An intensive rehabilitation program that focuses on improving upper extremity (UE) skills produced measurable benefits and was well received and safe in infants with perinatal arterial ischemic stroke (PAIS), early results of a phase 3 study suggested.

Babies who received the most intensive form of therapy — constraint-induced movement therapy (CIMT) — and were most adherent achieved the best overall outcomes.

CIMT has previously been shown to be effective in older children with cerebral palsy, but this study represents the first adequately powered trial of high-dose CIMT in infants and toddlers with perinatal stroke.

“For the first time, we now have a protocol, a treatment, which produces benefits [for children with PAIS],” study investigator Sharon Ramey, PhD, professor, Department of Psychiatry and Behavioral Medicine, Virginia Tech Carilion School of Medicine, and professor, Fralin Biomedical Research Institute, Roanoke, Virginia, told Medscape Medical News.

The findings were presented on February 6 at the International Stroke Conference (ISC) 2026 meeting.

Few Treatment Options

PAIS is among the more common forms of ischemic stroke in children, study co-investigator Warren D. Lo, MD, of Nationwide Children’s Hospital and The Ohio State University, Columbus, Ohio, told Medscape Medical News.

He noted that across the perinatal period, stroke from all causes occurs in approximately 1 in 1000 live births, with the majority of cases attributable to arterial ischemic stroke. Infants with PAIS are at risk for lifelong neuromotor impairment and epilepsy and frequently experience cognitive dysfunction.

The study was conducted to address the lack of evidence-based rehabilitation options for infants and toddlers with PAIS. While CIMT has demonstrated benefit in older children with unilateral cerebral palsy, it has not been evaluated in an adequately powered trial in very young children with perinatal stroke.

Investigators tested CIMT during early development, a period of heightened neuroplasticity when intervention may have the greatest impact on long-term motor outcomes.

The study included 167 infants with PAIS and hemiparesis, with roughly equal numbers of boys and girls across 15 US sites. The trial, known as the I-ACQUIRE study, included children aged 8-36 months, a period of rapid brain development with high neuroplastic potential, Ramey said.

She noted that children typically begin to develop voluntary control of their arms and hands at around 6-8 months of age. Unlike adults, infants with stroke do not lose previously acquired skills but are learning these motor functions for the first time.

At baseline, functional ability varied widely among the children. Participants were randomly assigned to one of three groups: usual care, typically consisting of 1 hour each of occupational and physical therapy per week; moderate-dose therapy (3 h/d, 5 d/wk for 4 weeks); or high-dose therapy (6 h/d, 5 d/wk for 4 weeks).

A Promising Approach

The intervention involved task-oriented therapy delivered by specially trained occupational or physical therapists, typically in the home, to improve motor function, coordination, and functional independence.

During most sessions, children wore a lightweight cast on the unimpaired arm and hand. Treatment goals were set with parents, who participated in one weekly session and completed a daily home practice program of about 45 minutes.

The primary outcome was the Emerging Behaviors Scale, which includes 30 UE items. A favorable outcome was a gain of at least 7 points at the end of treatment and at 6 months.

Parents rated how well and how often their child used their affected limb in performing various tasks including such things as manipulating toys and using gestures to communicate.

The proportion of children achieving a prespecified favorable outcome, defined as a gain of at least 7 points at both posttreatment assessments, was lower than anticipated. Ramey noted that while outcomes favored therapy, the magnitude of improvement was smaller than expected.

However, there were some positive signals. At the end-of-treatment assessment, both intervention groups outperformed the usual care group: The moderate-dose group gained a mean of 2.94 skills, and the high-dose group gained a mean of 3.30 skills compared with an average gain of one skill in the usual care group.

At 6 months, children in the high-dose group showed greater skill gains than those in the other two groups. Notably, the usual care group had caught up with the moderate-dose group by this timepoint.

Ramey said she was surprised and encouraged that children receiving approximately 2 hours of therapy per week still demonstrated benefit at 6 months. Children in the high-dose group who were compliant with the therapy benefited the most from the intervention.

Parents were generally very satisfied with their child’s progress, Ramey said, and described the therapy as highly impactful. Feedback included terms such as amazing, life-changing, very beneficial, and fantastic, with one parent noting that the experience was difficult to put into words.

The investigators plan to follow the children through 12 months and present additional findings at future meetings, Ramey said.

The researchers also want to identify clinical characteristics and biomarkers that may predict which children derive the greatest benefit from the intervention, as well as to examine which components of usual care were associated with the highest — and lowest — gains.

Parents and investigators alike want to understand which children benefit most from the intervention and why, and whether the therapy can be adapted to improve outcomes for those who derive less benefit, Ramey said.

Addressing a Critical Gap

Asked to comment on the research, ISC Chair Lauren Sansing, MD, professor of neurology at Yale School of Medicine in New Haven, Connecticut, said the study addresses a critical evidence gap in pediatric stroke, an area where high-quality data remain limited.

Sansing noted that conducting intensive rehabilitation studies in pediatric stroke is particularly challenging because participation requires commitment from the entire family, not just the child. She praised the investigators for successfully completing a complex and demanding trial.

Sansing noted that the improvement seen in the usual care group over the course of the study itself is an important finding. She said the results underscore the need to better understand which therapies and interventions contributed to those gains, with the goal of optimizing care for infants with stroke.

These findings are especially important in light of how common infant strokes are, added ISC Vice Chair Bijoy Menon, MD, neurologist in Calgary, Alberta, Canada.

Menon noted that demonstrating an effective intervention in this population is particularly meaningful. He added that although CIMT is not yet widely available, advances in the field may eventually allow it to be delivered virtually or with the support of robotic technologies.

“Once you have proof of principle that something like this potentially works, it opens up funding opportunities and new avenues of how this might actually be delivered.”

The study received funding from the National Institute of Neurological Disorders and Stroke, an institute in the National Institutes of Health. Ramey, Sansing, and Menon reported having no relevant disclosures.