Objectives: Osteoblast is known to regulate hematopoiesis according to preclinical studies but the causal rela tionship in human remains uncertain. We aimed to evaluate causal relationships of bone mineral density (BMD) with blood cell traits using genetic data. Methods: Summary statistics from the largest available genome-wide association study were retrieved for total body BMD (TBBMD), lumbar spine BMD (LSBMD), femoral neck BMD (FNBMD) and 29 blood cell traits including red blood cell, white blood cell and platelet-related traits. Using two-sample Mendelian randomization (MR) approach, inverse-variance weighted method was adopted as main univariable MR analysis. Multivariable MR (MVMR) analysis was conducted to evaluate whether the casual effect is independent of confounders. Results: BMD was positively associated with reticulocyte-related traits, including high light scatter reticulocyte count and percentage, immature reticulocyte fraction, reticulocyte count and percentage, with causal effect estimate (beta) ranging from 0.023 to 0.064. Conversely, inverse association of BMD with hematocrit, hemo globin, and red blood cell count was observed, with beta ranging from − 0.038 to − 0.019. The association remained significant in MVMR analysis after adjustment for confounders. For white blood cells, BMD was inversely associated with neutrophil count (beta: 0.029 to − 0.019) and white blood cell count (beta: 0.024 to − 0.02). Results across TBBMD, LSBMD, and FNBMD were consistent. Conclusions This study suggested bone metabolism had a causal effect on hematopoietic system in humans. Its causal effect on red blood cell traits was independent of confounders. Further studies on how improving bone health can reduce risk of hematological disorders are warranted.

Objectives: Osteoblast is known to regulate hematopoiesis according to preclinical studies but the causal rela tionship in human remains uncertain. We aimed to evaluate causal relationships of bone mineral density (BMD) with blood cell traits using genetic data. Methods: Summary statistics from the largest available genome-wide association study were retrieved for total body BMD (TBBMD), lumbar spine BMD (LSBMD), femoral neck BMD (FNBMD) and 29 blood cell traits including red blood cell, white blood cell and platelet-related traits. Using two-sample Mendelian randomization (MR) approach, inverse-variance weighted method was adopted as main univariable MR analysis. Multivariable MR (MVMR) analysis was conducted to evaluate whether the casual effect is independent of confounders. Results: BMD was positively associated with reticulocyte-related traits, including high light scatter reticulocyte count and percentage, immature reticulocyte fraction, reticulocyte count and percentage, with causal effect estimate (beta) ranging from 0.023 to 0.064. Conversely, inverse association of BMD with hematocrit, hemo globin, and red blood cell count was observed, with beta ranging from − 0.038 to − 0.019. The association remained significant in MVMR analysis after adjustment for confounders. For white blood cells, BMD was inversely associated with neutrophil count (beta: 0.029 to − 0.019) and white blood cell count (beta: 0.024 to − 0.02). Results across TBBMD, LSBMD, and FNBMD were consistent. Conclusions This study suggested bone metabolism had a causal effect on hematopoietic system in humans. Its causal effect on red blood cell traits was independent of confounders. Further studies on how improving bone health can reduce risk of hematological disorders are warranted.

Objectives: Osteoblast is known to regulate hematopoiesis according to preclinical studies but the causal rela tionship in human remains uncertain. We aimed to evaluate causal relationships of bone mineral density (BMD) with blood cell traits using genetic data. Methods: Summary statistics from the largest available genome-wide association study were retrieved for total body BMD (TBBMD), lumbar spine BMD (LSBMD), femoral neck BMD (FNBMD) and 29 blood cell traits including red blood cell, white blood cell and platelet-related traits. Using two-sample Mendelian randomization (MR) approach, inverse-variance weighted method was adopted as main univariable MR analysis. Multivariable MR (MVMR) analysis was conducted to evaluate whether the casual effect is independent of confounders. Results: BMD was positively associated with reticulocyte-related traits, including high light scatter reticulocyte count and percentage, immature reticulocyte fraction, reticulocyte count and percentage, with causal effect estimate (beta) ranging from 0.023 to 0.064. Conversely, inverse association of BMD with hematocrit, hemo globin, and red blood cell count was observed, with beta ranging from − 0.038 to − 0.019. The association remained significant in MVMR analysis after adjustment for confounders. For white blood cells, BMD was inversely associated with neutrophil count (beta: 0.029 to − 0.019) and white blood cell count (beta: 0.024 to − 0.02). Results across TBBMD, LSBMD, and FNBMD were consistent. Conclusions This study suggested bone metabolism had a causal effect on hematopoietic system in humans. Its causal effect on red blood cell traits was independent of confounders. Further studies on how improving bone health can reduce risk of hematological disorders are warranted.

Objectives: Osteoblast is known to regulate hematopoiesis according to preclinical studies but the causal rela tionship in human remains uncertain. We aimed to evaluate causal relationships of bone mineral density (BMD) with blood cell traits using genetic data. Methods: Summary statistics from the largest available genome-wide association study were retrieved for total body BMD (TBBMD), lumbar spine BMD (LSBMD), femoral neck BMD (FNBMD) and 29 blood cell traits including red blood cell, white blood cell and platelet-related traits. Using two-sample Mendelian randomization (MR) approach, inverse-variance weighted method was adopted as main univariable MR analysis. Multivariable MR (MVMR) analysis was conducted to evaluate whether the casual effect is independent of confounders. Results: BMD was positively associated with reticulocyte-related traits, including high light scatter reticulocyte count and percentage, immature reticulocyte fraction, reticulocyte count and percentage, with causal effect estimate (beta) ranging from 0.023 to 0.064. Conversely, inverse association of BMD with hematocrit, hemo globin, and red blood cell count was observed, with beta ranging from − 0.038 to − 0.019. The association remained significant in MVMR analysis after adjustment for confounders. For white blood cells, BMD was inversely associated with neutrophil count (beta: 0.029 to − 0.019) and white blood cell count (beta: 0.024 to − 0.02). Results across TBBMD, LSBMD, and FNBMD were consistent. Conclusions This study suggested bone metabolism had a causal effect on hematopoietic system in humans. Its causal effect on red blood cell traits was independent of confounders. Further studies on how improving bone health can reduce risk of hematological disorders are warranted.

Objectives: Osteoblast is known to regulate hematopoiesis according to preclinical studies but the causal rela tionship in human remains uncertain. We aimed to evaluate causal relationships of bone mineral density (BMD) with blood cell traits using genetic data. Methods: Summary statistics from the largest available genome-wide association study were retrieved for total body BMD (TBBMD), lumbar spine BMD (LSBMD), femoral neck BMD (FNBMD) and 29 blood cell traits including red blood cell, white blood cell and platelet-related traits. Using two-sample Mendelian randomization (MR) approach, inverse-variance weighted method was adopted as main univariable MR analysis. Multivariable MR (MVMR) analysis was conducted to evaluate whether the casual effect is independent of confounders. Results: BMD was positively associated with reticulocyte-related traits, including high light scatter reticulocyte count and percentage, immature reticulocyte fraction, reticulocyte count and percentage, with causal effect estimate (beta) ranging from 0.023 to 0.064. Conversely, inverse association of BMD with hematocrit, hemo globin, and red blood cell count was observed, with beta ranging from − 0.038 to − 0.019. The association remained significant in MVMR analysis after adjustment for confounders. For white blood cells, BMD was inversely associated with neutrophil count (beta: 0.029 to − 0.019) and white blood cell count (beta: 0.024 to − 0.02). Results across TBBMD, LSBMD, and FNBMD were consistent. Conclusions This study suggested bone metabolism had a causal effect on hematopoietic system in humans. Its causal effect on red blood cell traits was independent of confounders. Further studies on how improving bone health can reduce risk of hematological disorders are warranted.