Herpes zoster ophthalmicus (HZO) is caused by reactivation of varicella zoster virus (VZV), causing pathognomonic vesicular eruption in the distribution of the ophthalmic division of the trigeminal nerve; it is often accompanied by severe neuritic pain and neuralgia.[1] VZV infection may affect the trigeminal nerve in 10% to 20% of cases, and the ophthalmic branch (V1) is the most frequently involved site for herpes zoster (HZ),[2-4] followed by thoracic dermatomes. Consequently, the ophthalmic branch is one of the most common sites of postherpetic neuralgia (PHN) in the elderly,[2, 5] and this condition can severely impact the quality of life (QoL) in these patients.[6-8] Acute herpetic neuralgia (AHN) is defined as intense pain in the affected nerve within 30 days of acute zoster infection,[6] while PHN is defined as pain that persists for 90 days or more after rash onset.[9] Acute ophthalmic herpetic neuralgia (AOHN), which is a severe complication that is difficult to manage,[3, 10, 11] is likely to be due to an immediate nociceptive response. Local inflammation and tissue damage stimulate the afferent and subcutaneous fibers of V1, which are manifested neurologically as pain.[12] Neuritic pain should be controlled in the acute phase, and antiviral therapy significantly reduces the severity of infection, duration of the eruptive phase, and intensity of acute pain.[13] However, this therapy does not completely alleviate AHN.[14, 15] Therefore, knowledge of the optimal time and treatment option to control severe neuritic pain and neuralgia in the acute phase of HZO may be of clinical significance, not only in terms of improved QoL but also in preventing the development of PHN.[12, 16]

Some strategies for controlling local neuritis and preventing nerve damage might have potential for the treatment of AOHN. Cobalamin (Cbl, vitamin B₁₂), which is a neurotrophic agent,[17, 18] has a specific affinity for neural tissues.[17, 19, 20] Vitamin B₁₂ is not directly active in humans, but is converted into activate forms such as methylcobalamin (MeB₁₂) or adenosylcobalamin. Compared with other analogs, MeB₁₂is the most effective in terms of uptake by the subcellular organelles of neurons. Therefore, MeB₁₂ may be a better treatment for neuronal disorders through effective systemic or local delivery.[21, 22] MeB₁₂ therapy may have beneficial pharmacologic effects on neurologic function in a variety of disorders, such as diabetic neuropathy, glossopharyngeal neuralgia, autistic disorder, and Alzheimer’s disease-related cognitive decline.[23-29] In China, vitamin B1 and MeB₁₂ are the most common treatments for herpes zoster.[30] Clinical experience has shown that local MeB₁₂ injection is an effective treatment modality for subacute herpetic neuralgia[31] and PHN.[32] MeB₁₂ might be potentially relevant for AOHN in the eruptive phase, and local administration may have more significant neurotrophic effects than systemic administration on affected fibers and epidermal nerve endings. Topical lidocaine (5%) medicated plasters are recommended as a first-line therapy for PHN.[33, 34] On the basis of multiple randomized clinical trials that demonstrated the efficacy of lidocaine patches in patients with PHN, we hypothesized that lidocaine could be efficacious in relieving pain associated with AOHN.

A single-center, randomized, observer-blind clinical trial was conducted in patients with AOHN to explore the efficacy of local MeB₁₂administered in combination with lidocaine. The primary objective of the trial was to explore whether local MeB₁₂ in combination with lidocaine injection administered in the early stages of the onset of HZO could reduce the intensity and duration of AOHN and whether it had the potential to prevent or reduce the incidence and/or severity of PHN. The secondary objective was to evaluate the time-dependent effect on the pain relief and to ascertain the optimal therapeutic window to achieve alteration in disease progression toward recovery.

Although the mechanism of VZV reactivation in the neurons has not been elucidated fully, the intense pain associated with HZO is considered to be related to the neuroinflammation spread from the ganglion of Gasser to the ophthalmic branch of the trigeminal nerve.[40, 41] Cbl may exert antinociceptive and anti-inflammatory effects against acute and chronic pain.[42, 43] In the present study, analysis of the longitudinal data in mixed models indicated significant differences in the changes in the mean pain score between the two different methods of treatment. In the A0 and B0 groups treated with intramuscular MeB₁₂ in addition to local lidocaine injection, a significant response was observed up to 14 days relative to the baseline. The rate of change in pain score for the control group decreased to 0.49 and 0.47 time-point after the baseline for A0 and B0, respectively (P < 0.0001). However, only one of the 24 subjects (4.2% for A0) or four of the 25 subjects (16.0% for B0) who had received antiviral therapy had pain scores ≤ 3 at the end of the 14-day treatment. This is similar to the known natural history of pain decline in HZ, in which pain is resolved in approximately 16% of subjects within 15 to 30 days following rash onset.[15] Consequently, it is difficult to distinguish between the significant benefits of the intramuscular MeB₁₂ in addition to local lidocaine injection and spontaneous recovery. The present study showed that systemic MeB₁₂ administration did no significantly improved clinical outcomes during the first week and subsequent treatment did not show superiority, while local pain point lidocaine injection produced a transient (30 to 40 minutes) reduction in acute pain in the lesional area.

There was a significant difference in the trend of the general change in the mean pain scores between the groups in the same period since rash onset (P < 0.001) over time and showed an additional decrease of 0.39 (A1) and 0.27 (B1) per time-point for the local MeB₁₂combination with lidocaine injection (Tables 3 and 4). Compared with the control group, the mean pain score in the treatment group decreased significantly over time. A1 and B1 had mean pain scores of 2.6 ± 0.7 and 1.2 ± 0.8, respectively; only two of the 25 subjects in A1 and one of the 24 in B1 group had pain scores ˃ 3 but ˂ 4 after the 14-day treatment, and the numbers needed to treat (NNT) for the significant treatment effects of local MeB₁₂ injection on pain ≤ 3 at the 14-day endpoint were 1.14 (95% CI: 0.99 to 1.34) for the A groups (1 to 3 days) and 1.25 (95% CI: 1.03 to 1.58) for the B groups (4 to 7 days), respectively. In the treatment group, pain decreased substantially in the first week of treatment relative to the baseline, and the efficacy was continued over 12 months.

The strategy adopted in the present study was based on previously reported evidence. On the one hand, although Cbl deficiency is fairly common, particularly in elderly adults,[44-46] no single parameter can be used to diagnose this condition.[47] Some Cbl-deficient patients may have serious neurological dysfunction, even when they have no recognizable hematological abnormalities.[48, 49] Cbl has been applied mainly in the treatment of peripheral neuropathy for many years,[23, 50, 51] although the effectiveness of oral Cbl in reversing neurological abnormalities has yet to be established.[52] The traditional cornerstone of vitamin B₁₂ therapy has been oral medication; some studies have suggested that oral Cbl administration is better than parenteral supplementation.[53] However, due to limited uptake capacity, gastrointestinal degradation, or high background uptake by healthy tissues,[54] variability in dose responses to oral Cbl among individuals, especially across aging populations, may be attributed to slowed uptake.[55] A number of studies have provided strong evidence that vitamin B₁₂ has analgesic properties[56-60] in experimental animals. These studies have indicated a novel effect of the B vitamins on neuropathic pain and suggested the possible clinical utility of B vitamins for treating similar neuropathy at higher doses.[59-61] Clinical experience has shown that Cbl therapy improves the symptoms of both peripheral neuropathy and autonomic dysfunction,[62] while intrathecal high-dose MeB₁₂ can relieve symptoms of neuropathy.[63] The British Columbia Medical Association and Ministry of Health of Canada recommend oral replacement of B₁₂(1,000 to 2,000 μg/days) for most cases of vitamin B₁₂ deficiency, including pernicious anemia. For patients with neurologic symptoms, they recommend an initial B₁₂ injection (1,000 μg) followed by oral replacement.[64] As delayed therapy can lead to irreversible neurological dysfunction, parenteral Cbl therapy should be strongly considered.[52, 65] In clinical treatment of subacute herpetic neuralgia and PHN, the local high doses of MeB₁₂ were injected into the skin with damaged fibers and epidermal nerve endings.[31, 32] Such an injection had a significant analgesic effect and was not associated with toxicity in general. Therefore, MeB₁₂ can be recommended at doses much higher than the physiological level to achieve the best treatment effect.[21] The HZO usually invades one or more branches of the ophthalmic division of the trigeminal nerve, namely the supraorbital, lacrimal, and nasociliary branches,[66] resulting in zoster-associated pain often along the involved dermatome. Therefore, treatment strategies should be focused particularly on the local subcutaneous nerve fibers of the most painful area. The results of the current study demonstrate that MeB₁₂, a neurotrophic agent, had a significant and sustained analgesic effect on AOHN. These data indicated that local treatment allowed direct delivery of MeB₁₂ to topical HN-damaged subcutaneous and neuronal tissue to obtain rapid and effective neurological responses. Specifically, this approach was shown to have a more significant analgesic effect than systemic administration on the nerve fibers and endings affected by AOHN. These results are consistent with the conclusions of Dongre,[67] who reported that the overall reduction in the mean pain score of diabetic peripheral neuropathy patients treated over 14 days by pregabalin with MeB₁₂ was 72.3%. Local lidocaine could be used in combination with MeB₁₂ and could relieve fear and pain during injection into the irritated skin.

The eyelids are commonly involved in HZO[66] and may develop edema and inflammation in the surrounding tissues. This study also evaluated the rash, swelling healing time, and pain interference in terms of QoL. The times for swelling reduction and rash healing were significantly shortened in both A1 (286.4 ± 45.3; 385.7 ± 68.9) and B1 (162.0 ± 32.9; 239.3 ± 39.9) compared with those in A0 (440.0 ± 67.4; 540.0 ± 98.5) and B0 (338.9 ± 40.6; 419.0 ± 40.2), which implied that MeB₁₂ combined with lidocaine exerted anti-inflammatory and detumescence effects on the affected area. With regard to the disease course factor, there were no significant differences in the time for rash healing and swelling reduction between A0 and B0. The subjects in A1 (78.6 ± 7.5) and B1 (91.5 ± 9.4) experienced greater improvement in their QoL and better state of health than those in A0 (31.3 ± 19.6) and B0 (48.0 ± 22.3) at the end of treatment.

Although early active and aggressive management, including antiviral treatment during the acute phase of HZ promise better results,[68] this study showed that early and repeated local administration of MeB₁₂ did not provide significant pain relief when administered immediately after pain and rash onset. The KM curve indicated that the subjects in the B1 responded better to treatment than those in A1, and the median minimum intervention time was 6 days in B1 and 11 days in A1. No immediate effects of local MeB₁₂ injection were observed in A1, but B1 was associated with a significantly higher incidence of pain scores ≤ 3 within 21 days. The benefit ratio for B1 versus A1 was 8.67, indicating that this treatment is ineffective until the HZO reaches a certain stage in its progression. The mechanism underlying the differences in responses associated with the timing of treatment requires further investigation.

Most subjects complained of continuous neuroinflammatory pain (redness, swelling, burning, throbbing, aching, and shooting pain) in the acute phase within 7 days. A few had experienced more than one type of sensory symptoms or discomfort such as paroxysmal pain, tactile allodynia, tingling, numbness, and itching during treatment and subsequent follow-up. These results indicated that treatment by local MeB₁₂combination with lidocaine injection alters or mediates the pathological process from neuroinflammatory pain to neuropathic pain, with the incidence and intensity of positive symptoms such as various types of pain decreasing, while the incidence of itching increased.

PHN occurs in 5% to more than 30% of patients.[69] In the ARIZONA study, the percentage of subjects with PHN ranged from 6.0% to 15.2% in the 70 to 74 years to 85+ years age groups, respectively, at month 12.[70] It has been reported that PHN lasts longer than 1 year in a greater percentage of patients after HZO than HZ overall (50% of HZO vs. 37% of HZ, aged 60 to 69; 61% of HZO vs. 47% of HZ, aged ≥ 70 years).[71] This study results showed that the incidence of PHN was 2.04% (1/49) at 3 months for the treatment group as a whole, which was considerably lower than that reported by Rabaud et al.,[15] in which 6.0% of subjects with zoster in the general population had pain at month 12. Early treatment with MeB₁₂ not only provided an analgesic effect, but was also shown to prevent PHN. The proportion of subjects with reported pain, tingling, numbness, or itching scores ˃ 0 at month 12 ranged from 8.16% to 22.45% for the treatment group as a whole, which implied that local MeB₁₂ combined with lidocaine therapy did not completely eliminate all symptoms.

This study has some limitations that need to be taken into account when considering the study and its results. The first limitation was the difficulty of blinding local injections. Pain assessment was conducted by an independent rehabilitation physician who did not participate in clinical management and was kept blind to treatment allocation to prevent bias due to unblinding of the treatment. The second limitation concerning the primary evaluation criteria used for assessing pain by the patient during regular phone calls; it is possible that the patients’ perception of pain may have been different during a routine monitoring visit. However, patient-reported outcomes are pertinent to gather information in a real-life situation.

The findings of this 14-day single-center, observer-blind, randomized, controlled clinical trial suggest that MeB₁₂ has a significant analgesic effect on AOHN. MeB₁₂ combined with lidocaine not only improved detumescence and cutaneous healing of the affected area, but also exerted a significant and sustained analgesic effect on AOHN. Furthermore, the incidence of PHN was also significantly reduced. Local treatment allows direct delivery of MeB₁₂ to topical HN-damaged subcutaneous and neuronal tissue to obtain more rapid and effective responses innerve fibers affected by AOHN than are achieved by systemic administration. It can be speculated antineuroinflammatory, analgesic, and neuroprotective effects are involved in the mechanism of the functions of MeB₁₂. Local MeB₁₂ injection within 4 to 7 days may be an optimal therapeutic option, with more rapid and greater pain relief, improvement in QoL, and reduction in the incidence of PHN achieved by timely initiation of treatment. It should be stressed that first-line therapy for HN should include strategies using neurotrophic agents to protect the affected neural fibers and endings against VZV damage. Effective and long-lasting pain relief in AOHN and PHN remains a largely unmet medical need, and further studies are needed to confirm the findings of the present study.

We express our sincere appreciation to the patients who served as subjects for this study and Drs Yan Feng, Chaosheng Zhou, and Wen Li for collection of the clinical trial data.