When a Baby's Arm Does Not Move After Birth

Four recent studies on brachial plexus birth injury — what scans can and cannot show, which nerve operation works better, why therapy matters so much, and what children themselves say years later

A weak arm at birth, and the long road that follows

Some babies are born unable to move one arm properly, because the nerves running from the neck into the shoulder were stretched during birth. Most recover on their own within a few months, but a substantial minority do not, and those children may need scans, nerve surgery and years of therapy. Four studies published between late 2025 and mid-2026 look at four steps along that road, and together they change what families can reasonably be told at each one.

The nerves in question form a bundle called the brachial plexus, which leaves the spinal cord in the neck and fans out to supply the shoulder, arm and hand. When a shoulder is held up during delivery, that bundle can be stretched. If the nerves are only stretched, they usually recover. If a nerve is torn, or pulled clean off the spinal cord — surgeons call that an avulsion — it will not recover on its own. The whole difficulty of this condition is that these three situations look almost identical in a newborn — which is why the four studies below deal, in turn, with what a scan can show [1], which operation restores movement [2], what makes that operation succeed [3], and how the children are doing years afterwards [4].

How families used to face this, and what changed

For most of the twentieth century, a family's entire prognosis rested on watching and waiting. The pattern of which nerve roots were affected was described in a classification that is still in use [5], and doctors learned that whether a baby could bend the elbow by a certain age was the single most useful early sign, a finding turned into a formal score [6]. A standard 0-to-7 scale for grading how much a baby can move each part of the arm followed, and a grade of 6 — moving the arm through more than half its range against gravity — became the usual definition of a working shoulder [7].

Two discoveries shifted practice from waiting to acting. First, surgeons showed that babies whose elbow function had not returned by six months did better after nerve surgery than the natural course of the condition predicted [8]. Second, it became clear that a shoulder left with unbalanced muscles does not simply stay weak — the joint itself deforms and can slip out of position, which was documented in detail in imaging studies of affected children [9]. Once you know the joint is on a clock, referral to a specialist team stops being optional.

What has not been resolved is how common the condition is or how completely children recover. A survey covering the whole of the United Kingdom and Republic of Ireland found about 1 baby in every 2,300 affected, with a difficult shoulder at delivery in 60% of cases and more than half of the babies larger than average [10]. A Swedish study following every birth in one region found a higher rate, with roughly 1 in 2,000 babies left with lasting effects [11]. In the United States, the rate fell by almost half over sixteen years as caesarean births became more common [12]. And when one group followed children to nearly four years of age, only two-thirds had recovered completely — with the ability to rotate the arm outwards recovering last and least of all [13].

There is also an uncomfortable fact underneath all of this: the known risk factors do not explain who gets injured. A 2026 review suggests that some babies may simply be born with a slightly different anatomy in the neck that makes their nerves more vulnerable to ordinary birth forces [14]. Specialists have also long struggled to agree on what to measure, which makes results from different hospitals hard to compare — the reason an international group was convened to try to settle it [15], and a problem a recent overview of the field says is still unsolved [16].

What a scan can honestly tell you

The first new study, from a specialist brachial plexus centre in Leeds in the United Kingdom, asked a simple question: when a magnetic resonance imaging (MRI) scan says a nerve root has been torn off the spinal cord, is it right? The team studied 18 babies under one year old, born between January 2019 and May 2024, who had a scan and then surgery under the same anaesthetic, so the surgeon's direct view of the nerves could be used as the truth [1].

The answer was: often, but not reliably enough to make decisions on. Counting each baby as a whole, the scan got the picture wrong in nearly a third of cases. It was best at ruling an injury in — when the scan showed a particular tell-tale fluid pouch where a nerve had been pulled away, that finding was correct every single time. But it was poor at ruling injury out, and it was weakest at exactly the two nerve roots that are injured most often. One encouraging detail: scans done at younger ages agreed with the surgical findings better than scans done later, and the youngest baby scanned was just over nine weeks old. Nothing went wrong for any baby during scanning or surgery.

For families, the practical message is that the scan helps the surgeon plan, but the decision to operate is made on how the baby is moving, not on the pictures. A reassuring scan is not a guarantee.

Two ways to rewire a shoulder

The second study comes from Tampere in Finland, where surgeons reviewed children born between 1999 and 2021 who had reconstruction of the nerve that powers outward rotation of the shoulder [2]. There are two common ways to do this: bridge the gap with a piece of nerve borrowed from the leg, or borrow a working nerve from nearby — the one that supplies a neck muscle — and plug it in directly.

The direct transfer did better. Outward rotation improved by an average of 94 degrees after transfer, against 44 degrees after the graft. The difference families would notice most is in repeat operations: 11 of the 16 children who had the graft needed further surgery for a shoulder that still would not rotate, at around two years after the first operation, compared with 2 of the 15 who had the transfer. There is good news in this study for children already down the graft route — when the transfer was done later as a rescue operation, at an average age of just over two years, every one of those seven children reached a working shoulder. The window does not slam shut. A larger earlier study of 145 infants pointed the same way [17].

One thing to expect: recovery after the transfer took longer to arrive, on average three and a half years rather than under two. Slower is not worse here, but it is much easier to live through if you know it is coming.

The part nobody puts in the operation note

The third study is, for families, probably the most important. Two surgeons at two American hospitals — one in the Bronx in New York, one in Los Angeles — performed the same nerve transfer on 25 babies using the same technique. The only real difference was how therapy was organised afterwards [3].

At the hospital where the surgeon and experienced specialist therapists chose the therapy clinic together, booked the first appointment and kept talking to each other, 8 of 11 children reached a working shoulder. At the hospital where families were left to find their own therapy, 2 of 14 did. Within that second hospital, children who happened to receive therapy on site did considerably better than those who went elsewhere — which suggests the difference is about coordination and access, not about money or equipment.

This was a small study and the exact numbers should be treated cautiously. But the direction is striking: the way therapy was organised appeared to matter more than which nerve the surgeon used. If you are a parent, the question "who exactly will do the therapy, how often, and are they in contact with the surgical team?" deserves to be asked before the operation, not after. Other add-on therapies, such as electrical stimulation of the muscles, do not yet have strong evidence behind them [18].

What the children say

The fourth study asked 207 children aged 8 to 18, at four leading centres in the United States and Canada, how their lives were actually going [4]. Their overall health scores were slightly below those of other children their age — but only slightly, and, strikingly, children with the most extensive injuries scored no worse than those with the mildest.

What did track with how well children said they were doing had almost nothing to do with the physical examination. It was emotional wellbeing, relationships with family and friends, pain, and feeling stigmatised or self-conscious about how the arm looks. How far the arm could rotate barely registered. When the researchers modelled the relationships formally, physical function reached overall wellbeing only indirectly, through how the child felt about their own body [19]. This fits earlier work on mental health in these children [20], on what children themselves say matters to them [21], and on the loss of sensation that often goes unnoticed [22].

What this means for your family, and what comes next

If your baby has a weak arm, most of the important information will come from watching how they move over the coming months, not from a scan [1]. If surgery is recommended, ask about the nerve transfer, ask how long recovery is expected to take [2], and ask in detail about therapy — who, where, how often, and how the therapists and surgeons talk to each other [3]. And take seriously the finding that the things that shaped these children's lives were pain, confidence and friendships [4]. Researchers are now working on sharper scanning methods, on testing coordinated therapy properly, and on rebuilding the way outcomes are measured around what children actually care about. Families navigating the emotional side of a long NICU or specialist pathway may also find support for parent mental health useful.

References

  1. Hardie CM, Karia C, Wade RG, et al. Magnetic resonance imaging for detecting root avulsions in brachial plexus birth injuries. J Hand Surg Eur Vol. 2026;51(5):589-596. doi:10.1177/17531934251406957 ↩
  2. Hyttinen M, Rönkkö H, Paavilainen P, Jokihaara J. Comparison of Spinal Accessory Nerve Transfer versus C5 Grafting for Suprascapular Nerve Reinnervation in Brachial Plexus Birth Injury. Plast Reconstr Surg. 2026;158(1):76e-84e. doi:10.1097/PRS.0000000000012602 ↩
  3. Friedman AJ, Robbins V, Gotlieb-Horowitz M, et al. Multidisciplinary rehabilitation network enhances outcomes after nerve transfer in brachial plexus birth injury. Childs Nerv Syst. 2026;42(1):121. doi:10.1007/s00381-026-07216-w ↩
  4. Bauer AS, Manske MC, Dorich JM, et al. Children's Health-Related Quality of Life After Brachial Plexus Birth Injury. JAMA Netw Open. 2026;9(5):e2610237. doi:10.1001/jamanetworkopen.2026.10237 ↩
  5. Gilbert A, Whitaker I. Obstetrical brachial plexus lesions. J Hand Surg Br. 1991;16(5):489-491. doi:10.1016/0266-7681(91)90100-3 ↩
  6. Michelow BJ, Clarke HM, Curtis CG, et al. The natural history of obstetrical brachial plexus palsy. Plast Reconstr Surg. 1994;93(4):675-680. doi:10.1097/00006534-199404000-00001 ↩
  7. Curtis C, Stephens D, Clarke HM, Andrews D. The Active Movement Scale: an evaluative tool for infants with obstetrical brachial plexus palsy. J Hand Surg Am. 2002;27(3):470-478. doi:10.1053/jhsu.2002.32965 ↩
  8. Waters PM. Comparison of the natural history, the outcome of microsurgical repair, and the outcome of operative reconstruction in brachial plexus birth palsy. J Bone Joint Surg Am. 1999;81(5):649-659. doi:10.2106/00004623-199905000-00006 ↩
  9. Waters PM, Smith GR, Jaramillo D. Glenohumeral deformity secondary to brachial plexus birth palsy. J Bone Joint Surg Am. 1998;80(5):668-677. doi:10.2106/00004623-199805000-00007 ↩
  10. Evans-Jones G, Kay SPJ, Weindling AM, et al. Congenital brachial palsy: incidence, causes, and outcome in the United Kingdom and Republic of Ireland. Arch Dis Child Fetal Neonatal Ed. 2003;88(3):F185-F189. doi:10.1136/fn.88.3.F185 ↩
  11. Lagerkvist AL, Johansson U, Johansson A, Bager B, Uvebrant P. Obstetric brachial plexus palsy: a prospective, population-based study of incidence, recovery, and residual impairment at 18 months of age. Dev Med Child Neurol. 2010;52(6):529-534. doi:10.1111/j.1469-8749.2009.03479.x ↩
  12. DeFrancesco CJ, Shah DK, Rogers BH, Shah AS. The Epidemiology of Brachial Plexus Birth Palsy in the United States: Declining Incidence and Evolving Risk Factors. J Pediatr Orthop. 2019;39(2):e134-e140. doi:10.1097/BPO.0000000000001089 ↩
  13. Hoeksma AF, ter Steeg AM, Nelissen RG, et al. Neurological recovery in obstetric brachial plexus injuries: an historical cohort study. Dev Med Child Neurol. 2004;46(2):76-83. doi:10.1017/s0012162204000179 ↩
  14. Beaumont HO, Wiberg A, Brown H, Quick TJ. Anatomical predisposition and neurological vulnerability to brachial plexus birth injury: a contemporary narrative review. J Hand Surg Eur Vol. 2026;51(7):961-966. doi:10.1177/17531934261443088 ↩
  15. Pondaag W, Malessy MJA. Outcome assessment for brachial plexus birth injury. Results from the iPluto world-wide consensus survey. J Orthop Res. 2018;36(9):2533-2541. doi:10.1002/jor.23901 ↩
  16. Shackleford T, Little KJ. Brachial plexus birth injury in children and adolescents: A current concepts review. J Child Orthop. 2026;20(3):207-215. doi:10.1177/18632521261420152 ↩
  17. Manske MC, Kalish LA, Cornwall R, Peljovich AE, Bauer AS. Reconstruction of the Suprascapular Nerve in Brachial Plexus Birth Injury: A Comparison of Nerve Grafting and Nerve Transfers. J Bone Joint Surg Am. 2020;102(4):298-308. doi:10.2106/JBJS.19.00627 ↩
  18. Celbek B, Hoşbay Z, Urhun Keleş E, et al. Neuromuscular Electrical Stimulation in Brachial Plexus Birth Injury Rehabilitation: A Systematic Review. Medicina (Kaunas). 2026;62(6):1143. doi:10.3390/medicina62061143 ↩
  19. Kostanjsek N. Use of The International Classification of Functioning, Disability and Health (ICF) as a conceptual framework and common language for disability statistics and health information systems. BMC Public Health. 2011;11(Suppl 4):S3. doi:10.1186/1471-2458-11-S4-S3 ↩
  20. Ho ES, Ferdosi DV, Pourtousi A, Davidge KM, Anthony SJ. Mental health risk and protective factors associated with brachial plexus birth injuries: a scoping review. Disabil Rehabil. 2024;46(1):21-30. doi:10.1080/09638288.2022.2156628 ↩
  21. Sarac C, Bastiaansen E, Van der Holst M, et al. Concepts of functioning and health important to children with an obstetric brachial plexus injury: a qualitative study using focus groups. Dev Med Child Neurol. 2013;55(12):1136-1142. doi:10.1111/dmcn.12270 ↩
  22. Corkum JP, Kuta V, Tang DT, Bezuhly M. Sensory outcomes following brachial plexus birth palsy: A systematic review. J Plast Reconstr Aesthet Surg. 2017;70(8):987-995. doi:10.1016/j.bjps.2017.05.007 ↩