Secondary Logo

Journal Logo


Early postoperative neuropathic pain assessed by the DN4 score predicts an increased risk of persistent postsurgical neuropathic pain

Beloeil, Helene; Sion, Barthelemy; Rousseau, Chloe; Albaladejo, Pierre; Raux, Mathieu; Aubrun, Frederic; Martinez, Valeria for the SFAR research network

Author Information
European Journal of Anaesthesiology: October 2017 - Volume 34 - Issue 10 - p 652-657
doi: 10.1097/EJA.0000000000000634
  • Free


This article is accompanied by the following Invited Commentary:

Steyaert A, Lavand’homme P. Acute and chronic neuropathic pain after surgery. Still a lot to learn. Eur J Anaesthesiol 2017; 34:650–651.


Chronic postsurgical pain (CPSP) is traditionally defined as pain persisting longer than 2 to 3 months after surgery.1,2 The incidence of CPSP varies from 10 to 50% in the literature.3 Previously identified risk factors include genetic factors, preoperative pain, type of surgery, postoperative pain and hyperalgesia.4 A high proportion of CPSP is neuropathic in nature, [chronic postsurgical neuropathic pain (CPSNP)]5 which has been described in detail in a meta-analysis.6 Postoperative pain is traditionally classified as nociceptive pain and the more intense this pain on a numeric pain scale (NRS), the higher is the risk that pain will become chronic and that severe pain will last longer.7,8 However, acute neuropathic pain (ANP) is present in the postoperative setting when researchers specifically look for it.9,10 Searle et al.10 reported that 8% of patients after thoracotomy had acute postsurgical neuropathic pain (APSNP), and 22% had CPSNP. Data on the prevalence of ANP immediately after surgery are scarce, and no screening tool has been validated in this setting. Searle et al.10 used The Leeds assessment of neuropathic symptoms and signs score after thoracotomy whereas Martinez et al.11 used the DN4 questionnaire after iliac crest bone harvest. A Delphi survey among physicians showed that ANP is diagnosed differently than chronic neuropathic pain and encompasses items such as response to medications.12 The first objective of our multi-centre observational study was therefore to prospectively describe the incidence of APSNP in a large population using the DN4 questionnaire. The previous studies10,11 found a link between the presence of APSNP and CPSNP in specific postoperative settings, but our study aimed to further evaluate such a link in a large population undergoing a wide variety of surgical procedures.


This multi-centre prospective observational study was approved by the Institutional Ethics Committee of the Rennes University Hospital (number 14.02, 02/05/2014) ( identifier: NCT02826317). Investigators and centres were recruited through the SFAR (French Society of Anaesthesiology and Intensive Care) research network. Enrolment for the study took place during the same two consecutive days in all of the centres in France (16 June 2015). Each centre agreed to include as many patients as they could during this period. All patients gave oral informed consent. All adult patients who underwent inpatient scheduled or emergency surgery were eligible. The exclusion criteria were a patient decision to withdraw from the study, inability to answer questions and outpatient surgery. The study was solely observational and did not interfere with patient care. Preoperative assessment included demographic data (age, sex, BMI and type of surgery); assessment of preoperative chronic pain and its localisation; and preoperative opioid, gabapentin or pregabalin consumption. Postoperatively, we collected the following data: type of anaesthesia, intraoperative ketamine administration and type and total dose of opioid administered during surgery. Pain was evaluated preoperatively and at least 2 h after surgery on the same day (D0), on day two (D2) and at 1 and 2 months after surgery (M1 and M2). Pain was assessed using a 10-point NRS (numeric rate scale). If NRS more than 0, neuropathic pain was diagnosed on the basis of the DN4 questionnaire. Acute postsurgical pain was defined by D0 or D2 NRS more than 0. APSNP was defined by NRS more than 0 and D0 and/or D2 DN4 ≥ 4/10).13 Persistent postsurgical pain (PPSP) was defined by M2 NRS more than 0. Persistent postsurgical neuropathic pain (PPSNP) was defined by NRS more than 0 and M2 DN2 at least 3/7. The DN2 or seven-item DN4 questionnaire correspond to the first two questions of the DN4 questionnaire. The DN4 (clinician-administered) and DN2 (self-reported) questionnaires have an excellent consistency.14 The DN2 questionnaire sensitivity is 82%, and the specificity is 86% for a cut-off score of 3/7. The preoperative and D0 and D2 data were collected by one investigator in each centre. M1 and M2 assessments were via telephone contact by research nurses. If the patient was not reachable, three further attempts were made over 2 days.


The primary outcome was the rate of APSNP (DN4 ≥ 4/10) after surgery. Secondary outcomes were the rate of PPSNP (DN2 ≥ 3/7) at 2 months after surgery and identification of a possible link between APSNP and PPSNP. For continuous variables, mean ± SD is reported unless noted otherwise. For categorical variables, the number of patients in each category and the corresponding percentage [with numerator (number of patients with the studied characteristic) and denominator (total number minus missing data)] are given. NRS are presented as median and interquartile range. Each risk was expressed as a percentage with its 95% confidence interval (CI). The characteristics of patients were compared using χ2 or Fisher's exact test when appropriate for categorical variables. The probability of association was calculated using the Spearman's correlation with numerical data. Univariate analysis of possible risk factors of PPSNP at 2 months after surgery was performed with subsequent multi-variate logistic regression analysis using variables significantly associated with PPSNP in the univariate analysis. A P value less than 0.05 was considered significant. Statistical analysis was performed with SAS software, Version 9.4 (SAS Institute Inc., Cary, North Carolina, USA).


Sample characteristics

Twenty-seven centres enrolled patients during the two consecutive days of the study in June 2015. As shown in Fig. 1, 608 patients corresponding to 593 available data sets were included. Of these, 231 (38.6%) and 260 (43.8%) did not answer the telephone at M1 and M2, respectively. At M1, the characteristics of answering and non-answering patients were similar, in terms of male sex (50.3 vs. 51.3%), mean age [57 (±18) vs. 55 (±18) years], American society of anesthesiologists (ASA) I/II (83.3 vs. 79.5%) and scheduled surgery (93.3 vs. 90.1%), respectively. These proportions were similar at M2: male sex (50.3 vs. 51.9%), mean age [58 (±18) vs. 55 (±18) years], ASA I/II (82 vs. 81.6%) and scheduled surgery (94.8 vs. 88.4%).

Fig. 1
Fig. 1:
Flow chart.

A majority of the procedures were scheduled in ASA I or II patients. Orthopaedic procedures were the most common with the majority under general anaesthesia and 51.8% receiving ketamine during surgery (Table 1). Patients who received ketamine were not different from the population of the study [52.4% of men with a mean age of 55 (±17) years, ASA I/II (83.1%) undergoing scheduled procedures (94.1%) in orthopaedic (37.5%), abdominal surgery (16.4%) and urology (10.9%)].

Table 1
Table 1:
Patient characteristics

Pain characteristics

A substantial number of patients were in pain before surgery (n = 260/589, 44.1%) with 41.2 and 34.1% describing pain intensity higher than 0 or 3 on an NRS, respectively and 8.2% describing neuropathic pain. The pain was located at the site of the procedure for 72.1% (194/269) of the patients. Three per cent of the patients had gabapentinoids before surgery for chronic pain whereas 19.5% received gabapentinoids as premedication. The majority of the patients describing neuropathic pain before surgery were women (64.6%) with a mean age of 55 (±18) years, ASA I/II (87.2%) undergoing scheduled procedures (93.8%) in orthopaedic (39.6%), neurosurgery (16.7%) and ear nose throat surgery (12.5%).

After surgery, 72.2% on D0 and 71.3% on D2 experienced acute pain. Pain was higher than 3 on an NRS for 52.8% of the patients on D0 and 48.1% on D2. It was neuropathic in nature on D0 in 5.6% (95% CI, 3.6 to 8.3) of patients and in 12.9% (95% CI, 9.7 to 16.7) on D2. Patients with APSNP had a NRS of 4 (3 to 5.5) on D0 and 4 (2 to 5.5) on D2. When NRS was more than 0/10, the relative risk of having a DN4 at least 4/10 on D0 or D2 was 1.36 (95% CI, 1.06 to 1.76). It was 1.46 (95% CI, 1.09 to 1.96) when NRS was at least 3/10. One month after surgery, 51% of the patients described pain (NRS > 0). Pain was higher than 3 on a NRS in 28.6%, and 34.8% experienced a neuropathic component. Patients with a neuropathic component had a NRS of 3 (2 to 4) at M1. Two months after surgery, 39.2% described PPSP with 21.8% describing pain higher than 3 on a NRS. PPSNP was present in 33.3% of the patients. Patients with PPSNP had a NRS of 5 (3 to 6.5) at M2. Among the patients without APSNP on D0 and/or D2, 11.7% reported PPSNP at M2.

Link between acute postsurgical neuropathic pain and persistent postsurgical neuropathic pain

There was a weak positive correlation between the maximum DN4 score on D0 and/or D2 and the DN2 score after 1 month (Pearson's r = +0.2308, P = 0.0004) and after 2 months (Pearson's r = +0.2683, P = 0.0001). Patients with APSNP were more likely to have PPSNP 2 months later [RR = 2.08 (95% CI, 1.05 to 4.14)] (Table 2).

Table 2
Table 2:
Comparison of DN2 scores at M1 and M2 in patients with or without acute postsurgical neuropathic pain

Persistent postsurgical neuropathic pain risk factors

Ten variables were included in the univariate analysis (Table 3). Among them, preoperative DN4 at least 4/10, D0 and/or D2 DN4 at least 4/10, D2 DN4 at least 4/10 and preoperative pain were associated with PPSNP at M2. The type of anaesthesia and premedication with gabapentinoids had no influence on persistent neuropathic pain. Multivariate analysis showed that D0 and/or D2 DN4 at least 4/10 was a significant risk factor for PPSNP [odds ratios 4.22 (95% CI, 2.19 to 8.12)] (Table 4).

Table 3
Table 3:
Univariate analysis of risk factors of persistent postsurgical neuropathic pain (DN2 ≥ 3 at M2)
Table 4
Table 4:
Step-by-step multivariate analysis of risk factors of persistent postsurgical neuropathic pain (DN2 ≥ 3 at M2)

Preventive therapy

Chronic gabapentinoid therapy, but not single preoperative administration, was protective and associated with a decreased risk of neuropathic pain 2 months after surgery (P = 0.03).


After a wide range of surgical procedures performed in 27 different centres, a majority of the patients experienced acute postoperative pain which was neuropathic in nature in up to one in 12 cases on D0 and up to one in six cases at D2. At 2 months, PPSP was present in 39.2% of the 333 patients assessed at M2, and one in three of these patients exhibited PPSNP. Moreover, our results showed that APSNP appeared to be a significant risk factor for PPSNP.

The population in our study is representative of the target population when compared with other studies in France15 or in Europe.16 Our results are comparable with the literature describing CPSP: recent numbers vary from 12% after a wide range of procedures16 to 15% after outpatient surgery17 or general surgery18 to 18% in the general Norwegian population.19 The definition of chronic or persistent pain varies in these studies from 3 to 22 months. The neuropathic component of this CPSP is known to contribute to severity and is often difficult to treat.3,8 Recent publications have highlighted this neuropathic component in the postoperative setting: an incidence of CPSNP of 22% 3 months after thoracotomy,10 3% after laparoscopic herniorrhaphy,15 37% 6 months after breast surgery15 and 23% 6 months after iliac crest bone harvest.11 The PAIN-OUT study recorded CPSNP in 35% of the patients with moderate CPSP and 57% in patients with severe CPSP.16

To our knowledge, this study is the first to report both acute and PPSNP in a variety of procedures. Screening for ANP immediately after surgery is not routine and is rarely described in the literature. The incidence of APSNP in our study was comparable with that previously reported10 and early identification of a neuropathic component may have the potential to improve patient care. Indeed, patients with APSNP on D0 and/or D2 had more risk of developing PPSNP at 2 months in our study. This has already been reported in very specific surgical procedures known to be associated with high risks of CPSP and CPSNP.10,11 We have further demonstrated that APSNP leads to PPSNP after a variety of procedures. Although we used the DN4 or DN2 questionnaires as opposed to the LANNSS tool, all these scores have been validated as efficient tools in detecting chronic neuropathic pain13,20 and in postsurgical chronic neuropathic pain.21 However, the use of these scores immediately after surgery is relatively new and so far has only been used for research purposes. They have not yet been formerly validated in the immediate postoperative setting.22 One could thereby hypothesise that patients diagnosed with APSNP using the DN4 questionnaire could be treated with a different postoperative analgesic strategy with, for example, anti-neuropathic treatment immediately after surgery.

As previously described, preoperative pain was a risk factor for developing PPSNP in our study.16,23 Moreover, the neuropathic characteristic of preoperative pain was a risk factor for PPSNP. Very few previous publications have studied this potential link. Martinez et al.24 showed that preoperative pain and opioid use were higher in patients who developed CPSNP than in those without CPSNP. This suggests that preoperative nervous system sensitisation could increase the risk of developing CPSP after surgery. Our results are in line with other studies that suggest a link between the presence of preoperative neuropathic pain and PPSNP after spinal surgery,25 knee arthroplasty26 and general surgery.15 There is a growing body of evidence supporting the role of genetic factors in the development of neuropathic pain.27,28 It also explains why not all patients with nerve injury develop neuropathic pain. In the absence of routine genotype testing, screening for preoperative pain may be beneficial.25

Taken together, our results support the systematic use of the DN4 questionnaire in perioperative patients who are in pain. However, more studies are needed to demonstrate the benefit of a different therapeutic strategy in preventing chronic pain. Indeed, the DN4 questionnaire is not easy to perform. Recommending its routine daily use needs to be based on more scientific proof in terms of benefit for the patient.

Our study did not show that gabapentinoids used as premedication could prevent PPSNP. This point is in line with the most recent reviews.29,30 However, chronic gabapentinoid therapy before surgery was associated with a decreased risk of neuropathic pain at 2 months after surgery. The contrast between the results obtained with chronic or punctual gabapentinoid therapy may suggest two approaches – first, a focus on prevention only in patients with pain vulnerability and second, the need to cover a large period of time before surgery with antihyperalgesic medication to prevent sensitisation. To our knowledge, our study brings forward the first evidence concerning this point.

Our study has some limitations. First, our initial aim was to collect approximately 1000 complete data sets. We did not reach our objective mostly because centres overestimated their capacity to include patients, and two centres failed to include any patients. Another limitation is the missing data between D2 and M1 and/or M2. Patients who could not be reached the first time were called three other times within 2 days. Despite this insistence, some were never reached. The DN4 or DN2 questionnaires have not been definitively validated in the perioperative setting and APSNP assessment might be altered by concomitant inflammatory pain and negative interaction of one pain with another (inhibitory controls). However, these questionnaires have been used in large surveys estimating the prevalence of neuropathic pain in patients suffering from chronic pain14 as well as PPSP.16 Finally, we only evaluated the patients beyond 2 months after surgery and had no further information on their quality of life or pain treatment.


Our results suggest that acute postoperative pain could be an early warning of chronic neuropathic pain. Screening for neuropathic pain both before and after surgery could enable development of tailored clinical pathways and help reduce the burden of PPSP.

Acknowledgements relating to this article

Assistance with the study: none.

Financial support and sponsorship: this study was supported by the APICIL foundation and SFAR (French Society of Anaesthesiology and Intensive Care) France.

Conflicts of interest: none.

Presentation: this study has been presented as an abstract at the SFAR annual meeting in September 2016 in Paris.


1. Macrae WA. Chronic postsurgical pain: 10 years on. Br J Anaesth 2008; 101:77–86.
2. Treede RD, Rief W, Barke A, et al. A classification of chronic pain for ICD-11. Pain 2015; 156:1003–1007.
3. Martinez V, Baudic S, Fletcher D. Chronic postsurgical pain. Ann Fr Anesth Reanim 2013; 32:422–435.
4. Perkins FM, Kehlet H. Chronic pain as an outcome of surgery. A review of predictive factors. Anesthesiology 2000; 93:1123–1133.
5. Eisenberg E. Postsurgical neuralgia. Pain 2004; 111:3–7.
6. Haroutiunian S, Nikolajsen L, Finnerup NB, et al. The neuropathic component in persistent postsurgical pain: a systematic literature review. Pain 2013; 154:95–102.
7. McDermott AM, Toelle TR, Rowbotham DJ, et al. The burden of neuropathic pain: results from a cross-sectional survey. Eur J Pain 2006; 10:127–135.
8. Attal N, Lanteri-Minet M, Laurent B, et al. The specific disease burden of neuropathic pain: results of a French nationwide survey. Pain 2011; 152:2836–2843.
9. Power I. Recent advances in postoperative pain therapy. Br J Anaesth 2005; 95:43–51.
10. Searle RD, Simpson MP, Simpson KH, et al. Can chronic neuropathic pain following thoracic surgery be predicted during the postoperative period? Interact Cardiovasc Thorac Surg 2009; 9:999–1002.
11. Martinez V, Ben Ammar S, Judet T, et al. Risk factors predictive of chronic postsurgical neuropathic pain: the value of the iliac crest bone harvest model. Pain 2012; 153:1478–1483.
12. Searle RD, Howell SJ, Bennett MI. Diagnosing postoperative neuropathic pain: a Delphi survey. Br J Anaesth 2012; 109:240–244.
13. Bouhassira D, Attal N, Alchaar H, et al. Comparison of pain syndromes associated with nervous or somatic lesions and development of a new neuropathic pain diagnostic questionnaire (DN4). Pain 2005; 114:29–36.
14. Bouhassira D, Lanteri-Minet M, Attal N, et al. Prevalence of chronic pain with neuropathic characteristics in the general population. Pain 2008; 136:380–387.
15. Duale C, Ouchchane L, Schoeffler P, et al. Neuropathic aspects of persistent postsurgical pain: a French multicenter survey with a 6-month prospective follow-up. J Pain 2014; 15:24e21-24 e20.
16. Fletcher D, Stamer UM, Pogatzki-Zahn E, et al. Chronic postsurgical pain in Europe: an observational study. Eur J Anaesthesiol 2015; 32:725–734.
17. Hoofwijk DM, Fiddelers AA, Peters ML, et al. Prevalence and predictive factors of chronic postsurgical pain and poor global recovery 1 year after outpatient surgery. Clin J Pain 2015; 31:1017–1025.
18. Simanski CJ, Althaus A, Hoederath S, et al. Incidence of chronic postsurgical pain (CPSP) after general surgery. Pain medicine 2014; 15:1222–1229.
19. Johansen A, Schirmer H, Stubhaug A, et al. Persistent postsurgical pain and experimental pain sensitivity in the Tromso study: comorbid pain matters. Pain 2014; 155:341–348.
20. Bennett M. The LANSS Pain Scale: the Leeds assessment of neuropathic symptoms and signs. Pain 2001; 92:147–157.
21. Abdallah FW, Morgan PJ, Cil T, et al. Comparing the DN4 tool with the IASP grading system for chronic neuropathic pain screening after breast tumor resection with and without paravertebral blocks: a prospective 6-month validation study. Pain 2015; 156:740–749.
22. Sadler A, Wilson J, Colvin L. Acute and chronic neuropathic pain in the hospital setting: use of screening tools. Clin J Pain 2013; 29:507–511.
23. VanDenKerkhof EG, Hopman WM, Goldstein DH, et al. Impact of perioperative pain intensity, pain qualities, and opioid use on chronic pain after surgery: a prospective cohort study. Reg Anesth Pain Med 2012; 37:19–27.
24. Martinez V, Uceyler N, Ben Ammar S, et al. Clinical, histological, and biochemical predictors of postsurgical neuropathic pain. Pain 2015; 156:2390–2398.
25. Shamji MF, Mohanty C, Massicotte EM, et al. The Association of Cervical Spine Alignment with Neurologic Recovery in a Prospective cohort of patients with surgical myelopathy: analysis of a series of 124 cases. World Neurosurg 2016; 86:112–119.
26. Thomazeau J, Rouquette A, Martinez V, et al. Acute pain Factors predictive of postoperative pain and opioid requirement in multimodal analgesia following knee replacement. Eur J Pain 2016; 20:822–832.
27. Costigan M, Belfer I, Griffin RS, et al. Multiple chronic pain states are associated with a common amino acid-changing allele in KCNS1. Brain 2010; 133:2519–2527.
28. Nissenbaum J. From mouse to humans: discovery of the CACNG2 pain susceptibility gene. Clin Genet 2012; 82:311–320.
29. Chaparro LE, Smith SA, Moore RA, et al. Pharmacotherapy for the prevention of chronic pain after surgery in adults. Cochrane Database Syst Rev 2013; 7:CD008307.
30. Mishriky BM, Waldron NH, Habib AS. Impact of pregabalin on acute and persistent postoperative pain: a systematic review and meta-analysis. Br J Anaesth 2015; 114:10–31.

* K. Asehnoune, C. Tardy, M. Leguen, J. Kavafyan, G. Daccache, M. Garot, N. Schinkel, S. Tissier, C. Deswarte, A. Luzi, V. Piriou, V. Joindreau, I. Negre, H. Montcharmont, F. Vial, C. Aveline, P. Guedj, J. Landrin, M. Latiere-Paulhe, M. Carles, Y. Maghous, E. Marret.

© 2017 European Society of Anaesthesiology