DN4, painDETECT, and S-LANSS Instruments Have Little Utility in Polyneuropathy

In the setting of neuropathic pain, the clinician-rated Douleur Neuropathique 4 (DN4), painDETECT, and Self-completed Leeds Assessment of Neuropathic Symptoms and Signs (S-LANSS) instruments were not effective for evaluating symptoms of polyneuropathy. These findings were published in Pain.

This study was conducted at 5 hospitals in Norway between 2017 and 2021. Patients (N=729) with bilateral distal lower extremity pain who were referred for a polyneuropathy assessment in an outpatient setting were evaluated for symptoms using the DN4, painDETECT, and S-LANSS. Instrument outcomes were compared with the reference standard Neuropathic Pain Special Interest Group (NeuPSIG) criteria, as well as findings from a clinical examination and nerve conduction assessment. The 3 pain instruments used in this study were translated into Norwegian using international guideline protocols.

The mean age of study participants was 55 (standard deviation [SD], 11) years, 56% were women, 46% used pain medications, and the mean numerical rating scale (NRS) pain score during the prior 3 months was 5.5 (SD, 2.1) points. The most common causes for polyneuropathy were diabetes (21%) and small fiber neuropathy (15%). Idiopathic polyneuropathy was reported in 21% of study participants.

According to NeuPSIG, 21% of study participants were graded as having unlikely, 15% possible, 14% probable, and 50% definite neuropathic pain. Compared with NeuPSIG, DN4 was in fair agreement (κ, 0.38; P <.001) and S-LANSS (κ, 0.13; P =.001) and painDETECT (κ, 0.12; P =.002) were in no to slight agreement.

Stratified by having neuropathic pain (probable or definite grading; n=465) and nonneuropathic pain (unlikely or possible grading; n=264), the DN4 had the best discrimination for diagnosing neuropathic pain with a cutoff of ≥4 (sensitivity, 0.87; specificity, 0.50; positive predictive value [PPV], 0.75; negative predictive value [NPV], 0.68), followed by S-LANSS with a cutoff of ≥12 (sensitivity, 0.60; specificity, 0.54; PPV, 0.71; NPV, 0.42) and painDETECT with a cutoff of ≥19 (sensitivity, 0.50; specificity, 0.64; PPV, 0.73; NPV, 0.39).

The area under the receiver operating characteristic curve (AUROC) scores were 0.773 for DN4, 0.612 for S-LANSS, and 0.586 for painDETECT. These values indicated that DN4 had acceptable discrimination, S-LANSS had poor discrimination, and painDETECT was nondiscriminative. None of the tools reached a positive likelihood ratio >2, which meant that a positive result would not provide clinicians with important diagnostic information.

In sensitivity analysis that excluded patients who were graded by the NeuPSIG as having probable neuropathic pain, the diagnostic accuracy of the instruments was not significantly affected.

This study may have been limited by its unblinded design.

Study authors concluded, “The discriminative ability of DN4 was acceptable, while poorer results were observed for painDETECT and S-LANSS. […] The probability of getting a correct test result was three-quarters at the very best, and only two fifths at worst. Hence, neither tool is appropriate when trying to distinguish between neuropathic and nonneuropathic pain in patients referred to polyneuropathy assessment at neurological outpatient clinics.”