Chronic neuropathic pain after traumatic peripheral nerve injuries in the upper extremity: prevalence, demographic and surgical determinants, impact on health and on pain medication

2.1 Patients

Subjects included all patients admitted to the Hand Surgery Clinic, UUH with a diagnosis of traumatic peripheral nerve injuries between January 1, 2010, and June 1, 2018. Eligibility criteria included age more than 18 years, ability to speak Swedish. Individuals who had surgery earlier than 6 months before June 1, 2018 were not included in study.

2.2 Study design

S-LANSS questionnaire (Self report-Leeds Assessment of Neuropathic Symptoms and Signs), RAND-36 (Item short form health survey), QuickDASH (Disability of Shoulder, Arm and Hand) and additional questionnaires concerned medication and quality of life (Appendix 1) were collected from 1,051 patients with nerve injuries. A standardized data collection template was employed for all the patients. The template included demographic data (age, gender), pain diagnosis, type and level of injured nerve, date of the lesion and surgery (nerve suture, exploration, amputation, reconstruction, suture and amputation) or conservative treatment, reoperation, time since the procedure. The types of traumatic injury included median, ulnar, radial and digital nerve injuries in the forearm and hand. Regardless of mechanism, the resulting lesions were uniformly managed with identical surgical techniques. For the more proximal nerves, nerve suture was always performed in a standard way by tensionless epineural suture [13]. Autologous nerve grafting was performed in the presence of a nerve gap where end-to-end suture was not possible. In complex wound injuries during the initial exploration and debridement, the nerve ends were tagged for later repair and definitive reconstruction was then performed after 2–3 weeks [14]. With massive tissue destruction, amputation was performed.

An 11-point numeric pain rating scale (NRS) was used. The patients were divided into those experiencing pain and those without pain.

4.1 Estimated prevalence of persistent pain of neuropathic origin

Of the 669 patients who underwent surgical treatment, 337 (50.3%) subsequently experienced chronic pain and 246 of those (73%) had a S-LANSS score ≥12 indicating a neuropathic pain component.

4.2 Age and gender

The mean age of the patients was 56±2 years and a majority were men (74%). No gender differences were seen between the two groups with pain and without pain (p=0.9). A significant age difference (p<0.001) was seen between groups with younger patients in the group with persistent pain (mean age 46) in comparison with the group without pain (mean age 55).

4.3 Time from operation

The prevalence of chronic pain generally decreased with time after surgery (p<0.001). No differences were seen between the group with pain and the group without pain regarding time from injury to operation (p=0.119).

4.4 The site of injury

The patients with traumatic digital nerve injuries were predominant in the both groups (67.4% and 75%, respectively). The injuries in major nerves were associated with more pain (p=0.019) (Table 1). Complex injuries involving more than one nerve, vascular injury and fractures were found in 10.9% of cases (37/337 patients) with pain and 3.9% of cases (13/332 patients) without pain. Combined nerve injuries most commonly involved the median and ulnar nerves.

4.5 Surgical procedure (Table 2)

The patients with pain and without pain were compared according to different surgical approaches (Table 2). Nerve suture (neurorrhaphy) was the most common procedure (79.3%). In the group with pain, median nerve neurorrhaphy was performed in 50% (n=39) of the patients, radial nerve neurorrhaphy in 18% (n=15) and ulnar nerve neurorrhaphy in 31% (n=24). In the group without pain, nerve suture of the median nerve was performed in 39% (n=42) of the patients, radial nerve suture in 28% (n=30) and ulnar nerve suture in 31% (n=36). Combined injuries which required neurorrhaphy of more than one nerve were 8% in the group with pain (median and ulnar in most of the cases) and 2% in the patients without pain. We found that the amputated patients with pain and without pain were older (mean=70 years) than in the other procedures (p<0.01). Risk of chronic pain was related to the type of injured nerves. Complex injuries affecting the major nerves (radial, ulnar, median nerves) were associated with higher risk of chronic pain (n<0.003).

4.6 Patients on pain medication and the medications effects on pain (Table 1)

The patients who had tried pharmacological treatment responded poorly to medication. Of the patients with pain resulting from traumatic or surgical nerve injury, 88.1% had no pain medication despite the presence of pain greater than NRS 6 in more than 40% of the patients.

4.7 Pain affecting daily life (Table 1)

Pain affected daily life in 44% of the patients with pain who underwent surgery. The majority of the operated patients were pain free before surgery (80%) and developed pain after surgery (84%). RAND-36, a generic instrument measuring health related quality of life was used to measure health status between patients with pain and without pain (p<0.0001). The negative correlation of RAND-36 scores with pain intensity (Pearson r −0.330) indicates that patients with a better wellbeing tended to have lower intensity of pain (p=0.0046) (Fig. 2).

Fig. 2:

RAND-36 scores were negative correlated with intensity of pain (p=0.0046) (Pearson r −0.330).

4.8 QuickDASH (Quick Disability of the Arm, Shoulder, and Hand Questionnaire) scores

The average QuickDASH survey scores were 33.46 for patients with pain and 7.5 for patients without pain (p<0.0001). QuickDASH score were positively (Pearson r=0.7) correlated with the intensity of pain (p=0.483) (Fig. 3) or more disability was associated with higher intensity of pain.

Fig. 3:

QuickDASH score were positively correlated with the intensity of pain (p=0.483).

4.9 Pain intensity

Pain intensities in the group with surgical procedures as assessed by the numerical rating scale were NRS worst 5.1±2.4 and NRS best 2.1±2. Forty percent of the patients with pain (141/342) reported severe pain intensity (NRS>6).

4.10 Odds ratio (OR) for chronic pain (Fig. 4)

Fig. 4:

Forest plot for the association between persistent pain after surgery and no pain. Odds ratio (OR) and 95% Confidence interval (95% CI))]. The association of age, gender, years from the operation, type of injured nerves (digital or major nerves of the upper extremity), time from injury to operation, number of reoperations, type of surgery studied with ordinal logistic regression analyses are represented.

Partial proportional odds models were used to investigate the predictors for chronic pain for the patients who were operated. Among the factors investigated, the strongest predictors for chronic pain development were age OR 0.78 (CI 0.71–0.86) (p<0.001), combined injuries involving radial, ulnar, median nerves OR 1.68 (CI 1.18–2.38) (p<0.004) and surgery [nerve suture OR 6.17 (CI 2.08–18.25) (p=0.001), surgical exploration OR 6.09 (CI 1.90–19.53) (p<0.002), amputation OR 9.60 (CI 2.34–39.34) (p<0.002), amputation and nerve suture OR 10.88 (CI 2.84–41.67) (p<0.001)]. Multivariate analysis indicated that injury of a major nerve OR 1.6 (CI 1.10–2.31) (p=0.013), time after surgery OR 0.91(CI 0.86–0.96) (p=0.01), younger age OR 0.7 (CI 0.72–0.88) (p<0.001), were the main factors for predicting pain after surgery. Other independent predictors for pain were gender, surgical technique, time from injury to operation, reoperation. The type of the nerve injured was the strongest predictor for chronic pain.

5.1 The prevalence of chronic neuropathic pain after traumatic nerve injuries in the upper extremities

According to the grading system for neuropathic pain, possible neuropathic pain can be assessed with a specific questionnaire screening tool [15]. Higher diagnostic accuracy is achieved with the LANSS (The Leeds Assessment of Neuropathic Symptoms and Signs) scale and the DN4 (Douleur Neuropathique en 4 Questions) questionnaires given that their scores also reflect physical tests [16], [17]. The S-LANSS is a modified version of LANSS for self-administrated pain questionnaires [18] and is more suitable for large-scale research [19]. The role of this questionnaire remains only as an indicative tool [20] because despite an anatomically verified nerve lesion, the results of this study indicate that not all of the patients developed neuropathic pain after a nerve injury as detected with S-LANSS. Persistent pain after traumatic nerve injuries could be predominantly due to neuropathic mechanisms related to a nerve lesion or could be a result of underlying musculoskeletal nociceptive mechanisms or both. The complex injuries in 10.9% of the patients were possibly due to both pain originating from nerve injuries and from musculoskeletal origins. A previous study [21] with neuropathic pain originating from neck and upper limb demonstrated the limited ability of LANSS to identify patients with clinically defined NP (LANSS sensitivity was substantially lower than other accuracy measures 22%, specificity 88%). A diagnosis of definite neuropathic pain is based not only on clinical examination including accurate sensory examination, but also on neurophysiological methods [15]. Chronic neuropathic pain has been associated with a poor outcome in patients after nerve injury in the upper extremity [22]. This study confirms that nerve injury alone is not always responsible for the progression of acute to chronic pain in the absence of other predictors such as psychosocial risk factors and genetic predisposition [23], [24].

5.2 Time frame

The time from injury and operation was a significant predictor against persistent pain (p<0.001). This is consistent with other findings where chronic pain was reduced over time [25], [26]. Phillips et al., found that the prevalence of neuropathic pain decreased from its highest (27%) at 6-weeks postoperatively to 7% after 3 years’ post-surgery [27].

5.3 Age and gender, pain affecting daily activities, disability and health

In previous studies, the following factors for recovery after nerve repair and reconstruction in the upper extremity were mentioned: age, delay before repair or reconstruction, the level of injury as well as type of the injury [3], [28]. We found that younger age was a predictor for an increased prevalence of chronic pain. These findings contradict the previous studies which reported that neuropathic pain increases with the age at which nerve damage occurs [29]. In these studies, younger patients supposedly have stronger regenerative capacity and achieve better sensory recovery than older patients and are less likely to develop long-term chronic neuropathic pain syndromes following nerve injury [30] probaly because neuropathic pain depends on the maturity of the nervous system [31], [32]. It is known that nerve regeneration is poorest in patients over 50 years of age [33]. However, our study demonstrates that patients under 50 years are at higher risk of developing chronic pain. It was demonstrated previously that nerve injuries in the upper extremity were more frequent in young males [4] in whom the quality of life was impaired probably as a consequence of work loss [34]. Regarding our secondary study questions, both disability and health indicators correlated with intensity of pain.

5.4 Surgical timing/delay

It has been proposed that early repair is preferable since a number of neurobiological factors influencing healing diminish over time [23]. Therefore, earlier repair is associated with better outcome [8]. Timing of nerve repair is thought to be essential in order to improve functional outcome and to reduce health economic costs [35], [36]. However, Wang et al. demonstrated that an average of 3.5 days vs. 24 h delay in surgical repair of a transected nerve did not affect outcome as expressed in motor and sensory recovery [37]. Accordingly, surgical timing/delay and reoperation were not among the risk factors for chronic pain in this study.

5.5 Type of injured nerve

Nerve injuries most commonly affected the most distal nerves in the upper extremity such as digital nerves, followed by median, ulnar and radial nerves in the forearm. Excluding digital nerves, we found that the median nerve was more commonly injured. Other authors have found the same sites of injury [38] and others found that after digital nerves, ulnar nerve injury was more prominent [37]. A majority of patients with peripheral nerve injuries had surgical repair (95.4%). This observation has been seen previously [37], with 83% having surgical repair. The type of nerve injured was considered to be significant for outcome [39]. Poor outcome was seen after repair of a nerve at a proximal location while distal injuries such as digital nerve injury resulted in sensory loss in the fingertips but regenerated well and had better outcome [40]. The size of the injured nerves and chronic pain were subject to previous studies [8] which demonstrated that small nerves might be associated with a lower prevalence of persistent pain [41] in comparison with sectioning of a large nerve such as sciatic nerve or multiple intercostal nerves [42], [43]. In this study, a significant statistical difference (p<0.001) was seen between digital nerve injuries (47% chronic pain) and the patients with injuries of the major nerves (58% chronic pain). There are few controlled studies which compare the outcomes with different surgical techniques [44] but in the present study we did not identify statistically significant associations between the surgical technique and chronic pain.

5.6 Pain medication

Despite the wide use of medication for pain in patients with chronic pain generally and a dramatic escalation in the use of prescription opioids for the treatment of chronic non-cancer pain, [45] in this study only a few patients were prescribed pain medication (10.4%) and responded poorly to medication. In an observational study which included patients with traumatic neuropathic pain the authors concluded that we still do not have very successful medications for the treatment of neuropathic pain [46].