Abstract
Not all neonatal brachial plexus injuries should be deemed the fault of the accoucheur. However, there is a small (<10%) subset of neonatal brachial plexus injuries that are related to excessive traction by the accoucheur: permanent injuries to the anterior arm after SD. The position regarding posterior injuries remains predominantly the same; if the injury is to the posterior shoulder, the injury is likely to have been caused by maternal propulsion against the sacral promontory before the foetal head is delivered, rather than excessive and inappropriate traction. However, there is no reliable evidence that a combination of maternal propulsion and diagnostic traction alone causes significant and permanent injury to the anterior shoulder after shoulder dystocia. This was recognised in Deith vs. Lanarkshire where the judge found: that where there is a severe injury to an anterior arm after SD, excessive traction is overwhelmingly likely to be the cause.
Whilst we are not yet in the territory of res ipsa loquitur, the recent literature demonstrates that a small subset of brachial plexus injuries (BPIs), permanent injuries to an anterior arm after shoulder dystocia (SD), are almost completely preventable with reasonable care.
Claims for damages in respect of neonatal brachial plexus injuries (NBPIs) can raise significant evidential difficulties for those acting for both claimants and defendants, who must grapple with whether excessive or incorrect traction was applied, and if so whether it caused the injury. New data from both sides of the Atlantic should inform causation for a small subset of NBPIs.
At the heart of litigation cases is the extent to which it can be demonstrated that the likely cause of injury was excessive or incorrect traction, rather than maternal propulsion forces. Whilst the witness evidence of those present at the delivery is the starting point, the delivery often occurred many years previously, may be supported by thin or poor notes, and is the subject of some controversy between the parties. In the circumstances, it is important to look at any objective evidence that supports the conclusion that the injury relates to excessive or incorrect traction.
Medical theories and expert opinions about the causal relationships between birth, management of SD and NBPIs have ebbed and flowed over the last four decades along a spectrum of always excessive traction by the accoucheur, to almost never.
In 2005, the RCOG guidelines stated (emphasis added): Not all injuries are due to excess traction by the accoucheur and there is now a significant body of evidence that maternal propulsive forces may contribute to some of these injuries. Moreover, a substantial minority of brachial plexus injuries are not associated with clinically evident shoulder dystocia. In one series, 4% of injuries occurred after a caesarean section. Specifically, where there is Erb’s palsy, it is important to determine whether the affected shoulder was anterior or posterior at the time of delivery, because damage to the plexus of the posterior shoulder is considered not due to action by the accoucheur. Causation of obstetric brachial plexus injury is multifactorial; evidence suggests that while some cases are traction mediated, others may not be. There is growing acceptance in both the medical literature and case law that the propulsive forces of uterine contraction may play a part. The assumption that the presence of an injury is evidence that traction must have been applied is no longer valid. Injury may occur regardless of best efforts of the accoucheur. Diagnostic traction is acceptable and claimants now need to demonstrate factual evidence of the use of excessive force or other inappropriate management to succeed in arguing negligent management.
However, the Courts did find some assistance from a template which provided indicators for determining between a propulsion and iatrogenic injury, published in 2008. 1
In Croft, Mr Justice Hickinbottom described the template as follows: This checklist has been referred to in other cases, including Bennion and the recent judgment of His Honour Judge Oliver-Jones QC sitting as a Deputy High Court Judge in Birmingham District Registry in Blakeborough v Walsall Hospitals NHS Trust (17 February 2012, Unreported) to which I was also referred. It is no doubt a helpful checklist on factors that might be taken into account in any assessment of causation. Although on the evidence before me, some presence of some criteria can result in a claim failing (e.g. if the injured arm if posterior), the paper makes clear that that assessment requires judgment and consideration of whether the relevant factors apply and, if so, the weight to be given to each. The paper does not suggest that the presence of any one factor (or combination of factors) necessarily leads to a conclusion that there is an iatrogenic cause of the injury.
BPI at birth is associated with a heterogeneous constellation of factors, which makes blanket statements about causation for all NBPI unhelpful and potentially misleading. This is similar to cerebral palsy (CP), which also has a heterogeneous set of causes. Historically, it was thought that CP had a single cause: an acute intrapartum hypoxic event. However, we now understand that only 10% of CP is related to intrapartum management and there are a variety of other causal mechanisms for the remaining 90%. 5 This is likely to be similar for NBPI.
It was previously accepted that all NBPIs were deemed the fault of the accoucheur, who must have applied excessive traction during difficult delivery of the shoulders. However, over the past two decades, there have been increasing reports of OBPI in the absence of reported/coded SD and after delivery by caesarean section. Therefore, as with CP, there are likely to be different causal mechanisms for NBPI, e.g. anterior and posterior NBPI may have different causes; similarly, for NBPI with and without SD, as well as injuries for infants after a breech presentation and injuries after delivery by caesarean section. There also may be different causal mechanisms for temporary and permanent injuries.
A recent review of the causation of NBPI, published by the American College of Obstetrics & Gynaecology (ACOG), 6 recognises this complexity, and concluded: there were limited data that confirmed that NBPI can be caused only by a specific amount of applied force beyond that typically used by healthcare providers and experienced during a delivery without NBPI, i.e. NBPI is not always caused by excessive traction by accoucheurs.
However, the authors made no attempt to differentiate between anterior and posterior injuries, different modes of birth, the presence of SD and/or permanence of the injuries. This is unhelpful.
Nevertheless, the task force did recognize that: knowledge about NBPI is continually evolving, and since the task force finished their review, there have been at least two publications that should challenge their previous conclusion.2,3
Most experts agree that not all NBPI is caused by excessive traction, but there is accruing clinical data that for one particular subset of NBPI: (i) permanent injury; to the (ii) anterior arm (iii) after SD, there is likely to be a single cause: excessive traction.
NBPI after caesarean section, breech birth and with congenital hypotonia are unlikely to be preventable and this will not be discussed further.
Shoulder dystocia vs. no shoulder dystocia
It is recognised that a significant minority of BPIs complicate births without recorded SD;7,8 however, once again, the follow-up data are contradictory.
A recent review of 382 infants with BPI in a neurosurgical unit demonstrated that there was no difference between those infants with persistent BPI at one year of age born without or with recorded SD. 9 However, the authors recognise that there was a significant selection bias; in particular, temporary injuries may not have been referred to a neuro-surgical unit and furthermore there may have been recall bias for SD by the mothers. At least one study of maternal recall of events in labour concluded: studies that rely on patient recall of obstetric variables are likely to have high error rates. 10
Other obstetric teams have published data suggesting that non-SD-related permanent BPI is both uncommon and the majority of these BPIs are temporary. Infants with a BPI >12 months were 75× more likely to occur after SD (21.6%) than after a non-SD-related delivery (0.3%) in at least one US paper 11 and the authors concluded that more than 90% of permanent BPIs follow SD. 11
Furthermore, it is likely that only the injuries related to SD can be prevented as only the management of SD can be improved, rather than the changing the process/mechanics of labour itself.
Permanent vs. temporary injuries
There are a number of reports demonstrating reductions in transient NBPI rates following training,12–14 which suggests that at least a proportion of transient injuries are preventable with best care. However, transient injuries have been reported after best care of SD, 2 whereas permanent injuries have not.
One key issue for causation for permanent injuries is recognised by the ACOG report: Because stretch of a nerve is a recognized mechanism of brachial plexus rupture or avulsion during the birth process, it is important to explore what can cause sufficient stretch of the nerve to lead to this type of injury. From a biomechanical perspective, a necessary condition for significant stretch to the brachial plexus is the widening of the angle between the fetus’ head and neck.
The authors concluded that the persistent injuries were associated with a perceived higher level of downward traction on the head than the transient injuries. In addition, the transient injuries were associated with a perceived higher level of downward traction than used for a control set of uninjured infants. Downward traction of the fetal head was applied more often and with greater force in the group with persistent damage. The risk of persistent obstetric brachial plexus palsy at age 18 months depended on obstetric management and increased significantly with increasing force used in downward traction of the fetal head.
There have been criticisms of these data, particularly the widespread use of fundal pressure: 94% (17/18) of the cases of persistent NBPI were associated with the application of fundal pressure after the head delivered, compared with 8% of deliveries for the infants whose injury spontaneously recovered and 6% of the control deliveries. The ACOG task force therefore concluded that it was not possible to separate out the effect of increased traction from the effect of increased expulsive force induced by the application of fundal pressure. 6
This conclusion requires fundal pressure to be causative, indeed more causative than downward traction. There are a number of problems with this thesis: there is only a single paper relating to 13 births complicated by SD where fundal pressure was employed without any other manoeuvres. 16 Fundal pressure alone will increase the impaction but not widen the angle between the head and shoulder. Moreover, fundal pressure will not result in the birth of the baby, and therefore it is very likely that excessive traction was employed to overcome the SD in these cases as described by the accoucheurs, and this caused the injury. This is supported by an observation made by the ACOG task force 6 that that the rate of NBPI remained reasonably constant between the early 1980s and the 1990s, despite the widespread abandonment of fundal pressure. Fundal pressure is therefore likely not to be causative in isolation.
A different neuro-surgical team concluded, in their review of children with permanent NBPI from the Netherlands, that traction during SD was likely to be causative:
17
In this neurosurgical series of 206 OBPL infants, it is shown that the neurological severity of the OBPL correlated with the birthweight of the child. This disproportion probably reflects the severity of the shoulder dystocia and the probably stronger clinician-applied traction that occurred during birth.
Anterior vs. posterior arm injuries
It is widely recognised that either the anterior or the posterior brachial plexus can be injured and that the causal mechanism of injury is different although both are mediated through the same bio-mechanical widening of the angle between the head and the shoulder, thereby increasing the strain on the brachial plexus.
Both the 2008 clinical template 1 and the recent ACOG report 6 suggest that posterior NBPIs are likely to be caused by impaction of the posterior shoulder on the sacral promontory where the uterine forces continue to push the baby down the birth canal, which stretches the foetal brachial plexus. In particular, the head travelling along the curve of carus could cause the necessary widening of the angle between the shoulder and head to cause the injury. However, this is a function of labour not birth and injury is not caused by excessive traction once the head is delivered.
However, it is accepted that anterior arm injuries after SD are more likely to be related to accoucheurs pulling on the head, once again widening the angle between the shoulder and head. This traction is likely to be causative in different ways related to the force employed, its direction and also its nature. Force in a downward direction appears to be particularly injurious 18 as is force of a ‘jerking’ nature 19 ; both of which are avoidable with reasonable care.
Alternative ‘endogenous’ mechanisms of injury
There are a plethora of data derived from computer modelling of births 20 and also physical models21–23 demonstrating that the myometrium in normal labour can generate the 100 N force that may be the threshold for NBPI. 24 This has led many ‘experts’ to conclude that it would therefore be possible for the normal forces involved during labour to cause an injury. This is likely to be the case for posterior NBPI and transient injuries; however, permanent NBPI is less likely to be related to compression and requires widening of the head–shoulder angle to sufficiently stretch the brachial plexus nerves to permanently damage them.
There is at least one report, with a video record of the birth, of a temporary NBPI without traction (or SD). 25
There is also a single case report of a permanent BPI after SD without excessive and downward traction. 26 However, since its publication, a federal court judge in Massachusetts disputed the findings.
Conclusion
Not all NBPI should be deemed the fault of the accoucheur. However, there is a small (<10%) subset of NBPIs that are related to excessive traction by the accoucheur: permanent injuries to the anterior arm after SD.
The publication of clinical templates for CP has helped provide a framework to understand causation for affected individuals more accurately,5,27 and there was a similar template published for NBPI in 2008. 1 However, since the ACOG document, new data have changed our understanding of causation for permanent, anterior arm NBPI after SD and the application of the template should also be reconsidered and updated.
The position regarding posterior injuries remains predominantly the same; if the injury is to the posterior shoulder, the injury may have been caused by maternal propulsion against the sacral promontory before the foetal head is delivered, rather than excessive and inappropriate traction.
However, there is no reliable evidence that a combination of maternal propulsion and diagnostic traction alone causes significant and permanent injury to the anterior shoulder after SD. This was recognised in Deith vs. Lanarkshire where the judge found: that where there is a severe injury to an anterior arm after SD, excessive traction is overwhelmingly likely to be the cause.d
Whilst we are not yet in the territory of res ipsa loquitur, the recent literature demonstrates that permanent injuries to an anterior arm after SD are almost completely preventable with reasonable care.
Footnotes
Declaration of conflicting interests
The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Tim Draycott is a senior advisor for maternity for NHS resolution. He has also acted as a consultant for Limbs & Things and Laerdal.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.