests were measuredwith a fluorescence polarization immunoassay (AbbottLaboratories, Chicago, IL). Serum FTI was calculatedby dividing the serum T4 value by the T-uptake test.
Controls
119
40:7945 ± 8
62-154
76-1520.8-3.0
0.23-3.7
187 (17-1808) [244, 3661b
1.24 ± 0.75 [1.10, 1.38]
5.22 ± 1.00 (5.03, 5.39]1.24 ± 0.34 [1.18,1.31)
n
M:FAge, yearsT4, nmoVLFriT3, pmoVLTSH, mIU/LApo(a), lU/L
TG, mmoVLTC, mmoVLHDL-C,
mmoi/L
LDL-C,mmoVL
Apo Al, g/L
ApoB,g/L
Hyperthyroid subjects40
11:29
50 ± 10
254 ± 598
302 ± 8085.0 ± 2.08
<0.05
81(17-614) [90, 170]a,b
1.18 ± 0.44 [1.05, 1.33]4.07 ± 0.80 [3.80, 4.34r1.05 ± 0.33(0.96, 1.16]
3.40 ± 0.88 (3.24, 3.57] 2.47 ± 0.89 [2.19, 2.75r
1.46 ± 0.28 (1.41, 1.51)
1.13 ± 0.34(1.06,1.1912.07 ± 0.42 (1.84, 2.3r
0.66 ± 0.23 (0.59, o.73r
CLINICAL CHEMISTRY, Vol. 41, No. 2, 1995 227
Serum total triiodothyronine (T3) was determined byRIA (Amersham, Bucks., UK) and TSH by micropar-tide capture enzyme immunoassay (Abbott).
The Lp(a) concentration was determined by measur-
ing serum apo(a) by an IRMA (Pharmacia, Uppsala,Sweden). The assay is based on the direct sandwich
technique, in which two monoclonal antibodies aredirected against separate antigenic determinants on
the apo(a) molecule. The intraassay variations at low
and high apo(a) concentrations were 2.6% and 1.4%,respectively, and interassay variations were 4.8% and3.8%, respectively. Total cholesterol (TC) and triglyc-eride (TG) were determined enzymatically (BoehringerMannheim, Mannheim, Germany) on a Hitachi 717analyzer. High-density lipoprotein cholesterol (HDL-C)was quantified by the same enzymatic method afterprecipitation of very-low-density lipoprotein (VLDL)and LDL with phosphotungstic acid. LDL-C was calcu-lated according to the Friedewald equation. Apo A-Iand apo B were measured by rate nephelometric im-munoassay with the Beckman Array system (BeckmanInstruments, Brea, CA).
Statistical Methods
Two-sample t-test and confidence interval analyses
were used to compare results for patients and controls.Apo(a) concentrations were log-transformed beforeanalysis because of their skewed distribution. The
longitudinal analysis for each variable was achieved (a)by comparing the mean value at each time point withthe pretreatment mean value by the two-sample t-testand (b) by correlation analysis. Pearson’s correlationanalysis was performed between (i) the interval changefor each variable between time points after treatmentand (ii) the percentage interval change analogous to (i).
P <0.05 was considered significant.
Results
Thyroid Function Tests
All 40 patients had increased serum T4, V1’I, and T3and suppressed TSH, as shown in Table 1. After RAT,the percentage of subjects achieving normalization of
serum T4, FTI, and T3 at 1, 2, 4, and 6 months were57.5%, 77.5%, 92.5%, and 87.5%, respectively. Fivepatients showed a recurrence of hyperthyroidism at 6months after RAT after a period of euthyroidism. Fivesubjects had transient hypothyroidism, whereas twohad permanent hypothyroidism (defined biochemicallyby low T4, FTI, and T3 with increased TSH) lasting for>3 months. Apart from the few subjects who hadhypothyroidism, the TSH concentrations remainedsuppressed (<0.05 mIUfL) in 87.5%, 80%, 67.5%, and55% of the patients at 1, 2, 4, and 6 months, respec-tively.
Lipids
Compared with controls, thyrotoxic patients had
significantly lower amounts of apo(a), TC, LDL-C,HDL-C, and apo B, but higher apo A-I (Table 1). Serum
Table 1. Lipid concentrations (mean ± SD) ofhyperthyroid patients and controls.
#{149}Significant at p <0.001.b Geometric mean (and range); [95% confidence intervalsi.
TG was similar in the patients and the controls. Thereis no sex difference in apo(a) concentrations in bothgroups. The lipoprotein concentrations of the post-menopausal women were not statistically differentfrom those of premenopausal subjects, perhaps becauseof the small number of patients studied.
Because at each time point after treatment some of
the patients were still hyperthyroid, the data of the
subjects who were euthyroid at each time point were
analyzed and compared with the baseline values. WithRAT therapy, there was a progressive and significantincrease in apo(a), TC, LDL-C, and apo B concentra-tions (Table 2). Also, a small but significant increase in
TG was observed. The most significant percentageinterval changes of the lipid concentration from base-line value occurred at 2-4 months after therapy, whenthe serum thyroid hormones returned to the euthyroidrange. Euthyroidism was associated with rapid nor-malization of serum TC, LDL-C, and ape B, which was
achieved within 1 month after treatment. However,apo(a) concentrations took 4 months to normalize com-pletely, whereas HDL-C remained persistently lowerand apo A-I persistently higher even at 6 months afterRAJ (Table 2). Fig. 1 shows the changes in LDL and
apo(a) concentrations of the 28 patients who reachedeuthyroidism at 6 months. A delay in normalization inapo(a) as compared with LDL was evident. Fig. 2 showsthe typical reciprocal changes of apo(a) and LDL withFTI after treatment. Patient 1 had normalization of?I’I within 1 month after RAT. The effect on LDL ofachieving euthyroidism was evident earlier than that
on apo(a), although both showed a parallel increase.Patient 2 had transient hypothyroidism at 2-4 monthsafter RAT; associated with this was a surge in theapo(a) and LDL-C concentrations.
The relation between the serum thyroid hormones
and the lipid concentrations of all 40 patients duringtreatment of hyperthyroidism was also studied. Serum
n
FriApo(a), lU/LTG, mmoVLTC, mmoVLHDL-C, mmoVLLDL-C, mmoVLApe A-I,g/L
ApoB,g/L
0
40302 ± 80
81 (17-614r1.18 ± 0.444.07 ± 0.80
1.05 ± 0.33
2.47 ± 0.77
2.07 ± 0.420.66 ± 0.23
Geometric mean (and range).b Values in brackets are percentage changes compared with pretreatment values.c.dSignificant at c P <0.05 and d p <o.oi vs pretreatment value.
B
300
600 F-
0I I I
0 1 2 3Months
C
L J J
0 1 2 3Months
0I I
0 1 2 3Months
Fig. 1. Serial results of FTI (A), apo(a) (B), and LDL-C (C) of the 28I I I patients who reached euthyroidism at 6 months.4 5 6 Dotted lines in (B)and (C) represent the 95% confidence interval of the mean
of the controls.
228 CLINICAL CHEMISTRY, Vol. 41, No. 2,1995
Table 2. Changes in lipoprotein concentrations with time after treatment of hyperthyroid patients in patients whobecame euthyroid at each time point after RAI therapy.
lime, months
21122 ± 28
90 (17_515)8 [#{247}11.11k’1.47 ± 0.69 [+24.514.86 ± 0.99 [± 19.411.04 ± 0.39 [-1.0]3.15 ± 1.11 [+25.512.18 ± 0.49 [+5.3]0.83 ± 0.19 [+25.7]
2
25113 ± 33
90 (17-435r [+11.1]1.62 ± 0.69 [+37.3]5.30 ± 1.33 [+30.2]1.03 ± 0.41 (-2.01
3.52 ± 1.36 [+41.7]2.20 ± 0.47 (+6.2]0.90 ± 0.29 [+ 36.3]
4
28115 ± 28
148 (17-435r [#{247}82.7]
1.47 ± 0.66 [+ 24.5]5.55 ± 1.28 (+35.1]
0.97 ± 0.36 [-8.7]3.91 ± 1.39 [+58.3]2.14 ± 0.46 [#{247}3.0]
0.98 ± 0.30 [+48.5]
6
28112 ± 30
119(17-765r [+46.9]
1.56± 0.708[+32.2]
5.45 ± 1.63’ [+33.9)0.97 ± 0.24 [-8.7]3.81 ± 166d [+54.2]
2.24 ± 0.49 (+8.2]0.97 ± 028d [+46.9]
600
500
400
200
100 I-
7
6
5
-J
E-j3
-J
2
1
4 5 6
1000
-J
00.
800 F-
400
200
0
4 5 6
apo(a), TC, LDL-C, and apo B were negatively corre-lated with serum T4, FF1, and T3 and positively corre-lated with TSH during the transitional period fromhyperthyroidism to euthyroidism at 1,2, and 4 monthsafter RAT. Correlation coefficients of these measure-
ments at 4 months after RAT are shown in Table 3.Once euthyroidism had been achieved, no correlationwas apparent between the serum lipids and the thyroidhormones. Furthermore, interval differences (differ-ence of the value compared with that at the preceding
1
0 1 2Months
4 6
5
4OO 4-J
o3OO E
0
1200 29
100 1
0 0
800 8
700 7
600 6
500 5
-Ja400
300 3I 0
I -J
.200 2
100 1
0 0
I I I
1 2 4 6
Months
Fig. 2. Serial results for FT1, apo(a), and LDL-C in two patients afterRAI therapy.Top, patient with normalization of Ffl1 month after treatment. The effect ofachieving euthyroidism on LDL was evident earlier than for apo(a) althoughboth showed a parallel increase. Bottom, patient with transIent hypothyroid-ism after RAI. A surge in LDL was followed by a parallel change in apo(a)concentrations.
time point) of apo(a) were correlated with those of TSHbut not T4, FF1, or T3. Otherwise, the interval differ-ences of TG, TC, LDL-C, apo A-I, and apo B werecorrelated negatively with that of the thyroid hormonesand positively with that of TSH at each time pointduring the transitional period from hyperthyroidism toeuthyroidism. The correlation coefficients for theserelations at 2-4 months after RAT are shown in Table4. Similar correlations were observed between thepercentage changes (i.e., compared with the values atthe preceding time point) in these lipid measurementsand in the thyroid hormones (data not shown). These
L0
600 6
CLINICAL CHEMISTRY, Vol. 41, No. 2, 1995 229
Table 3. Correlations (r) between thyroid hormonesand lipoproteins during treatment of hyperthyroidism
at 4 months after RAI therapy.T4 FTI T3 TSH
Apo(a) -0.398 0.358 +0.47c
TG NS NS NS NSTC -0.42’ -0.398 +0.608LDL-C 041b +0.608HDL-C NS NS NS NSApoA-l NS NS NS NSApo B +0.57C
Significant at P <0.05, b P <0.01, and C P <0.005.NS, not significant.
correlations suggested parallel changes of lipid mea-surements and thyroid hormones with time after treat-ment of hyperthyroidism. Because only a few patientsdeveloped permanent hypothyroidism after treatment,we were unable to analyze the effect of hypothyroidism
on the lipid concentrations.
Discussion
The results demonstrate that thyroid hormones mod-ulate Lp(a) and other lipoprotein concentrations. Ourstudy design is unique in that serum lipoproteins andthyroid hormones were monitored serially duringtreatment of hyperthyroidism; thus, correlations couldbe properly assessed between the thyroid hormonesand the lipoproteins. We observed that, first, thyroidhormones were directly correlated with apo(a) duringtreatment of hyperthyroidism and, second, treatmentof hyperthyroidism was associated with parallel in-creases in apo(a), TC, LDL-C, and ape B that were ofsimilar magnitude (56.5%, 33.9%, 54.3%, and 46.9%,respectively). Conflicting results have been reportedwith regard to Lp(a) changes in thyroid disorders.Klausen et al. (11) and Spandrio et al. (12) reported alack of statistically significant influence of thyroidhormones on Lp(a) concentration, whereas de Bruin etal. (7) and Engler and Riesen (13) reported findingssimilar to our present study. The failure of detection ofcorrelation in some of the studies may be related to thesmall number of subjects studied and lack of serialmeasurements. The changes in TC, LDL-C, HDL-C,and apo B with hyperthyroidism were similar to datapublished previously (14). There was a small increasein the TG concentrations after treatment in our study.In contrast, Muls et al. found no significant change andnoted that variable effects had been found in earlierstudies (15).
The effect of thyroid hormones on LDL in humans is
well known. T4 increases LDL receptor-mediated clear-ance of LDL, thus lowering LDL concentrations, aneffect most evident in practice during treatment ofhypothyroidism (16). The way in which the observedreduction of Lp(a) by excess thyroid hormones comesabout is less clear. Plasma concentrations of Lp(a) vary
widely between euthyroid individuals and are to alarge extent genetically determined. In general, lower-
Month 1-2 Month 2-4
Table 4. CorrelatIons (r) of interval change of lipoproteins and thyroid hormones in 40 patientsperiod from hyperthyroidism to euthyroidism after RAI therapy.
during transitional
iFTI hiT3 iTSH if4 LiFTI hiT3 5TSH
A Apo(a) NS NS NS +0.47C NS NS NS +0.33a
A TG 041b NS -0.3 o.32 NS +0.32a
A TC +0.47C -0 42C -O 46C -0 348 +0 338
A LDL-C 0.37 +0.358 -0.298 #{247}0.358AHDL-C NS NS NS NS NS NS NS NSA Apo A-I NS #{247}042b 04b NS +0.40k’
A Apo B O.38 -0.42k’ NS +0.388 -0.398 NS -1-0.398
Significant at a p <0.05, b P <0.01, and c P <0.005.NS, not significant.
230 CLINICAL CHEMISTRY, Vol. 41, No. 2, 1995
molecular-mass isoforms of apo(a) are associated withhigher concentrations (17). Precisely how the differentalleles affect apo(a) production, association with LDLto form Lp(a), and catabolism of Lp(a) is not clear atthis time (18).
Transgenic mouse experiments indicate that human
apo(a) is expressed independently of LDL (19). Meta-bolic studies of the effect of nicotinic acid, which lowersLp(a) markedly, on Lp(a) turnover indicate that thisagent decreases the synthetic rate rather than increas-ing catabolism (20). Thus the effect of T4 could be viaan alteration of apo(a) and consequently Lp(a) produc-tion. Alternatively, because T4 alters LDL receptor
function, it could alter Lp(a) removal via receptors.Although the evidence is still somewhat conflicting, itseems unlikely, from studies in patients with LDLreceptor defects (i.e., familial hypercholesterolemia),that LDL receptors are involved in the catabolism ofLp(a) itself, although this might depend on the partic-ular LDL receptor mutation involved (21, 22). In ourstudy the effects of reduced T4 production after RAItreatment were in the same direction for LDL andLp(a) but the time courses were different. The effect onLDL was evident much earlier, whereas for Lp(a) itwas evident after 4 months. This is consistent with arapid effect on LDL receptor function affecting LDLconcentrations and a much slower effect, primarily onapo(a) production, affecting Lp(a) concentrations. Theknown dramatic effect of hydroxymethylglutaryl-CoAreductase inhibitors on LDL, via receptor upregulation,in the face of no change in apo(a) concentrations in thistreatment, also supports independent mechanisms ofLDL and apo(a) regulation. Further specific studies onapo(a) and Lp(a) kinetics in thyroid disease are neededto give the definitive answer.
We observed in this study that apo A-I concentra-tions were higher in untreated hyperthyroid subjectsthan in normal controls, and the concentrations re-mained high even after treatment. This differs fromthe finding of Muls et a!. (15), who reported that apoA-I concentrations in hyper- or hypothyroid patientswere similar to those in normal controls. Markedincreases in apo A-I concentrations had been observedin rats chronically treated with T4 injection (23). Bothliver and intestinal apo A-I production were increased
by T4, with the action being mediated at both transcrip-tional and posttranscriptional levels. Liver apo A-ImRNA was increased threefold with T4 injection, theincrease caused by an increased stability of nuclearapoA-I RNA precursors in chronic hyperthyroidism(23-25). The persistent increase of apo A-I after treat-ment in our patients may be related to the prolongedperiod of hyperthyroidism before RAT therapy. Sincehyperthyroidism is associated with decreased HDL-Cbut increased apo A-I concentrations, probably changethe structure and density of the HDL-C particle. This,however, needs verification by determination of HDL2
and HDL3 concentrations in these patients. Altered
lipoprotein composition together with decreased cho-
lesteryl ester transfer activity has been observed in
hypothyroidism (26). Change in the cholesterol/cho-lesteryl ester molar ratio in HDL was inversely related
to that of cholesteryl ester transfer activity during
treatment of hypothyroidism, which accounts forhigher HDL-C and free cholesterollcholesteryl esterratio in the hypothyroid state. This finding supportsour observation of decreased HDL-C in hyperthyroidpatients.
In conclusion, thyroid hormones modulate lipopro-
teins, particularly Lp(a). Amounts of apo(a) are low in
hyperthyroidism and revert to normal more slowly
than do LDL concentrations, suggesting an effect on
apo(a) production rather than LDL removal.
We thank Gloria Chan for nursing assistance, Monica Lam forsecretarial support, and J. D. Robinson for assistance in collectionof control samples.
References1. Berg K. Lp(a) lipoprotein: an overview. In: Scanu AM, ed.Lipoproteins(a). New York: Academic Press, 1990:1-23.2. Utermann G. The mysteries of lipoprotein (a) [Review]. Sci-ence 1989;246:904-10.3. McLean JW, Tomlinson JT, Kuang WJ, Eaton DL, Chen EY,Flees GM, et al. cDNA sequence of apolipoprotein (a) is homolo-gous to plasminogen. Nature 1987;300:132-7.4. Berg K. Genetics of atherogenic lipoprotein (a) and its relationto other atherosclerotic risk factors. Ann Intern Med 199 1;15:209-14.5. Hajjar KA, Gavish D, Breslow JL, Nachman RL. Lipoprotein(a) modulation of endothelial cell surface fibrinolysis and itspotential role in atherosclerosis. Nature 1989;339:303-10.6. Utermann GH, Menzel HJ, Kraft HG, Duba HC, Kemmler HG,
CLINICAL CHEMISTRY, Vol. 41, No. 2, 1995 231
Seitz C. Lp(a) glycoprotein phenotypes. Inheritance and relationto Lp(a)-lipoprotein concentrations in plasma. J Chin Invest1987;80:458-65.7. De Bruun TWA, van Barhngen H, van Linde-Sibenius, Trip M,van Vuurst de Vries ARR, Akveld MJ, et al. Lipoprotein (a) andapolipoprotein B plasma concentrations in hypothyroid, euthy-roid and hyperthyroid subjects. J Clin Endocrinol Metab 1993;76:121-6.S. Henriksson P, Angelin B, Bergiund L. Hormonal regulation ofserum Lp(a) levels: opposite effects after estrogen treatment andorchidectomy in males with prostatic carcinoma. J Clin Invest1992;89:1161-5.9. Eden S, Wiklund 0, Oscarsson J, Rosen T. Treatment ofgrowth hormone deficient adults results in a marked increase inLp(a) and HDL-cholesterol concentrations. Arterioscler Thromb1993;13:296-9.10. Lam KSL, Pang RWC, Janus ED, Kung AWC, Wang CCL.Serum apolipoprotein (a) correlates with growth hormone levelsin Chinese patients with acromegaly. Atherosclerosis 1993;104:183-8.11. Klausen IC, Nielsen FE, Hegedus L, Gerdes LU, Charles P.Faergeman 0. Treatment of hypothyroidism reduces low-densitylipoproteuns but not lipoprotein (a). Metabolism 1992;41:911-4.12. Spandrio S. Sleiman I, Scalvini T, Salvi A, Stefano OD,Pagliaini R, Balestrieri GP. Lipoprotein (a) levels in hypothyroidand hyperthyroid patients before and after treatment. HormMetab Res 1993;25:586-9.13. Engler H, Riesen WF. Effect of thyroid function on concen-tration of lipoprotein(a). Chin Chem 1993;39:2466-9.14. Hoch FL. Lipids and thyroid hormones [Reviewl. Prog LipidRes 1988;27:169-270.15. Muls E, Blaton V, Rosseneu M, Lesaifre E, Lamberigts G, deMoor P. Serum lipids and apolipoproteuns A-I, A-lI and B inhyperthyroidism before and after treatment. J Cliii EndocrinolMetab 1982;55:459-64.16. Thompson GR, Soutar AK, Spengel FA, Jadhav A, GaviganSJP, Myant NB. Defects of receptor-mediated low density lipopro-
teun catabolism in homozygous familial hypercholesterolaemia andhypothyroidism in vivo. Proc Natl Acad Sci USA 1981;78:2951-5.17. Boerwinkle E, Leffert CC, Lin J, Lackner C, Chiesa G, HobbsH. Apolipoprotein (a) accounts for greater than 90% of thevariation in plasma lipoprotein(a) concentrations. J Chin Invest1992;90:52-60.18. Wade DP. Lipoprotein (a). Curr Opin Lipidol 1993;4:244-9.19. Chiesa G, Hobbs H, Koschinsky ML, Lawn RM, Maika SD,Hammer RE. Reconstitution of lipoprotein (a) by infusion of humanlow density lipoprotein into transgenic mice expressing humanapolipoprotein (a). J Biol Chem 1992;267:24369-74.20. Seed M, O’Conner B, Perombolon N, O’Dennell M, ReaveleyD, Knight BL. The effect of nicotinic acid and acipimox onhipoprotein (a) concentration and turnover. Atherosclerosis 1993;101:61-8.21. Knight BL, Perombelon YFN, Soutar AK, Wade DP, Seed M.Catabolism of lipoprotein (a) in familial hypercholesterolaemicsubjects. Atherosclerosis 1991;87:227-37.22. LeitersdorfE, Friedlander Y, Bard JM, Fruchart JC, Eisen-berg 5, Stein Y. Diverse effect of ethnicity on plasma lipopro-
tein(a) levels in heterozygote patients with familial hypercholes-terolaemia. J Lipid Res 1991;32:1513-9.23. Steels B, van Tol A, Chan L, Will H, Verhoven G, Auwerx J.Alterations in thyroid status modulate apolipoprotein, hepatictriglyceride lipase, and low density lipoprotein receptor in rats.Endocrinology 1990;127:1144-52.24. Strobl W, Gorder NL, Lin-Lee YC, Gotto AM Jr, Patsch W.Role of thyroid hormones in apolipoprotein A-I gene expression inrat liver. J Cliii Invest 1990;85:659-67.25. Wilcox HG, Tso P, Heimberg M. Stimulation of intestinalsecretion of apolipoprotein A-I by triiodothyronine. Biochem Bio-phys Res Commun 1988;153:606-11.26. Dullaart RP, Hoogenberg K, Greener JE, Dikkeschei LD,Erkelens DW, Doorenbos H. The activity of cholesteryl estertransfer protein is decreased in hypothyroidism: a possible con-tribution to alterations in high-density lipoproteins. Eur J ChinInvest 1990;20:581-7.