Randomized trials have shown the benefits of injecting bone marrow-derived mesenchymal stem cells (BmMSCs) after standard hip decompression in patients with osteonecrosis of the femoral head. However, the combination of BmMSCs and platelet-rich plasma (PRP) injected into the femoral head after decompression has not been reported previously. This study reports the results in a preliminary series of patients with osteonecrosis of the femoral head treated with BmMSCs plus PRP. (1) What is the survivorship free from reoperation, hip arthroplasty, and femoral head collapse in a preliminary series of patients with osteonecrosis of the femoral head treated with BmMSCs plus PRP? (2) Is there a change in the degree of femoral head involvement based on modified Kerboul angle? (3) What were the scores observed for pain and function at last followup? (4) Was there a difference in survivorship free from reoperation as a function of in vitro MSC count and viability? Twenty-two consecutive patients (35 hips; 11 men and 11 women) with corticosteroid-induced osteonecrosis who met study inclusion criteria were enrolled; none declined participation, and none was lost to followup, although one patient (two hips) died within a year of the procedure for reasons unrelated to it, and five patients (seven hips) did not undergo MRI at the 1-year followup. All patients had precollapse osteonecrosis, rated either University of Pennsylvania Stage 1 (n = 4) or Stage 2 (n = 31 hips). Mean age and body mass index were 43 years and 31 kg/m, respectively. Patients underwent pre- and postoperative radiographs and MRI to assess femoral head involvement using the modified Kerboul angle. Absolute cell count and colony-forming unit (CFU) assays were used to assess MSC abundance and viability of the bone marrow obtained at the time of surgery. Patients were followed at regular intervals to assess clinical response to treatment with a mean followup of 3 years (range, 2-4 years). The change in femoral head involvement was assessed with the modified Kerboul angle; the Harris hip score was used to assess clinical outcome; and conversion to THA, reoperation, and survivorship free from femoral head collapse were analyzed with the Kaplan-Meier method on a per-hip basis. Survivorship free from THA, any procedure, and femoral head collapse was 84% (95% confidence interval [CI], 75%-93%), 67% (95% CI, 55%-79%), and 93% (95% CI, 76%-98%), respectively, at 3 years postoperatively; two patients (four hips) underwent a second decompression and MSC injection for persistent pain without signs of radiographic collapse. All patients with collapse underwent THA. The mean modified Kerboul angle improved from 205° ± 47° to 172° ± 48° postoperatively (mean change -30° ± 6°, p = 0.01). A greater proportion of patients who underwent an additional procedure had a modified Kerboul grade of 3 or 4 preoperatively (80% [four of five] versus 13% [four of 30 Grade 1 or 2; odds ratio, 26; 95% CI, 2-296; p = 0.005). Preoperatively the mean Harris hip score was 57 ± 12, which improved to 85 ± 15 (mean change 28 ± 3, p < 0.001) at most recent followup. Patients undergoing a reoperation or THA had a lower mean concentration of nucleated cells/mL (5.5 x 10 ± 2.8 x 10 cells/mL versus 2.3 x 10 ± 2.2 x 10 cells/mL, p = 0.02) and lower mean CFUs (13 ± 6 versus 19 ± 7, p = 0.04) compared with those who did not. Core hip decompression with injection of concentrated bone marrow plus PRP improved pain and function; > 90% of hips in this series were without collapse at a minimum of 2 years. In this preliminary study, successful results were seen when nucleated cell count was high and modified Kerboul grade was low. Further randomized studies are needed to determine this procedure's efficacy versus core decompression or nonoperative treatment alone. Level II, therapeutic study.