Hemoglobin
Hemoglobin is the oxygen-carrying protein in red blood cells. WHO defines anemia as below 12.0 g/dL in women and below 13.0 g/dL in men, while US survey ranges and other proposals differ by sex, age and ancestry. Cohorts link both low and high levels to more deaths. Readings run higher at altitude and in smokers.
- Status
- Part of the complete blood count, ordered by a clinician and also sold direct to consumers by some US laboratories. FDA classifies automated hemoglobin systems and automated cell counters, which may also measure hemoglobin and hematocrit, as class II devices [1][2].
- Measures
- The concentration of hemoglobin, the oxygen-carrying protein of red blood cells, in whole blood. Hematocrit, reported beside it, is the share of blood volume taken up by red cells.
- Also called
- Hb, Hgb, HGB, hemoglobin, haemoglobin, blood hemoglobin, hemoglobin concentration, CBC hemoglobin
- Tested for
- anemia, fatigue, iron deficiency, blood loss, heavy periods, pregnancy check, blood donation eligibility, polycythemia, living at altitude, general checkup
Strong several reviews agree · Moderate one review or consistent trials · Limited one or a few small studies · Untested no study in people, listed with what it rests on
Linked in research4
- StrongShown in kidney disease trials: aiming for a higher hemoglobin with erythropoietin-type drugs raised stroke, blood pressure and access clotting, and did not lower deaths.
- ModerateLinked to more deaths at both low and high levels in three cohorts of middle-aged and older adults, with the lowest risk in the middle of the range.
- ModerateLinked to more low birth weight, preterm birth and stillbirth at low maternal hemoglobin, and to more stillbirth at high levels, in pooled studies.
- LimitedLimited evidence: one Dutch cohort linked both low and high hemoglobin to more dementia over 12 years.
Reported risks0
No risks sourced yet.
Reference ranges
Each range is the one its source published, in g/dL. Also reported in g/L. Where sources differ, each is shown.
World Health Organization guideline, 2024, nonpregnant women
Nonpregnant women aged 15 to 65 · [3]
| Range | What the source calls it |
|---|---|
| below 8 g/dL | "severe anaemia" A doctor, immediately |
| 8 to 10.9 g/dL | "moderate anaemia" |
| 11 to 11.9 g/dL | "mild anaemia" |
| 12 or above g/dL | "no anaemia" |
WHO prints its cutoffs in g/L (below 120 g/L for any anemia). The cutoff is based on the 5th percentile of hemoglobin in apparently healthy people, mostly from higher-income settings. WHO kept its earlier severity bands, citing insufficient evidence to change them. It gives no upper cutoff, so this block does not classify high results. WHO adjusts the cutoffs for altitude and smoking and makes no adjustment for ancestry [3].
World Health Organization guideline, 2024, men
Men aged 15 to 65 · [3]
| Range | What the source calls it |
|---|---|
| below 8 g/dL | "severe anaemia" A doctor, immediately |
| 8 to 10.9 g/dL | "moderate anaemia" |
| 11 to 12.9 g/dL | "mild anaemia" |
| 13 or above g/dL | "no anaemia" |
Printed as below 130 g/L for any anemia. WHO notes that data in adult men may suggest a higher cutoff of 135 g/L (13.5 g/dL), and that its guideline group kept 130 g/L given uncertainty in the evidence. No upper cutoff is given [3].
World Health Organization guideline, 2024, pregnancy, first trimester
Pregnant women in the first trimester · [3]
| Range | What the source calls it |
|---|---|
| below 7 g/dL | "severe anaemia" A doctor, immediately |
| 7 to 9.9 g/dL | "moderate anaemia" |
| 10 to 10.9 g/dL | "mild anaemia" |
| 11 or above g/dL | "no anaemia" |
Printed as below 110 g/L for any anemia. No upper cutoff is given [3]. The US Centers for Disease Control and Prevention published the same first trimester cutoff of 11.0 g/dL in 1998 [4].
World Health Organization guideline, 2024, pregnancy, second trimester
Pregnant women in the second trimester · [3]
| Range | What the source calls it |
|---|---|
| below 7 g/dL | "severe anaemia" A doctor, immediately |
| 7 to 9.4 g/dL | "moderate anaemia" |
| 9.5 to 10.4 g/dL | "mild anaemia" |
| 10.5 or above g/dL | "no anaemia" |
Printed as below 105 g/L for any anemia. WHO describes this as the existing cutoff for the second trimester, kept in this guideline. No upper cutoff is given [3]. The 1998 US cutoff for the second trimester is also 10.5 g/dL [4].
World Health Organization guideline, 2024, pregnancy, third trimester
Pregnant women in the third trimester · [3]
| Range | What the source calls it |
|---|---|
| below 7 g/dL | "severe anaemia" A doctor, immediately |
| 7 to 9.9 g/dL | "moderate anaemia" |
| 10 to 10.9 g/dL | "mild anaemia" |
| 11 or above g/dL | "no anaemia" |
Printed as below 110 g/L for any anemia. No upper cutoff is given [3]. The 1998 US cutoff for the third trimester is also 11.0 g/dL [4].
World Health Organization guideline, 2024, children aged 6 to 23 months
Children aged 6 to 23 months · [3]
| Range | What the source calls it |
|---|---|
| below 7 g/dL | "severe anaemia" A doctor, immediately |
| 7 to 9.4 g/dL | "moderate anaemia" |
| 9.5 to 10.4 g/dL | "mild anaemia" |
| 10.5 or above g/dL | "no anaemia" |
Printed as below 105 g/L for any anemia. This is the one age group whose cutoff WHO changed in 2024. No upper cutoff is given [3].
World Health Organization guideline, 2024, children aged 24 to 59 months
Children aged 24 to 59 months · [3]
| Range | What the source calls it |
|---|---|
| below 7 g/dL | "severe anaemia" A doctor, immediately |
| 7 to 9.9 g/dL | "moderate anaemia" |
| 10 to 10.9 g/dL | "mild anaemia" |
| 11 or above g/dL | "no anaemia" |
Printed as below 110 g/L for any anemia. No upper cutoff is given [3].
World Health Organization guideline, 2024, children aged 5 to 11
Children aged 5 to 11 years · [3]
| Range | What the source calls it |
|---|---|
| below 8 g/dL | "severe anaemia" A doctor, immediately |
| 8 to 10.9 g/dL | "moderate anaemia" |
| 11 to 11.4 g/dL | "mild anaemia" |
| 11.5 or above g/dL | "no anaemia" |
Printed as below 115 g/L for any anemia. No upper cutoff is given [3].
World Health Organization guideline, 2024, children aged 12 to 14
Boys and nonpregnant girls aged 12 to 14 years · [3]
| Range | What the source calls it |
|---|---|
| below 8 g/dL | "severe anaemia" A doctor, immediately |
| 8 to 10.9 g/dL | "moderate anaemia" |
| 11 to 11.9 g/dL | "mild anaemia" |
| 12 or above g/dL | "no anaemia" |
Printed as below 120 g/L for any anemia. WHO lists boys and nonpregnant girls of this age on separate rows with the same cutoffs. No upper cutoff is given [3].
US Centers for Disease Control and Prevention recommendations, 1998, men
Men aged 18 and older · [4]
| Range | What the source calls it |
|---|---|
| below 13.5 g/dL | "anemia" |
| 13.5 or above g/dL | No name given |
Based on the 5th percentile in the third US National Health and Nutrition Examination Survey, after excluding people with a high likelihood of iron deficiency. The matching hematocrit cutoff is below 39.9%. The same table gives separate cutoffs for children, adolescents and each stage of pregnancy, which are not shown as bands here. No upper cutoff is given [4].
US Centers for Disease Control and Prevention recommendations, 1998, women
Nonpregnant and lactating women aged 18 and older · [4]
| Range | What the source calls it |
|---|---|
| below 12 g/dL | "anemia" |
| 12 or above g/dL | No name given |
Derived the same way as the cutoff for men. The matching hematocrit cutoff is below 35.7%. No upper cutoff is given [4].
US national survey laboratory procedure manual, 2017 to 2018, men aged 19 to 65
Men aged 19 to 65, as the reference range used by the survey laboratory · [5]
| Range | What the source calls it |
|---|---|
| below 13.1 g/dL | No name given |
| 13.1 to 17.5 g/dL | "reference range" |
| above 17.5 g/dL | No name given |
The 2.5th to 97.5th percentiles of everyone measured in the US national survey from 1999 to 2004, calculated by ranking, without screening out people with iron deficiency or illness. The hematocrit range in the same table is 38.7% to 51.4% [5].
US national survey laboratory procedure manual, 2017 to 2018, men aged 66 and older
Men aged 66 and older, as the reference range used by the survey laboratory · [5]
| Range | What the source calls it |
|---|---|
| below 11.4 g/dL | No name given |
| 11.4 to 17.1 g/dL | "reference range" |
| above 17.1 g/dL | No name given |
Derived the same way from the 1999 to 2004 survey sample. The hematocrit range is 33.9% to 50.9% [5].
US national survey laboratory procedure manual, 2017 to 2018, women aged 19 to 65
Women aged 19 to 65, as the reference range used by the survey laboratory · [5]
| Range | What the source calls it |
|---|---|
| below 10.6 g/dL | No name given |
| 10.6 to 15.6 g/dL | "reference range" |
| above 15.6 g/dL | No name given |
Derived the same way from the 1999 to 2004 survey sample, which was not screened for iron deficiency. The lower limit sits 1.4 g/dL under the WHO anemia cutoff for women [3][5]. The hematocrit range is 32.0% to 45.9% [5].
US national survey laboratory procedure manual, 2017 to 2018, women aged 66 and older
Women aged 66 and older, as the reference range used by the survey laboratory · [5]
| Range | What the source calls it |
|---|---|
| below 10.9 g/dL | No name given |
| 10.9 to 15.9 g/dL | "reference range" |
| above 15.9 g/dL | No name given |
Derived the same way from the 1999 to 2004 survey sample. The hematocrit range is 32.8% to 47.0% [5].
US national survey laboratory procedure manual, 2017 to 2018, boys aged 1 to 5
Boys aged 1 to 5, as the reference range used by the survey laboratory · [5]
| Range | What the source calls it |
|---|---|
| below 10.7 g/dL | No name given |
| 10.7 to 14.2 g/dL | "reference range" |
| above 14.2 g/dL | No name given |
Derived the same way from the 1999 to 2004 survey sample. The hematocrit range is 32.1% to 41.7% [5].
US national survey laboratory procedure manual, 2017 to 2018, boys aged 6 to 18
Boys aged 6 to 18, as the reference range used by the survey laboratory · [5]
| Range | What the source calls it |
|---|---|
| below 11.9 g/dL | No name given |
| 11.9 to 16.9 g/dL | "reference range" |
| above 16.9 g/dL | No name given |
Derived the same way from the 1999 to 2004 survey sample. The hematocrit range is 35.3% to 49.9% [5].
US national survey laboratory procedure manual, 2017 to 2018, girls aged 1 to 5
Girls aged 1 to 5, as the reference range used by the survey laboratory · [5]
| Range | What the source calls it |
|---|---|
| below 11 g/dL | No name given |
| 11 to 14.2 g/dL | "reference range" |
| above 14.2 g/dL | No name given |
Derived the same way from the 1999 to 2004 survey sample. The hematocrit range is 32.5% to 41.9% [5].
US national survey laboratory procedure manual, 2017 to 2018, girls aged 6 to 18
Girls aged 6 to 18, as the reference range used by the survey laboratory · [5]
| Range | What the source calls it |
|---|---|
| below 11.2 g/dL | No name given |
| 11.2 to 15.1 g/dL | "reference range" |
| above 15.1 g/dL | No name given |
Derived the same way from the 1999 to 2004 survey sample. The hematocrit range is 33.5% to 44.6% [5].
Nordic Reference Interval Project, 2004, women
Women aged 18 to 90 in Denmark, Finland, Iceland, Norway and Sweden · [6]
| Range | What the source calls it |
|---|---|
| below 11.7 g/dL | No name given |
| 11.7 to 15.3 g/dL | "reference interval" |
| above 15.3 g/dL | No name given |
The 2.5th to 97.5th percentiles in 1,826 healthy men and women, measured on 12 types of analyzer in 60 laboratories. Printed as 117 to 153 g/L. The hematocrit interval for women is 34.8% to 45.9% [6].
Nordic Reference Interval Project, 2004, men
Men aged 18 to 90 in Denmark, Finland, Iceland, Norway and Sweden · [6]
| Range | What the source calls it |
|---|---|
| below 13.4 g/dL | No name given |
| 13.4 to 17 g/dL | "reference interval" |
| above 17 g/dL | No name given |
Printed as 134 to 170 g/L, from the same healthy sample. The hematocrit interval for men is 39.5% to 50.0% [6].
Beutler and Waalen, proposed lower limits, 2006, white men aged 20 to 59
White men aged 20 to 59 · [7]
| Range | What the source calls it |
|---|---|
| below 13.7 g/dL | "below the lower limit of normal" |
| 13.7 or above g/dL | No name given |
The 5th percentile in a California health screening database, after excluding people with a transferrin saturation below 16% or ferritin below 10 ng/mL, and older people with raised creatinine or inflammation markers, with a US national survey used as confirmation. The authors propose it in place of the WHO cutoff of 13.0 g/dL, which they describe as based on very few data. No upper limit is proposed [7].
Beutler and Waalen, proposed lower limits, 2006, white men aged 60 and older
White men aged 60 and older · [7]
| Range | What the source calls it |
|---|---|
| below 13.2 g/dL | "below the lower limit of normal" |
| 13.2 or above g/dL | No name given |
Derived the same way. No upper limit is proposed [7].
Beutler and Waalen, proposed lower limits, 2006, white women
White women aged 20 and older · [7]
| Range | What the source calls it |
|---|---|
| below 12.2 g/dL | "below the lower limit of normal" |
| 12.2 or above g/dL | No name given |
Derived the same way. The authors give the same limit for women aged 20 to 49 and 50 and older. No upper limit is proposed [7].
Beutler and Waalen, proposed lower limits, 2006, Black men aged 20 to 59
Black men aged 20 to 59 · [7]
| Range | What the source calls it |
|---|---|
| below 12.9 g/dL | "below the lower limit of normal" |
| 12.9 or above g/dL | No name given |
Derived the same way. The authors attribute part of the difference from white adults to the high frequency of alpha-thalassemia, which they chose not to exclude [7]. WHO reports a mean about 1 g/dL lower in Black than in White individuals and states that cutoffs are not to be adjusted for ancestry, citing insufficient evidence [3].
Beutler and Waalen, proposed lower limits, 2006, Black men aged 60 and older
Black men aged 60 and older · [7]
| Range | What the source calls it |
|---|---|
| below 12.7 g/dL | "below the lower limit of normal" |
| 12.7 or above g/dL | No name given |
Derived the same way, from a smaller sample. WHO does not adjust its cutoffs for ancestry [3][7].
Beutler and Waalen, proposed lower limits, 2006, Black women
Black women aged 20 and older · [7]
| Range | What the source calls it |
|---|---|
| below 11.5 g/dL | "below the lower limit of normal" |
| 11.5 or above g/dL | No name given |
Derived the same way. The authors give the same limit for women aged 20 to 49 and 50 and older. WHO does not adjust its cutoffs for ancestry [3][7].
World Health Organization classification of blood cancers, 2016, polycythemia vera criterion, men
Men being assessed for polycythemia vera · [8]
| Range | What the source calls it |
|---|---|
| 16.5 or below g/dL | No name given |
| above 16.5 g/dL | "major criterion" |
One of three major criteria for diagnosing polycythemia vera, a bone marrow disorder. The others are a bone marrow biopsy finding and a JAK2 gene mutation, so a result above the threshold is not a diagnosis. The same criterion can be met by a hematocrit above 49%. The authors report criticism that the threshold will send many people without the disorder for testing [8].
World Health Organization classification of blood cancers, 2016, polycythemia vera criterion, women
Women being assessed for polycythemia vera · [8]
| Range | What the source calls it |
|---|---|
| 16 or below g/dL | No name given |
| above 16 g/dL | "major criterion" |
One of three major criteria, as for men. The same criterion can be met by a hematocrit above 48% [8].
What has been shown to move it
Shown to lower it5
Small drops in blood counts and albumin
Small drops in hematocrit, hemoglobin, white cells and albumin were seen on terazosin in trials, which the label attributes to blood dilution. PSA did not change.
The numbers and sources
- How common
- How often is not known: the sources give no rate.
- Effect
- Small but statistically significant decreases in hematocrit, hemoglobin, white blood cells, total protein and albumin were seen in controlled trials. Treatment for up to 24 months had no significant effect on PSA [19].
- Population
- People in terazosin controlled trials [19].
- Sources
- [19]
- Caveats
- The label gives no size for the drops.
Fenofibrate: blood counts
The label reports mild to moderate drops in hemoglobin, hematocrit and white cells after starting fenofibrate that then stabilized, with rare low platelets or white cells.
The numbers and sources
- How common
- How often is not known: the sources give no rate.
- Effect
- Mild to moderate decreases in hemoglobin, hematocrit and white blood cells were observed after starting fenofibrate and stabilized with long-term use. Thrombocytopenia and agranulocytosis have been reported [7].
- Population
- People starting fenofibrate [7].
- Sources
- [7]
- Caveats
- The label gives no sizes or rates.
Hemolysis in G6PD deficiency
Case reports: red blood cells broke down after high-dose IV vitamin C, and every reported patient with G6PD deficiency developed hemolysis.
The numbers and sources
- How common
- How often is not known: the sources give no rate.
- Effect
- Two patients with colorectal cancer given 1 g per kg of vitamin C developed hemolytic anemia with hemoglobin below 50 g/L, and recovered after blood replacement. Their review of published cases found that all patients with G6PD deficiency developed hemolysis after high-dose IV vitamin C [13]. A systematic review lists hemolysis in G6PD deficiency as a major risk and suggests screening first [9].
- Population
- Two patients [13]. Published case reports [13].
- Sources
- [13], [9]
- Caveats
- Case reports at cancer-care doses, far above the 10 g used in the fatigue trial. G6PD deficiency is an inherited enzyme deficiency.
Hemolytic anemia in G6PD deficiency
The label lists G6PD deficiency as a contraindication because of the risk of hemolytic anemia.
In the Alzheimer's trial, dose-dependent falls in hemoglobin were the most common laboratory abnormality, described as not clinically significant.
The numbers and sources
- How common
- How often is not known: the sources give no rate.
- Effect
- The label lists G6PD deficiency as a contraindication because of the risk of hemolytic anemia, and lists hemolytic anemia among its warnings [6]. In the Alzheimer's trial, dose-dependent falls in hemoglobin were the most common laboratory abnormality, described as not clinically significant [1].
- Population
- Label [6]; 891 trial participants [1].
- Sources
- [6], [1]
- Caveats
- [1] studied LMTM, not methylene blue.
Mouth sores, gut effects and blood count changes
Small trials found mouth sores, rash and gut problems in a few older adults, and falls in red cell counts without clinical effects.
Hemoglobin, hematocrit and red cell count fell on rapamycin, without clinical effects over the study.
The numbers and sources
- How common
- Very common: more than 1 in 10 people.
- Effect
- In the pilot, adverse effects on rapamycin were a facial rash (1), mouth sores (1) and gastrointestinal problems (2), against mouth sores (1) on placebo. Hemoglobin, hematocrit and red cell count fell on rapamycin, without clinical effects over the study [2]. In LAM, the label lists mouth sores, diarrhea, abdominal pain, nausea, acne and swelling among reactions in 20% or more [8].
- Population
- 25 older adults [2]; LAM patients in the label [8].
- Sources
- [2], [8]
- Caveats
- Small numbers. In PEARL, adverse events did not differ across groups [1].
Shown to raise it8
Raising iron stores
Shown to raise ferritin in iron-deficient adults and athletes without anemia, most in those starting with very low ferritin.
The numbers and sources
- Effect
- In iron-deficient non-anemic endurance athletes, iron raised serum ferritin (Hedges' g 1.088, 95% CI 0.914 to 1.263) and hemoglobin (g 0.695) [1]. In 13 trials of athletes, oral iron raised ferritin (SMD 1.27, 95% CI 0.44 to 2.10) and did not significantly change hemoglobin (SMD 1.31, 95% CI -0.29 to 2.93) [2]. In non-anemic iron-deficient adults, ferritin rose by 9.23 (95% CI 6.48 to 11.97) and hemoglobin by 4.01 g/L [3].
- Population
- 17 studies of endurance athletes [1]; 449 athletes in 13 RCTs [2]; 1,170 patients in 18 trials [3].
- Sources
- [1], [2], [3]
- Caveats
- In [2], the ferritin rise was evident in athletes starting at 12 µg/L or below, with minimal effect at higher starting ferritin.
Anemia with fibroids: relugolix combination
Shown to raise hemoglobin by more than 2 g/dL in 44% and 55% of anemic women with fibroids, against 17% and 6% on placebo, in two trials.
Compared with placebo [2]
The numbers and sources
- How common
- Very common: more than 1 in 10 people.
- Effect
- Among women with hemoglobin of 10.5 g/dL or less at the start, 44.2% (L1) and 55.0% (L2) on relugolix combination therapy gained more than 2 g/dL by week 24, against 17.2% and 5.7% on placebo [2]. A meta-analysis reported hemoglobin improved in 46.3% of patients [3].
- Population
- 43 and 40 anemic women on treatment against 29 and 53 on placebo, within 770 women [2]; 511 participants in 6 studies [3].
- Sources
- [2], [3]
- Caveats
- Iron was required for women with hemoglobin of 8 to 10 g/dL, and women under 8 g/dL were excluded [2].
Anemia with fibroids: elagolix with add-back
Shown to raise hemoglobin by more than 2 g/dL in 50% and 62% of anemic women with fibroids, against 16% and 21% on placebo, in two trials.
The numbers and sources
- How common
- Very common: more than 1 in 10 people.
- Effect
- Among women with hemoglobin of 10.5 g/dL or less, 62% (UF-1) and 50% (UF-2) on elagolix with add-back gained more than 2 g/dL by month 6, against 16% and 21% on placebo. Over 90% of these women took iron [6]. A meta-analysis found more women met the hemoglobin end point on elagolix than on placebo [8].
- Population
- 52 and 48 anemic women on treatment against 31 and 24 on placebo [6]; 1,217 patients in 4 trials [8].
- Sources
- [6], [8]
- Caveats
- Nearly all anemic women also took iron, so the rise reflects the drug on top of iron [6].
Anemia before fibroid surgery: leuprolide and other GnRH agonists
Shown to raise hemoglobin before fibroid surgery, by about 0.9 g/dL against placebo or no treatment, and with iron 75% reached target levels against 49% on iron alone.
Compared with iron alone [10]; placebo or no pretreatment [11]
The numbers and sources
- How common
- Very common: more than 1 in 10 people.
- Effect
- In women with hematocrit of 30% or less or hemoglobin of 10.2 g/dL or less, Lupron Depot 3.75 mg monthly with iron raised hemoglobin by a mean 4.2 g/dL and hematocrit by 10.1% at 3 months. At week 12, 75% on leuprolide with iron reached hematocrit of 36% and hemoglobin of 12 g/dL, against 49% on iron alone [10]. Across 10 trials of GnRH agonists before surgery, preoperative hemoglobin rose 0.88 g/dL more than with placebo or no pretreatment (95% CI 0.68 to 1.08, I² 0%, moderate certainty) [11].
- Population
- 104 women on leuprolide with iron and 98 on iron alone [10]; 834 premenopausal women in 10 trials [11].
- Sources
- [10], [11]
- Caveats
- The Cochrane review pools GnRH agonists as a class, not leuprolide alone [11]. The label notes some women may benefit from iron alone or from 1 to 2 months of treatment [10].
Anemia in pregnancy
Shown to halve anemia risk in non-anemic pregnant women, with no change in birth weight or preterm birth.
The numbers and sources
- Effect
- In 23 studies of 4,492 non-anemic pregnant women, oral iron raised hemoglobin by 6.95 g/L (95% CI 4.81 to 9.09) and ferritin by 12.22 ng/mL, and halved the risk of anemia (RR 0.50, 95% CI 0.34 to 0.74, high certainty, number needed to treat 10). Birth weight, preterm birth and cesarean section did not differ [7].
- Population
- Non-anemic pregnant women in RCTs and observational studies [7].
- Sources
- [7]
- Caveats
- Heterogeneity was high for hemoglobin (I² 91%) and ferritin (I² 87%). Reporting of harms was inconsistent, and the authors found too little evidence to judge them [7].
Hemoglobin before fibroid surgery
Shown to raise hemoglobin before fibroid surgery by about 0.9 g/dL more than placebo, in women taking SPRMs including ulipristal.
Compared with placebo [6]; GnRH agonists [6]
The numbers and sources
- Effect
- SPRMs raised preoperative hemoglobin by 0.93 g/dL against placebo (95% CI 0.52 to 1.34, 2 studies, 173 women, I² 0%, moderate certainty). Preoperative hemoglobin did not differ between GnRH agonists and other medicines including ulipristal (MD -0.02 g/dL, 95% CI -0.41 to 0.37, 242 women) [6].
- Population
- Premenopausal women before fibroid surgery in randomized trials [6].
- Sources
- [6]
- Caveats
- Two pooled trials, pooled across SPRMs. In one trial all women received iron [1].
Fibroid size and heavy bleeding
One small trial found green tea extract shrank fibroids by about a third over 4 months in women with symptomatic fibroids. An uncontrolled study found no size change.
Compared with placebo capsules of brown rice [13]; before treatment, in the same women [14][15]
In the green tea group, hemoglobin rose 0.7 g/dL (P 0.02) and average blood loss fell from 71 to 45 mL a month (P 0.001).
The numbers and sources
- Effect
- In a placebo-controlled trial, 800 mg a day of green tea extract (45% EGCG) for 4 months reduced total fibroid volume by 32.6%, while volume rose 24.3% on placebo (P 0.0001). Fibroid symptom severity fell 32.4% and quality of life rose 18.53% against placebo. In the green tea group, hemoglobin rose 0.7 g/dL (P 0.02) and average blood loss fell from 71 to 45 mL a month (P 0.001). No adverse effects or endometrial changes were seen in either group [13]. In 25 women who took green tea extract capsules for 6 months with no control group, fibroid size, symptom scores and laboratory values did not change, while the physical score of a quality of life questionnaire rose [14]. In 16 premenopausal women over 40 given 300 mg of EGCG with vitamin D and vitamin B6 for 90 days, mean fibroid size fell 17.8% and bleeding lasted 0.9 days less, with no change in flow intensity [15].
- Population
- 39 women aged 18 to 50 with symptomatic fibroids, 33 of them analyzed [13]; 25 women, median age 45, with few or no fibroid symptoms [14]; 16 women over 40 with fibroids [15].
- Sources
- [13], [14], [15]
- Caveats
- One small trial, with 11 women finishing on placebo [13]. The hemoglobin and blood loss changes are reported within the green tea group, not as differences from placebo [13]. [14] and [15] had no control group, and [15] combined EGCG with vitamin D and vitamin B6. The trial dose supplies about 360 mg of EGCG a day [13], within the 140 to about 1,000 mg range in liver injury case reports [8].
Heavy bleeding with fibroids (hormonal IUD and combined pill)
Limited evidence: in women with fibroids, the hormonal IUD lowered bleeding and raised hemoglobin more than the combined pill in one small trial, without clearly shrinking fibroids.
Compared with a low-dose combined pill and oral norethisterone acetate [27]; the levonorgestrel IUD and placebo [28]; comparisons in 11 studies the source abstract does not describe [29]
The numbers and sources
- Effect
- In one trial of 44 women with fibroids, the levonorgestrel IUD reduced bleeding on the alkaline hematin test by 77.50% more than a low-dose combined pill (95% CI 70.44 to 84.56) and raised hemoglobin by 1.50 g/dL more (95% CI 0.85 to 2.15) at 12 months. Fibroid size did not clearly differ (MD 1.90%, 95% CI -12.24 to 16.04). Against oral norethisterone acetate, the IUD improved a visual bleeding score and raised hemoglobin by 10.14% more at 6 months, and spotting was more common (64.3% against 30%) [27]. A review of 2 trials found the combined pill did less than the levonorgestrel IUD for bleeding, quality of life and hemoglobin, and in a single trial at high risk of bias it shrank fibroids more than placebo [28]. A review of 11 studies of 10 to 104 women each found the levonorgestrel IUD reduced menstrual blood loss and raised hemoglobin, ferritin and hematocrit, with no evidence that fibroids shrank. Expulsion was uncommon and linked to fibroids larger than 3 cm [29].
- Population
- 221 premenopausal women with fibroids in 4 trials [27]; women with fibroids and heavy bleeding in 2 trials [28]; premenopausal women with symptomatic fibroids in 11 studies [29].
- Sources
- [27], [28], [29]
- Caveats
- The Cochrane reviewers rated all of this evidence very low quality and wrote that they were uncertain of the effects [27]. The combined pill findings rest on single trials [28]. The 11-study review abstract does not describe the studies' designs [29].
How it compares with the reference method
Mixed results2
Finger-prick and point-of-care testing against a laboratory analyzer
Mixed results: most point-of-care methods averaged within 7% of a laboratory analyzer, while single finger-prick drops varied by about 1.6 g/dL in either direction.
The numbers and sources
- Compared with
- Venous blood measured on an automated hematology analyzer
- Effect
- In a review of method comparisons, finger-prick blood usually gave higher hemoglobin than venous blood, and single drops gave lower values than pooled capillary blood. Mean bias against the analyzer was 0.1 to 1.6 g/dL for the HemoCue Hb-201 and 0.05 to 0.6 g/dL for the Hb-301, and most comparisons fell within the 7% limit used by US laboratory regulators [17]. In a study of 149 children and adults, a HemoCue 201+ read 0.31 g/dL high on venous blood. After correcting for that, single finger-prick drops still read 0.42 g/dL high, with single results varying by about 1.6 g/dL in either direction, against 0.7 g/dL or less for pooled capillary or venous blood [18]. WHO's statement on measurement names venous blood and automated analyzers as the method for assessing hemoglobin, and says it cannot give a correction for capillary results because the size of the difference has not been confirmed [3].
- Population
- Method comparison studies from clinical and field settings [17]. 49 children, 50 adult women and 50 older adults in Mexico [18].
- Sources
- [17], [18], [3]
- Caveats
- Results differed by device model, blood source and field conditions [3][17]. The single-drop study tested one device model [18].
Variation from one draw to the next in one person
Shown to vary about 2.7% from one draw to the next within the same person, in a pooled estimate from a European laboratory federation.
The numbers and sources
- Compared with
- The same person's results on repeated venous draws
- Effect
- The European Federation of Clinical Chemistry and Laboratory Medicine database gives a pooled within-person variation of 2.7% for hemoglobin, with lower and upper limits of 1.7% and 3.6%, from 10 included estimates. Variation between people is 6.2%. For hematocrit the within-person figure is 2.8% [19]. At a hemoglobin of 14.0 g/dL, 2.7% is about 0.4 g/dL.
- Population
- Healthy adults in the studies the database pools [19].
- Sources
- [19]
- Caveats
- A summary across studies in healthy volunteers. The database states no threshold for agreement, so the result is given as numbers only [19].
Did not match the reference method1
Altitude and smoking
Shown to read higher with altitude and smoking, by 0.4 g/dL at 500 to 999 m, 1.4 g/dL at 2,000 to 2,499 m and 0.3 to 0.6 g/dL in smokers.
The numbers and sources
- Compared with
- Hemoglobin in people living near sea level who do not smoke
- Effect
- WHO states that hemoglobin is to be adjusted for the elevation at which a person lives and for smoking before it is compared with an anemia cutoff. Its adjustments, subtracted from the measured value, are 0 below 500 m, 0.4 g/dL at 500 to 999 m, 0.8 at 1,000 to 1,499 m, 1.1 at 1,500 to 1,999 m, 1.4 at 2,000 to 2,499 m, 1.8 at 2,500 to 2,999 m, 2.1 at 3,000 to 3,499 m, 2.5 at 3,500 to 3,999 m, 2.9 at 4,000 to 4,499 m and 3.3 at 4,500 to 4,999 m. For smokers the adjustments are 0.3 g/dL for fewer than 10 cigarettes a day or an unknown amount, 0.5 for 10 to 19 and 0.6 for more than 20, and the two adjustments add together [3]. In the analysis behind them, smokers averaged 0.33 g/dL higher than nonsmokers (95% CI 0.25 to 0.40), and the authors found that earlier adjustments were too small at lower elevations and too large at higher ones [16][3]. The 1998 US recommendations start adjusting at 3,000 feet, by 0.2 g/dL, rising to 2.0 g/dL at 10,000 to 11,000 feet, and adjust by 0.3 to 0.7 g/dL for smoking [4].
- Population
- 68,193 preschool children and nonpregnant women with altitude data and 19,826 women with smoking data, from 13 population surveys and 1 cohort study [16]. A combined analysis of 3,440,932 people for the 2024 adjustments [3].
- Sources
- [3], [16], [4]
- Caveats
- WHO rates the altitude evidence as moderate quality, notes that data above 3,000 m come mainly from Central and South America, and reports that the effect of smoking was larger in women than in men. Smoking was self-reported [3].
What studies tie it to
Linked in research4
Death from any cause
Linked to more deaths at both low and high levels in three cohorts of middle-aged and older adults, with the lowest risk in the middle of the range.
The numbers and sources
- Effect
- In 5,888 US adults aged 65 and older followed 11.2 years, the lowest fifth of hemoglobin (below 13.7 g/dL in men, below 12.6 g/dL in women) and the highest fifth had adjusted hazard ratios for death of 1.33 (95% CI 1.15 to 1.54) and 1.17 (1.01 to 1.36) against the fourth fifth. Anemia by WHO criteria had a hazard ratio of 1.38 (1.19 to 1.54) [9]. In 17,030 Canadian adults aged 66 and older, hemoglobin below 11.0 g/dL in people with normal kidney function had a hazard ratio for death of 4.29 (3.55 to 5.12), and risk was lowest at 13.0 to 15.0 g/dL in women and 14.0 to 17.0 g/dL in men [10]. In 292,194 Korean adults over 40, the association with cardiovascular and all-cause death was U-shaped or J-shaped. Men whose anemia had resolved by a second exam had a hazard ratio for death of 0.67 (0.59 to 0.77) against men whose anemia persisted [11].
- Population
- 5,888 community-dwelling adults aged 65 and older in 4 US communities [9]. 17,030 community-dwelling adults aged 66 and older in Calgary, 1,983 deaths over about 3 years [10]. 170,078 men and 122,116 women without cardiovascular disease, median follow-up 8 years [11].
- Sources
- [9], [10], [11]
- Caveats
- Observational. Illness that lowers or raises hemoglobin before death cannot be ruled out, although adjustment for kidney function, inflammation and frailty did not remove the association in the US cohort [9].
Stroke, clots and death in kidney disease trials that raised hemoglobin
Shown in kidney disease trials: aiming for a higher hemoglobin with erythropoietin-type drugs raised stroke, blood pressure and access clotting, and did not lower deaths.
The numbers and sources
- Effect
- In a meta-analysis of 9 trials, a higher hemoglobin target carried a higher risk of death from any cause (risk ratio 1.17, 95% CI 1.01 to 1.35) and of clotting in the dialysis access (1.34, 1.16 to 1.54) than a lower target [12]. A later meta-analysis of 27 trials found that a higher target raised stroke (relative risk 1.51, 95% CI 1.03 to 2.21), hypertension (1.67, 1.31 to 2.12) and access clotting (1.33, 1.16 to 1.53). Differences in death (1.09, 0.99 to 1.20) and serious cardiovascular events (1.15, 0.98 to 1.33) did not reach significance [13].
- Population
- 5,143 patients with anemia from chronic kidney disease in 9 randomized trials [12]. 10,452 patients in 27 trials [13].
- Sources
- [12], [13]
- Caveats
- These trials raised hemoglobin with erythropoiesis-stimulating drugs in people with kidney disease, so they do not show what a higher hemoglobin reached another way does in other people. The second analysis could not separate the effect of the drug dose from the effect of the hemoglobin reached [13].
Dementia
Limited evidence: one Dutch cohort linked both low and high hemoglobin to more dementia over 12 years.
The numbers and sources
- Effect
- The association was U-shaped. Against the middle fifth of hemoglobin, the hazard ratio for dementia was 1.29 (95% CI 1.09 to 1.52) in the lowest fifth and 1.20 (1.00 to 1.44) in the highest. Anemia, present in 6.1%, went with a 34% higher risk of dementia (95% CI 11% to 62%) [14].
- Population
- 12,305 people without dementia in the Rotterdam Study, mean age 64.6, 1,520 cases over a mean 12.1 years [14].
- Sources
- [14]
- Caveats
- One observational cohort in one city. It cannot show that hemoglobin itself changes dementia risk [14].
Birth outcomes by hemoglobin in pregnancy
Linked to more low birth weight, preterm birth and stillbirth at low maternal hemoglobin, and to more stillbirth at high levels, in pooled studies.
The numbers and sources
- Effect
- In a meta-analysis of 95 studies, low maternal hemoglobin had pooled odds ratios of 1.42 (95% CI 1.31 to 1.55) for low birth weight, 1.36 (1.26 to 1.46) for preterm birth and 1.49 (1.15 to 1.92) for stillbirth. Odds rose as the cutoff fell: for low birth weight, 1.42 at 11.0 g/dL or below and 2.97 (1.85 to 4.75) at 7.0 g/dL or below. High maternal hemoglobin had odds ratios of 1.88 (1.21 to 2.91) for stillbirth and 1.22 (1.08 to 1.37) for a baby small for gestational age, and was not significantly associated with low birth weight or preterm birth [15].
- Population
- 95 studies in the meta-analysis, within a systematic review of 272 [15].
- Sources
- [15]
- Caveats
- Observational. Few studies separated anemia from iron deficiency from other anemia, and results varied with the trimester of measurement. The authors judge the data insufficient to say whether cutoffs need revising [15].
Related
Sources
- US Food and Drug Administration. 21 CFR 864.5620, Automated hemoglobin system. Electronic Code of Federal Regulations. https://www.ecfr.gov/current/title-21/section-864.5620
- US Food and Drug Administration. 21 CFR 864.5200, Automated cell counter. Electronic Code of Federal Regulations. https://www.ecfr.gov/current/title-21/section-864.5200
- World Health Organization. Guideline on haemoglobin cutoffs to define anaemia in individuals and populations. Geneva: World Health Organization, 2024. ISBN 978-92-4-008854-2. https://www.who.int/publications/i/item/9789240088542
- Centers for Disease Control and Prevention. Recommendations to Prevent and Control Iron Deficiency in the United States. MMWR Recommendations and Reports. 1998;47(RR-3). https://www.cdc.gov/mmwr/preview/mmwrhtml/00051880.htm
- Centers for Disease Control and Prevention, National Center for Health Statistics. National Health and Nutrition Examination Survey 2017-2018 Laboratory Procedure Manual: Complete Blood Count, whole blood, Beckman Coulter DxH 800 analyzer. Performed by the NHANES Mobile Examination Center. https://wwwn.cdc.gov/nchs/data/nhanes/public/2017/labmethods/CBC-J-MET-508.pdf
- Nordin G, Mårtensson A, Swolin B, et al. A multicentre study of reference intervals for haemoglobin, basic blood cell counts and erythrocyte indices in the adult population of the Nordic countries. Scandinavian Journal of Clinical and Laboratory Investigation. 2004. https://doi.org/10.1080/00365510410002797
- Beutler E, Waalen J. The definition of anemia: what is the lower limit of normal of the blood hemoglobin concentration? Blood. 2006. https://doi.org/10.1182/blood-2005-07-3046
- Barbui T, Thiele J, Gisslinger H, et al. The 2016 WHO classification and diagnostic criteria for myeloproliferative neoplasms: document summary and in-depth discussion. Blood Cancer Journal. 2018. https://doi.org/10.1038/s41408-018-0054-y
- Zakai NA, Katz R, Hirsch C, et al. A Prospective Study of Anemia Status, Hemoglobin Concentration, and Mortality in an Elderly Cohort: The Cardiovascular Health Study. Archives of Internal Medicine. 2005. https://doi.org/10.1001/archinte.165.19.2214
- Culleton BF, Manns BJ, Zhang J, et al. Impact of anemia on hospitalization and mortality in older adults. Blood. 2006. https://doi.org/10.1182/blood-2005-10-4308
- Lee G, Choi S, Kim K, et al. Association of Hemoglobin Concentration and Its Change With Cardiovascular and All-Cause Mortality. Journal of the American Heart Association. 2018. https://doi.org/10.1161/JAHA.117.007723
- Phrommintikul A, Haas SJ, Elsik M, Krum H. Mortality and target haemoglobin concentrations in anaemic patients with chronic kidney disease treated with erythropoietin: a meta-analysis. The Lancet. 2007. https://doi.org/10.1016/S0140-6736(07)60194-9
- Palmer SC, Navaneethan SD, Craig JC, et al. Meta-analysis: Erythropoiesis-Stimulating Agents in Patients With Chronic Kidney Disease. Annals of Internal Medicine. 2010. https://doi.org/10.7326/0003-4819-153-1-201007060-00252
- Wolters FJ, Zonneveld HI, Licher S, et al. Hemoglobin and anemia in relation to dementia risk and accompanying changes on brain MRI. Neurology. 2019. https://doi.org/10.1212/WNL.0000000000008003
- Young MF, Oaks BM, Tandon S, et al. Maternal hemoglobin concentrations across pregnancy and maternal and child health: a systematic review and meta-analysis. Annals of the New York Academy of Sciences. 2019. https://doi.org/10.1111/nyas.14093
- Sharma AJ, Addo OY, Mei Z, Suchdev PS. Reexamination of hemoglobin adjustments to define anemia: altitude and smoking. Annals of the New York Academy of Sciences. 2019. https://doi.org/10.1111/nyas.14167
- Whitehead RD, Mei Z, Mapango C, Jefferds MED. Methods and analyzers for hemoglobin measurement in clinical laboratories and field settings. Annals of the New York Academy of Sciences. 2019. https://doi.org/10.1111/nyas.14124
- De la Cruz-Góngora V, Méndez-Gómez-Humarán I, Gaona-Pineda EB, et al. Drops of Capillary Blood Are Not Appropriate for Hemoglobin Measurement with Point-of-Care Devices: A Comparative Study Using Drop Capillary, Pooled Capillary, and Venous Blood Samples. Nutrients. 2022. https://doi.org/10.3390/nu14245346
- European Federation of Clinical Chemistry and Laboratory Medicine. EFLM Biological Variation Database, meta-analysis estimates, read 2026-11-13. https://biologicalvariation.eu/