Neonatal brachial plexus palsy (NBPP) is a perinatal condition with an incidence of 0.9 to 1.5 per 1000 live births that results in weakness and sensory disturbances in the upper extremity.^1-3 More than 30% of patients develop lifelong impairments,⁴ including impaired extremity function, behavioral challenges, musculoskeletal and cosmetic deformity, and socioeconomic limitations.^{5, 6} Most children present with Erb palsy, reflecting upper trunk and nerve root injury, but nearly all patients with NBPP have some degree of impairment of shoulder and elbow movement.

Management of these children ranges from rehabilitation and therapy alone to surgical nerve or muscle reconstruction. Historically, the paradigm that ‘all NBPP recovers’ guided treatment. However, with the advent of safe and efficacious nerve reconstruction options, surgery has become a potential treatment for NBPP. While indications for surgery are controversial,^{7, 8} most surgeons view the goal of surgical intervention as exceeding the outcomes achieved by conservative means, thus maximizing functional recovery and limiting persistent disability. Notably, there is a paucity of published studies regarding outcomes after conservative management; the studies that are available are limited by their retrospective nature, small overall numbers, and short follow-up times.^9-11 In their systematic review aimed at determining the natural history of nonoperative NBPP, Pondaag et al.² concluded that ‘although the natural history of [NBPP] has been described in a large number of studies, no study had the width of scope necessary to answer the question posed’.

Optimal timing of referral of patients to multidisciplinary specialty centers, which provide access to occupational therapy, rehabilitation providers, and nerve surgeons, is also unclear, particularly if the child's primary physician is waiting for and anticipating spontaneous recovery. Several groups have advocated for early referral, but to date the impact of early management at a specialty center on functional recovery has not been demonstrated.^{3, 12-14}

In this study, we interrogated a prospectively collected database of conservatively managed children with NBPP evaluated or treated at a single multidisciplinary specialty NBPP center over a 15-year period to demonstrate the trajectory of recovery of active range of motion (AROM) at the shoulder and elbow over time. Specifically, we tested the hypothesis that nonsurgically treated patients with NBPP spontaneously recover full AROM with time, and we ask whether early referral to a specialty center alters the trajectory or maximal extent of recovery (or both).

METHOD

RESULTS

Six hundred and eighty-five patients were evaluated for NBPP at a single tertiary referral center between November 2005 and November 2020. Of these, 100 patients underwent nerve surgery and were excluded from the analysis. One hundred and fifty-four patients were seen initially after 2 years of age and two patients had no AROM recorded, leaving a total of 429 patients (220 males, 209 females) representing 441 injured arms (Figure S1).

To determine the long-term impact of age at presentation on recovery of AROM of the upper extremities, four age cutoffs were modeled: presentation to a specialty clinic before and after 30, 60, 90, and 120 days. The 30-day cutoff was the most predictive of differences in long-term outcomes (Figures S2 and S3); additional analyses focused on this cohort.

One hundred and thirty-one children were initially evaluated early (≤30 days), while 298 were evaluated later (>30 days). The mean (SD) follow-up was 23.44 (35.92) months. There was no difference in mean follow-up between the group who presented to the clinic at 30 days or earlier versus the group who presented to the clinic later than 30 days (25.84 months vs 22.39 months, p = 0.37). There were no differences in sex, ethnic grouping, side of injury, and extent of injury (Narakas classification) between the two groups (Table 1).

A total of 13 139 AROM measurements were collected for shoulder abduction, forward flexion, and external rotation and elbow flexion, extension, supination, and pronation and plotted against patient age for the group who presented to the clinic early (n = 4572) and the group who presented to clinic later (n = 8567). A histogram was plotted to demonstrate the distribution of patient age at the time of AROM measurement (Figure 1).

The plots of normalized AROM against patient age for shoulder abduction, forward flexion, and external rotation are shown in Figure 2 (Tables S2–S4). For the overall group, AROM recovery for shoulder abduction and forward flexion largely improved in a parallel fashion as patient age increased, although there was some tapering at the later time points. Importantly, despite a lower starting AROM, there was a statistically significant difference in long-term AROM between the two groups, with 27.8% more AROM in shoulder abduction (162° vs 113°, p = 0.01) and 23.1% more AROM in shoulder forward flexion (156° vs 114°, p = 0.049) for the early group than for the late group. Regarding shoulder external rotation, the greatest improvement in AROM for the overall group occurred within the first 20 months, again with some tapering and stabilization of recovery over time. The group who presented to the clinic early had significantly better AROM than the group who presented to the clinic later, between 28 and 33 months, 40 and 48 months, and 61 and 73 months, although significance did not persist at the latest time points.

Figure 3 shows the plots of normalized AROM for elbow flexion, elbow extension, forearm supination, and forearm pronation (Tables S5–S8). For the overall cohort, AROM for elbow flexion demonstrated near-complete recovery by approximately 5 years of age. Forearm supination AROM also improved with age, although to a lesser extent than elbow flexion, while the AROM for elbow extension and forearm pronation decreased over time. For elbow extension, there was a statistically significant difference in long-term AROM between the early and late groups, with 21.2% more AROM for the early group than for the late group (−3.2° vs −22.2°, p < 0.001). For forearm supination, there was a statistically significant difference in AROM between the groups from 25.8 to 31.9 months of age (67.9° vs 50.2°, p = 0.02), 40.7 to 51.6 months (74.7° vs 53.6°, p = 0.003), 51.6 to 66.4 months (78.8° vs 56.5°, p = 0.008), and 66.4 to 90.9 months (78° vs 52.6°, p = 0.002). However, the significance did not persist at the latest measured time points. For elbow flexion, the late group outperformed that early group from 1.5 months of age to 2.4 months of age (127.7° vs 113.5°, p = 0.05) but there was no significant difference between the groups at the later time points. For forearm pronation, there was no significant difference between the groups at any time point.

DISCUSSION

In this study, we analysed the AROM for seven discrete and relevant upper-extremity movements measured during clinic visits over the course of 15 years for 429 children with nonoperative NBPP and illustrated the trajectory of recovery of AROM of each movement for the overall cohort over time. The overall cohort had nearly complete recovery of elbow flexion AROM, while recovery to approximately 60% to 80% of full AROM was seen for shoulder abduction, forward flexion, and external rotation, and in forearm supination. Loss of AROM was also noted for elbow extension and forearm pronation. These findings refute the hypothesis that patients with conservatively managed NBPP eventually achieve full AROM. However, we found that children referred to a specialty clinic within 30 days of birth had better overall recovery of AROM for shoulder abduction, forward flexion, and external rotation, and forearm supination, with reduced loss of elbow extension AROM (Figures 2 and 3).

Historically, recovery of shoulder function has lagged behind recovery at the elbow. However, a recent study of adults with brachial plexus injury showed a direct correlation between shoulder AROM and real-world use of the arm as measured using wearable technology, emphasizing the importance of adequate shoulder movement to spontaneous, patient-initiated use of the arm in everyday life.^{18, 19} These studies underscore the importance of our findings that children evaluated at a specialty center within 30 days have significantly better long-term AROM in shoulder abduction and forward flexion. While not statistically significant at the latest time point, shoulder external rotation AROM for the early group was consistently better than for the later group.

We were not surprised to see that elbow extension AROM decreased for the overall cohort over time. Elbow flexion contracture is a well-known complication of NBPP, with prevalence reported between 25% and 89%.²⁰ Elbow extension has an important role in stabilizing the arm while lifting, carrying, pushing, pulling, and throwing; contractures greater than 30 degrees interfere with activities of daily living.^{21, 22} Additionally, children and adolescents with NBPP have reported concerns about the cosmetic appearance of elbow flexion contracture.²³ The etiology of elbow flexion contracture is controversial; however, recent evidence from mouse models suggests that denervation of flexor muscles in the arm results in decreased growth of muscle sarcomeres and shortening of the muscle, which in turn leads to musculoskeletal contracture.²⁰

However, we were very surprised to see that children seen within 1 month had almost no loss of elbow extension over time. Taken together, these data suggest that early specialized rehabilitation may help prevent potentially disabling elbow flexion contracture, possibly by preventing sarcomere shortening of the biceps muscle.

We also noted loss of forearm pronation for the overall cohort. Like elbow flexion contracture, supination contracture is a known complication of NBPP, with an overall incidence in one series of 7% that rises to 23% in patients with injury involving the entire brachial plexus.²⁴ Our data argue for vigilance in recognizing and addressing the development of supination contractures in this patient population.