Sustained ITP remodels the BM microenvironment. ITP progression was associated with blood vessel structural changes and reversible changes in stromal cell number. Mice were treated with IgG (control) or anti-CD41 (ITP) for 4 weeks. An additional group of ITP mice were allowed to recover for an additional 4 weeks after the last injection of anti-CD41, during which platelet count returned to normal by day 18 (ITP recovered group). All data shown from controls and mice with ITP are representative of 3 independent experiments; the data from ITP recovered mice is representative of 1 independent experiment. (A) Representative confocal images of control, ITP, and ITP recovered diaphysis BM from WT femurs. (B) Relative numbers of LepR⁺ stromal cells were inferred by comparing total LepR staining between groups. Staining was quantified by exporting the LepR channel images as TIFF images and quantifying total staining relative to controls using ImageJ software (n = 4-9 mice; data are mean ± SD, with an average of 10 images analyzed per mouse). Control vs ITP *P = .018, control vs ITP recovered P = .91, ITP vs ITP recovered **P = .0021. P values were calculated by using a Kruskal-Wallis test with Dunn’s multiple comparison test. (C) Vessel analysis showing (i) sum vessel area (control vs ITP ****P < .0001, control vs ITP recovered ****P < .0001, ITP vs ITP recovered ***P = .0001), (ii) average vessel area (control vs ITP ****P < .0001, control vs ITP recovered ****P < .0001, ITP vs ITP recovered P = .066), and (iii) vessel number (control vs ITP P > .99, control vs ITP recovered ***P = .0008, ITP vs ITP recovered ***P = .0007). Vessel information was quantified using StrataQuest analysis software (TissueGnostics). P values were calculated by a Kruskal-Wallis test with Dunn’s multiple comparison test (n = 44-55 images). Violin plots show median values and upper and lower quartiles, with an average of 10 images analyzed per mouse. All data are shown as mean ± SD.