Background and purpose Assessment of the severity of chronic peripheral neuropathy during oxaliplatin treatment is based on symptoms. Efforts to adjust the total dose of oxaliplatin to prevent severe neuropathy can be complicated by the worsening of neuropathy symptoms following treatment. Objective measures of the structure and function of peripheral nerves during early phases of treatment may aid in determining the optimal oxaliplatin dose in individual patients. Intraepidermal nerve fibre density (IENFD) has been suggested as an early marker of peripheral neuropathy. Methods Sixty patients were examined before treatment and following 25% and 50% of the total planned oxaliplatin dose. Fifty‐five of them were also examined at completion of chemotherapy and 6 months later. IENFD in skin biopsies from the distal leg, nerve conduction studies and quantitative sensory testing at the dorsum of the foot were performed. Forty‐six healthy subjects were examined at baseline and after 6 and 52 weeks for comparison. Results Intraepidermal nerve fibre density was not reduced during treatment. Sural nerve amplitude and conduction velocity, vibration detection thresholds, mechanical detection threshold and cold detection threshold were significantly reduced during treatment. Compared to reference values and spontaneous changes in healthy subjects, the largest proportions of patients with deterioration were found for vibration detection thresholds followed by nerve conduction studies, mechanical detection threshold, cold detection threshold and IENFD. Conclusions Significant changes were most pronounced for measures of large nerve fibre function, especially vibration sensation. Skin biopsies do not seem to provide a clinically relevant objective measure of peripheral nerve deterioration during oxaliplatin treatment.

Background: Oxaliplatin-induced peripheral neuropathy negatively affects the quality of life for patients with gastrointestinal cancers and may cause neuropathic pain. Measures of peripheral nerve structure or function, such as intraepidermal nerve fiber density (IENFD) during treatment could reduce neuropathy severity through individualized dose reduction. Objective: The aim was to evaluate the predictive values of IENFD, quantitative sensory testing (QST), and nerve conduction studies (NCS) for significant neuropathy and neuropathic pain. Methods: Fifty-five patients were examined prospectively before, during, and six months following treatment using skin biopsies, QST and NCS. Clinically significant neuropathy six months after treatment was defined as reduced Total Neuropathy Score of more than five and neuropathic pain was assessed according to International Association for the Study of Pain criteria. Results: Thirty patients had a clinically significant neuropathy, and 14 had neuropathic pain. Vibration detection threshold (VDT) before treatment was correlated with clinically significant neuropathy six months after treatment (OR 0.54, p = 0.01) and reductions in cold detection threshold (CDT) after 25% of treatment (OR 1.38, p = 0.04) and heat pain threshold (HPT) after 50% of treatment (OR 1.91, p = 0.03) with neuropathic pain. Cut off values of 5 for baseline VDT and changes of more than –0.05 °C and –0.85 °C in CDT and HPT were estimated. Sensitivity and specificity was low to moderate. There was no correlation between changes in IENFD or NCS and significant neuropathy or neuropathic pain. Conclusions: Vibration detection thresholds and thermal detection thresholds may be useful for prediction of clinically significant and painful neuropathy, respectively. However, low to moderate sensitivity and specificity may limit the predictive value in clinical practice.

Background and aims: Polyneuropathy is a common neurological disorder with many potential causes. An essential part in screening, diagnosis, and follow-up evaluation of polyneuropathy is testing of the sensory function including vibratory sensation. The graduated Rydel-Seiffer tuning fork and the biothesiometer have been developed to quantify vibratory sensation through detection thresholds. The aim of this study is to compare the vibration detection thresholds determined by the two instruments regarding intraindividual temporal changes, interindividual variation in healthy subjects and comparison of the diagnostic value in patients with a clinical suspicion of polyneuropathy.Methods: Ninety-four healthy subjects, 98 patients with and 97 patients without a diagnosis of polyneuropathy were included. Quantitative sensory testing including biothesiometry, structured clinical examination, and nerve conduction studies were performed three times during 52 weeks in healthy subjects and once in patients.Results: There were no significant changes over time for neither the Rydel-Seiffer tuning fork nor the biothesiometer, and both had larger between-subject variation than within-subject variation. Relative intertrial variability was largest for the biothesiometer. Diagnostic value (sensitivity, specificity, positive predictive value, and negative predictive value) was moderate for both methods (Rydel-Seiffer tuning fork: 58%, 74%, 70%, 64%; biothesiometer: 47%, 77%, 68%, 59%). Interpretation:The Rydel-Seiffer tuning fork and the biothesiometer have a low testretest and time dependent variation. They perform almost equally as diagnostic tools in patients with suspected polyneuropathy with a tendency toward better performance of the tuning fork.