The favorable effects of long-term selenium supplementation on regression of cervical tissues and metabolic profiles of patients with cervical intraepithelial neoplasia: a randomised, double-blind, placebo-controlled trial – Expression of concern

Abstract

The Editor-in-Chief has been alerted to concerns about the integrity of the above article and is investigating the claims. The concerns have also been referred to the Iranian National Committee for Ethics in Biomedical Research. This statement will be updated when both investigations have been completed and the authors have been given the opportunity to respond to the outcomes of those investigations.

Scientific opinion on the tolerable upper intake level for selenium

“…(2015) Iran RCT G1, placebo: 35/35 G1, 200 μg Se/d: 35/35 Duration: 8 wk Aged 18–40 yr, with PCOS, before menopause; no use of Se supplements and metformin in the last 3 mo; non‐smokers; no diabetes or hypothyroidism; no special diet, oral conceptive, ovulation induction agents Sex: F Age (yr) G1: 25.7 ± 4.8 G2: 25.4 ± 5.1 BMI (kg/m 2 ) G1: 25.2 ± 4.1 G2: 25.0 ± 3.7 Ethnicity: NR Serum/plasma Se: NR Se dietary intake (μg/day, mean ± SD) (by 3‐d dietary records measured at week 2, 4, 6) G1: 58.5 ± 8.0 G2: 56.1 ± 10.5 Selenised yeast (200 μg Se/d) vs placebo Adherence, pills count (%): > 90% Fasting plasma glucose, fasting serum insulin, HOMA‐IR, HOMA‐B at baseline and at 8 wk Fasting plasma glucose (mmol/L) G1: 5.15 ± 0.39 G2: 4.91 ± 0.52 Fasting insulin (pmol/L) G1: 73.58 ± 59.50 G2: 80.69 ± 42.28 HOMA‐IR G1: 2.78 ± 2.25 G2: 3.00 ± 1.69 HOMA‐B G1: 2.78 ± 2.25 G2: 3.00 ± 1.69 Change from baseline (adjusted for baseline value, age and BMI) Fasting plasma glucose (mmol/L) G1: 0.05 ± 0.09 G2: −0.29 ± 0.09 p = 0.010 Fasting insulin (pmol/L) G1: 7.32 ± 9.73 G2: −28.09 ± 9.73 p = 0.012 HOMA‐IR G1: 0.38 ± 0.39 G2: −1.11 ± 0.39 p = 0.010 HOMA‐B G1: 3.06 ± 5.97 G2: −17.59 ± 5.97 p = 0.017 Karamali et al. (2015) Iran RCT G1, placebo: 28/28 G2, 200 μg Se/d: 28/28 Duration: 6 mo Aged 18–55 yr, with cervical intraepithelial neoplasia grade 1; no history of cervical cancer or other cancers of the lower genital tract; no history of hysterectomy or destructive therapy of the cervix; not pregnant. Sex: F Age (yr) G1: 38.3 ± 9.2 G2: 38.3 ± 9.1 BMI (kg/m 2 ) G1: 28.7 ± 3.9 G2: 28.6 ± 4.0 Ethnicity: NR Serum/plasma Se: NR Se intake: NR Selenised yeast (200 μg Se/d) vs placebo Adherence, pills count (%): > 90% Fasting plasma glucose, serum insulin, HOMA‐IR, HOMA‐B at baseline and 6 mo Fasting plasma glucose (mg/dL) G1: 89.4 ± 8.3 G2: 94.5 ± 12.1 Fasting insulin (μIU/mL) G1: 12.7 ± 4.4 G2: 13.9 ± 4.5 HOMA‐IR G1: 2.8 ± 1.0 G2: 3.3 ± 1.4 HOMA‐B G1: 48.1 ± 18.2 G2: 48.9 ± 14.0 Change from baseline (adjusted for baseline value, age and BMI) (mean ± SE) Fasting plasma glucose (mg/dL) G1: 0.3 ± 1.7 G2: −5.9 ± 1.7 p = 0.01 Fasting insulin (μIU/mL) G1: 1.9 ± 1.1 G2: −4.5 ± 1.1 p < 0.001 HOMA‐IR G1: 0.3 ± 0.2 G2: −1.2 ± 0.2 p < 0.001 HOMA‐B G1: 9.7 ± 4.9 G2: −14.7 ± 4.9 p = 0.001 Hosseinzadeh et al. (2016) Iran RCT G1, placebo: 30/27 G2, 200 μg Se/d: 30/26 Duration: 12 wk Aged 18–42 yr, with PCOS; non‐smokers; no congenital adrenal hyperplasia, Cushing's syndrome, androgen‐secreting tumours and hyperprolactinemia; no diabetes, hypo‐ or hyperthyroidism, renal dysfunction, liver disease or cardiovascular disease; no use of medications affecting metabolic and hormonal profile; no use of Se supplement in past 3 mo. Sex: F Age (yr, mean ± SE) G1: 28.90 ± 1.17 G2: 29.23 ± 0.96 BMI (kg/m 2 , mean ± SE) G1: 28.39 ± 0.72 G2: 27.4 ± 0.88 Ethnicity: NR Serum/plasma Se: NR Se intake (μg/d, mean ± SE) (6‐d 24‐h recall) G1: 36.4 ± 2.5 G2: 28.3 ± 3.1 Selenised yeast (200 μg Se/d) vs placebo Adherence, pills count (%): > 90% Fasting serum glucose, fasting insulin, HOMA‐IR at baseline and 12 wk Fasting serum glucose (mg/dL, mean ± SE) G1: 87.11 ± 1.90 G2: 87.34 ± 2.45 Fasting insulin (mU/L, mean ± SE) G1: 10...…”

“…Measures of glucose tolerance were investigated as the primary outcome in three trials (Jamilian et al., 2015; Hosseinzadeh et al., 2016; Raygan et al., 2018) and as secondary outcome in five trials (Richie et al., 2014; Karamali et al., 2015; Mesdaghinia et al., 2017; Jacobs et al., 2019; Tamtaji et al., 2019). Two trials measured FBG among other blood parameters (including markers of antioxidant or inflammatory status, blood lipids) (Hawkes et al., 2008; Navas‐Carretero et al., 2011).…”

“… Intervention studies investigating the effect of selenium supplementation vs placebo on fasting glucose and fasting insulin, excluding: Karamali et al. (2015), Mesdaghinia et al. (2017), Raygan et al.…”

“…Ten eligible RCTs reported on the effect of selenium supplementation on measures of glucose tolerance. FBG concentrations were measured in nine trials, among which seven also reported measures of FI concentration (Navas‐Carretero et al., 2011; Jamilian et al., 2015; Karamali et al., 2015; Hosseinzadeh et al., 2016; Mesdaghinia et al., 2017; Raygan et al., 2018; Tamtaji et al., 2019), while two studies did not (Hawkes et al., 2008; Richie et al., 2014). The supplemental intake of selenium ranged from 30 to 300 μg/day and the mean/median study intervention period ranged from 8 to 48 weeks.…”

“…Among the RCTs which reported on measures of glucose tolerance, seven studies also assessed changes in the HOMA2‐IR (Navas‐Carretero et al., 2011; Jamilian et al., 2015; Karamali et al., 2015; Hosseinzadeh et al., 2016; Mesdaghinia et al., 2017; Raygan et al., 2018; Tamtaji et al., 2019), among which four also HOMA2‐β. A statistically significant decrease in HOMA2‐IR was found in six trials (30–200 μg/day selenium for 8–26 weeks, Navas‐Carretero et al., 2011; Jamilian et al., 2015; Karamali et al., 2015; Mesdaghinia et al., 2017; Raygan et al., 2018; Tamtaji et al., 2019), with a consistent decrease in HOMA2‐β in the three trials reporting this parameter (Jamilian et al., 2015; Karamali et al., 2015; Mesdaghinia et al., 2017). In contrast, a statistically significant increase in HOMA2‐IR was found in one trial (200 μg/day for 12 weeks; Hosseinzadeh et al., 2016) (Appendix D.9.2).…”