Long-Term Outcomes following Pediatric Peripheral Nerve Injury Repair

Abstract

Introduction  Peripheral nerve injuries in children are uncommon and can be challenging to diagnose. There is a paucity of data on long-term sensorimotor and functional outcomes following surgical repair. We present a 12-year retrospective analysis of pediatric peripheral nerve repair with long-term functional outcomes.

Materials and Methods  We performed a retrospective analysis of pediatric patients with peripheral nerve injury requiring surgical repair. Clinical records were analyzed for procedure type, time to surgery, mechanism of injury, postoperative recovery, and complications.

Results  A total of 108 patients were identified and 87 patients were included. Out of 87 patients, 83 (95.4%) had partial or complete sensorimotor recovery at final follow-up and 4 did not improve. Minor complications occurred in 10.3% of patients, all resolved with conservative management. Mechanisms of injury were predominantly lacerations with sharp objects or crush injuries. Age at time of injury was inversely correlated with sensorimotor recovery, and time to surgical repair was not.

Conclusion  Surgical repair with long-term hand therapy results in excellent functional outcomes following pediatric peripheral nerve injury. A low threshold for exploration and repair should be used in instances of diagnostic uncertainty. Timing of surgical repair is dependent on a patient’s clinical presentation; however, repair within 48 hours is sufficient for optimal sensorimotor recovery.

Introduction

Peripheral nerve injuries in the pediatric population are uncommon and complex to manage; they may not be immediately obvious on examination, and there can be communication and examination barriers, particularly in young children. Due to these challenges, a low threshold for surgical exploration should be used where there is a possibility of underlying nerve injury.

Peripheral nerve injuries in the pediatric population occur commonly due to trauma, often associated with broken glass, 1 but they may also be caused by fractures, dislocations, and crush injuries. 2 3 Nerve regeneration in children has been associated with superior outcomes compared with adults, and some small studies have suggested this particularly declines after the first 10 years of life. 4 To determine the presence and severity of a peripheral nerve injury, a full clinical examination along with imaging is necessary. It is, however, important to maintain a high index of suspicion for an underlying nerve injury, as the outcome of the clinical examination is often dependent on the cooperation of the child.

The prognosis for surgical intervention after injury depends on its degree of severity. Peripheral nerve injuries can be classified into six degrees, where first- and second-degree injuries have the best prognosis with spontaneous recovery, whereas third- to sixth-degree injuries involve scar formation, making spontaneous recovery less likely and thus indicating the need for surgical repair. Primary surgical repairs should be attempted where possible; however, contamination, bony instability, and compromised tissue viability may preclude this, requiring delayed secondary repair. 5

Due to the infrequency of this type of injury in the pediatric population, data on long-term outcomes and prognostic indicators are lacking in the literature. Studies on the adult population have found that increasing age and smoking are factors associated with poor recovery following nerve repair; however, these data are not available for the pediatric population and some adult prognostic indicators are not applicable to children. 6

Robust outcomes data are required to support evidence-based clinical decisions, particularly in instances of diagnostic uncertainty. This study aims to identify long-term outcomes and prognostic indicators in the pediatric population after peripheral nerve repair with a retrospective analysis of 108 patients who underwent peripheral nerve repair over a period of 12 years.

Materials and Methods

We conducted a retrospective analysis of our center’s trauma database to identify all patients aged < 18 years presenting with peripheral nerve injury who underwent surgical repair, with nerve injury being confirmed intraoperatively. Our unit provides trauma care to a large urban center population. Electronic medical records were reviewed to confirm procedure type, time to surgery, mechanism of injury, injuries identified intraoperatively, postoperative complications, postoperative sensorimotor recovery, and postoperative follow-up period. Patients were excluded if their operative records were unavailable or if they failed to attend postoperative hand therapy follow-up.

Patients received postoperative follow-up in hand therapy clinic with the goal of providing physiotherapy to rehabilitate hand function and to monitor sensorimotor recovery. These electronic records were reviewed for postoperative sensorimotor recovery, including whether sensorimotor function improved from their preoperative baseline assessment and whether full sensorimotor recovery was achieved within the follow-up period.

Two-point discrimination thresholds were used to assess sensory recovery, with active movement to assess motor function. Assessment of patients aged < 5 years is often challenging due to communication constraints. In such instances, functional assessment was utilized by experienced hand therapists. Full sensory recovery was defined as recovery to normal two-point discrimination in the absence of paraesthesia. Full motor recovery was defined as full active range of motion at each joint with no impairment in strength.

Data were analyzed using IBM SPSS statistics version 24. Descriptive statistics were reported, and statistical analysis was performed where sufficient data were present. Statistical analysis of two independent, non-normally distributed groups was conducted using the Mann-Whitney U test, with ANOVA (one-way analysis of variance) used for comparing more than two groups. Relationships between variables were investigated using logistic regression analysis. Two-tailed p values are reported, and a p value < 0.05 was considered statistically significant.

Results

Total 108 patients were identified over a 12-year period (2006–2018) from our retrospective dataset analysis. Of the 108 patients, 13 were excluded due to missing records or being subsequently found to not fit the inclusion criteria. Of the 95 patients remaining, 7 were lost to follow-up and thus were excluded due to the absence of long-term outcomes data. Of the 87 patients included in our study, 60 were male and 27 were female.

Sensorimotor Recovery

The mean follow-up time postoperatively was 15.36 weeks (1–63 weeks) ( Fig. 1 ). Of the 87 patients with follow-up data, 4 (4.6%) had no improvement in sensorimotor function following surgical nerve repair, whereas 35 (40.2%) demonstrated incomplete recovery and 48 (55.2%) demonstrated complete recovery of sensorimotor function ( Fig. 2 ). Complete recovery was defined as no sensorimotor deficit at final follow-up; incomplete recovery was defined as incomplete recovery of sensorimotor function postoperatively or complete recovery with the presence of paraesthesia.

Fig. 1.

 Length of time between surgical repair and the final documented hand therapy clinic attendance for all included patients.

Fig. 2.

 Postoperative sensorimotor recovery status for all included patients at the final hand therapy clinic attended. Incomplete recovery was defined as incomplete recovery of sensorimotor function postoperatively or complete recovery with the presence of paraesthesia.

Of those with incomplete sensorimotor recovery at final follow-up, the majority had improving but reduced sensory thresholds (76.5%). The remainder were due to either suboptimal range of movement (14.7%) or paraesthesia (8.8%). Patients with incomplete recovery received longer follow-up at a mean of 17.2 weeks (1–52 weeks); six of the incompletely recovered patients were lost to follow-up within 1 month of surgical repair.

Age at Injury

The median age at time of injury was 7.06 years (0–16) ( Fig. 3 ). Patient age at time of injury was inversely correlated with complete sensorimotor recovery during the follow-up period ( p = 0.002, regression coefficient −0.169, 95% CI: −0.274, −0.064).

Fig. 3.

 Age of all included patients at the time of nerve injury.

Previous studies have suggested that recovery declines significantly after age 10 years. We found that the patients aged > 10 years ( n = 25) had significantly inferior recovery compared with those aged < 10 ( p = 0.006, two tailed, n = 62).

Time to Repair

The mean time to surgery following peripheral nerve injury was 2.3 days (0–8 days) ( Fig. 4 ). Time to surgery was inversely associated with complete sensorimotor recovery; however, this was not statistically significant ( p = 0.335, regression coefficient −0.140, 95% CI: −0.424, 0.144).

Fig. 4.

 Time between the patient’s nerve injury and completion of the surgical repair.

Nerve Injured

Most pediatric patients presented with injury to the digital nerves (79.3%). The remaining patients sustained injuries to the ulnar (6.9%), median (5.7%), superficial branch of the radial (4.6%), superficial branch of the ulnar (2.3%), and the common digital (1.1%) nerves. A small percentage of patients presented with injury to more than one nerve (9.2%).

A subgroup analysis was performed to compare outcomes following injury to thin sensory nerves and larger nerves with both sensory and motor function; there was no significant difference in long-term recovery between these two groups ( p = 0.189, two tailed).

Subgroup analysis with thin sensory only nerves was consistent with the primary analysis—age ( p = 0.001, regression coefficient −0.182, 95% CI: −0.286, −0.078), time to repair ( p = 0.126, regression coefficient −0.226, 95% CI: −0.516, 0.064). Subgroup analysis with the larger sensorimotor nerves did not reach statistical significance when analyzing the relationship between patient age and recovery. Given the small size of this subgroup ( n = 11), the validity of this test is questionable—age ( p = 0.961, regression coefficient −0.009, 95% CI: −0.359, 0.342), time to repair ( p = 0.745, regression coefficient 0.128, 95% CI: −0.643, 0.899).

Tendon Injury

Coexistence of the tendon and motor nerve injures can complicate the assessment of apparent motor deficits. Of our cohort, four patients presented with both motor nerve and tendon injury, of whom one patient failed to reach full sensorimotor recovery during hand therapy follow-up, with reduced two-point discrimination and weak finger abduction.

Complications

Complication rates were extremely low ( Table 1 ). Nine (10.3%) patients reported minor complications that all resolved with conservative management. One patient had a documented contracture of the right thumb web space; however, this was a predictable complication of the patient’s extensive soft tissue injury at presentation. There were no documented wound infections.

Table 1. Documented complications for all included patients.

Complication	Frequency	Outcome
Moderate wound edema	9	All resolved with conservative management
Contracture of thumb webspace	1	Contracture secondary to extensive soft tissue injury at presentation. No further operative intervention required

Mechanism of Injury

The mechanisms of injury were predominantly lacerations (94.3%) with sharp objects (e.g., broken glass and knives), the remainder being due to crush injuries (4.6%) or degloving injury (1.1%). There was no difference in the sensorimotor recovery for sharp and crush injuries ( p = 0.558, two tailed).

Complete nerve transection occurred in 79.3% of injuries, with the remaining 20.3% being incomplete. The precise extent of incomplete nerve injuries was not recorded in our database. In 90.8% of patients, only one injured nerve was found intraoperatively, with the remaining 9.2% of patients having two or more simultaneously injured nerves. There were no significant differences between the sensorimotor recovery of patients based on the number or extent of nerve injury ( p = 0.451, p = 0.388 respectively, two tailed).

Discussion

Peripheral nerve injuries are uncommon in the pediatric population and present unique challenges in their diagnosis and management. High-quality studies with large patient cohorts evaluating surgical repair for this type of injury in the pediatric population are lacking. Although studies have been published evaluating long-term prognosis following peripheral nerve repair in the adult population, 6 the findings of these studies are not reliably applicable to the pediatric population. Our study adds significantly to the literature by demonstrating the long-term outcomes following peripheral nerve repair in a large pediatric population.

Our study demonstrates that sensorimotor recovery following primary nerve repair in pediatric patients is excellent. Nearly all patients (95.4%) recovered partial or complete sensorimotor function following surgery, with a mean follow-up time of 15.36 weeks (1–63 weeks). Within their follow-up period, most patients demonstrated complete sensorimotor recovery, and of those who had incomplete recovery, 76.5% were categorized as such due to reduced, but improving sensory thresholds. Patients with incomplete recovery were followed up for longer on average; however, some of these patients were lost to follow-up within a relatively short period (six patients within the first postoperative month). It would be reasonable to expect that as they were improving clinically, a proportion of these patients would also go on to complete sensorimotor recovery.

Our study demonstrates that age at time of injury is inversely correlated with the recovery of sensorimotor function following peripheral nerve injury. This finding is consistent with the current literature in smaller studies; Chemnitz et al demonstrated in a study of 45 patients that sensory recovery was superior in their cohort aged 0 to 11 years than their cohort aged 12 to 20 years. 7 Age has been demonstrated as a negative predictive factor of sensory recovery in adult patients as well; Weinzweig et al found in a cohort of 96 adult patients that those aged > 40 years had inferior sensory recovery than those aged < 40 years. 8

Direct comparison between adult and pediatric sensorimotor recovery is challenging due to heterogeneity in study design and outcome measures; however, in the study by Weinzweig et al, normal two-point discrimination was achieved in only 19% of adults aged 21 to 30 years and 11% of those aged > 40 years. 8

Lundborg and Rosén have suggested that sensory recovery following nerve repair declines precipitously after approximately the age of 10 years, with a subsequent plateau; however, the strength of this conclusion is limited by only 6 of their 54 patients being aged < 15, with a wide variance in sensory recovery scores. 9 Our study supports the relationship suggested by Lundborg and Rosén in a larger patient cohort, with our patients aged < 10 ( n = 62) having superior sensorimotor recovery compared with those aged > 10 ( n = 25); however, it is difficult to ascertain whether this is a nonlinear decline in recovery as they suggest.

The optimal time to primary repair is unknown. Some studies have suggested that repair within 24 hours of injury is optimal, with others finding no association between time and patient outcome. 5 8 Our study did not find a statistically significant relationship between time to repair and sensorimotor recovery. Given that our patients received primary repair at a mean of 2.3 days (0–8), it is reasonable to assert that repair within 48 hours of injury is sufficient for optimal recovery.

The complication rate following surgery was extremely low, with only 10.3% experiencing minor complications following surgical repair, all off which resolved with conservative management. The one patient with a documented contracture of the thumb web space was identified to have an extensive soft tissue injury at presentation; thus, this contracture is most likely an expected complication of the initial trauma.

Our study demonstrates that the most common mechanisms of injury are lacerations with sharp objects followed by crush injuries, in keeping with the findings of previous studies. 1 2 3 Some studies have correlated sharp laceration with superior recovery when compared with mechanisms such as crush. 8 In our patient cohort, there was no such association; however, due to the small number of patients presenting due to crush injury, this study may be underpowered to detect a statistical difference.

Inevitably there are limitations to our study. There was a small loss to follow-up; however, this represents a small percentage of the overall cohort and is in keeping with typical postoperative follow-up. Sensory nerve injures predominate in our cohort, and whereas subgroup analysis did not demonstrate any objective difference with larger motor nerve injury, further investigation is required to corroborate this. Finally, the retrospective nature of our study limits our analysis to data recorded during follow-up, and thus we are unable to ascertain whether those patients with improving but incomplete recovery at their final follow-up subsequently continued to improve.

Peripheral nerve injuries can have profound impacts on a patient’s quality of life, particularly so in the developing child where sensorimotor interaction is a fundamental part of learning and development. Outcomes in these patients are poorly evidenced at present and based on short periods of follow-up, yet clinical decision making and informed patient consent are dependent on this, particularly in cases of diagnostic uncertainty. Our study provides clear evidence that surgical repair with long-term postoperative hand therapy results in excellent functional outcomes, so a low threshold for exploration and repair should be used by clinicians in cases of diagnostic uncertainty. The optimal timing of surgical repair is dependent on a patient’s clinical presentation; however, repair within 48 hours is sufficient for optimal recovery.