Abstract
Objective
Neutropenia is a complication of Graves' disease (GD), but there is currently no means by which to predict its occurrence. This study aimed to investigate the risk factors for the development of neutropenia in untreated GD.
Methods
This was a retrospective cohort study. Between January 1, 2010, and July 31, 2020, 1000 patients with new-onset or relapsing GD without treatment were enrolled in the study and divided into two groups: neutropenia group (neutrophil count < 2 × 109/L) and non-neutropenia group (neutrophil count ≥ 2 × 109/L). Clinical characteristics of subjects were compared between the two groups, and logistic regression analysis was applied to determine risk factors for neutropenia. To further explore the correlation of radioactive iodine uptake (RAIU) with neutropenia, subjects were first classified according to quartile of 3 h RAIU and 24 h RAIU prior to logistic regression analysis.
Results
Of all patients recruited, 293 (29.6%) were diagnosed with neutropenia. Compared with non-neutropenic patients, those with neutropenia had a higher level of free thyroxine (FT4) (56.64 ± 31.80 vs 47.64 ± 39.64, P = 0.001), 3 h RAIU (55.64 ± 17.04 vs 49.80 ± 17.21, P < 0.001) and 24 h RAIU (67.38 ± 12.54 vs 64.38 ± 13.58, P < 0.001). Univariate logistic regression analysis revealed that FT4, 3 h RAIU, 24 h RAIU, creatinine, and low-density lipoprotein were risk factors for development of neutropenia in GD. After adjusting for confounding factors of age, BMI, and sex, we determined that 3 h RAIU and 24 h RAIU (Model 1: OR = 1.021, 95% CI: 1.008–1.033, P = 0.001; Model 2: OR = 1.023, 95% CI: 1.007–1.039, P = 0.004), but not FT4, were associated with the development of neutropenia.
Conclusions
RAIU is associated with neutropenia in patients with untreated GD.
Introduction
Graves' disease (GD), also known as an organ-specific autoimmune disorder, is characterized by high radioactive iodine uptake (RAIU) and hyperthyroidism (1). Excessive production of thyroid hormones results in a substantially adverse impact on a patient's quality of life and serious symptoms such as weight loss, muscle weakness, tremors, neuropsychiatric symptoms, and rarely cardiovascular collapse and death. The synthesis of thyroid hormone is a complex procedure that involves multiple proteins such as thyroid peroxidase (TPO), thyroglobulin (TG), and sodium iodide symporter (NIS) (2, 3). NIS is a key protein that mediates active iodine uptake in the thyroid follicular epithelial cells. This procedure is the first critical step in thyroid hormone synthesis.
Methimazole (MMI) is an established first-line therapy for GD in China and MMI-induced granulocytosis is dose-dependent (4). Antithyroid drug (ATD)-associated agranulocytosis is a rare but life-threatening event (1). Nonetheless, there has been little study of neutropenia prior to the commencement of ATD therapy. A small retrospective study by Aggarwal et al. of 206 newly diagnosed untreated GD patients revealed that the prevalence of neutropenia was as high as 14.1% (5). Thyroid function and race were associated with neutropenia with non-Caucasians more at risk. In addition, free triiodothyronine (FT3) was negatively correlated with neutrophil count (5). These study findings were limited by the small number of participants (only 29 of GD individuals had neutropenia) and may not have carried sufficient statistical power to determine potential risk factors for the development of neutropenia.
In our study, 1000 patients diagnosed with GD were classified as neutropenic or non-neutropenic according to neutrophil count. We then evaluated the clinical and biochemical factors associated with neutropenia to determine risk factors for its development in GD.
Methods
Study population
A retrospective cohort study was carried out between January 1, 2010, and July 31, 2020, at Shanghai Fifth People's Hospital, Fudan University and included 1112 patients who had newly diagnosed or relapsed GD and were not receiving ATDs. Finally, 1000 patients were enrolled and they chose RAI treatment. Among them, some GD patients suffered from neutropenia, recurrence after a course of ATD, cardiac arrhythmias, hepatotoxicity, and thyrotoxic periodic paralysis. Other patients without complications chose RAI treatment because they were concerned about the potential side effects of ATDs or the rapid effect of RAI treatment. The diagnosis is based on the initial biochemical evaluation, diagnostic testing, and clinical presentation including goiter, ocular symptoms, rapid heartbeat, poor tolerance of heat, and weight loss. Baseline data for each participant were extracted from the Hospital's Health Information System and included sex, age, BMI (weight (kg)/height squared (m2)), blood cell count, thyroid function (free thyroxine (FT4), thyrotropin (TSH), TPO antibodies (TPOAb), TG antibodies (TGAb), and thyrotropin receptor antibodies (TRAb)), and RAIU (3 h RAIU and 24 h RAIU). A flowchart of the study is shown in Fig. 1. Patients were excluded from the study if they had any of the following: i) previous diagnosis of malignancy; ii) any infectious disease; iii) were pregnant or lactating; iv) any chronic autoimmune disease; v) severe cardiovascular or cerebrovascular events; and vi) other causes of neutropenia such as drug, hematological disorder, vitamin B12 deficiency, congenital causes, and so on. After exclusions, 1000 patients were enrolled in our study (Fig. 1) and were then grouped according to the presence or absence of neutropenia. Neutropenia was defined as a neutrophil count less than 2.0 × 109/L. The study protocol was approved by the Medical Ethics Committee of the Fifth People’s Hospital of Shanghai, Fudan University (NO. 2017-029).
Flowchart of study. GD, Graves' disease.
Citation: Endocrine Connections 12, 2; 10.1530/EC-22-0474
Assays
Full blood count was analyzed using an automatic blood cell analyzer (XN9000, Sysmex, Kobe, Japan) and thyroid function using an automatic biochemical analyzer (Cobas 8000, Roche Diagnostics). Radioactive iodine uptake was measured at 3 h and 24 h after oral administration of 3 μci iodine-131. RAIU was detected by thyroid function instrument (ANHUI USTC Zonkia Scientific Instruments Co. Ltd, Hefei, China). The following reference ranges were applied: serum TSH 0.27–4.2 mIU/L, FT4 12–22 pmol/L, TPOAb was considered positive if it exceeded 34 IU/mL, TGAb positive if >115 IU/mL, TRAb positive if >1.75 U/L, white cell count 4.0–10.0 × 109/L, neutrophils 2.0–7.0 × 109/L. The cut-off points for increased iodine uptake rate were 3 h RAIU > 20% and 24 h RAIU > 45%.
Flow cytometry
Nthy-ori 3-1 (a normal human thyroid follicular epithelial cell line) was ordered from Shanghai Royal Industrial Co. Ltd (Shanghai, China). Nthy-ori 3-1 cells were grown in RPMI-1640 (Gibco) supplemented with 10% fetal bovine serum (FBS) (Gibco), 100 IU/mL penicillin, and 100 mg/mL streptomycin sulfate. Cells from different group were resuspended in PBS containing 2% FBS and 1 mL eBioscience™ Foxp3/Transcription Factor Staining Buffer Set (Invitrogen) for permeabilized conditions with 1 μL anti-NIS (Proteintech, Wuhan, China). After washing, cells were incubated with donkey anti-rabbit IgG H&L (Abcam). The fluorescence of 107 cells per tube was assayed using the fluorescence-activated cell sorting (FACS) technique (BD Bio-Sciences, New York, NY, USA). Data were analyzed with FlowJo software (Tree Star, Ashland, Kentucky, USA).
Statistical analysis
Participants were categorized into two groups based on neutrophil count: neutrophils < 2 × 109/L or neutrophils ≥ 2 × 109/L. The Kolmogorov–Smirnoff test was used to test for normal distribution. Mean ± s.d. was calculated for normally distributed variables, and median was calculated for data with skewed distribution. General characteristics (continuous and categorical variables subject to normal distribution) of patients with or without neutropenia were compared using the Student t-test and those with skewed distribution were compared using a nonparametric test. To further understand the relationship between variables and granulocytopenia, univariate logistic regression analysis was applied. Subsequently, different variables (Model 1 further adjusted for the age, BMI, sex, FT4, and 3 h RAIU; Model 2 further adjusted for age, BMI, sex, FT4, and 24 h RAIU) were entered into a logistic regression to predict the risk of neutropenia development prior to ATDs treatment. To validate the association of RAIU with neutrophil count, patients were divided into four groups according to quartile of RAIU; quartile of 3 h RAIU: Q1 (<37.32%), Q2 (37.32–50.72%), Q3 (50.72–64.08%), and Q4 (>64.08%); and quartile of 24 h RAIU: Q1 (<55.59%), Q2 (55.59–65.83%), Q3 (65.83–74.77%), and Q4 (>74.77%). Kruskal–Wallis test was applied to compare multiple groups. Logistic regression was applied to determine the correlation of RAIU with neutropenia. P < 0.05 was considered statistically significant. All statistical analyses were performed using SPSS version 23.0 for Windows (IBM).
Results
Clinical characteristics of participants
The percentage of neutropenia in 1000 patients with new-onset or relapsing GD without treatment was 29.7%. Participants underwent RAI treatment after providng both written informed and verbal consent on all aspects of efficacy and potential side-effects of ATDs and RAI therapy. All patients had undertaken RAIU evaluation before RAI therapy. Patients were classified according to neutrophil count: neutropenia group (neutrophil count < 2 × 109/L) or non-neutropenia group (neutrophil count ≥ 2 × 109/L). Only seven patients had non-specific symptoms, such as fatigue and poor appetite due to their neutrophil count < 1 × 109/L. Agranulocytosis (neutrophil count < 0.5 × 109/L) is uncommon but life-threatening. However, in our study, there were no patients who developed agranulocytosis. Neutrophil count was between 1 × 109/L and 2 × 109/L in the vast majority of the neutropenia group. Considering RAI treatment of GD can cause a transient exacerbation of hyperthyroidism, about 60–80% patients received Leucogen tablets especially when neutrophil count was below 1 × 109/L. Baseline characteristics of subjects are shown in Table 1. Compared with non-neutropenic patients, those in the neutropenia group had a much higher level of FT4, 3 h RAIU, and 24 h RAIU and much lower level of creatinine (Cr), total cholesterol (TC), and low-density lipoprotein (LDL). Nonetheless sex, age, BMI, TSH, TPOAb, TgAb, TRAb, alanine transaminase (ALT), alanine transaminase (AST), blood urea nitrogen (BUN), and high-density lipoprotein (HDL) were similar for both groups.
Clinical characteristics of subjects according to the different neutrophil count.
| Variables | Neutropenia group | Non-neutropenia group | P |
|---|---|---|---|
| n | 296 | 704 | - |
| Sex (M/F) | 87/209 | 237/466 | 0.183 |
| Age (year) | 42 ± 13 | 41 ± 13 | 0.126 |
| BMI (kg/m2) | 21.17 ± 2.72 | 21.20 ± 3.01 | 0.909 |
| FT4 (pmol/L) | 56.64 ± 31.80 | 47.64 ± 39.64 | 0.001 |
| TSH (mIU/L) | 0.005 (0–0.010) | 0.005 (0–0.010) | 0.330 |
| TPOAb (IU/L) | 145.5 (22.6–443.2) | 131.0 (22.6–397.5) | 0.500 |
| TGAb (IU/L) | 129.45 (38.60–386.58) | 99.60 (28.80–340.05) | 0.857 |
| TRAb (IU/L) | 11.82 (6.40–23.04) | 12.54 (7.42–25.85) | 0.275 |
| 3 h RAIU (%) | 55.64 ± 17.04 | 49.80 ± 17.21 | <0.001 |
| 24 h RAIU (%) | 67.38 ± 12.54 | 64.38 ± 13.58 | 0.001 |
| ALT (U/L) | 29 (20–44) | 28 (19–43) | 0.538 |
| AST (U/L) | 22.4 (16.6–30.8) | 23.0 (17.0–32.0) | 0.527 |
| BUN (mmol/L) | 4.9 (4.1–5.7) | 4.9 (4.0–5.7) | 0.758 |
| Cr (μmol/L) | 41.0–(33.0–49.0) | 44.5 (36.0–54.0) | 0.001 |
| UA (μmol/L) | 297.0 (251.0–351.0) | 308.5 (260.7–364.0) | 0.134 |
| TC (mmol/L) | 3.01 ± 0.65 | 3.19 (2.77–-3.76) | <0.000 |
| TG (mmol/L) | 0.84 (0.65–1.10) | 0.90 (0.70–1.24) | 0.094 |
| HDL (mmol/L) | 1.51 (1.25–1.86) | 1.18 (0.98–1.39) | 0.134 |
| LDL (mmol/L) | 1.70 (1.34–2.21) | 1.72 (1.35–2.18) | <0.000 |
ALT, alanine transaminase; AST, alanine transaminase; BUN, blood urea nitrogen; Cr, creatinine; FT4, free thyroxine; HDL, high-density lipoprotein; LDL, low-density lipoprotein; M/F, men/female; RAIU, radioactive iodine uptake rate; TC, total cholesterol; TG, triglyceride; TgAb, thyroglobulin antibodies; TPOAb, thyroid peroxidase antibodies; TRAb, thyrotropin receptor antibody; TSH, thyrotropin; UA, uric acid.
RAIU was an independent risk factor for the development of neutropenia in untreated GD patients
Determination of risk factors for patients with untreated GD complicated by neutropenia using univariate logistic regression (Table 2) revealed that FT4, 3 h RAIU, 24 h RAIU, Cr, TC, and LDL were associated with the development of neutropenia. Multivariable logistic regression analysis revealed that 3 h RAIU and 24 h RAIU were associated with neutropenia after adjusting for age, sex, BMI, and FT4 (OR = 1.021, 95% CI: 1.008–1.033, in Model 1; OR = 1.023, 95% CI: 1.007–1.039, in Model 2) (Table 3).
Univariable logistic regression analysis to determine the risk factors for the development of neutropenia in the study.
| Outcome: neutropenia | ||
|---|---|---|
| Variables | OR ( 95% CI ) | P |
| Sex (M/F) | 0.820 (0.611–1.102) | 0.188 |
| Age (year) | 1.008 (0.998–1.018) | 0.126 |
| BMI (kg/m2) | 0.996 (0.937–1.059) | 0.909 |
| FT4 (pmol/L) | 1.006 (1.002–1.010) | 0.002 |
| 3 h RAIU (%) | 1.020 (1.012–1.028) | <0.001 |
| 24 h RAIU (%) | 1.017 (1.007–1.028) | 0.001 |
| TPOAb (IU/L) | 1 (1–1) | 0.928 |
| TGAb (IU/L) | 1 (1–1) | 0.303 |
| TRAb (IU/L) | 0.998 (0.97–1.009) | 0.698 |
| ALT (U/L) | 1.002 (0.997–1.006) | 0.451 |
| AST (U/L) | 1 (0.992–1.008) | 0.964 |
| BUN (mmol/L) | 0.979 (0.87–1.103) | 0.732 |
| Cr (μmol/L) | 0.975 (0.963–0.988) | <0.001 |
| UA (μmol/L) | 0.998 (0.996–1.001) | 0.150 |
| TC (mmol/L) | 0.584 (0.444–0.767) | <0.001 |
| TG (mmol/L) | 0.725 (0.486–1.082) | 0.116 |
| HDL (mmol/L) | 0.615 (0.335–1.130) | 0.117 |
| LDL (mmol/L) | 0.467 (0.321–0.678) | <0.001 |
ALT, alanine transaminase; AST, alanine transaminase; BUN, blood urea nitrogen; Cr, creatinine; FT4, free thyroxine; HDL, high-density lipoprotein; LDL, low-density lipoprotein; M/F, men/female; RAIU, radioactive iodine uptake rate; TC, total cholesterol; TG, triglyceride; TgAb, thyroglobulin antibodies; TPOAb, thyroid peroxidase antibodies; TRAb, thyrotropin receptor antibody; TSH, thyrotropin; UA, uric acid.
Multivariable logistic regression analysis to determine the risk factors for the development of granulocytopenia in the study.
| Outcome: neutropenia | |||
|---|---|---|---|
| Variables | OR ( 95% CI ) | P | |
| Model 1 | Sex (M/F) | 0.818 (0.609–1.099) | 0.183 |
| Age (year) | 1.008 (0.993–1.024) | 0.276 | |
| BMI (kg/m2) | 0.995 (0.928–1.067) | 0.886 | |
| FT4 (pmol/L) | 1.003 (0.998–1.009) | 0.255 | |
| 3 h RAIU (%) | 1.021 (1.008–1.033) | 0.001 | |
| Model 2 | Sex (M/F) | 0.879 (0.575–1.343) | 0.551 |
| Age (year) | 1.006 (0.991–1.021) | 0.441 | |
| BMI (kg/m2) | 0.998 (0.930–1.071) | 0.963 | |
| FT4 (pmol/L) | 1.004 (0.999–1.010) | 0.141 | |
| 24 h RAIU (%) | 1.023 (1.007–1.039) | 0.004 | |
FT4, free thyroxine; M/F, men/female; RAIU, radioactive iodine uptake rate.
Patients were divided into four groups based on quartile of 3 h RAIU (Q1 (<37.32%), Q2 (37.32–50.72%), Q3 (50.72–64.08%), and Q4 (>64.08%)) or quartile of 24 h RAIU (Q1 (<55.59%), Q2 (55.59–65.83%), Q3 (65.83–74.77%), and Q4 (>74.77%)). There was a gradual decrease in neutrophil count as 3 h RAIU or 24 h RAIU increased, while FT4 and TRAb levels increased gradually (Tables 4 and 5). We then analyzed the risk factors for neutropenia incidence in GD patients using logistic regression with different quartiles of 3 h RAIU or 24 h RAIU as dependent variables in fully adjusted models. Remarkably, the risk of neutropenia occurrence increased 2.241-fold in the 3 h RAIU Q4 vs 3 h RAIU Q1 group (95% CI: 1.243–4.041, P = 0.007), and 2.054-fold in the 24 h RAIU Q4 vs 24 h RAIU Q1 group (95% CI: 1.156–3.649, P = 0.014), regardless of age, BMI, sex, or FT4 (Table 6).
Clinical characteristics of subjects when patients were divided into four groups based on quartiles of 3 h RAIU.
| Variables | Q1 | Q2 | Q3 | Q4 | P |
|---|---|---|---|---|---|
| n | 250 | 250 | 250 | 250 | - |
| Sex (M/F) | 153/97 | 162/88 | 168/82 | 190/60 | 0.001 |
| Age (year) | 45 (35–55) | 43 (31–53) | 40 (31–49) | 35 (28–45) | <0.001 |
| BMI (kg/m2) | 21.54 ± 3.38 | 21.08 (19.46–23.40) | 21.48 ± 2.67 | 20.19 (18.65–22.31) | 0.137 |
| Neutrophil (×109/L) | 2.69 (2.12–3.15) | 2.43 (1.95–3.17) | 2.27 (1.83–3.03) | 2.20 (1.68–3.02) | <0.001 |
| FT4 (pmol/L) | 33.76 (19.81–52.38) | 48.47 (23.58–63.59) | 48.19 (29.21–77.05) | 63.10 (35.62–93.12) | <0.001 |
| TSH (mIU/L) | 0.005 (0.00–0.010) | 0.005 (0.000–0.010) | 0.005 (0.000–0.010) | 0.005 (0.00–0.010) | 0.993 |
| TPOAb (IU/L) | 142.15 (24.35–454.13) | 130.55 (23.65–3434.12) | 151.90 (43.25–408.25) | 205.0 (31.15–518.7) | 0.622 |
| TGAb (IU/L) | 140.95 (39.80–433.70) | 153.85 (40.60–444.42) | 113.05 (39.62–406.95) | 82.7 (29.82–348.20) | 0.267 |
| TRAb (IU/L) | 6.56 (3.59–16.87) | 9.65 (4.82–20.80) | 11.53 (6.41–22.50) | 18.39 (10.20–34.81) | <0.001 |
| 24 h RAIU (%) | 52.6 ± 8.9 | 62.19 ± 9.06 | 69.77 (64.99–75.71) | 76.93 (71.33–83.17) | <0.001 |
| ALT (U/L) | 25 (16–40) | 30 (21–46) | 28 (19–45) | 37 (19–41) | 0.053 |
| AST (U/L) | 23 (17–31) | 24 (19–32) | 23 (16–33) | 22 (16–31) | 0.401 |
| BUN (mmol/L) | 5 (4–6) | 5 (4–6) | 5 (4–6) | 5 ± 1 | 0.405 |
| Cr (μmol/L) | 45 (39–59) | 46 (38–54) | 41 (34–51) | 39 (31–49) | <0.001 |
| UA (μmol/L) | 304 (254–372) | 316 ± 79 | 311 ± 78 | 305 (253–359) | 0.730 |
| TC (mmol/L) | 3.32 ± 0.84 | 3.14 (2.80–3.72) | 3.13 (2.75–3.55) | 2.98 (2.60–3.38) | 0.016 |
| TG (mmol/L) | 0.93 (0.73–1.37) | 0.95 (0.71–1.35) | 0.88 (0.67–1.17) | 0.81 (0.66–1.07) | 0.134 |
| HDL (mmol/L) | 1.16 ± 0.32 | 1.18 (0.99–1.39) | 1.16 (0.98–1.39) | 1.16 (1.00–1.37) | 0.562 |
| LDL (mmol/L) | 1.79 (1.36–2.33) | 1.69 (1.33–2.22) | 1.64 (1.32–2.08) | 1.54 (1.26–1.84) | 0.014 |
Data are means ± s.d. or median (interquartile range).
ALT, alanine transaminase; AST, alanine transaminase; BUN, blood urea nitrogen; Cr, creatinine; FT4, free thyroxine; HDL, high-density lipoprotein; LDL, low-density lipoprotein; M/F, men/female; RAIU, radioactive iodine uptake rate; TC, total cholesterol; TG, triglyceride; TgAb, thyroglobulin antibodies; TPOAb, thyroid peroxidase antibodies; TRAb, thyrotropin receptor antibody; TSH, thyrotropin; UA, uric acid.
Clinical characteristics of subjects when patients were divided into four groups based on quartiles of 24 h RAIU.
| Variables | Q1 | Q2 | Q3 | Q4 | P |
|---|---|---|---|---|---|
| n | 250 | 250 | 250 | 250 | - |
| Sex (M/F) | 147/103 | 149/101 | 178/72 | 202/48 | <0.001 |
| Age (year) | 42 (33–52) | 41 (32–54) | 42 (30–52) | 37 (28–48) | 0.145 |
| BMI (kg/m2) | 21.79 ± 3.40 | 21.08 ± 2.68 | 20.83 (18.90–22.86) | 20.74 ± 2.55 | 0.209 |
| Neutrophil (×109/L) | 2.67 (2.06–3.43) | 2.55 (1.93–3.49) | 2.22 (1.80–2.81) | 2.26 (1.77–2.95) | <0.001 |
| FT4 (pmol/L) | 34.78 (31.88–57.54) | 44.80 (15.47–68.02) | 54.80 (33–76.57) | 55.73 (35.13–76.90) | <0.001 |
| TSH (mIU/L) | 0.005 (0.000–0.010) | 0.005 (0.000–0.010) | 0.005 (0.000–0.010) | 0.005 (.000–0.010) | 0.252 |
| TPOAb (IU/L) | 155.90 (30.90–410.97) | 130.55 (22.60–439.07) | 158.70 (28.75–370.72) | 161.90 (43.47–525.40) | 0.939 |
| TGAb (IU/L) | 153.85 (10.02–412.07) | 136.9 (32.35–532.80) | 116.80 (38.00–385.05) | 92.50 (32.10–340.30) | 0.730 |
| TRAb (IU/L) | 8.39 (3.64–21.61) | 10.21 (4.53–22.12) | 12.37 (7.00–23.58) | 13.4 (6.95–27.51) | 0.009 |
| 3 h RAIU (%) | 32.92 (27.58–41.20) | 43.54 (36.11–52.12) | 56.70 ± 11.73 | 68.27 ± 13.33 | <0.001 |
| ALT (U/L) | 26 (17–40) | 29 (20–45) | 30 (21–43) | 27 (19–43) | 0.059 |
| AST (U/L) | 23 (17–30) | 23 (17–31) | 24 (18–33) | 22.47 (16.17–32.85) | 0.057 |
| BUN (mmol/L) | 5 (4–6) | 5 ± 2 | 5 ± 1 | 5 (4–6) | 0.510 |
| Cr (μmol/L) | 46 (41–57) | 44 (36–54) | 40 (33–49) | 41 (33–51) | <0.001 |
| UA (μmol/L) | 305 (253–361) | 320 ± 81 | 290 (251–337) | 324 ± 78 | 0.021 |
| TC (mmol/L) | 3.28 ± 0.82 | 3.12 ± 0.80 | 3.19 (2.80–3.51) | 3.09 (2.76–3.53) | 0.542 |
| TG (mmol/L) | 0.89 (0.74–1.35) | 0.85 (0.66–1.22) | 0.87 (0.68–1.22) | 0.91 (0.69–1.12) | 0.484 |
| HDL (mmol/L) | 1.09 (0.93–1.39) | 1.11 (0.95–1.32) | 1.24 ± 0.31 | 1.19 (1.02–1.37) | 0.150 |
| LDL (mmol/L) | 1.74 (1.35–2.29) | 1.59 (1.22–2.02) | 1.66 (1.39–2.01) | 1.63 (1.31–2.00) | 0.773 |
Data are means ± s.d. or median (interquartile range).
ALT, alanine transaminase; AST, alanine transaminase; BUN, blood urea nitrogen; Cr, creatinine; FT4, free thyroxine; HDL, high-density lipoprotein; LDL, low-density lipoprotein; M/F, men/female; RAIU, radioactive iodine uptake rate; TC, total cholesterol; TG, triglyceride; TgAb, thyroglobulin antibodies; TPOAb, thyroid peroxidase antibodies; TRAb, thyrotropin receptor antibody; TSH, thyrotropin; UA, uric acid.
Logistic regression evaluating the association of neutropenia with RAIU in the models.
| OR | 95% CI | P | OR | 95% CI | P | ||||
|---|---|---|---|---|---|---|---|---|---|
| 3 h RAIU | 24 h RAIU | ||||||||
| Model 1 | Q1 | Reference | - | - | Model 2 | Q1 | Reference | - | - |
| Q2 | 1.648 | 0.899–3.023 | 0.106 | Q2 | 1048 | 0.569–1.931 | 0.879 | ||
| Q3 | 1.681 | 0.933–3.031 | 0.084 | Q3 | 1.731 | 0.971–3.088 | 0.063 | ||
| Q4 | 2.241 | 1.243–4.04 | 0.007 | Q4 | 2.054 | 1.156–3.649 | 0.014 |
Model 1 further adjusted for the age, BMI, sex, FT4, and 3 h RAIU.
Model 2 further adjusted for the age, BMI, sex, FT4, and 24 h RAIU.
FT4, free thyroxine; RAIU, radioactive iodine uptake rate.
CD40 increased expression of NIS
NIS is a key protein that mediates active iodine uptake in the thyroid follicular epithelial cells and reflects the iodine uptake capacity. Cells were unstimulated 0 or treated with 2 μg/mL and 10 μg/mL anti-CD40 for a period of 48 h. Membrane surface protein NIS was analyzed by flow cytometry. The study shows that with the increase of anti-CD40 stimulation concentration, NIS expression increased.
Discussion
We retrospectively assessed the percentage of neutropenia in patients with new-onset or relapsing GD without treatment. Among them, 29.7% had neutropenia. We enrolled the patients who were willing to choose RAI treatment, but not in natural population. Among them, many GD patients suffered from neutropenia. This may result in a significant increase in GD patients with neutropenia according to the population selection bias. Considering this was a retrospective case–control study, the population selection bias did not affect the aim of the study.
The mechanism of GD in patients with neutropenia is not completely clear. The current data cannot determine the relationship between the RAIU and the functional status of immune cells. Granulocyte colony-stimulating factor (G-CSF) influences the survival, proliferation, and differentiation of neutrophils and is an effective treatment for ATD-induced agranulocytosis (6). To maintain a normal number of granulocytes, GD patients may require a higher level of G-CSF. A study by Iitaka et al. (7) revealed that serum G-CSF level in patients with untreated GD was significantly higher compared with healthy controls. Nonetheless, those with neutropenia had a lower serum G-CSF level than other patients with untreated GD and no neutropenia (7). Gao et al. revealed that neutropenia in GD was associated with increased levels of E-selectin that promoted leukocyte adhesion on vessel walls and leukocyte aggregation at the edge of the pool (8). Weitzman reported antineutrophil autoantibodies in some patients with GD that induced immune neutropenia (9, 10). The current data cannot determine the relationship between the RAIU and the functional status of immune cells. Kawa et al. reported thyroid hormone receptor gene especially thyroid hormone receptor (TRα1) gene expression in bone marrow CD34+-enriched progenitor cells. The percentage of apoptotic cells among human bone marrow CD34+-enriched progenitor cells increased gradually with an increase in T3 concentration. Furthermore, hyperthyroidism significantly affected the proliferative potential of human hematopoietic stem/progenitor cells by decreasing the expression of pro- and anti-apoptotic genes, such as B-cell lymphoma-2 (BCL-2), B-cell lymphoma-XL (BCL-xL), and BCL2-associated X protein (BAX) (11, 12). In our study, the current data cannot determine the relationship between the RAIU and the functional status of immune cells. Patients in the neutropenia group had a much higher level of FT4 compared with the non-neutropenia group. It appears that thyroid hormone played an important role in influencing blood cell formation.
It is established that granulocytopenia is common in GD, but it is not present in all patients, even those with severely abnormal thyroid function. Our study revealed that RAIU, not thyroid function, was closely related to the development of neutropenia after adjusting for confounding factors. As we know, iodine uptake was the first step in thyroxine synthesis. So thyroid hormones increased after pathogenic TRAb stimulated iodine uptake in the thyroid follicular epithelium in GD patients. NIS is a key protein involved in iodine uptake in the thyroid follicular epithelium, mainly distributed in the thyroid, mammary and salivary glands, lacrimal gland sac, gastrointestinal tract, placenta, and other organs (2). Iodine uptake rate and 131I are widely used in clinical diagnosis and treatment of thyrotoxicosis due to the characteristics of NIS in the absorption of iodine. NIS is regulated by a number of factors. TSH binds TSHR to activate CREB phosphorylation of the cAMP pathway (13, 14), which can bind to the upstream enhancer region of the NIS gene to promote NIS expression (15). In addition, many inflammatory factors such as interleukin (IL)-1α and IL-1β show concentration-dependent negative regulation of NIS (16, 17). IL1 is secreted mainly by neutrophils (15). Previous studies have found that peripheral blood IL-1β decreases in patients with hyperthyroidism. We speculated that neutropenia and IL-1 expression decreased in patients with GD and that this weakened the dose-dependent inhibition of NIS and further increased the iodine uptake. Mavroudi et al. found that the combination of CD40 and CD40 ligand in bone marrow promoted apoptosis of granulocytes at various stages by up-regulating the expression of Fas in the presence of inflammation (18). We confirmed in in vitro experiments that after co-culture of CD40L and thyroid cells, by interacting with CD40 on the thyroid cell surface, NIS expression was significantly increased (Supplementary Figure 1, see section on supplementary materials given at the end of this article). Considering elevated soluble CD40 ligand in newly diagnosed GD and the high expression of CD40 in the thyroid tissue of patients with GD (19), we speculated that the chronic inflammatory state and CD40 may influence RAIU and neutropenia in GD patients although basic research were needed to prove it.
There were some limitations of this study. First, it was a single-center study and not representative of the entire GD population. Second, the determination of iodine uptake rate is common in the diagnosis of hyperthyroidism. We did not evaluate the iodine uptake rate in patients with GD and neutropenia after treatment. We did not measure serum IL1 or CD40L in GD patients before or after granulocytopenia treatment. The correlation between the iodination rate and granulocytes needs further study. In our study, the neutrophil count returned to normal within 1–2 months in most neutropenic GD patients. We will do further research to enlarge the sample size and prolong the follow-up time.
Conclusions
Since neutropenia is common in patients with GD, a baseline assessment should be considered prior to initiation of ATD therapy. This study showed that patients with significantly increased iodine uptake rate are susceptible to the development of granulocytopenia. Our finding cannot be directly applied to clinical practice because further studies are needed to elucidate exactly how RAIU are related to the pathogenesis of neutropenia and modifications their effects may have on neutropenia development.
Supplementary materials
This is linked to the online version of the paper at https://doi.org/10.1530/EC-22-0474.
Declaration of interest
There is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
This work was supported by the Natural Science Research Funds of Minghang District, Shanghai (No. 2020MHZ077.), Grant sponsor: Medical Key Faculty Foundation of Shanghai(Grant number: ZK2019B15) and Natural Science Foundation of Shanghai (Grant number: 22ZR1448700).
Author contribution statement
Qian Yang, Wencai Ke, Fanfan Pan: Conceptualization, Methodology, Data curation, Writing- Original draft preparation; Qian Yang, Wencai Ke, Xinmei Huang: Software; Qian Yang, Bingbing Zha: Writing- Reviewing and Editing; Jun Liu, Bingbing Zha: Visualization, Supervision.
Acknowledgements
The authors thank Wei Zhang from Department of Biostatistics, School of Public Health, Fudan University for his help in statistical analysis.
References
- 1↑
Ross DS, Burch HB, Cooper DS, Greenlee MC, Laurberg P, Maia AL, Rivkees SA, Samuels M, Sosa JA & Stan MN et al.2016 American Thyroid Association guidelines for diagnosis and management of hyperthyroidism and other causes of thyrotoxicosis. Thyroid 2016 26 1343–1421. (https://doi.org/10.1089/thy.2016.0229)
- 2↑
Ravera S, Reyna-Neyra A, Ferrandino G, Amzel LM, Carrasco N. The sodium/iodide symporter (NIS): molecular physiology and preclinical and clinical applications. Annual Review of Physiology 2017 79 261–289. (https://doi.org/10.1146/annurev-physiol-022516-034125)
- 3↑
Czarnocka B Thyroperoxidase, thyroglobulin, Na(+)/I(-) symporter, pendrin in thyroid autoimmunity. Frontiers in Bioscience-Landmark 2011 16 783–802. (https://doi.org/10.2741/3720)
- 4↑
Takata K, Kubota S, Fukata S, Kudo T, Nishihara E, Ito M, Amino N, Miyauchi A. Methimazole-induced agranulocytosis in patients with Graves' disease is more frequent with an initial dose of 30 mg daily than with 15 mg daily. Thyroid 2009 19 559–563. (https://doi.org/10.1089/thy.2008.0364)
- 5↑
Aggarwal N, Tee SA, Saqib W, Fretwell T, Summerfield GP, Razvi S. Treatment of hyperthyroidism with antithyroid drugs corrects mild neutropenia in Graves' disease. Clinical Endocrinology 2016 85 949–953. (https://doi.org/10.1111/cen.13133)
- 6↑
Wang Y, Li X, Yang Q, Wang W, Zhang Y, Liu J, Zheng L, Zha B. Granulocyte-colony-stimulating factor effectively shortens recovery duration in anti-thyroid-drug-induced agranulocytosis: a systematic review and meta-analysis. Frontiers in Endocrinology 2019 10 789. (https://doi.org/10.3389/fendo.2019.00789)
- 7↑
Iitaka M, Noh JY, Kitahama S, Fukasawa N, Miura S, Kawakami Y, Kawasaki S, Yamanaka K, Ishii J & Katayama S et al.Elevated serum granulocyte colony-stimulating factor levels in patients with Graves' disease. Clinical Endocrinology 1998 48 275–280. (https://doi.org/10.1046/j.1365-2265.1998.00422.x)
- 8↑
Gao Y, Shen HJ, Zhou P, Hu H, Tang JL, Peng LL, Tong J. Relationship between leukopenia and intercellular adhesion molecules in Graves' disease. West Indian Medical Journal 2014 63 601–604. (https://doi.org/10.7727/wimj.2013.113)
- 9↑
Zheng P, Chang X, Lu Q, Liu Y. Cytopenia and autoimmune diseases: a vicious cycle fueled by mTOR dysregulation in hematopoietic stem cells. Journal of Autoimmunity 2013 41 182–187. (https://doi.org/10.1016/j.jaut.2012.12.011)
- 10↑
Afeltra A, Paggi A, De Rosa FG, Manfredini P, Addessi MA, Amoroso A. Antineutrophil cytoplasmic antibodies in autoimmune thyroid disorders. Endocrine Research 1998 24 185–194. (https://doi.org/10.1080/07435809809135527)
- 11↑
Kawa MP, Grymula K, Paczkowska E, Baskiewicz-Masiuk M, Dabkowska E, Koziolek M, Tarnowski M, Klos P, Dziedziejko V & Kucia M et al.Clinical relevance of thyroid dysfunction in human haematopoiesis: biochemical and molecular studies. European Journal of Endocrinology 2010 162 295–305. (https://doi.org/10.1530/EJE-09-0875)
- 12↑
Grymula K, Paczkowska E, Dziedziejko V, Baskiewicz-Masiuk M, Kawa M, Baumert B, Celewicz Z, Gawrych E, Machalinski B. The influence of 3,3’,5-triiodo-L-thyronine on human haematopoiesis.' Cell Proliferation 2007 40 302–315. (https://doi.org/10.1111/j.1365-2184.2007.00435.x)
- 13↑
Schmitt TL, Espinoza CR, Loos U. Characterization of a thyroid-specific and cyclic adenosine monophosphate-responsive enhancer far upstream from the human sodium iodide symporter gene. Thyroid 2002 12 273–279. (https://doi.org/10.1089/10507250252949388)
- 14↑
Shaywitz AJ, Greenberg ME. CREB: a stimulus-induced transcription factor activated by a diverse array of extracellular signals. Annual Review of Biochemistry 1999 68 821–861. (https://doi.org/10.1146/annurev.biochem.68.1.821)
- 15↑
Jang D, Morgan SJ, Klubo-Gwiezdzinska J, Banga JP, Neumann S, Gershengorn MC. Thyrotropin, but not thyroid-stimulating antibodies, induces biphasic regulation of gene expression in human thyrocytes. Thyroid 2020 30 270–276. (https://doi.org/10.1089/thy.2019.0418)
- 16↑
Spitzweg C, Joba W, Morris JC, Heufelder AE. Regulation of sodium iodide symporter gene expression in FRTL-5 rat thyroid cells. Thyroid 1999 9 821–830. (https://doi.org/10.1089/thy.1999.9.821)
- 17↑
Rasmussen AK, Bendtzen K, Feldt-Rasmussen U. Thyrocyte-interleukin-1 interactions. Experimental and Clinical Endocrinology and Diabetes 2000 108 67–71. (https://doi.org/10.1055/s-2000-5797)
- 18↑
Mavroudi I, Papadaki HA. The role of CD40/CD40 ligand interactions in bone marrow granulopoiesis. The Scientific World Journal 2011 11 2011–2019. (https://doi.org/10.1100/2011/671453)
- 19↑
Smith TJ, Sciaky D, Phipps RP, Jennings TA. CD40 expression in human thyroid tissue: evidence for involvement of multiple cell types in autoimmune and neoplastic diseases. Thyroid 1999 9 749–755. (https://doi.org/10.1089/thy.1999.9.749)
