Traumatic peripheral nerve injuries: a classification proposal

Abstract

Background

Peripheral nerve injuries (PNIs) include several conditions in which one or more peripheral nerves are damaged. Trauma is one of the most common causes of PNIs and young people are particularly affected. They have a significant impact on patients’ quality of life and on the healthcare system, while timing and type of surgical treatment are of the utmost importance to guarantee the most favorable functional recovery. To date, several different classifications of PNIs have been proposed, most of them focusing on just one or few aspects of these complex conditions, such as type of injury, anatomic situation, or prognostic factors. Current classifications do not enable us to have a complete view of this pathology, which includes diagnosis, treatment choice, and possible outcomes. This fragmentation sometimes leads to an ambiguous definition of PNIs and the impossibility of exchanging crucial information between different physicians and healthcare structures, which can create confusion in the choice of therapeutic strategies and timing of surgery.

Materials

The authors retrospectively analyzed a group of 24 patients treated in their center and applied a new classification for PNI injuries. They chose (a) five injury-related factors, namely nerve involved, lesion site, nerve type (whether motor, sensory or mixed), surrounding tissues (whether soft tissues were involved or not), and lesion type—whether partial/in continuity or complete. An alphanumeric code was applied to each of these classes, and (b) four prognostic codes, related to age, timing, techniques, and comorbidities.

Results

An alphanumeric code was produced, similar to that used in the AO classification of fractures.

Conclusions

The authors propose this novel classification for PNIs, with the main advantage to allow physicians to easily understand the characteristics of nerve lesions, severity, possibility of spontaneous recovery, onset of early complications, need for surgical treatment, and the best surgical approach.

Level of evidence: according to the Oxford 2011 level of evidence, level 2.

Introduction

Traumatic peripheral nerve injuries (PNIs) include several conditions producing damage of one or more peripheral nerves, together with possible loss of motor or sensory functions. Trauma is one of the most common causes of PNI in the general population and the most common in young people, with an incidence estimated between 1.46% and 2.8%, especially in the upper extremities [1,2,3]. Nerve injuries that occur during specific sports account for less than 0.5%, but recent studies suggest a higher rate in the USA [4].

In the upper limbs, the radial nerve is the most frequently involved, followed by the ulnar nerve, and the median nerve. In the lower limbs, the sciatic nerve is the most frequently damaged, followed by the peroneal nerve [2, 5]. Among the forces responsible for injuries, traction, transection, radiation, compression, thermal, and electrical forces must be mentioned, as they produce the same pathophysiological effects, including demyelination and Wallerian degeneration [6].

When approaching PNI, it is crucial to consider the significant impact they have on patients’ quality of life and on healthcare systems. PNIs, indeed, affect mostly young and economically active people, who suffer from various degrees of disability and from neuropathic pain [7, 8], which is a chronic and especially challenging condition for patients with PNI, leading to a reduction in autonomy during activities of daily living (ADLs) and precluding job opportunities and forcing people to adopt chronic use of painkillers and other drugs [9,10,11]. Moreover, orthoses, largely used to reduce pain, may be very uncomfortable and aesthetically unpleasant for patients with PNI [12, 13].

The most favorable recovery can be achieved through early diagnosis, correct timing, and type of surgery [14], but can be affected by several factors such as age [15], gender [11], comorbidities, type and level of injury, and the presence of concomitant injuries [16]. With regard to time, in our experience, a delay in identification and treatment of PNIs is often seen, especially in road accident victims. Late referral to specialized centers or concomitant other life-threatening conditions, such as head or thoracic traumas, produce a crucial delay in PNI diagnosis and treatment, and irreversibly affect the outcome [17,18,19,20,21].

Correct diagnosis is paramount for surgical planning, as bone, vessel and muscle, and multiple nerve injuries may also be associated [2, 22, 23], and a single surgery may not be enough to restore satisfactory function; in some cases, multiple surgeries are required [23,24,25]. Moreover, revision surgery is not infrequent, especially for patients with delayed diagnosis [26,27,28]. Identifying precise clinical and surgical data through the application of a univocal language could produce more accurate exchanges between hospitals and surgeons and allow surgeons to better plan revision surgeries.

In this paper, we describe a novel anatomo-topographical and prognostic classification, in which two distinct types of alphanumeric code can be applied to all PNIs to guide physicians in the decision process. We introduced this classification in our center to verify its feasibility, and it has been adopted among physiatrists, physiotherapists, neurologists, and surgeons in a multidisciplinary setting.

Materials and methods

We chose the following five injury-related factors to create the alphanumeric code:

• Nerve involved: axillary or circumflex (A),   suprascapular (Su), long thoracic (LT), thoracodorsal (TD), musculocutaneous (Mu), median (Me), radial (R), ulnar (U), common digital (CD), sciatic (Sc), peroneal (P), tibial (T)

• Lesion site: 1 shoulder/pelvis, 2 arm/thigh, 3 elbow/knee, 4 forearm/leg, 5 wrist–hand/ankle–foot

• Nerve type: 1 motor, 2 sensory, 3 mixed

• Surrounding tissues: open (O), closed (C)

• Lesion type: partial/in continuity (P), complete (C) (modifiers: 1 clean, 2 crushed, 3 loss of tissue);

Regarding the prognostic alphanumeric code, we considered the following four factors:

• Age: aged (A—more than 60), young (B)

• Timing: 1 immediate, 2 delayed, 3 secondary

• Technique: 1 suture, 2 graft ≤ 6 cm, 3 graft > 6 cm.

• Comorbidities (smoking, diabetes, etc.): yes (Y), no (N)

Each alphanumeric code can unambiguously be associated with a management strategy and specific treatment options. The prognostic alphanumeric code, calculated on patient- and treatment-related factors, such as age, comorbidities, and surgical technique, guides the physician in predicting time of recovery and prognosis.

A complete description of our classification is summarized in Table 1.

Table 1 Peripheral nerve injury code description

Extensive application of the alphanumeric code was carried out in 24 patients treated in our center (Table 2), also reporting the surgical treatment we chose for each patient.

Table 2 Table of patients

Results

Surgery was decided because of the above-mentioned anatomo-topographical and prognostic factors, and a multidisciplinary team was able to retrospectively propose the same therapeutic strategies thanks to the clear definition of the lesion.

Every alphanumeric code could unambiguously be associated with a management strategy and specific treatment options. The prognostic alphanumeric code, calculated on patient- and treatment-related factors, such as age, comorbidities, and surgical technique, guided the physician in predicting time of recovery and prognosis.

Discussion

As for spine injuries [29] and fractures to the appendicular skeleton [30], our classification with its alphanumerical system moves a step forward in describing PNI through the improvement in inter- and intraobserver reliability, helping to address both surgeons’ decision process and prognosis prediction.

Historically, in the context of PNI, the first attempt to classify nerve injury was made by Seddon [31], considering the correlation between pathological evidence and outcome. Since nerve regeneration was seen only in class I and II injuries corresponding to neuroapraxia and axonotmesis, respectively, a surgical option was justified in the early period in class III injuries (neurotmesis) where the surrounding connective tissue of the nerve was disrupted. Following Seddon’s classification, Sunderland created a five-point grading system of PNI severity in ascending order, giving a more detailed description of surrounding connective tissue damage [32]. Connective tissue is spared in grade II injury, while increased involvement of the connective layers surrounding the nerve fibers, namely endoneurium, perineurium, and epineurium, defines a progressively worse injury requiring surgical intervention. Mackinnon [33] added a grade VI to Sunderland’s classification, to include PNIs with mixed pattern and, consequently, to better reflect clinical practice. There is no reference to other elements in the cited classifications that may influence the outcome, such as the traumatic mechanism of the nerve lesion.

Among general classifications, Millesi et al. [34] proposed an analysis to guide surgeons during neurolysis procedures. The authors focused on the site of fibrosis that could occur at different levels after a traumatic injury. They explained different pathogenetic mechanisms through which scar tissue in different layers could impair the nerve and may affect the clinical picture. Specifically, the authors defined 4 types of fibrosis correlating to Sunderland’s grade, requiring specific neurolytic procedures: fibrosis of the epifascicular epineurium (type A) needs epifascicular epineurotomy, when the scar tissue involves the interfascicular epineurium (type B) it is necessary to remove the epineurium layer performing an epineurectomy and, to free deeper layers, it is often associated with partial interfascicular epineurectomy, and in type C fibrosis the endoneurium is involved and the presence of neurolysis aids in the making of a diagnosis. In the original article, a type D was reported corresponding to loss of fascicular pattern observed during neurolysis. This classification represents a first attempt to categorize the nerve fibrosis that could occur after a trauma—or after surgery—and to guide surgeons during surgical decision making. However, considering just the “effect” of the lesion without the underlying “mechanism” and level of injury, the information is partial and incomplete for correct surgical management.

We could say that Seddon, Sunderland, MacKinnon, and Millesi described the injury per se as far as it concerns the nerve trunk.

Other recent studies attempted a classification of single nerve injuries as far as it concerns the nerve’s regional anatomy, with their surgical treatment proposals. Ghoraba et al. [35] have recently proposed an algorithm to assess ulnar injuries considering four anatomical zones: distal to the proximal hiatus of Guyon’s canal (zone I), from the proximal hiatus of Guyon’s canal to the proximal border of the pronator quadratus (zone II), from the proximal border of the pronator quadratus to the first motor branch of the ulnar nerve (zone III), and proximal to the first motor branch of the ulnar nerve (zone IV). They managed injuries to zone I and II, primarily, with neurorrhaphy or with an autologous nerve graft, and more proximal lesions (zone III and IV) with anterior transposition. Although the study shows good results, the proposed classification still remains limited to a single anatomical compartment.

In the case of brachial plexus injuries (BPIs), Millesi et al. [36] identified four anatomical sites of injury, namely (I) supraganglionic/preganglionic, (II) infraganglionic/postganglionic, (III) trunk, and (IV) cord. Indeed, this simplification for surgical management obtained good correlation with outcome. Improvement and diffusion of imaging techniques brought Yang et al. [37] to characterize five types of BPIs with the aid of magnetic resonance imaging (MRI). MRI was used to locate BPIs in relation to the preganglionic nerve root and postganglionic spinal nerve, to guide the surgical strategy and to formulate a provisional prognosis. Unfortunately, despite specificity, none of these classifications takes into account certain preoperative factors that can affect functional outcomes after treatment, and few classifications prioritize the management and prognosis of specific nerve injuries.

There are many works that have emphasized that loss of nerve substance, local ischemia, and extensive tissue damage [5, 38, 39], as well as other patient-related factors such as age [15] or smoking [40], could all be possible causes of impaired nerve regeneration. Other elements that have been investigated as influencing factors to the nerve repairing process include the mechanisms of injury [41,42,43], with evidence of better sensory and motor recovery for clean-cut injuries compared with crush and avulsion injuries. Additional factors, such as operative delay (despite the heterogeneity of results in literature [20, 32, 44, 45]), the level of injury, and operative timing, were analyzed showing better recovery for distal compared with proximal injuries [46,47,48], while there is no consensus in the literature on the role of operative delay in the outcome of PNIs [11, 38].

Given all these limits, a first attempt proposing a more complete classification was made by Goubier et al. [49]: they considered several preoperative factors, such as type of injury, delay of motor nerve repair, level of motor nerve injury, age of patient, perioperative smoking, and management in microsurgical unit, which have been demonstrated to have some impact on the outcome, to predict the final prognosis of peripheral nerve lesions. However, because of the absence of validation of this scale, it is of limited use in surgeons’ decision-making process for therapeutic strategies and surgical timing.

Our classification can improve the lack of information of the previously published attempts, as it can describe the type of lesion in a complete and exhaustive way, particularly concerning the localization and the extension of injuries and the surrounding tissues.

In fact, injuries to long nerves, such as median, radial, ulnar, and sciatic nerves, running through different limb segments, and also in the case of multiple levels of injury, can appropriately be described (our class is “lesion site”), and the type of injuries and the conditions of the surrounding tissues (our class defines whether close or open) are also clearly ascribable as well.

The following practical examples can show how the classification can be applied extensively and give complete information.

The alphanumeric code can in fact be applied to multiple-level lesions by sequentially describing the different levels of the lesion after repeating the code for the nerve. For example, R42CC/A22 R23CC/A22 is a code representing a lesion of the radial nerve in an elderly subject on two different levels in the limb. In this way, the classification applies to many complex traumas with lesions of the same nerve on several levels.

As another example, R23C3B2Y is the alphanumeric prognostic code associated with an injury of the radial nerve (R) at the arm level (2), where the nerve is mixed (3), the injury is now presenting as closed (C) with loss of substance (3), in a young patient (B), with delayed treatment (2), treated with a graft > 6 cm (2), and the patient presents with comorbidities (Y). As in this case, even limited information about the type of surgical reconstruction and anamnestic data could be enough to help nerve surgeons and clinicians in predicting a prognosis for a particular PNI case.

Regarding inveterate lesions, applying this code, for example, a complete inveterate sciatic nerve lesion, localized at proximal third of the thigh, with a 8 cm gap, is associated with the alphanumeric code Sc23CC3 (Sc sciatic nerve, 2 thigh, 3 lesion of a mixed nerve, C surrounding tissue closed, C3 complete with loss of tissue). The clear definition of the lesion leads to adopting the treatment of neurolysis and reconstruction with nerve (sural) grafts.

For inveterate PNIs, a unique clinical assessment might be helpful in a multidisciplinary environment [50]. Physiatrists could draft a rehabilitation program according to the “code” of nerve injury, finalize it after clinical evaluation, and train physiotherapists accordingly. Electrophysiologists could set up their equipment and detect a nerve lesion more critically, if the clinical question is well described: a clear dialogue between specialists leads to more efficient management of patients over time.

For acute trauma cases with PNI, remote consultation is rarely applicable, and time is life changing. Early ultrasound imaging and electrophysiologic tests are operator-dependent, whereas MRI may suffer from some limitations for acute nerve injuries [51]. Clinical evaluation in these cases is one of the most important elements to obtain a correct diagnosis and proper clinical management. An “identity (ID) code” for nerve lesions provides clear, immediate, and unambiguous information about a specific clinical condition and the related preferable surgical treatment, when applicable. If shared, this classification might easily permit smart dialogue between hospitals, especially when patients need to be treated in a center that is different from the one where a PNI diagnosis was originally made.

Outcome evaluation always depends on correct categorization of nerve lesions [52]. The more a univocal classification is shared among healthcare centers, the lower the data bias registered during follow-up evaluations. Results in different patients—and also in a single patient over time—would be directly comparable, leading to more accurate clinical management of patients and a higher standard of care. In clinical practice, nerve injuries are treated in specialized centers, which are often different from the center where a PNI diagnosis is made. An all-embracing classification for these particular traumatic lesions allows clinicians to speak the same language between different healthcare centers. Even a non-experienced physician could classify PNI cases correctly to provide complete and clear information to the reference center.

Our classification might indirectly suggest the first surgical approach, by giving physicians unambiguous information about nerve lesions, nature of injury, and connective tissue involvement. The PNI code leads to more precise, careful and realistic surgical planning, which translates into more complete, straight-forward information for patients and families. With the proposal of an alphanumeric classification, we attempted to create an all-encompassing classification of PNI that can help physicians and healthcare workers to efficiently manage these injuries in daily practice. Even the most complex, multilevel injuries, such as those involving all the plexuses and the facial nerve, can be described by this method. The anatomical site of injury in each nerve course is also included in the classification, which helps in single nerve exploration. Indeed, this advantage should also be kept in mind as we consider that single classifications of nerve injury have been proposed for the radial nerve [53], and for facial nerve injury [54] based on their anatomy. Those classifications focus on the single nerve injury and fit the lesion well, but our classification also may include this “single” classification. Regarding lesion type, with the help of this classification, we can predict outcome and adapt to evolving clinical/instrumental findings, which are crucial in neuropraxia. Here the condition may change over time, and a clinical and instrumental reappraisal is needed.

Moreover, it is often very difficult to code, even via a detailed physical examination, NCV/EMG, and image study. All surgeons of the peripheral nerves know that a definite diagnosis needs intraoperative exploration. We believe that in these special cases, the classification and the coding variations may be useful, also as a retrospective tool documenting any changes in a complete way. The final purpose is to spread knowledge and awareness for nerve lesion cases among healthcare personnel and to encourage communication and data exchange between different medical centers, to guarantee the best possible treatment and care for patients with PNI.

Conclusions

Like so many other pathological entities, PNIs need a correct clinical assessment and categorization to supply all the necessary information for the precise understanding and management of these patients. Our PNI-code represents a clear, all-embracing classification, able to ensure that physicians are speaking the same language when approaching an acute or chronic nerve injury. We believe that sharing basic information can assure good clinical practice, even for complex cases or with non-experienced workers, maintaining the same high standards of care between different medical centers.

Our practical attempt to use this classification for 24 patients showed its feasibility in clinical practice. We strongly believe that a clinical and anamnestic PNI-code will not only be useful in defining single cases or case series, but could also represent a valid tool to suggest indications and define prognosis. The real situation might be further described by finding out whether single PNI-code subcategorizations correlate with different clinical outcomes. A good lesion definition may guarantee correct data analysis, thereby reducing bias.

The weakness of our study lies in the retrospective nature of our data, together with the limited and heterogeneous number of cases taken into consideration, which does not allow us to relate the application of our classification to the definition of treatments and prognosis in this study.

The correlation between the classification of PNIs and prognosis, therefore, goes beyond the scope of this study, which is limited to the definition of a descriptive classification that is made to be as complete as possible, and also includes those in the literature that are recognized as the main prognostic factors.

However, the utility of our alphanumeric classification might grow enormously with its spread and acceptance in the scientific community, especially in the matter of revision and second surgeries, where its utility can be easily appreciated even with a small number of cases.

On the other hand, coding using multiple alphabets and numbers is difficult to memorize as it needs to cover the entire clinical scenario in its current format, but we are dealing with a complex matter, and complexity cannot be reduced if one must have complete information. Emergency department and all physicians involved in the treatment of peripheral nerve injuries may use our classification chart, just as burn centers use the Lund and Browder chart.

Thus, new, larger studies are needed to improve the outcome prediction reliability and accuracy of this classification in management and treatment recommendations.