Neurogenic heterotopic ossification (NHO) after traumatic brain injury may involve osteogenic synthesis of a type I collagen-rich matrix. Procollagen type I N-terminal propeptide (P1NP) reflects type I collagen formation. This single-center retrospective cohort study examined whether baseline P1NP was associated with radiographic NHO within 12 weeks after traumatic brain injury. We reviewed 124 adults transferred to physical and rehabilitation medicine within 21 days after injury who completed 26-34 days of inpatient rehabilitation and had 12-week follow-up. The primary Firth penalized logistic regression model included P1NP per 50 ng/ml, acute care length of stay per 5 days, overall Modified Ashworth Scale score per 5 points, and neutrophil-to-lymphocyte ratio. Twelve-week NHO occurred in 19 (15.3%) patients. Higher P1NP was associated with NHO in the primary model (odds ratio: 6.11, 95% confidence interval: 1.34-33.97; P = 0.013). Longer acute care length of stay, higher overall Modified Ashworth Scale score, and higher neutrophil-to-lymphocyte ratio were also associated with NHO. In a supplementary model replacing acute care length of stay with the Disability Rating Scale, P1NP did not reach statistical significance (odds ratio: 4.63, 95% confidence interval: 0.85-30.18; P = 0.062). In adults with traumatic brain injury undergoing early inpatient rehabilitation, higher baseline P1NP was associated with 12-week NHO in the primary model. Prospective studies with standardized imaging and serial sampling are needed to confirm its predictive value.
Upper-limb recovery after stroke may be affected by conditions that restrict the use of the nonparetic limb during early rehabilitation. We evaluated whether nonparetic forearm fracture was associated with early manual function outcomes after inpatient stroke rehabilitation. This retrospective 1 : 3 matched cohort study included adults with ischemic or hemorrhagic stroke admitted between January 2018 and November 2025. Patients with imaging-confirmed nonparetic forearm fracture treated with conservative immobilization were matched 1 : 3 to patients without forearm fracture. The primary outcome was the 4-week manual function test score. Manual function test scores at rehabilitation admission, 4 weeks, and 4-week change were summarized descriptively. The primary analysis used matched-set fixed-effects regression with cluster-robust standard errors. The matched cohort included 244 patients, comprising 61 with fracture and 183 without fracture. At 4 weeks, the fracture group had lower scores than the no-fracture group (18.6 ± 5.2 vs. 22.5 ± 6.0; P < 0.001). In matched-set fixed-effects analysis, fracture status was associated with a lower 4-week manual function test score (β = -3.72, 95% confidence interval: -4.75 to -2.69; P < 0.001). Nonparetic forearm fracture was associated with worse early manual function after inpatient stroke rehabilitation.
