Peripheral Neuropathy After Viral Eradication With Direct-acting Antivirals in Chronic HCV Hepatitis

A Prospective Study

Maria M. Zanone; Claudia Marinucci; Alessia Ciancio; Dario Cocito; Federica Zardo; Emanuela Spagone; Bruno Ferrero; Cristina Cerruti; Lorena Charrier; Franco Cavallo; Giorgio M. Saracco; Massimo Porta


Liver International. 2021;44(11):2611-2621. 

In This Article


HCV-related Neuropathy

Ninety-four patients were recruited at baseline (T0) and their clinical characteristics are summarised in Table 1. Forty-seven patients (50%) were treated with Sofosbuvir/Velpatasvir (Epclusa®) for 12 weeks, 29 (31%) with Glecaprevir/Pibrentasvir (Maviret®) for 8 weeks, and the remaining ones (19%) with Elbasvir/Grazoprevir (Zepatier®) for 12 weeks, according to clinical and virological variable.[32]

At T0, nerve conduction studies disclosed moderate to severe, length-dependent, sensori-motor or predominantly sensory axonal neuropathy, prevalent symmetrical on the lower limbs, in 22 patients (23%), with no significant differences in age, duration of infection, creatinine, viral load, viral genotype, grade of liver fibrosis, or presence of concomitant pathologies compared to the patients with normal electrophysiological results (Figure 1). Thirty-two (34%) showed impaired deep tendon reflexes (Table 2), with higher prevalence in patients with altered electroneurography, compared to those with normal results (18/22, 82% vs 14/72, 19%, P < .01). Questionnaire scores are shown in Table 2. Thirty-seven patients (39%) complained of neuropathic pain at DN4, 15 (16%) had severe sensory symptoms and 24 (26%) severely impaired muscle strength.

Figure 1.

Electrophysiological data at T0. Mean±SD values of CAMP amplitude, MCV and DL for motor nerves (A), and of SNAP amplitude, SCV and DL for sensory nerves (B) in 72 patients with normal parameters (white) and 19 patients with neuropathy (black) detected at T0. *P < .05 vs patients with normal parameters. § P = .051 vs patients with normal parameters

There was a trend towards higher prevalence of altered nerve conduction studies, worse amplitude parameters (significantly different for median nerve CMAP amplitude, 6.7 ± 2.4 vs 8.4 ± 1.9 mV, P = .001, and peroneal nerve CMAP amplitude, 4.8 ± 2.2 vs 6.1 ± 2.1 mV, P = .012) and worse neurological symptoms in CG+compared to CG- patients, with only SSS reaching a statistically significant difference (P = .031) (Table 3).

Considering the prognostic index, patients with Child Pugh B showed significantly worse MRC (53 ± 25.3 vs 73.9 ± 7.5, P = .008), DN4 (5.4 ± 1.8 vs 3 ± 2.1, P = .0408), NPSI (26.5 ± 19.3 vs 11.4 ± 13.2, P = .033), SSS (11 ± 5.5 vs 4.1 ± 4.1, P = .014), and INCAT (4.3 ± 3.3 vs 1.2 ± 1.3, P = .009) scores compared to grade A, while lower Euro-QoL approached significance (0.003 ± 1.1 vs 0.71 ± 0.3, P = .06).

Re-assessment After DAA Treatment

At the end of treatment, all patients but one showed undetectable levels of HCV-RNA. Ninety-one patients completed the 10 months follow-up after the end of treatment. At T1, 3 of the patients with abnormal electroneurography dropped out (1 sudden death, 2 developed HCC), while 4 of the 19 patients with sensory-motor neuropathy completing the study (4/19, 21%) showed a normal electroneurography (Figure S1), and 2 had a marked improvement of nerve amplitude parameters (1 patient with absent peroneal nerve CAMP recovered motor response, although still abnormal, and 1 patient approached normal parameters for sural SNAP).

In general, there was a trend towards an increase of motor and sensory amplitude parameters in the whole subgroup, reaching or approaching statistical significance for median, peroneal nerve CMAP and sural nerve SNAP amplitudes, mainly for CG- patients (Table 2 and Table 4). Deep tendon reflexes remained impaired in only 11 patients (11/91, 12%) (P < .000 compared to T0), and 10 patients (10/91, 11%) still complained neuropathic pain at DN4 (P < .000 compared to T0). There was a statistically significant improvement for all neurological scores, indicating significant improvements for pain, limb disability, muscle strength, and global impression of quality of life (P < .000 compared to T0) (Table 2).

When considering the presence of cryoglobulinemia, a significant improvement was detected in both groups (CG+and CG-) from T0 to T1 for all neurological scores. The improvement in the CG+patient subgroup was also significantly higher than the one in the CG- subgroup for neuropathic pain at DN4 (Table 5).