CLINICAL SCIENCE SESSION (CSS)

*Kepaniteraan Klinik Senior/ G1A220080

**Pembimbing/ dr. Rr. Achirini WR, M. Ked (ClinPath), Sp. PK

Hubungan antara Serum Thyroxine Bebas dan Anemia pada Orang

Dewasa Eutiroid: Sebuah Studi Nasional

Oleh:
Hanna Saskia, S.Ked*

Pembimbing:
dr. Rr. Achirini WR, M. Ked (ClinPath), Sp. PK **

PENDIDIKAN PROFESI DOKTER

BAGIAN ILMU PATOLOGI KLINIK
RSUD RADEN MATTAHER JAMBI
FAKULTAS KEDOKTERAN DAN ILMU KESEHATAN
UNIVERSITAS JAMBI
2022
 LEMBAR PENGESAHAN
CLINICAL SCIENCE SESSION (CSS)

Hubungan antara Serum Thyroxine Bebas dan Anemia pada Orang

Dewasa Eutiroid: Sebuah Studi Nasional

Disusun Oleh:
Hanna Saskia, S.Ked
G1A220080

Telah diterima dan dipresentasikan sebagai salah satu tugas

Bagian Ilmu Patologi Klinik RSUD Raden Mattaher Jambi
Program Studi Pendidikan Kedokteran Universitas Jambi

Laporan ini telah diterima dan dipresentasikan

Jambi, September 2022

Pembimbing

dr. Rr. Achirini WR, M. Ked (ClinPath), Sp. PK

ii
 KATA PENGANTAR

Segala puji dan syukur penulis ucapkan kehadirat Tuhan Yang Maha Esa atas berkat
dan rahmat-Nya sehingga penulis dapat menyelesaikan Clinical Science Session yang berjudul
“Hubungan antara Serum Thyroxine Bebas dan Anemia pada Euthyroid Dewasa:
Sebuah Studi Nasional” sebagai kelengkapan persyaratan dalam mengikuti Kepaniteraan
Klinik Senior Bagian .
Penulis mengucapkan terima kasih kepada dr. Rr. Achirini WR, M. Ked (Clinpasth),
Sp, PK, yang telah meluangkan waktu dan pikirannya sebagai pembimbing sehingga penulis
dapat menyelesaikan laporan ini.
Penulis menyadari bahwa laporan ini masih jauh dari sempurna. Oleh karena itu kritik
dan saran yang membangun dari berbagai pihak sangat diharapkan. Selanjutnya, penulis
berharap semoga laporan ini dapat bermanfaat dan menambah ilmu bagi para pembaca.

Jambi, September 2022

Penulis,

Hanna Saskia

3
 Hubungan antara Serum Thyroxine Bebas dan Anemia pada Orang Dewasa Eutiroid:
Sebuah Studi Nasional

Mijin Kim1, Bo Hyun Kim1, Hyungi Lee2, Min Hee Jang1, Jeong Mi Kim1, Eun Heui Kim1,
Yun Kyung Jeon1, Sang Soo Kim1, In Joo Kim1

1
Department of Internal Medicine, 2Academic Research Organization, Clinical Trial Center,
Biomedical Research Institute, Pusan National University Hospital, Busan, Korea

Latar belakang:Studi tentang hubungan antara fungsi tiroid dan anemia pada kisaran
eutiroid masih langka. Kami bertujuan untuk mengevaluasi hubungan antara anemia dan
serum tiroksin bebas (fT4) dan thyrotropin (TSH) pada orang dewasa eutiroid.
Metode:Data dari 5.352 peserta berusia ≥19 tahun diperoleh dari Korea National Health
and Nutrition Examination Survey VI (2013 hingga 2015). Anemia didefinisikan sebagai
hemoglobin (Hb) <13 dan <12 g/dL untuk pria dan wanita, masing-masing.
Hasil:Secara keseluruhan, 6,1% peserta menderita anemia, dan lebih banyak wanita
(9,9%) menderita anemia daripada pria (2,8%,P<0,001). Dalam analisis multivariat, kadar
fT4 serum, tetapi tidak TSH, secara positif terkait dengan kadar Hb serum pada kedua
jenis kelamin (P<0,001, masing-masing). Kadar Hb serum berkurang secara linear di
seluruh kelompok kuartil fT4 serum yang menurun pada kedua jenis kelamin (P<0,001,
masing-masing). Setelah disesuaikan untuk faktor pembaur potensial, peserta dengan fT4
rendah-normal memiliki 4,4 (P=0,003) dan 2,8 kali (P<0,001) berisiko lebih tinggi untuk
anemia dibandingkan mereka dengan FT4 normal tinggi di antara pria dan wanita,
masing-masing. Ketika peserta dibagi menjadi dua kelompok pada usia 50 tahun, pada
peserta yang lebih muda, pria dan wanita dengan kuartil pertama berisiko lebih tinggi
terkena anemia dibandingkan pria dengan kuartil kedua (rasio odds [OR], 3,3;P=0,029)
dan wanita dengan kuartil keempat (OR, 3.2;P<0,001), masing-masing. Asosiasi ini tidak
diamati pada peserta yang lebih tua.
Kesimpulan:Hasil ini menunjukkan bahwa kadar fT4 serum yang rendah-normal
dikaitkan dengan kadar Hb serum yang lebih rendah dan risiko anemia yang lebih tinggi
pada orang dewasa eutiroid, terutama pada peserta yang lebih muda.
Kata kunci:Tes fungsi tiroid; Tirotropin; Tiroksin; Anemia; hemoglobin

4
PENGANTAR
Anemia adalah kondisi multifaktorial yang mempengaruhi 25% populasi dunia,
terhitung 8,8% dari total beban penyakit [1,2], dan telah diakui sebagai komplikasi
disfungsi tiroid [3-5]. Prevalensi anemia yang lebih tinggi hingga 57% diidentifikasi pada
pasien dengan hipertiroidisme atau hipotirodisme lebih nyata dibandingkan dengan
mereka yang eutiroidisme [4-6]. Kombinasi anemia dan disfungsi tiroid dapat dikaitkan
dengan penurunan kualitas hidup, hasil klinis yang lebih buruk, dan peningkatan biaya
perawatan kesehatan [7]. Karena prevalensinya yang tinggi dan keterkaitan yang erat, baik
anemia maupun penyakt tiroid merupakan masalah klinis yang signifikan yang sering
dihadapi oleh dokter [4,8,9].
Meskipun bukti yang mendukung hubungan antara anemia dan penyakit tiroid yang nyata,
studi yang menilai kondisi subklinis atau eutiroid terbatas [6,10-12]. Sebuah studi kohort besar
melaporkan bahwa bahkan pada pasien eutiroid, kadar tiroksin (fT4) bebas berhubungan positif
dengan indeks eritrosit seperti hemoglobin (Hb), hematokrit, dan jumlah eritrosit, dengan
korelasi negatif antara kadar tirotropin (TSH) dan kadar serum. saturasi besi dan transferin
[10]. Asosiasi ini memiliki dampak klinis dalam menentukan fungsi tiroid yang optimal dan
target terapi. Karena mayoritas orang yang berisiko anemia memiliki kadar hormon tiroid
dalam kisaran referensi, secara klinis penting untuk menentukan apakah fungsi tiroid juga
terkait dengan anemia dalam kisaran eutiroid. Dengan demikian, studi ini bertujuan untuk
mengevaluasi assosiasi antara anemia dan kadar serum fT4 dan TSH pada orang dewasa
eutiroid.

METODE
Sumber Data
Kami memperoleh data yang tersedia untuk umum dari Korean National Health and
Nutrition Examination Survey (KNHANES VI, 2013 hingga 2015), survei cross-sectional
nasional yang menggunakan multistage cluster sampling yang hierarkis untuk mewakili
populasi Korea. KNHANES telah dilakukan setiap tahun sejak tahun 1998 oleh Pusat
Pengendalian dan Pencegahan Penyakit Korea untuk menilai status kesehatan dan gizi
orang Korea [13]. Informed consent tertulis diperoleh dari semua peserta sebelum survei
dan protokol penelitian ini telah disetujui oleh Institutional Review Board Rumah Sakit
Universitas Nasional Pusan, Busan, Korea (No. 1908-022-082).

5
Populasi Studi
Secara keseluruhan, 6.037 peserta berusia≥19 tahun memiliki data tes fungsi tiroid dan kadar
Hb serum. Diantaranya, partisipan yang memiliki riwayat penyakit tiroid, mereka yang sedang
mengonsumsi obat-obatan yang dapat mempengaruhi fungsi tiroid (n=251), dan ibu hamil
(n=23) dikeluarkan. Selain itu, peserta dikeluarkan jika mereka memiliki TSH abnormal (<0,62
atau > 6,86 mIU/L) dan/atau kadar fT4 abnormal (<0,89 atau >1,76 ng/ mL) (n=411) [13].
Pengukuran Laboratorium
Kadar serum TSH, fT4, dan antibodi peroksidase anti-tiroid (TPOAb) diukur menurut protokol
yang disetujui, seperti yang dilaporkan sebelumnya [13]. Sampel darah diambil dari setiap
peserta di pagi hari setelah puasa minimal 8 jam, didinginkan dengan benar, dan dikirim ke
fasilitas pengujian. Kadar serum TSH, fT4, dan TPOAb diukur dengan immunoassay
elektrokimia (Roche Diagnostics, Mannhein, Jerman). Kadar TSH serum diukur dengan kit E-
TSH (Roche Diagnostics), dengan kisaran referensi 0,35 hingga 5,50 mIU/ L. Kadar fT4 serum
diukur dengan kit E-Free T4 (Roche Diagnostics), dengan kisaran referensi 0,89 hingga 1,76
ng/ mL. Tingkat TPOAb diukur dengan kit E-Anti-TPO (Roche Diagnostics), dengan kisaran
referensi <34,0 IU/mL pada manusia. Kadar serum TSH, fT4, dan TPOAb yang diukur
memenuhi kontrol kualitas dan program jaminan dari College of American Pathology [13].
Serum Hb diukur dengan XN-9000 (Sysmex, Kobe, Jepang). Kreatinin serum (Cr) diukur
menggunakan Jaffe atau metode kompensasi rate-blanked (Hitachi Automatic Analyzer 7600-
240, Hitachi, Tokyo, Jepang). Perkiraan laju filtrasi glomerulus (eGFR) dihitung dari
persamaan yang dikembangkan menggunakan studi Modifikasi Diet pada Penyakit Ginjal:
eGFR (mL/min/1,73 m2 )=175×(serum Cr mg/dL)–1.154×usia–0.203× (0,742 untuk wanita)
[14]. Konsentrasi yodium urin (UIC) diukur menggunakan spektrometri massa plasma yang
digabungkan secara induktif (ICP-MS, Perkin Elmer, Waltham, MA, USA) dengan yodium
sebagai standar (Anorganic Venture, Christiansburg, VA, USA). UIC disesuaikan
menggunakan kadar Cr untuk mengkompensasi tingkat ekskresi air yang tidak merata selama
pengumpulan spesimen urin.
Definisi
Anemia didefinisikan menurut kriteria Organisasi Kesehatan Dunia sebagai tingkat Hb
<13 g/dL untuk pria dan <12 g/dL untuk wanita tidak hamil [15]. Eutiroidisme
didefinisikan sebagai serum TSH dan fT4 dalam kisaran referensi normal masing-masing
0,62-6,86 mIU/L dan 0,89-1,76 ng/mL, dari data KNHANES VI [13].
Diabetes didefinisikan sebagai berikut: (1) didiagnosis diabetes oleh dokter; (2) sedang
mengonsumsi obat antidiabetes; atau (3) glukosa plasma puasa ≥126 mg/dL. Hipertensi

6
didiagnosis ketika tekanan darah sistolik ≥140 mmHg, tekanan darah diastolic ≥90
mmHg, atau jika pasien menggunakan obat antihipertensi. Peserta diklasifikasikan
memiliki penyakit ginjal kronis ketika eGFR <60 mL/min/1.73 m2 atau proteinuria ≥1+.
Rheumatoid arthritis, riwayat keganasan, merokok dan status ekonomi dan pembatasan
aktivitas fisik dinilai berdasarkan kuesioner yang dilaporkan sendiri.
Analisis Statistik
Analisis statistic dilakukan dengan menggunakan R program versi 3.5.1 (R Foundation for
Statistical Computing, Wina, Austria; http://www.R-project.org). Data dievaluasi
menggunakan bobot yang diberikan kepada individu yang dijadikan sampel untuk mewakili
orang Korea. Variabel kontinu disajikan sebagai rata-rata dengan kesalahan standar rata-
rata±kesalahan standar mean atau median dengan rentang interkuartil (IQR) dan dianalisis
menggunakan Student's t test. Variabel kategori disajikan sebagai frekuensi dengan persentase
dan dianalisis menggunakan uji chisquare. Kami melakukan analisis regresi linier berganda
untuk menilai temuan bahwa kadar fT4 serum secara independen terkait dengan Hb serum.
Kecenderungan serum Hb menurut kuartil serum fT4 dievaluasi dengan memeriksa P untuk
semua tren berdasarkan analisis regresi linier berganda. Analisis regresi logistik ganda
dilakukan untuk memperkirakan odds ratios (OR) dengan interval kepercayaan 95% (CI)
untuk anemia menurut kategori kuartil serum fT4. P<0,05 dianggap signifikan secara
statistik.

HASIL
Karakteristik Dasar
Setelah mengecualikan 685 peserta, 5.352 (2.747 laki-laki dan 2.605 perempuan) dimasukkan
dalam analisis ini. Karakteristik dasar peserta tercantum dalam Tabel 1. Usia rata-rata adalah
44,0 ±0,218 tahun, dan pria lebih muda dari wanita (P= 0,024). Secara keseluruhan, 6,1%
peserta menderita anemia, dan lebih banyak perempuan (9,9%) mengalami anemia
dibandingkan laki-laki (2,8%,P<0,001). Rata-rata kadar serum Hb dan fT4 masing-masing
adalah 14,4 g/dL dan 1,24 ng/dL, dan 16,7% dan 7,6% lebih tinggi pada pria daripada wanita
(P<0,001 danP<0,001), sedangkan kadar TSH serum lebih rendah 8,9% (P<0,001). Prevalensi
keseluruhan TPOAb positif adalah 5,2%, dan proporsi TPOAb positif lebih tinggi pada wanita
dibandingkan pria (7,3% vs 3,3%,P <0,001). UIC median adalah 280,7 g/L (IQR, 150,4 hingga
650,5 g/ L), dan tidak ada perbedaan UIC antara pria dan wanita.
Persentase peserta dengan kebiasaan merokok di masa lalu atau saat ini, dan mereka yang
menderita diabetes dan hipertensi secara signifikan lebih tinggi pada pria daripada wanita

7
 (P<0,001, masing- masing). Sementara wanita menunjukkan prevalensi rheumatoid arthritis
yang lebih tinggi dibandingkan dengan pria (P<0,001). Proporsi peserta dengan CKD,
keganasan, dan keterbatasan aktivitas masing-masing adalah 2,4%, 1,9%, dan 5,5%, tanpa
perbedaan yang signifikan antara pria dan Wanita.

Tabel 1.Karakteristik Dasar Subyek Eutiroid (≥19 Tahun) Yang Menjalani Tes Fungsi Tiroid dan Pengukuran
Kadar

Jumlah (n=5.352) Pria (n=2,747) Wanita (n=2,605)

44.0±0.218 43.7±0.270 44.5±0.295 0,024
334 (6.1) 83 (2.8) 251 (9.9) <0.001
14.4±0,023 15.4±0,026 13.2±0,025 <0.001
1.24±0,003 1.28±0,004 1.19±0,004 <0.001
Serum TSH, mIU/L 2.46±0,019 2.36±0,026 2.57±0,028 <0.001
94.5±0,292 94.3±0.354 94.7±0,444 0,439
294 (5.2) 92 (3.3) 202 (7.3) <0.001
280,7 (150.4–650.5) 272.1 (156.4-609.9) 273.9 (139.1–635.6) 0,091

1.030 (19,5) 883 (31.8) 147 (5.4) <0.001

1.248 (23.6) 1.080 (39.2) 168 (5.8)

469 (8.9) 290 (10.3) 179 (7.3) <0.001

1.224 (22.9) 742 (25.6) 482 (19.7) <0.001
119 (2.4) 72 (2.6) 47 (2.2) 0,356
70 (1.4) 23 (0.7) 47 (2.3) <0.001
108 (1.9) 54 (1.6) 54 (2.1) 0.297

715 (12.8) 327 (11,0) 388 (15.0) <0.001

1.424 (26.4) 708 (26.0) 716 (26,8)

1.583 (29.9) 827 (30.6) 756 (29.1)

1.630 (30,9) 885 (32,4) 745 (29.2)
297 (5.5) 137 (4,9) 160 (6.1) 0,071

8
Hubungan antara Hb serum dan serum fT4
Analisis regresi linier berganda dilakukan untuk mengevaluasi faktor-faktor yang secara
independen terkait dengan kadar Hb serum (Tabel 2). Pada pria, Hb serum dikaitkan secara
positif dengan fT4 (β = 0,865,P<0,001), merokok saat ini (β=0,201,P<0,001) dan hipertensi
(β=0,135,P=0,038), sedangkan berbanding terbalik dengan umur (β= – 0,020,P<0,001), eGFR
(β=–0,005,P=0,005), CKD (β= – 0,478,P=0,019), rheumatoid arthritis (β=–0,587,P=0,010),
dan keganasan (β=–1.259,P<0,001). Pada wanita, serum fT4 (β=0,842,P<0,001), merokok saat
ini (β=0,431,P<0,001), dan hipertensi (β=0,203,P=0,003) berhubungan positif dengan Hb
serum, dan eGFR berbanding terbalik dengan Hb serum (β=-0,005,P=0,004). Tidak ada
korelasi independen antara TSH serum dan Hb serum pada keduanya, baik pria (P= 0,067) dan
wanita ( P=0,247) (Tabel Tambahan S1).
Kadar fT4 serum dibagi menjadi empat kelompok berdasarkan kuartil pada setiap jenis kelamin
untuk menguji apakah hubungan antara fT4 serum dan kadar Hb serum memiliki efek ambang
batas. Pada kedua jenis kelamin, kadar Hb serum menurun secara linear di seluruh kelompok
kuartil fT4 serum yang menurun (P untuk tren <0,001 pada pria dan P untuk tren <0,001 pada
wanita) (Gbr. 1).
Risiko berkembangnya anemia menurut kadar fT4 serum
Analisis multivariat dilakukan untuk mengidentifikasi OR untuk anemia menurut kelompok
kuartil serum fT4 (Tabel 3). Pada pria, prevalensi anemia adalah 5,6%, 3,2%, 1,7%, dan 1,0%
di kuartil 1, 2, 3, dan 4. Dalam model yang tidak disesuaikan, OR untuk anemia adalah 6,0 kali
lebih tinggi di antara peserta pada kuartil terendah dibanding pada kuartil tertinggi (95% CI,
2,42 hingga 14,68;P<0,001). Setelah disesuaikan dengan usia, asupan yodium, TPOAb positif,
dan status merokok, kelompok kuartil terendah memiliki risiko anemia yang jauh lebih tinggi
dibandingkan dengan kelompok kuartil tertinggi (OR, 3,9; 95% CI, 1,46-10,24; P=0,007 dalam
model 1). Pada model 2, yang disesuaikan dengan penyakit penyerta, tingkat pendapatan, dan
pembatasan aktivitas pada Model 1, OR untuk anemia masih lebih tinggi secara signifikan pada
kelompok kuartil terendah dibandingkan pada kelompok kuartil tertinggi (OR, 4,4; 95% CI,
1,66 hingga 11.73;P=0,003).
Pada wanita, 15,4%, 8,7%, 8,7%, dan 7,0% di kuartil 1, 2, 3, dan 4, masing-masing, menderita
anemia. Dalam model yang tidak disesuaikan, peserta dengan kuartil terendah memiliki OR
2,4 kali lebih tinggi untuk anemia dibandingkan mereka dengan kuartil tertinggi (95% CI, 1,58-
3,70;P< 0,001). Dengan analisis multivariat yang disesuaikan dengan usia, asupan yodium,
TPOAb positif, status merokok, penyakit penyerta, tingkat pendapatan, dan pembatasan
aktivitas, peserta dengan kuartil terendah berada pada risiko anemia yang lebih besar secara

9
 signifikan dibandingkan dengan kuartil tertinggi ([OR, 2,8; 95% CI, 1,78 hingga 4,48;P<0,001
dalam model 1] dan [OR, 2,8; 95% CI, 1,78 hingga 4,51;P<0,001 dalam model 2]).

Tabel 2. Analisis Regresi Linier Berganda untuk Menentukan Kovariat manakah termasuk kadar fT4 serum
yang secara independent terkait dengan kadar Hb serum

Variabel terikat; Hb serum

b SE

0.865 0.154 <0.001

0,002 <0.001
eGFR 0,002 0,005
UIC <0.001 0,096
TPOAb (>34 IU/mL) 0.107 0,989
Merokok Saat Ini 0,201 0,044 <0.001

0,093 0,678
0.135 0,065 0,038
0,202 0,019
0.228 0,010
0.223 <0.001
0,001 0,025 0,978
0,126 0,066

0.842 0.184 <0.001

0.003 0.002 0.139
1.24±0,003 1.28±0,004 <0.001
2.46±0,019 2.36±0,026 <0.001
94.5±0,292 94.3±0.354 0,439
294 (5.2) 92 (3.3) <0.001

469 (8.9) 290 (10.3) <0.001

1.224 (22.9) 742 (25.6) <0.001
119 (2.4) 72 (2.6) 0,356
70 (1.4) 23 (0.7) <0.001
108 (1.9) 54 (1.6) 0.297
715 (12.8) 327 (11,0) <0.001
297 (5.5) 137 (4,9) 0,071

Nilai dinyatakan sebagai disesuaikan untuk semua variabel lain dalam tabel. fT4, tiroksin bebas; Hb,
hemoglobin; SE, standard error; eGFR, perkiraan laju filtrasi glomerulus; UIC, konsentrasi yodium
urin; TPOAb, antibodi anti-tiroid peroksidase.

10
Gambar 1.Kadar hemoglobin (Hb) serum menurut kuartil tiroksin (T4) serum bebas pada
(A) pria eutiroid dan (B) wanita. Nilai disajikan sebagai perkiraan rata-rata dengan
kesalahan standar rata-rata.

Hubungan antara serum fT4 dan anemia pada peserta yang lebih muda dan lebih tua
Prevalensi anemia pada setiap kelompok umur 19-29, 30-39, 40-49, 50-59, 60-69, dan≥70
adalah 1,1%, 0,5%, 1,8%, 3,8%, 7,5%, dan 10,7% pada pria, masing-masing, dan 8,9%, 15,4%,
14,6%, 4,5%, 5,9%, dan 10,6% pada wanita, masing-masing (Tabel Tambahan S2). Kami
selanjutnya menganalisis OR untuk anemia menurut kelompok kuartil serum fT4 pada peserta
yang lebih muda dan lebih tua (Tabel 4). Pada partisipan dengan usia <50 tahun, prevalensi

11
anemia kuartil 1, 2, 3, dan 4 adalah 3,2%, 1,6%, 0%, dan 0% pada pria, masing-masing, dan
20,3%, 12,9%, 10,7%, dan 7,7% pada wanita. Dalam analisis multivariat, pria yang lebih muda
di kuartil pertama memiliki risiko anemia yang lebih tinggi secara signifikan dibandingkan
dengan mereka yang berada di kuartil kedua (OR, 3,3; 95% CI, 1,13 hingga 9,75;P=0,029).
Juga, wanita yang lebih muda dengan fT4 normal rendah memiliki OR 3,2 kali lebih tinggi
dibandingkan mereka dengan fT4 normal tinggi (95% CI, 1,89 hingga 5,49;P<0,001). Namun,
asosiasi ini tidak diamati pada pria dan wanita yang lebih tua.

Kuartil fT4 Prevalensi, %a ORb(95% CI) P value

Lebih Muda (usia <50 thn)
Pria (n=1,658)
Q1 (0,89–1,18) 3.2 3.3 (1.13–9.75) 0,029
Q2 (1.19–1.30) 1.6 1.0 (referensi)
Q3 (1.31–1.41) 0 NA <0.001
Q4 (1.42–1.76) 0 NA <0.001
Wanita (n=1,619)
Q1 (0,89–1,09) 20.3 3.2 (1.89–5.49) <0.001
Q2 (1.10–1.19) 12.9 1,5 (0,83–2,64) 0,186
Q3 (1.20–1.30) 10.7 1.1 (0.61–2.05) 0,725
Q4 (1.31–1.76) 7.7 1.0 (referensi)
Lebih tua (usia≥50 tahun)
Pria (n=1.089
7.5
Q1 (0,89–1,10) 2.0 (0.81–5.19) 0,131
Q2 (1.11–1.20) 5.9 1,8 (0,74–4,42) 0.198
Q3 (1.21–1.31) 5.6 1,4 (0,53–3,71) 0,503
Q4 (1.32–1.76) 4.1 1.0 (referensi)
Wanita (n=986)
Q1 (0,89–1,07) 8.4 1,8 (0,74–4,60) 0.193
Q2 (1.08–1.16) 4.1 0,8 (0,29–2,39) 0,732
Q3 (1.17–1.26) 5.1 1.1 (0.37–3.19) 0.859
Q4 (1.27–1.76) 4.4 1.0 (referensi)

12
DISKUSI
Dalam studi cross-sectional nasional ini, kami menunjukkan hubungan yang signifikan antara
kadar fT4 serum dan anemia pada pria dan wanita eutiroid. Prevalensi anemia secara signifikan
lebih tinggi pada wanita dibandingkan pada pria. Setelah disesuaikan untuk parameter
pengganggu utama, kadar fT4 serum, tetapi tidak TSH, memiliki hubungan positif yang
independen dengan kadar Hb serum pada kedua jenis kelamin. Pada kedua jenis kelamin, kadar
Hb serum menurun secara linear dengan penurunan kelompok kuartil fT4 serum, dan peserta
dengan fT4 normal rendah memiliki risiko anemia yang lebih tinggi secara signifikan
dibandingkan dengan fT4 normal tinggi. Ketika peserta dibagi menjadi kelompok yang lebih
muda dan lebih tua pada usia 50 tahun, pada peserta eutiroid yang lebih muda, pria dan wanita
dengan kuartil pertama memiliki risiko anemia yang lebih tinggi daripada pria dengan kuartil
kedua dan wanita dengan kuartil keempat, masing-masing.
Namun, Meskipun bukti mendukung hubungan antara penyakit tiroid yang nyata atau subklinis
dan anemia, penelitian yang menargetkan kondisi eutiroid terbatas [6,10-12], meskipun
hubungan antara hormon tiroid dan indeks eritrosit pada subjek eutiroid telah dilaporkan akhir-
akhir ini [10-12]. Menurut sebuah penelitian yang dilakukan pada 1.011 peserta eutiroid yang
lebih tua, kadar fT4 memiliki hubungan positif linier dengan kadar Hb dan indeks eritrosit [10].
Demikian pula, berdasarkan kohort dari 708 peserta eutiroid yang lebih tua, fT4 secara
signifikan terkait dengan kadar Hb dan parameter eritrosit, sedangkan TSH tidak, setelah
disesuaikan untuk parameter pengganggu utama [11]. Hubungan antara fungsi tiroid dan indeks
eritrosit dievaluasi lebih lanjut oleh Lippi et al. [12] dalam kohort retrospektif dari 1.050
peserta yang lebih tua eutiroid. Meskipun penelitian tersebut gagal untuk mengidentifikasi
hubungan antara fungsi tiroid dan Hb atau hematokrit, itu adalah yang pertama melaporkan
hubungan positif yang signifikan antara lebar distribusi sel darah merah dan nilai hormon tiroid
[12]. Konsisten dengan penelitian sebelumnya, kami juga menemukan bahwa kadar fT4 serum,
tetapi bukan kadar TSH, memiliki pengaruh yang independent hubungan positif dengan kadar
Hb serum pada kedua jenis kelamin setelah disesuaikan dengan usia, UIC, TPOAb positif,
status merokok, komorbiditas, status ekonomi, dan pembatasan aktivitas.
Hormon tiroid memiliki peran penting dalam hematopoiesis, terutama dalam eritropoiesis [8].
Namun, mekanisme dimana hormon tiroid memodulasi produksi sel darah merah tidak
sepenuhnya dipahami [3,16-18]. Tampaknya ada beberapa mekanisme dimana hormon tiroid
mengatur produksi sel darah merah, termasuk produksi dan reaktivitas eritropoietin, dan
mempengaruhi transportasi dan pemanfaatan besi [3,16-18]. Tindakan ini dapat dimediasi baik
melalui TSH melalui reseptor TSH yang diusulkan untuk hadir dalam eritrosit dan prekursor

13
eritrosit, atau melalui efek metabolik yang dimediasi oleh hormon tiroid [8,11,19]. Dalam
penelitian ini, fT4, bukan TSH, secara signifikan terkait dengan kadar Hb dan peningkatan
risiko anemia, menunjukkan bahwa dampak eritropoiesis hormon tiroid dikendalikan oleh
fungsi langsung atau tidak langsung fT4 daripada mediasi reseptor TSH, terutama pada
individu eutiroid. Dengan demikian, fT4 mungkin merupakan penanda fisiologis status hormon
tiroid yang lebih tepat daripada TSH pada populasi eutiroid.
Kami menggunakan analisis multivariat untuk menyesuaikan faktor pembaur potensial lainnya
termasuk usia, status asupan yodium, autoimunitas, status merokok, adanya komorbiditas
(diabetes, hipertensi, CKD, rheumatoid arthritis, dan keganasan), status ekonomi, dan
aktivitas fisik. Faktor-faktor ini berhubungan dengan penyakit tiroid dan anemia. Sebenarnya,
prevalensi penyakit tiroid dan anemia meningkat seiring bertambahnya usia [20,21]. Pasien
dengan penyakit tiroid autoimun berada pada risiko yang lebih tinggi untuk mengembangkan
berbagai bentuk anemia, dan anemia pernisiosa sering muncul bersamaan dengan penyakit
tiroid autoimun [9]. Yodium juga diperlukan untuk sintesis hormon tiroid, dan asupan yodium
yang tinggi berhubungan dengan tiroiditis autoimun dan hipotiroidisme [22]. Anemia adalah
komplikasi CKD dan fungsi ginjal yang terkenal, direpresentasikan sebagai eGFR, juga terkait
dengan fungsi tiroid [23,24]. Merokok, diabetes, hipertensi, rheumatoid arthritis, keganasan,
dan status sosial ekonomi juga berhubungan dengan perkembangan anemia [2,25-28]. Dalam
penelitian ini, kami mencoba untuk memilih perancu yang tepat untuk analisis multivariat
dengan meminimalkan pengaruh faktor lain selain fungsi tiroid dan anemia.
Beban anemia juga berbeda menurut usia dan jenis kelamin. Prevalensi anemia jauh lebih
tinggi pada wanita dibandingkan pria di bawah 50 tahun, sedangkan pada pria dan wanita di
atas 50 tahun sama. Penyebab utama anemia pada dewasa muda berkaitan dengan kondisi
ginekologi yang sering muncul sebagai defisiensi besi [2]. Ketika kami membagi peserta
menjadi dua kelompok menggunakan batas usia 50 tahun, 87% wanita mengalami menopause.
Penelitian ini melakukan analisis subkelompok berdasarkan usia 50 tahun dengan pengecualian
ibu hamil untuk meminimalkan pengaruh faktor ginekologi yang mempengaruhi anemia. Pada
orang dewasa yang lebih tua, insufisiensi ginjal, penyakit kronis, keganasan, dan kekurangan
gizi adalah kontributor terbesar beban anemia total pada kedua jenis kelamin [29]. Dalam
penelitian kami, hubungan antara serum fT4 dan anemia tidak diamati pada peserta yang lebih
tua. Temuan ini menunjukkan bahwa banyak faktor selain serum fT4 dapat berkontribusi pada
perkembangan anemia pada orang tua.
Penelitian ini memiliki beberapa keterbatasan. Pertama, desain cross-sectional tidak memberi
kita informasi untuk menarik kesimpulan tentang kausalitas. Kedua, kami tidak mengukur

14
feritin dan transferin tambahan untuk membedakan anemia defisiensi besi dan anemia penyakit
kronis. Ketiga, informasi tentang faktor pengganggu nutrisi termasuk suplemen zat besi atau
vitamin tidak tersedia untuk sebagian besar peserta. Keempat, tidak ada peserta dengan anemia
di antara laki-laki muda dengan FT4 tinggi-normal. Meskipun demikian, sepengetahuan kami,
ini adalah analisis nasional pertama yang mengevaluasi risiko berkembangnya anemia menurut
fungsi tiroid pada orang dewasa eutiroid.
Kesimpulannya, data yang dikumpulkan dari survei cross-sectional yang representatif secara
nasional menunjukkan bahwa kadar fT4 serum normal yang rendah terkait dengan kadar Hb
serum yang lebih rendah dan risiko anemia yang lebih tinggi pada orang dewasa eutiroid,
terutama pada orang dewasa yang lebih muda. Temuan ini menunjukkan bahwa rentang
referensi fT4 mungkin tidak sesuai dari sudut pandang anemia dan bahwa menghindari kadar
serum fT4 normal yang rendah dapat membantu menghindari anemia pada pria dan wanita.
Studi lebih lanjut diperlukan untuk menentukan kelayakan peningkatan parameter hematologi
melalui pengobatan levothyroxine pada subjek dengan anemia dan serum fT4 normal rendah.

KONFLIK KEPENTINGAN
Tidak ada potensi konflik kepentingan yang relevan dengan artikel ini yang dilaporkan.

15
REFERENSI
1. de Benoist B, McLean E, Egli I, Cogswell M. WHO global database on anaemia. Geneva:
World Health Organization; 2008.
2. Kassebaum NJ, Jasrasaria R, Naghavi M, Wulf SK, Johns N, Lozano R, et al. A systematic
analysis of global anemia bur- den from 1990 to 2010. Blood 2014;123:615-24.
3. Fein HG, Rivlin RS. Anemia in thyroid diseases. Med Clin North Am 1975;59:1133-45.
4. M’Rabet-Bensalah K, Aubert CE, Coslovsky M, Collet TH, Baumgartner C, den Elzen WP,
et al. Thyroid dysfunction and anaemia in a large population-based study. Clin Endo- crinol
(Oxf) 2016;84:627-31.
5. Lippi G, Montagnana M, Salvagno GL, Guidi GC. Should women with abnormal serum
thyroid stimulating hormone undergo screening for anemia? Arch Pathol Lab Med 2008;
132:321-2.
6. Omar S, Hadj Taeib S, Kanoun F, Hammami MB, Kamoun S, Ben Romdhane N, et al.
Erythrocyte abnormalities in thy- roid dysfunction. Tunis Med 2010;88:783-8.
7. Wopereis DM, Du Puy RS, van Heemst D, Walsh JP, Brem- ner A, Bakker SJL, et al. The
relation between thyroid func- tion and anemia: a pooled analysis of individual participant data.
J Clin Endocrinol Metab 2018;103:3658-67.
8. Kawa MP, Grymula K, Paczkowska E, Baskiewicz-Masiuk M, Dabkowska E, Koziolek M,
et al. Clinical relevance of thyroid dysfunction in human haematopoiesis: biochemical and
molecular studies. Eur J Endocrinol 2010;162:295-305.
9. Szczepanek-Parulska E, Hernik A, Ruchała M. Anemia in thyroid diseases. Pol Arch Intern
Med 2017;127:352-60.
10. Bremner AP, Feddema P, Joske DJ, Leedman PJ, O’Leary PC, Olynyk JK, et al. Significant
association between thy- roid hormones and erythrocyte indices in euthyroid subjects. Clin
Endocrinol (Oxf) 2012;76:304-11.
11. Schindhelm RK, ten Boekel E, Heima NE, van Schoor NM, Simsek S. Thyroid hormones
and erythrocyte indices in a co- hort of euthyroid older subjects. Eur J Intern Med 2013;24:
241-4.
12. Lippi G, Danese E, Montagnana M. Thyroid hormone levels are associated with
anisocytosis in a cohort of euthyroid older outpatients. Eur J Intern Med 2014;25:e4-5.
13. KimWG,KimWB,WooG,KimH,ChoY,KimTY,etal. Thyroid stimulating hormone
reference range and preva- lence of thyroid dysfunction in the Korean population: Ko- rea
National Health and Nutrition Examination Survey 2013 to 2015. Endocrinol Metab (Seoul)
2017;32:106-14.

16
14. Levey AS, Coresh J, Greene T, Stevens LA, Zhang YL, Hendriksen S, et al. Using
standardized serum creatinine values in the modification of diet in renal disease study equation
for estimating glomerular filtration rate. Ann Intern Med 2006;145:247-54.
15. Beutler E, Waalen J. The definition of anemia: what is the lower limit of normal of the
blood hemoglobin concentra- tion? Blood 2006;107:1747-50.
16. Touam M, Guery B, Goupy C, Menoyo V, Drueke T. Hypo- thyroidism and resistance to
human recombinant erythropoi- etin. Nephrol Dial Transplant 2004;19:1020-1.
17. Christ-Crain M, Meier C, Huber P, Zulewski H, Staub JJ, Muller B. Effect of restoration
of euthyroidism on peripheral blood cells and erythropoietin in women with subclinical
hypothyroidism. Hormones (Athens) 2003;2:237-42.
18. Cinemre H, Bilir C, Gokosmanoglu F, Bahcebasi T. Hema- tologic effects of levothyroxine
in iron-deficient subclinical hypothyroid patients: a randomized, double-blind, con- trolled
study. J Clin Endocrinol Metab 2009;94:151-6.
19. Perrin MC, Blanchet JP, Mouchiroud G. Modulation of hu- man and mouse erythropoiesis
by thyroid hormone and reti- noic acid: evidence for specific effects at different steps of the
erythroid pathway. Hematol Cell Ther 1997;39:19-26.
20. Gaskell H, Derry S, Andrew Moore R, McQuay HJ. Preva- lence of anaemia in older
persons: systematic review. BMC Geriatr 2008;8:1.
21. Cooper DS, Biondi B. Subclinical thyroid disease. Lancet 2012;379:1142-54.
22. Jeon MJ, Kim WG, Kwon H, Kim M, Park S, Oh HS, et al. Excessive iodine intake and
thyrotropin reference interval: data from the Korean National Health and Nutrition Exami-
nation Survey. Thyroid 2017;27:967-72.
23. Wolf G, Muller N, Hunger-Battefeld W, Kloos C, Muller UA. Hemoglobin concentrations
are closely linked to renal function in patients with type 1 or 2 diabetes mellitus. Kid- ney
Blood Press Res 2008;31:313-21.
24. Sun MT, Hsiao FC, Su SC, Pei D, Hung YJ. Thyrotropin as an independent factor of renal
function and chronic kidney disease in normoglycemic euthyroid adults. Endocr Res
2012;37:110-6.
25. Nordenberg D, Yip R, Binkin NJ. The effect of cigarette smoking on hemoglobin levels
and anemia screening. JAMA 1990;264:1556-9.
26. Wolfe F, Michaud K. Anemia and renal function in patients with rheumatoid arthritis. J
Rheumatol 2006;33:1516-22.
27. Gaspar BL, Sharma P, Das R. Anemia in malignancies: patho- genetic and diagnostic
considerations. Hematology 2015;20: 18-25.

17
28. Styszynski A, Mossakowska M, Chudek J, Puzianowska- Kuznicka M, Klich-RaczkaA,
Neumann-Podczaska A, et al. Prevalence of anemia in relation to socio-economic factors in
elderly Polish population: the results of PolSenior study. J Physiol Pharmacol 2018;69:75-81.
29. Guralnik JM, Eisenstaedt RS, Ferrucci L, Klein HG, Wood- man RC. Prevalence of anemia
in persons 65 years and old- er in the United States: evidence for a high rate of unex- plained
anemia. Blood 2004;104:2263-8.

18
 Original Endocrinol Metab 2020;35:106-114
https://doi.org/10.3803/EnM.2020.35.1.106
Article pISSN 2093-596X · eISSN 2093-5978

Association between Serum Free Thyroxine and Anemia in

Euthyroid Adults: A Nationwide Study
Mijin Kim1, Bo Hyun Kim1, Hyungi Lee2, Min Hee Jang1, Jeong Mi Kim1, Eun Heui Kim1, Yun Kyung Jeon1,
Sang Soo Kim1, In Joo Kim1
1
Department of Internal Medicine, 2Academic Research Organization, Clinical Trial Center, Biomedical Research Institute,
Pusan National University Hospital, Busan, Korea

Background: Studies on the relationship between thyroid function and anemia in the euthyroid range are scarce. We aimed to evalu-
ate the association between anemia and serum free thyroxine (fT4) and thyrotropin (TSH) in euthyroid adults.
Methods: Data on 5,352 participants aged ≥19 years were obtained from the Korea National Health and Nutrition Examination
Survey VI (2013 to 2015). Anemia was defined as hemoglobin (Hb) <13 and <12 g/dL for men and women, respectively.
Results: Overall, 6.1% of participants had anemia, and more women (9.9%) had anemia than men (2.8%, P<0.001). In multivariate
analysis, serum fT4 levels, but not TSH, were positively associated with serum Hb levels in both sexes (P<0.001, each). Serum Hb
levels linearly reduced across decreasing serum fT4 quartile groups in both sexes (P<0.001, each). After adjusting for potential con-
founding factors, participants with low-normal fT4 had 4.4 (P=0.003) and 2.8 times (P<0.001) higher risk for anemia than those
with high-normal fT4 among men and women, respectively. When participants were divided into two groups at 50 years of age, in
younger participants, men and women with the first quartile were at higher risk of anemia than men with the second quartile (odds
ratio [OR], 3.3; P=0.029) and women with the forth quartile (OR, 3.2; P<0.001), respectively. This association was not observed in
older participants.
Conclusion: These results suggest that a low-normal level of serum fT4 was associated with a lower serum Hb level and a higher
risk of anemia in euthyroid adults, especially in younger participants.

Keywords: Thyroid function tests; Thyrotropin; Thyroxine; Anemia; Hemoglobins

INTRODUCTION those with overt hypothyroidism than in those with euthyroid-

Anemia is a multifactorial condition that affects 25% of the
world’s population, accounting for 8.8% of the total burden of
the disease [1,2], and has been recognized as a complication of
thyroid dysfunction [3-5]. A higher anemia prevalence of up to
57% was identified in patients with overt hyperthyroidism or in
ism [4,6]. The combination of anemia and thyroid dysfunction
can be associated with diminished quality of life, worse clinical
outcome, and increased health care cost [7]. Because of their
high prevalence and close interrelation, both anemia and thyroid
disease are significant clinical problems often encountered by
physicians [4,8,9].

Received: 4 September 2019, Revised: 25 November 2019,
Accepted: 23 December 2019
Corresponding author: Bo Hyun Kim
Division of Endocrinology and Metabolism, Department of Internal Medicine,
Biomedical Research Institute, Pusan National University Hospital, 179 Gudeok-
ro, Seo-gu, Busan 49241, Korea
Tel: +82-51-240-7236, Fax: +82-51-254-3237, E-mail: pons71@pusan.ac.kr
Copyright © 2020 Korean Endocrine Society
This is an Open Access article distributed under the terms of the Creative Com­
mons Attribution Non-Commercial License (https://creativecommons.org/
licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribu­
tion, and reproduction in any medium, provided the original work is properly
cited.

106 www.e-enm.org

Despite evidence supporting an association between anemia
and overt thyroid disease, studies assessing subclinical or euthy-
roid conditions are limited [6,10-12]. A large cohort study re-
ported that even in euthyroid patients, free thyroxine (fT4) lev-
els are positively associated with erythrocyte indices such as he-
moglobin (Hb), hematocrit, and erythrocyte count, with a nega-
tive correlation between thyrotropin (TSH) levels and the serum
level of iron and transferrin saturation [10]. This association has
a clinical impact on defining optimal thyroid function and thera-
peutic targets. Because the majority of people at risk of anemia
have thyroid hormone levels within the reference range, it is
clinically important to determine whether thyroid function is
also associated with anemia in the euthyroid range. Thus, this
study aimed to evaluate the association between anemia and se-
rum fT4 and TSH levels in community-dwelling euthyroid
adults.
METHODS
Data source
We obtained publicly available data from the sixth Korea Na-
tional Health and Nutrition Examination Survey (KNHANES
VI, 2013 to 2015), a nationwide cross-sectional survey that used
hierarchical multistage cluster sampling to represent the Korean
population. The KNHANES has been conducted every year
since 1998 by the Korea Centers for Disease Control and Pre-
vention to assess the health and nutritional status of Koreans
[13]. Written informed consent was obtained from all partici-
pants prior to the survey and the protocol of this study was ap-
proved by the Institutional Review Board of the Pusan National
University Hospital, Busan, Korea (No. 1908-022-082).
Study population
Overall, 6,037 participants aged ≥19 years had data on thyroid
function tests and serum Hb level. Among them, participants
who had a history of thyroid disease, those who were taking
medicines that could influence thyroid function (n=251), and
pregnant women (n=23) were excluded. In addition, partici-
pants were excluded if they had abnormal TSH (<0.62 or
>6.86 mIU/L) and/or abnormal fT4 (<0.89 or >1.76 ng/mL)
levels (n=411) [13].
Laboratory measurements
Serum TSH, fT4, and anti-thyroid peroxidase antibody (TPOAb)
levels were measured according to an approved protocol, as
previously reported [13]. A blood sample was collected from
Copyright © 2020 Korean Endocrine Society
Thyroid Function and Anemia
each participant in the morning after fasting for at least 8 hours,
properly refrigerated, and sent to the testing facility. Serum
TSH, fT4, and TPOAb levels were measured by an electroche-
miluminescence immunoassay (Roche Diagnostics, Mannhein,
Germany). Serum TSH levels were measured by an E-TSH kit
(Roche Diagnostics), with a reference range of 0.35 to 5.50
mIU/L. Serum fT4 levels were measured by an E-Free T4 kit
(Roche Diagnostics), with a reference range of 0.89 to 1.76 ng/
mL. TPOAb levels were measured by an E-Anti-TPO kit
(Roche Diagnostics), with a reference range of <34.0 IU/mL in
humans. The measured serum TSH, fT4, and TPOAb levels met
the quality control and assurance program of the College of
American Pathology [13].
Serum Hb was measured by an XN-9000 (Sysmex, Kobe, Ja-
pan). Serum creatinine (Cr) was measured using Jaffe or the
rate-blanked compensated method (Hitachi Automatic Analyzer
7600-240, Hitachi, Tokyo, Japan). The estimated glomerular fil-
tration rate (eGFR) was calculated from the equation developed
using the Modification of Diet in Renal Disease study: eGFR
(mL/min/1.73 m2)=175×(serum Cr mg/dL)–1.154×age–0.203×
(0.742 for women) [14]. Urinary iodine concentrations (UICs)
were measured using inductively coupled plasma mass spec-
trometry (ICP-MS, Perkin Elmer, Waltham, MA, USA) with io-
dine as the standard (Inorganic Venture, Christiansburg, VA,
USA). UICs were adjusted using Cr levels to compensate for
uneven water excretion rates during the spot urine specimen
collection.
Definitions
Anemia was defined according to the World Health Organization
criteria as an Hb level <13 g/dL for men and <12 g/dL for non-
pregnant women [15]. Euthyroidism was defined as serum TSH
and fT4 within the normal reference range of 0.62 to 6.86 mIU/L
and 0.89 to 1.76 ng/mL, respectively, from the KNHANES VI
data [13].
Diabetes was defined as follows: (1) diagnosed with diabetes
by doctor; (2) currently taking anti-diabetic medications; or (3)
fasting plasma glucose ≥126 mg/dL. Hypertension was diag-
nosed when systolic blood pressure was ≥140 mm Hg, diastolic
blood pressure ≥90 mm Hg, or if the patient was using anti-hy-
pertensive medications. Participants were classified as having
chronic kidney disease (CKD) when eGFR <60 mL/min/1.73 m2
or proteinuria ≥1+. Rheumatoid arthritis, malignancy history,
smoking and economic status and limitation of physical activity
were assessed based on self-reported questionnaires.
www.e-enm.org  107
 Kim M, et al.

Statistical analysis
Statistical analyses were performed using the R program ver-
sion 3.5.1 (R Foundation for Statistical Computing, Vienna,
Austria; http://www.R-project.org). Data were evaluated using
the weights assigned to individuals sampled to represent the
Koreans. Continuous variables were presented as means with
standard error of the mean±standard error of the mean or medi-
ans with interquartile ranges (IQRs) and were analyzed using
the Student’s t test. Categorical variables were presented as fre-
quencies with percentages and were analyzed using the chi-
square test. We performed multiple linear regression analysis to
assess the finding that serum fT4 level is independently associ-
ated with serum Hb. The tendency of the serum Hb according to
serum fT4 quartiles was evaluated by examining P for any
trends based on multiple linear regression analyses. Multiple lo-
gistic regression analysis was done to estimate the odds ratios
(ORs) with 95% confidence intervals (CIs) for anemia accord-
ing to serum fT4 quartile categories. P<0.05 was considered
statistically significant.
RESULTS
Baseline characteristics
After excluding 685 participants, 5,352 (2,747 men and 2,605
women) were included in this analysis. The baseline character-
istics of the participants are listed in Table 1. The mean age was
44.0±0.218 years, and men were younger than women (P=
0.024). Overall, 6.1% of participants had anemia, and more

Table 1. Baseline Characteristics of Euthyroid Subjects (≥19 Years) Who Underwent Thyroid Function Tests and Serum Hemoglobin
Levels Measurements in KNHANES VI (2013 to 2015)
Characteristic Total (n=5,352) Men (n=2,747) Women (n=2,605) P value
Age, yr 44.0±0.218 43.7±0.270 44.5±0.295 0.024
Anemiaa 334 (6.1) 83 (2.8) 251 (9.9) <0.001
Serum hemoglobin, g/dL 14.4±0.023 15.4±0.026 13.2±0.025 <0.001
Serum free T4, ng/dL 1.24±0.003 1.28±0.004 1.19±0.004 <0.001
Serum TSH, mIU/L 2.46±0.019 2.36±0.026 2.57±0.028 <0.001
eGFR, mL/min/1.73 m2 94.5±0.292 94.3±0.354 94.7±0.444 0.439
TPOAb positivity a,b
294 (5.2) 92 (3.3) 202 (7.3) <0.001
UIC, μg/L 280.7 (150.4–650.5) 272.1 (156.4–609.9) 273.9 (139.1–635.6) 0.091
Smoking status b

Past smoking 1,030 (19.5) 883 (31.8) 147 (5.4) <0.001

Current smoking 1,248 (23.6) 1,080 (39.2) 168 (5.8)
Comorbidityb
Diabetes 469 (8.9) 290 (10.3) 179 (7.3) <0.001
Hypertension 1,224 (22.9) 742 (25.6) 482 (19.7) <0.001
Chronic kidney disease 119 (2.4) 72 (2.6) 47 (2.2) 0.356
Rheumatoid arthritis 70 (1.4) 23 (0.7) 47 (2.3) <0.001
Malignancy 108 (1.9) 54 (1.6) 54 (2.1) 0.297
Economic status b

Low 715 (12.8) 327 (11.0) 388 (15.0) <0.001

Middle-low 1,424 (26.4) 708 (26.0) 716 (26.8)
Middle-high 1,583 (29.9) 827 (30.6) 756 (29.1)
High 1,630 (30.9) 885 (32.4) 745 (29.2)
Activity limitationb 297 (5.5) 137 (4.9) 160 (6.1) 0.071

Values are expressed as mean±standard error of the mean, number (%), or median (interquartile range).
KNHANES, Korea National Health and Nutrition Examination Survey; T4, thyroxine; TSH, thyrotropin; eGFR, estimated glomerular filtration rate;
TPOAb, anti-thyroid peroxidase antibody; UIC, urinary iodine concentration.
a
TPOAb positivity was defined as levels >34 IU/mL; bWeighted percentage.

108 www.e-enm.org Copyright © 2020 Korean Endocrine Society

 Thyroid Function and Anemia

women (9.9%) had anemia than men (2.8%, P< 0.001). Mean
Table 2. Multiple Linear Regression Analysis to Determine
serum Hb and fT4 levels were 14.4 g/dL and 1.24 ng/dL, re- Which Covariates Including Serum fT4 Level Are Independent-
spectively, and were 16.7% and 7.6% higher in men than in ly Associated with Serum Hb Level
women (P<0.001 and P<0.001), whereas serum TSH level was Dependent variable: serum Hb
8.9% lower (P<0.001). The overall prevalence of TPOAb posi- Independent variable
β SE P value
tivity was 5.2%, and the proportion of TPOAb positivity was
Men
higher in women than in men (7.3% vs. 3.3%, P< 0.001). The
fT4 0.865 0.154 <0.001
median UIC was 280.7 μg/L (IQR, 150.4 to 650.5 μg/L), and
Age –0.020 0.002 <0.001
there was no difference in UICs between both men and women.
eGFR –0.005 0.002 0.005
The percentages of participants with a past or current smok-
UIC <–0.001 <0.001 0.096
ing habit, and those with diabetes and hypertension were signif-
TPOAb (>34 IU/mL) –0.001 0.107 0.989
icantly higher in men than in women (P<0.001, each). While
Current smoking 0.201 0.044 <0.001
women presented with a higher prevalence of rheumatoid ar-
Comorbidity
thritis compared with men (P<0.001). The proportion of partic-
Diabetes –0.039 0.093 0.678
ipants with CKD, a malignancy, and activity limitation was
Hypertension 0.135 0.065 0.038
2.4%, 1.9%, and 5.5%, respectively, with no significant differ-
Chronic kidney disease –0.478 0.202 0.019
ence between men and women.
Rheumatoid arthritis –0.587 0.228 0.010
Malignancy –1.259 0.223 <0.001
Association between serum Hb and serum fT4
Economic status 0.001 0.025 0.978
Multiple linear regression analyses were conducted to evaluate
Activity limitation –0.231 0.126 0.066
the factors independently associated with serum Hb levels (Ta-
Women
ble 2). In men, serum Hb was associated positively with fT4
fT4 0.842 0.184 <0.001
(β=0.865, P<0.001), current smoking (β=0.201, P<0.001) and
Age 0.003 0.002 0.139
hypertension (β=0.135, P=0.038), while inversely with age (β=
eGFR –0.005 0.002 0.004
–0.020, P<0.001), eGFR (β=–0.005, P=0.005), CKD (β=
UIC <0.001 <0.001 0.957
–0.478, P=0.019), rheumatoid arthritis (β=–0.587, P=0.010),
TPOAb (>34 IU/mL) 0.154 0.101 0.128
and malignancies (β=–1.259, P<0.001). In women, serum fT4
Current smoking 0.431 0.087 <0.001
(β=0.842, P<0.001), current smoking (β=0.431, P<0.001),
Comorbidity
and hypertension (β=0.203, P=0.003) were positively associat-
Diabetes 0.043 0.107 0.690
ed with serum Hb, and eGFR was inversely associated with se-
Hypertension 0.203 0.067 0.003
rum Hb (β=–0.005, P=0.004). There was no independent cor-
Chronic kidney disease –0.406 0.259 0.118
relation between serum TSH and serum Hb in both men (P=
0.067) and women (P=0.247) (Supplemental Table S1). Rheumatoid arthritis –0.040 0.179 0.823

Serum fT4 levels were divided into four groups based on the Malignancy 0.002 0.192 0.999

quartiles in each sex in order to examine whether the associa- Economic status 0.001 0.028 0.961

tion between serum fT4 and serum Hb levels has a threshold ef- Activity limitation 0.102 0.091 0.264

fect. In both sexes, serum Hb levels linearly decreased across Values are expressed as adjusted for all other variables in table.
decreasing serum fT4 quartile groups (P for trend <0.001 in fT4, free thyroxine; Hb, hemoglobin; SE, standard error; eGFR, esti-
mated glomerular filtration rate; UIC, urinary iodine concentration;
men and P for trend <0.001 in women) (Fig. 1). TPOAb, anti-thyroid peroxidase antibody.

Risk for developing anemia according to serum fT4 levels

Multivariate analysis was performed to identify ORs for anemia
according to serum fT4 quartile groups (Table 3). In men, the
prevalence of anemia was 5.6%, 3.2%, 1.7%, and 1.0% in quar-
tiles 1, 2, 3, and 4, respectively. In an unadjusted model, the OR
for anemia was 6.0 times higher among participants in the low-
est quartile than among those in the highest quartile (95% CI,
2.42 to 14.68; P<0.001). After being adjusted for age, iodine
intake, TPOAb positivity, and smoking status, the lowest quar-
tile group had a significantly higher risk of anemia compared
with the highest quartile group (OR, 3.9; 95% CI, 1.46 to 10.24;

Copyright © 2020 Korean Endocrine Society www.e-enm.org  109

 Kim M, et al.

Men Women
16.0 13.5
P for trend <0.001 P for trend <0.001

15.5 13.0
Serum Hb (g/dL)

Serum Hb (g/dL)
15.0 12.5

14.5 12.0

0.5 0.5

0 0
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4
(0.89−1.15) (1.16−1.26) (1.27−1.38) (1.39−1.76) (0.89−1.08) (1.09−1.18) (1.19−1.28) (1.29−1.76)
Serum free T4 quartiles A Serum free T4 quartiles B

Fig. 1. Serum hemoglobin (Hb) levels according to serum free thyroxine (T4) quartiles in (A) euthyroid men and (B) women. Values are
presented as the estimated mean with standard error of the means.

Table 3. Multiple Logistic Regression Analyses to Determine the ORs of Having Anemia According to Serum fT4 Quartile Categories

Prevalence, Unadjusted Model Model 1 Model 2

fT4 quartiles
%a OR (95% CI) P value OR (95% CI) P value OR (95% CI) P value
Men
Q1 (0.89–1.15) 5.6 6.0 (2.42–14.68) <0.001 3.9 (1.46–10.24) 0.007 4.4 (1.66–11.73) 0.003
Q2 (1.16–1.26) 3.2 3.3 (1.36–8.16) 0.009 2.7 (1.06–6.92) 0.038 3.0 (1.16–7.62) 0.024
Q3 (1.27–1.38) 1.7 1.8 (0.64–4.97) 0.274 1.5 (0.55–4.34) 0.417 1.7 (0.58–4.73) 0.344
Q4 (1.39–1.76) 1.0 1.0 (reference) 1.0 (reference) 1.0 (reference)
Women
Q1 (0.89–1.08) 15.4 2.4 (1.58–3.70) <0.001 2.8 (1.78–4.48) <0.001 2.8 (1.78–4.51) <0.001
Q2 (1.09–1.18) 8.7 1.3 (0.79–2.04) 0.326 1.3 (0.76–2.11) 0.367 1.3 (0.75–2.10) 0.389
Q3 (1.19–1.28) 8.7 1.3 (0.78–2.04) 0.351 1.2 (0.70–1.95) 0.543 1.2 (0.71–1.96) 0.512
Q4 (1.29–1.76) 7.0 1.0 (reference) 1.0 (reference) 1.0 (reference)

Model 1 was adjusted for age, iodine intake, autoimmunity, and smoking. Model 2 was adjusted by Model 1, comorbidities (diabetes, hypertension,
chronic kidney disease, rheumatoid arthritis, cancer history), income level, and activity limitation.
OR, odds ratio; fT4, free thyroxine; CI, confidence interval.
a
Weighted percentage.

P=0.007 in model 1). In Model 2, which was adjusted for co-
morbidities, income level, and activity limitation in Model 1,
the OR for anemia was still significantly higher in the lowest
quartile group than in the highest quartile group (OR, 4.4; 95%
CI, 1.66 to 11.73; P=0.003).
In women, 15.4%, 8.7%, 8.7%, and 7.0% in quartiles 1, 2, 3,
and 4, respectively, had anemia. In the unadjusted model, partici-
pants with the lowest quartile had 2.4 times higher OR for anemia
than those with the highest quartile (95% CI, 1.58 to 3.70; P<
0.001). With the multivariate analysis adjusted for age, iodine in-
take, TPOAb positivity, smoking status, comorbidities, income
level, and activity limitation, participants with the lowest quartile

110 www.e-enm.org Copyright © 2020 Korean Endocrine Society

 Thyroid Function and Anemia

0%, and 0% in men, respectively, and 20.3%, 12.9%, 10.7%,

Table 4. ORs of Having Anemia According to Serum fT4 Quar-
tile Categories in Younger and Older Participants and 7.7% in women, respectively. In the multivariate analysis,
younger men in the first quartile had a significantly higher risk
fT4 quartiles Prevalence, %a ORb (95% CI) P value
of anemia compared to those in the second quartile (OR, 3.3;
Younger (age <50 yr)
95% CI, 1.13 to 9.75; P=0.029). Also, younger women with
Men (n=1,658)
low-normal fT4 had a 3.2 times higher OR than those with the
Q1 (0.89–1.18) 3.2 3.3 (1.13–9.75) 0.029
high-normal fT4 (95% CI, 1.89 to 5.49; P<0.001). However,
Q2 (1.19–1.30) 1.6 1.0 (reference)
these associations were not observed in older men and women.
Q3 (1.31–1.41) 0 NA <0.001
Q4 (1.42–1.76) 0 NA <0.001
DISCUSSION
Women (n=1,619)
Q1 (0.89–1.09) 20.3 3.2 (1.89–5.49) <0.001
In this nationwide cross-sectional study, we showed a signifi-
Q2 (1.10–1.19) 12.9 1.5 (0.83–2.64) 0.186
cant association between serum fT4 levels and anemia in both
Q3 (1.20–1.30) 10.7 1.1 (0.61–2.05) 0.725
euthyroid men and women. The prevalence of anemia was sig-
Q4 (1.31–1.76) 7.7 1.0 (reference)
nificantly higher in women than in men. After adjusting for ma-
Older (age ≥50 yr)
jor confounding parameters, serum fT4 levels, but not TSH, had
Men (n=1,089)
an independently positive association with serum Hb levels in
Q1 (0.89–1.10) 7.5 2.0 (0.81–5.19) 0.131
both sexes. In both sexes, serum Hb levels decreased linearly
Q2 (1.11–1.20) 5.9 1.8 (0.74–4.42) 0.198
with decreasing serum fT4 quartile groups, and participants
Q3 (1.21–1.31) 5.6 1.4 (0.53–3.71) 0.503 with low-normal fT4 had significantly higher risk of anemia
Q4 (1.32–1.76) 4.1 1.0 (reference) than those with high-normal fT4. When participants were divid-
Women (n=986) ed into younger and older groups at 50 years of age, in younger
Q1 (0.89–1.07) 8.4 1.8 (0.74–4.60) 0.193 euthyroid participants, men and women with the first quartile
Q2 (1.08–1.16) 4.1 0.8 (0.29–2.39) 0.732 were at higher risk of anemia than men with the second quartile
Q3 (1.17–1.26) 5.1 1.1 (0.37–3.19) 0.859 and women with the forth quartile, respectively. However, this
Q4 (1.27–1.76) 4.4 1.0 (reference) study did not find any association between the fT4 levels and
OR, odds ratio; fT4, free thyroxine; CI, confidence interval; NA, not ap- anemia in older euthyroid participants.
plicable. Despite evidence supporting the relationship between overt or
a
Weighted percentage; bAdjusted for iodine intake, autoimmunity, smok-
ing, comorbidities (diabetes, hypertension, chronic kidney disease, rheu- subclinical thyroid disease and anemia, studies targeting euthy-
matoid arthritis, cancer history), income level, and activity limitation. roid conditions are limited [6,10-12], although the relationship
between thyroid hormones and erythrocyte indices in euthyroid
subjects has been reported lately [10-12]. According to a study
were at a significantly greater risk for anemia than those with the conducted in a 1,011 euthyroid older participants, fT4 levels
highest quartile ([OR, 2.8; 95% CI, 1.78 to 4.48; P<0.001 in mod- had a linear positive association with Hb levels and erythrocyte
el 1] and [OR, 2.8; 95% CI, 1.78 to 4.51; P<0.001 in model 2]). indices [10]. Similarly, based on a cohort of 708 euthyroid older
participants, fT4 was significantly associated with Hb levels
Association between serum fT4 and anemia in younger and erythrocyte parameters, whereas TSH was not, after adjust-
and older participants ing for major confounding parameters [11]. The association be-
The prevalence of anemia in each age group of 19 to 29, 30 to tween thyroid function and erythrocyte indices was further
39, 40 to 49, 50 to 59, 60 to 69, and ≥70 was 1.1%, 0.5%, 1.8%, evaluated by Lippi et al. [12] in a retrospective cohort of 1,050
3.8%, 7.5%, and 10.7% in men, respectively, and 8.9%, 15.4%, euthyroid older participants. Although that study failed to iden-
14.6%, 4.5%, 5.9%, and 10.6% in women, respectively (Sup- tify a relationship between thyroid function and Hb or hemato-
plemental Table S2). We further analyzed ORs for anemia ac- crit, it was the first to report a significantly positive relationship
cording to the serum fT4 quartile groups in younger and older between red blood cell distribution width and thyroid hormone
participants (Table 4). In participants with aged <50 years, the values [12]. Consistent with previous studies, we also found that
prevalence of anemia of quartiles 1, 2, 3, and 4 was 3.2%, 1.6%, serum fT4 levels, but not TSH levels, had an independently

Copyright © 2020 Korean Endocrine Society www.e-enm.org  111

 Kim M, et al.
positive association with serum Hb levels in both sexes after ad-
justing for age, UIC, TPOAb positivity, smoking status, comor-
bidities, economic status, and activity limitation.
Thyroid hormones have a crucial role in hematopoiesis, espe-
cially in erythropoiesis [8]. However, the mechanisms by which
thyroid hormones modulate red blood cell production are not
completely understood [3,16-18]. It seems that there are several
mechanisms by which thyroid hormones regulate red blood cell
production, including erythropoietin production and reactive-
ness, and affect iron transport and utilization [3,16-18]. These
actions can be mediated either through TSH via the TSH recep-
tor proposed to be present in erythrocytes and erythrocyte pre-
cursors, or through metabolic effects mediated by thyroid hor-
mones [8,11,19]. In this study, fT4, not TSH, was significantly
related with Hb levels and an increased risk of anemia, indicat-
ing that the erythropoiesis impact of thyroid hormones is con-
trolled by direct or indirect functions of fT4 rather than TSH re-
ceptor mediation, particularly in euthyroid individuals. Thus,
fT4 may be a more appropriate physiological marker of thyroid
hormone status than TSH in the euthyroid population.
We used multivariate analysis to adjust for other potential
confounding factors including age, iodine intake status, autoim-
munity, smoking status, presence of comorbidities (diabetes,
hypertension, CKD, rheumatoid arthritis, and malignancy), eco-
nomic status, and physical activity. These factors are associated
with both thyroid disease and anemia. Actually, the prevalence
of thyroid disease and anemia increases with age [20,21]. Pa-
tients with autoimmune thyroid disease are at higher risk of de-
veloping various forms of anemia, and pernicious anemia fre-
quently coexists with autoimmune thyroid disease [9]. Iodine is
also necessary for the synthesis of thyroid hormone, and high
iodine intake is related to autoimmune thyroiditis and hypothy-
roidism [22]. Anemia is a well-known complication of CKD
and renal function, represented as eGFR, is also related to thy-
roid function [23,24]. Smoking, diabetes, hypertension, rheu-
matoid arthritis, malignancies, and socioeconomic status are
also associated with the development of anemia [2,25-28]. In
this study, we attempted to select appropriate confounders for
multivariate analysis by minimizing the influence of other fac-
tors aside from thyroid function and anemia.
Anemia burden also differed by age and sex. The prevalence
of anemia was much higher in women than men under 50 years
of age, while similar in men and women over 50 years of age.
The major cause of anemia in young adults is related to gyneco-
logic conditions that often present as iron deficiency [2]. When
we divided participants into two groups using a cutoff age of 50
112 www.e-enm.org
years, 87% of women underwent menopause. This study con-
ducted a subgroup analysis based on age 50 years with the ex-
ception of pregnant women to minimize the effect of gyneco-
logical factors affecting anemia. In older adults, renal insuffi-
ciency, chronic disease, malignancies, and nutritional deficiency
are the biggest contributors to total anemia burden in both sexes
[29]. In our study, the association between serum fT4 and ane-
mia was not observed in older participants. This finding demon-
strates that many factors other than serum fT4 may contribute to
the development of anemia in the elderly.
This study has several limitations. First, the cross-sectional
design did not provide us with the information to draw conclu-
sions about causality. Second, we did not additionally measure
ferritin and transferrin to distinguish iron deficiency anemia and
anemia of chronic disease. Third, information on nutritional
confounding factors including iron or vitamin supplements was
not available for most participants. Forth, there was no partici-
pant with anemia among younger male with high-normal fT4.
Nonetheless, to our knowledge, this is the first nationwide anal-
ysis to evaluate the risk of developing anemia according to thy-
roid function in euthyroid adults.
In conclusion, data collected from a nationally representative
cross-sectional survey showed that a low-normal serum fT4
level was related to lower serum Hb level and a higher risk of
anemia in euthyroid adults, especially in younger adults. These
findings demonstrate that the reference range of fT4 may not be
appropriate from the viewpoint of anemia and that avoiding
low-normal serum fT4 levels may help to avoid anemia in both
men and women. Further studies are required to determine the
feasibility of improving hematologic parameters through levo-
thyroxine treatment in subjects with anemia and low-normal se-
rum fT4.
CONFLICTS OF INTEREST
No potential conflict of interest relevant to this article was re-
ported.
AUTHOR CONTRIBUTIONS
Conception or design: M.K., B.H.K., H.L., Y.K.J., S.S.K., I.J.K.
Acquisition, analysis, or interpretation of data: M.K., B.H.K.,
H.L. Drafting the work or revising: M.K., B.H.K. Final approv-
al of the manuscript: M.K., B.H.K., H.L., M.H.J., J.M.K., E.
H.K., Y.K.J., S.S.K., I.J.K.
Copyright © 2020 Korean Endocrine Society

ORCID
Mijin Kim https://orcid.org/0000-0002-1538-8859
Bo Hyun Kim https://orcid.org/0000-0001-9632-9457
REFERENCES
1. de Benoist B, McLean E, Egli I, Cogswell M. WHO global
database on anaemia. Geneva: World Health Organization;
2008.
2. Kassebaum NJ, Jasrasaria R, Naghavi M, Wulf SK, Johns N,
Lozano R, et al. A systematic analysis of global anemia bur-
den from 1990 to 2010. Blood 2014;123:615-24.
3. Fein HG, Rivlin RS. Anemia in thyroid diseases. Med Clin
North Am 1975;59:1133-45.
4. M’Rabet-Bensalah K, Aubert CE, Coslovsky M, Collet TH,
Baumgartner C, den Elzen WP, et al. Thyroid dysfunction
and anaemia in a large population-based study. Clin Endo-
crinol (Oxf) 2016;84:627-31.
5. Lippi G, Montagnana M, Salvagno GL, Guidi GC. Should
women with abnormal serum thyroid stimulating hormone
undergo screening for anemia? Arch Pathol Lab Med 2008;
132:321-2.
6. Omar S, Hadj Taeib S, Kanoun F, Hammami MB, Kamoun
S, Ben Romdhane N, et al. Erythrocyte abnormalities in thy-
roid dysfunction. Tunis Med 2010;88:783-8.
7. Wopereis DM, Du Puy RS, van Heemst D, Walsh JP, Brem-
ner A, Bakker SJL, et al. The relation between thyroid func-
tion and anemia: a pooled analysis of individual participant
data. J Clin Endocrinol Metab 2018;103:3658-67.
8. Kawa MP, Grymula K, Paczkowska E, Baskiewicz-Masiuk
M, Dabkowska E, Koziolek M, et al. Clinical relevance of
thyroid dysfunction in human haematopoiesis: biochemical
and molecular studies. Eur J Endocrinol 2010;162:295-305.
9. Szczepanek-Parulska E, Hernik A, Ruchała M. Anemia in
thyroid diseases. Pol Arch Intern Med 2017;127:352-60.
10. Bremner AP, Feddema P, Joske DJ, Leedman PJ, O’Leary
PC, Olynyk JK, et al. Significant association between thy-
roid hormones and erythrocyte indices in euthyroid subjects.
Clin Endocrinol (Oxf) 2012;76:304-11.
11. Schindhelm RK, ten Boekel E, Heima NE, van Schoor NM,
Simsek S. Thyroid hormones and erythrocyte indices in a co-
hort of euthyroid older subjects. Eur J Intern Med 2013;24:
241-4.
12. Lippi G, Danese E, Montagnana M. Thyroid hormone levels
are associated with anisocytosis in a cohort of euthyroid
Copyright © 2020 Korean Endocrine Society
Thyroid Function and Anemia
older outpatients. Eur J Intern Med 2014;25:e4-5.
13. Kim WG, Kim WB, Woo G, Kim H, Cho Y, Kim TY, et al.
Thyroid stimulating hormone reference range and preva-
lence of thyroid dysfunction in the Korean population: Ko-
rea National Health and Nutrition Examination Survey 2013
to 2015. Endocrinol Metab (Seoul) 2017;32:106-14.
14. Levey AS, Coresh J, Greene T, Stevens LA, Zhang YL,
Hendriksen S, et al. Using standardized serum creatinine
values in the modification of diet in renal disease study
equation for estimating glomerular filtration rate. Ann Intern
Med 2006;145:247-54.
15. Beutler E, Waalen J. The definition of anemia: what is the
lower limit of normal of the blood hemoglobin concentra-
tion? Blood 2006;107:1747-50.
16. Touam M, Guery B, Goupy C, Menoyo V, Drueke T. Hypo-
thyroidism and resistance to human recombinant erythropoi-
etin. Nephrol Dial Transplant 2004;19:1020-1.
17. Christ-Crain M, Meier C, Huber P, Zulewski H, Staub JJ,
Muller B. Effect of restoration of euthyroidism on peripheral
blood cells and erythropoietin in women with subclinical
hypothyroidism. Hormones (Athens) 2003;2:237-42.
18. Cinemre H, Bilir C, Gokosmanoglu F, Bahcebasi T. Hema-
tologic effects of levothyroxine in iron-deficient subclinical
hypothyroid patients: a randomized, double-blind, con-
trolled study. J Clin Endocrinol Metab 2009;94:151-6.
19. Perrin MC, Blanchet JP, Mouchiroud G. Modulation of hu-
man and mouse erythropoiesis by thyroid hormone and reti-
noic acid: evidence for specific effects at different steps of
the erythroid pathway. Hematol Cell Ther 1997;39:19-26.
20. Gaskell H, Derry S, Andrew Moore R, McQuay HJ. Preva-
lence of anaemia in older persons: systematic review. BMC
Geriatr 2008;8:1.
21. Cooper DS, Biondi B. Subclinical thyroid disease. Lancet
2012;379:1142-54.
22. Jeon MJ, Kim WG, Kwon H, Kim M, Park S, Oh HS, et al.
Excessive iodine intake and thyrotropin reference interval:
data from the Korean National Health and Nutrition Exami-
nation Survey. Thyroid 2017;27:967-72.
23. Wolf G, Muller N, Hunger-Battefeld W, Kloos C, Muller
UA. Hemoglobin concentrations are closely linked to renal
function in patients with type 1 or 2 diabetes mellitus. Kid-
ney Blood Press Res 2008;31:313-21.
24. Sun MT, Hsiao FC, Su SC, Pei D, Hung YJ. Thyrotropin as
an independent factor of renal function and chronic kidney
disease in normoglycemic euthyroid adults. Endocr Res
2012;37:110-6.
www.e-enm.org  113
 Kim M, et al.
25. Nordenberg D, Yip R, Binkin NJ. The effect of cigarette
smoking on hemoglobin levels and anemia screening. JAMA
1990;264:1556-9.
26. Wolfe F, Michaud K. Anemia and renal function in patients
with rheumatoid arthritis. J Rheumatol 2006;33:1516-22.
27. Gaspar BL, Sharma P, Das R. Anemia in malignancies: patho-
genetic and diagnostic considerations. Hematology 2015;20:
18-25.
28. Styszynski A, Mossakowska M, Chudek J, Puzianowska-
114 www.e-enm.org
Kuznicka M, Klich-RaczkaA, Neumann-Podczaska A, et al.
Prevalence of anemia in relation to socio-economic factors in
elderly Polish population: the results of PolSenior study. J
Physiol Pharmacol 2018;69:75-81.
29. Guralnik JM, Eisenstaedt RS, Ferrucci L, Klein HG, Wood-
man RC. Prevalence of anemia in persons 65 years and old-
er in the United States: evidence for a high rate of unex-
plained anemia. Blood 2004;104:2263-8.
Copyright © 2020 Korean Endocrine Society