Abstract
Dear Editor,
We read with interest the narrative review by Beaumont et al. (2026), which proposes developmental anatomical predisposition and neurologic vulnerability as an additional explanation for brachial plexus birth injury (BPBI). We welcome attempts to understand why some infants sustain BPBI while many exposed to recognised obstetric risk factors do not. The hypothesis is biologically interesting and worthy of study. However, we are concerned that the conclusions and proposed clinical implications extend beyond the evidence currently available.
The central difficulty is the distinction between plausibility, association and causation. The authors cite several strands of evidence, including cervical ribs, variant brachial plexus anatomy, foetal skeletal development, thoracic outlet syndrome and computational modelling. Taken together, these support the possibility that anatomical variation might modify the local mechanical environment of the neonatal plexus. They do not, however, establish that such variation is a major causal mechanism for BPBI, nor that it should yet influence peripartum decision-making or primary nerve surgery.
The cervical rib evidence illustrates this problem. Becker et al. (2002) reported complete cervical ribs in five of 42 surgically treated infants with obstetric brachial plexus lesions. This is an important observation, but it arose from a highly selected operative cohort and lacked matched neonatal controls. It therefore cannot establish the prevalence of cervical ribs in all BPBI, nor determine whether the rib was causal, contributory or incidental. Desurkar et al. (2011) described two children with congenital lower plexus palsy and non-ossified cervical ribs as the only identified abnormality. These cases are valuable because they remind clinicians that not every neonatal plexus palsy is caused by intrapartum traction. However, rare case reports cannot sustain a broad aetiological model.
The foetal cervical rib literature also requires careful interpretation. Bots et al. (2011) found cervical ribs in approximately 40% of electively aborted foetuses but explicitly questioned whether that sample was informative about the general population. Schut et al. (2026) subsequently found poor agreement between prenatal and postnatal assessment of cervical ribs, and concluded that prenatal and postnatal agreement for cervical ribs and vertebral pattern was insufficient. These findings are difficult to reconcile with any near-term recommendation for prenatal anatomical screening as a reliable BPBI risk stratification tool.
The biomechanical literature is similarly supportive of plausibility, but not of clinical causation. Wright and Grimm (2024) developed a two-dimensional finite element model and showed that small angular variations may influence stress distribution at the nerve roots. The authors themselves describe this as a first step towards a more complete three-dimensional model. Iaconianni et al. (2024) modelled shoulder dystocia manoeuvres and showed that appropriate manoeuvres reduced both the required delivery force and brachial plexus strain. This supports the relevance of obstetric mechanics and suggests that applied force, direction of force and manoeuvre sequence remain central to injury risk.
A further question is how a predominantly intrinsic vulnerability theory accounts for epidemiology. Mollberg et al. (2005), in a Swedish population-based study of more than 1.2 million births, identified shoulder dystocia and birthweight of 4500 g or more as the strongest risk factors for obstetric brachial plexus palsy, with caesarean section associated with reduced risk. DeFrancesco et al. (2025) reported that shoulder dystocia remained the strongest risk factor for BPBI in contemporary US data and that caesarean section was protective across birthweight classes.
Most importantly, Draycott et al. (2008) provide direct clinical evidence that BPBI risk is modifiable through obstetric training. After mandatory practical shoulder dystocia training, the shoulder dystocia rate was unchanged, yet management improved substantially. McRoberts’ position, suprapubic pressure, internal rotational manoeuvres and delivery of the posterior arm were used more frequently, documented excessive traction decreased, neonatal injury after shoulder dystocia fell from 9.3 to 2.3%, and brachial plexus injury at birth fell from 7.4 to 2.3% (Draycott et al., 2008). A congenital anatomical trait present before labour should not change over short historical periods with simulation training, altered manoeuvre use or reduced traction, except through interaction with extrinsic mechanical factors.
We therefore suggest that developmental anatomy should currently be framed as a possible modifier of injury threshold, not as an explanatory model capable of displacing established obstetric mechanisms. This distinction matters. It is reasonable to hypothesize that a narrowed supraclavicular corridor, cervical rib or variant root angle could lower the strain threshold for injury. It is a much greater leap to imply that such anatomy explains BPBI in otherwise normal deliveries, supports prenatal screening or should lead surgeons to replace neuroma resection and grafting with decompression or neurolysis.
We agree that further research is warranted. The necessary studies should be prospective, controlled and lesion-specific, integrating systematic neonatal imaging, operative findings, detailed obstetric data, birthweight, shoulder dystocia documentation, manoeuvre sequence and long-term neurological outcome. Until such evidence exists, the proposed anatomical vulnerability model should be regarded as an interesting research hypothesis rather than a clinical explanation for BPBI.
