If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
The treatment of traumatic brachial plexus injury (BPI) is time-sensitive, and early nerve reconstruction is associated with superior nerve recovery. The objective of this study was to determine the rate of delayed referral to our centers for traumatic BPI, identifiable causes of delayed referral, and factors associated with delayed referral to a brachial plexus surgeon.
Methods
We identified 84 patients with traumatic BPI referred to and evaluated by brachial plexus surgeons at 2 tertiary care referral centers from 2015 to 2019. Delayed referral was defined as more than 3 months from the date of injury to the date of initial evaluation by a brachial plexus surgeon. Causes of delayed referral were identified by review of the medical record. Bivariate analysis and multivariable logistic regression were used to identify factors associated with delayed referral.
Results
Mean age of the 84 patients in the study was 45 years; 69% were male. The most common pattern of BPI was global palsy (39%), followed by upper-trunk palsy (23%) and infraclavicular palsy (15%). Median time from injury to surgical evaluation was 2 months (interquartile range, 2–4 months). Thirty-seven patients had a delayed referral (44%). Multivariable logistic regression analysis showed that the hospital to which the patient was referred, Medicare insurance, and motorcycle accident as the mechanism of injury were associated with a delayed referral.
Conclusions
Nearly half of traumatic BPI patients evaluated at 2 tertiary referral centers in a large metropolitan area in the United States presented in a delayed time frame. Both modifiable and nonmodifiable associations with delayed referral were identified. Patients with Medicare insurance had increased odds of delayed referral.
Clinical relevance
Establishment of multidisciplinary BPI specialty centers, outreach to local and regional hospitals, and development of referral algorithms and pathways may improve timeliness of referrals.
Although the optimal treatment algorithm for traumatic BPI is controversial, there is general consensus that time from injury is an important consideration when approaching surgical reconstruction. Primary nerve reconstruction is an option only until eventual deterioration of the denervated motor end plates.
Early referral to a brachial plexus surgeon may result in a more expeditious workup, a broader array of available reconstruction options, more favorable timing of brachial plexus reconstruction, and thus a better functional outcome. Unfortunately, delays from time of injury to surgical evaluation are seen in BPI patients across the world.
Many BPI patients are initially evaluated by a brachial plexus surgeon too late for optimal brachial plexus reconstruction, and some patients are never referred.
A database study demonstrated that nearly 30% of BPI patients in the United States underwent BPI surgery more than 1 year after injury, well outside the optimal treatment window.
The objectives of this study were to determine the rate of delayed referral for traumatic BPI, causes of delayed referral, and factors associated with delayed referral to a brachial plexus surgeon. In this study, we defined delayed referral as more than 3 months from the date of injury to the date of initial evaluation by a brachial plexus surgeon. The null hypothesis was that no identifiable associated factor would exist for delayed referral.
Materials and Methods
Study design and patient selection
This study was performed with institutional review board approval. A retrospective chart review was conducted of all traumatic BPI at 2 tertiary care referral centers in a single metropolitan area from January 1, 2015 to December 31, 2019. The hospital billing records database was queried using International Classification of Diseases, 10th Revision code S14.3 (injury of brachial plexus) for patients treated within the study period.
The initial query yielded 456 patients. We excluded 243 patients for miscoded diagnoses, 64 for nontraumatic BPI, 21 for late sequelae of remote BPI, 18 for treatment before the study period, 14 for brachial plexus birth palsies, and one for primary neoplasm of the brachial plexus. In addition, 11 patients never presented to a brachial plexus surgeon for evaluation. Six of these 11 patients were managed by an orthopedic trauma surgeon or shoulder surgeon and were never referred to a BPI surgeon, or had never followed up with one. Three of these 11 patients were evaluated by a consulting resident or fellow in the emergency department and never followed up with a BPI surgeon. Two of these 11 patients were managed by neurologists with serial examinations with or without neuropathic pain medication and were never referred to a BPI surgeon. A final cohort of 84 patients who were referred to and evaluated by a brachial plexus surgeon were included in the study (Fig. 1).
A total of 84 traumatic BPI treated by 22 surgeons composed the cohort. Nine surgeons treated 26 patients at hospital 1 and 13 surgeons treated 58 patients at hospital 2. Hospital 1 is the second largest American College of Surgeons Level I trauma center in the metropolitan area, with 804 inpatient beds. Hospital 2 is the largest American College of Surgeons Level I trauma center in the metropolitan area, with 1,017 inpatient beds. We leniently defined brachial plexus surgeon as any staff surgeon who operated on the brachial plexus, without reference to specialty, training, or experience. The response variable was delayed referral, which we defined as more than 3 months from the date of injury to the date of initial evaluation by a brachial plexus surgeon.
The following explanatory variables were studied: age, sex, race, ability to speak English, primary health insurance, in-state residence, distance traveled, social deprivation, injury laterality, body mass index (BMI), diabetes mellitus, smoking status, American Society of Anesthesiologists Physical Status Classification at the time of evaluation, hospital, mechanism of injury, closed head injury, rib fracture, chest injury, abdominal injury, and extremity fracture or dislocation. Race was designated as Caucasian or non-Caucasian owing to the small number of patients per group who identified as African American, Hispanic, Asian, or mixed race. Body mass index closest to the date of surgery, within 1 year before or after treatment, was used for analysis. Distance traveled was measured using both the straight-line distance and the driving distance in kilometers from the self-reported address of residence to the treating hospital. Social deprivation was measured by the Area Deprivation Index, a validated tool for assessing social deprivation that accounts for poverty, housing, employment, and education,
using self-reported address of residence. We reported the Area Deprivation Index as a US national percentile to provide context. Medical comorbidities were assessed by review of the electronic medical record. Pattern of brachial plexus injury was classified as global (C5–T1), upper trunk (C5–C6), extended upper trunk (C5–C7), lower trunk (C8–T1), and infraclavicular (BPI at the level of the divisions and cords).
Statistical analysis
Descriptive statistics for explanatory variables were calculated for the study cohort. All explanatory variables had 100% data completeness with the exception of 94% data completeness for social deprivation and BMI. All variables were analyzed using available data; missing data were excluded (Table 1). Bivariate analysis was used to screen for factors associated with delayed referral. Student t test was used for continuous variables, Mann-Whitney U test for ordinal variables, and Fisher exact test for categorical variables. We included variables with P < .1 in the multivariable logistic regression model for the response variable of delayed referral, dichotomously defined as evaluation by a brachial plexus surgeon less than or more than 3 months from the date of injury.
Data were partially available for area deprivation index (n = 79) and BMI (n = 79). Driving distance was not applicable for 2 patients referred from the island of Bermuda.
Variable
Total Cohort
Age, y (mean [SD])
45 (18)
Area Deprivation Index (US percentile)
30 (24)
BMI (mean [SD])
27 (6)
ASA (median [interquartile range])
2 (1–2)
Travel distance
Straight line, km
38 (14–148)
Driving, km
46 (24–157)
Male, n (%)
58 (69)
Caucasian race, n (%)
58 (69)
English-speaking, n (%)
81 (96)
In-state residence, n (%)
56 (67)
Insurance, n (%)
Private
48 (57)
Medicare
17 (20)
Medicaid or state equivalent
11 (13)
Workers’ compensation
3 (4)
Self-pay
3 (4)
Motor vehicle insurance
1 (1)
International
1 (1)
Laterality, n (%)
Left
40 (48)
Right
44 (52)
Diabetes mellitus, n (%)
5 (6)
Smoker, n (%)
11 (13)
Hospital, n (%)
Hospital 1
26 (31)
Hospital 2
58 (69)
Mechanism of injury, n (%)
Motorcycle accident
19 (23)
Motor vehicle accident
18 (21)
Ground-level fall
14 (17)
Sports injury
10 (12)
Snowmobile accident
8 (10)
Fall from height
6 (7)
Penetrating trauma
6 (7)
Blunt trauma
3 (4)
Closed head injury, n (%)
12 (14)
Rib fracture, n (%)
13 (16)
Chest injury, n (%)
10 (12)
Abdominal injury, n (%)
5 (6)
Extremity fracture or dislocation, n (%)
51 (61)
Pattern of brachial plexus injury, n (%)
Global (C5–T1)
33 (39)
Upper trunk (C5–C6)
19 (23)
Extended upper trunk (C5–C7)
10 (12)
Lower trunk (C8–T1)
9 (11)
Infraclavicular
13 (15)
∗ Data were partially available for area deprivation index (n = 79) and BMI (n = 79). Driving distance was not applicable for 2 patients referred from the island of Bermuda.
The standard significance criterion of α = 0.05 and standard power criterion of (1 – β) = 0.80 was employed for all statistical tests. A convenience sample was used. With a sample size of 84, using an assumption of a 45% rate of a positive response variable, we had greater than 80% power to detect an odds ratio (OR) of 5 for categorical variables in the multivariable logistic regression analysis.
Results
Patient demographics
Mean age of the 84 patients was 45 years, mean BMI was 27, and 69% were male. Median American Society of Anesthesiologists Physical Status Classification was 2. Six percent had diabetes mellitus and 13% were active smokers at the time of injury. Most patients were English-speaking (96%) and Caucasian (69%) and had private insurance (57%). Two-thirds of patients lived in-state. Mean Area Deprivation Index was in the 30th national percentile, representing a relatively affluent cohort. Mean Area Deprivation Index of patients treated at hospital 1 was in the 29th national percentile, and mean Area Deprivation Index of patients treated at hospital 2 was in the 31st percentile. Median straight line travel distance was 18.2 km to hospital 1 and 52.2 km to hospital 2. Median driving travel distance was 28.6 km to hospital 1 and 64.7 km to hospital 2.
The most common mechanism of injury was a motorcycle accident (23%), followed by a motor vehicle accident (21%) and a ground-level fall (17%). The most common pattern of BPI was global palsy (39%), followed by upper-trunk palsy (23%) and infraclavicular palsy (15%). Associated extremity fracture or dislocation was seen in 61% of patients, rib fracture in 16%, closed head injury in 14%, chest injury in 12%, and abdominal injury in 6% (Table 1).
Factors associated with delayed referral
Median time from injury to surgical evaluation was 2 months (interquartile range, 2–4 months; mean, 3 months; range, 0–17 months). Thirty-seven patients had a delayed referral and were first evaluated by a surgeon more than 3 months after the date of injury (44%) (Fig. 2).
Figure 2Number of traumatic BPI patients referred for surgical evaluation, by months from injury.
The most common cause of a delay in evaluation by a brachial plexus surgeon was delayed referral by the local hospital (n = 21). The second most common cause was a misdirected referral, such as to a shoulder surgeon, spine surgeon, or trauma surgeon (n = 6). Other identified causes of delay included management of polytrauma (n = 4), patient incarceration (n = 2), management of medical comorbidity (n = 1), active psychiatric illness (n = 1), acquisition of electrodiagnostic testing (n = 1), and the patient seeking multiple opinions (n = 1).
Bivariate analysis showed that patients injured in motorcycle collisions, those referred from out of state, those traveling from greater straight-line and driving distances, and those treated at hospital 2 were more likely to have a delayed referral (P < .05). In addition, left-sided laterality of injury, closed head injury, and Medicare insurance met criteria for inclusion in the regression model. Because of covariation between straight-line and driving distances, only driving distance was included in the multivariable analysis.
Multivariable logistic regression analysis showed that treatment at hospital 2 (OR = 8.7; 95% confidence interval [CI], 2.0–38), Medicare insurance (OR = 6.0; 95% CI, 1.4–27), and motorcycle accident mechanism of injury (OR = 5.9; 95% CI 1.4–24) were associated with a delayed referral (Table 2).
Table 2Final Multivariable Logistic Regression Model for Factors Significantly Associated With Delayed Referral for Traumatic BPI
Traumatic BPI are rare, devastating injuries that result in considerable upper-extremity motor deficits, sensory deficits, and neuropathic pain. Primary nerve reconstruction, with nerve grafting and/or nerve transfers, has the potential to improve motor function; however, the results of delayed or late nerve reconstruction are often inferior if the time of nerve regeneration and muscle reinnervation is greater than the survival time of the motor end plates once denervated.
Outcomes of single versus double fascicular nerve transfers for restoration of elbow flexion in patients with brachial plexus injuries: a systematic review and meta-analysis.
As such, early referral to a brachial plexus surgeon is paramount for a good functional outcome.
In this study, we have shown that at 2 tertiary referral centers in a large metropolitan area in the United States, nearly half of patients evaluated for traumatic BPI were referred in a delayed time frame. Our findings are disappointing, but similar findings have been reported in other parts of the world. In a study from Brazil, Faglioni et al
reported an average delay of 7 months from injury to surgical reconstruction, which the authors attributed to delayed referrals. In a study from Ireland, Dolan et al
found that nearly 30% of BPI patients underwent surgery more than a year from the time of injury, again likely owing to the timing of referral. In the current study, we found an average delay of 3 months from injury to evaluation, but many referrals were delayed, up to 17 months from injury. We identified insurance status as a significant determinant of access to care for this patient population. Misdirected referral, such as to a shoulder surgeon, spine surgeon, or trauma surgeon, was common, which highlights the importance of early accurate diagnosis of the underlying neurologic injury.
In our study, we showed that even within the same metropolitan area in the United States, the institution accepting the referral had a significant effect on the timeliness of the referral. One explanation for this inconsistency is differing communication pathways with local and regional hospitals and differing referral patterns. Alternatively, catchment areas may differ among regional referral centers; some hospitals accept patients with more severe injuries from longer distances. Finally, traumatic BPI are injuries treated by orthopedic surgeons, plastic surgeons, and neurosurgeons. Issues of triage at the individual hospital level may further compound delays in referrals. These potential causes of delayed referral argue for multidisciplinary specialty centers for traumatic BPI to expedite regional referrals and treatment.
Brachial plexus injury patients with Medicare insurance are at risk for delayed referral to a brachial plexus surgeon. The association of insurance type with delayed referral may be related to barriers to access of care with certain types of insurance. Insurance type has been shown to be a potential barrier to ambulatory hand surgery evaluation for an array of diagnoses, including pediatric trigger thumb,
Prior literature on the impact of health insurance and socioeconomic disparities on treatment in traumatic BPI has been mixed. A retrospective analysis of the National Inpatient Sample database from 2009 to 2014 by Bucknor et al
demonstrated that Caucasian men with private insurance were more likely to undergo brachial plexus surgery in the elective setting. By contrast, in a retrospective analysis of administrative databases from 3 states, Dy et al
did not find an association between insurance type and delay to surgery more than 365 days. The current study was performed at 2 tertiary referral centers and provides greater granularity of data than that achievable by administrative database analyses. By studying the date of initial evaluation by a brachial plexus surgeon, rather than the date of surgery, we are able to address the question of risk factors more directly for delayed referrals. Contrary to prior literature,
we did not find that travel distance was associated with delayed referrals after controlling for the confounding effects of insurance, treating hospital, and mechanism of injury. We are unable to explain the association between motorcycle accident as the mechanism of injury and delayed referrals; this association may result from an unobserved confounding variable.
We defined delayed referral as more than 3 months from the date of injury for the purposes of this study, but this is a point of debate. Some authors advocate surgical reconstruction even earlier in certain cases.
Although important diagnostic tests such as electrodiagnostic studies, magnetic resonance imaging, and computed tomography myelography can be obtained as early as 4 weeks from injury,
most brachial plexus surgeons believe that 3 to 6 months from the time of injury is often optimal for primary nerve reconstruction, allowing some time for spontaneous recovery but not at the expense of timely neurotization.
We chose 3 months as a cutoff point for delayed referral because of increasing evidence supporting improved outcomes with earlier surgical intervention. In a systematic review of the timing of surgery for traumatic BPI, Martin et al
found the highest percentage of Medical Research Council (MRC) grade 3 motor recovery in patients who underwent surgery within 3 months from the time of injury. Moreover, patients with at least MRC grade 3 recovery underwent surgery at a median delay of 4 months, whereas patients with less than MRC grade 3 recovery underwent surgery at a median delay of 7 months.
Although it is outside the main objectives of this study, we incidentally found that 11 patients known to our hospital system (12% of the potentially eligible cohort) were never formally evaluated by a brachial plexus surgeon. The cause of this was likely multifactorial, resulting from patient nonadherence with follow-up and selection bias. It may be that more patients in this group had early spontaneous recovery such that the treating surgeon or neurologist did not think a referral to a brachial plexus specialist was warranted.
This study had several limitations. First, the study was limited by its retrospective design. Causes of delayed referral may be multifactorial and unknown at the time of retrospective chart review. Second, we had access to medical records only from within our hospital systems. Because medical records from referring hospitals were unavailable to us, we are unable to comment on factors that may have influenced delayed referral in most cases. It is possible that patients were evaluated by brachial plexus surgeons from their local or regional hospital or another referral center before referral to our study hospitals. Third, it is possible that our retrospective query by diagnosis code was not exhaustive of BPI patients referred to our centers. The large number of exclusions for miscoded diagnoses in our study suggests inherent inaccuracies in identifying BPI by diagnosis code. Fourth, we used a sample of convenience, and the study was powered to detect relatively large effects. Although significant associations were identified, large CIs preclude conclusions about precise effect sizes. Finally, this study was performed at 2 tertiary referral centers in one metropolitan area in the United States; our findings may not be generalizable to other settings. The study sample was older on average; a sizable minority was injured by lower-energy mechanisms than previously described.
Moreover, the study population was relatively affluent, almost entirely English-speaking, largely Caucasian, and with mostly private insurance, which does not reflect all practice settings. Nevertheless, even in a socially advantaged patient population, the prevalence of delayed referral for BPI is high.
The treatment of traumatic BPI is time-sensitive, and early nerve reconstruction is associated with superior nerve recovery.
Within a relatively affluent US population, we demonstrated that insurance type and the institution accepting the referral are important factors in the timeliness of specialist evaluation for these injuries. Although surgeons or hospital administrators cannot modify the circumstances of the injury, such as the mechanism of injury, regional referral patterns and insurance approval for initial consultation are potentially modifiable. Efforts should be focused on clearing modifiable barriers to care for traumatic BPI patients. Strategies to promote timely referrals of traumatic BPI patients for specialist care might include establishing multidisciplinary specialty centers that care for traumatic BPI patients, providing outreach, education, and communication with local and regional hospitals, and developing referral algorithms and pathways.
Outcomes of single versus double fascicular nerve transfers for restoration of elbow flexion in patients with brachial plexus injuries: a systematic review and meta-analysis.
00098-8&id=gr1.jpg){kind=link}
00098-8&id=gr2.jpg){kind=link}