Abstract
Background
There is a growing body of literature documenting local herpes zoster outbreak following procedures. The mechanism underlying these outbreaks remains elusive. We present a case of zoster following onabotulinumtoxinA (BTX) for migraine and a literature review.
Methods
Chart and literature review.
Case
A 72-year-old woman with chronic migraine received BTX injections for 3 years without incident. She had a history of thoracic zoster with subsequent post-herpetic neuralgia. In August 2013, 48 hours after receiving BTX injections, she developed a painful rash in the right V1 distribution consistent with herpes zoster ophthalmicus. One week later the rash had resolved without treatment.
Literature review
We identified 65 (including 2 from Juel-Jenson) cases of zoster reactivation following minor procedures. These cases tend to be in young patients without specific risk factors. Outbreaks characteristically occur at the level of exposure to local trauma.
Discussion
Our review suggests that local trauma, regardless of the nature of stimuli, may be sufficient for zoster reactivation. We hypothesize that the stressors in these reported cases exert a local epigenetic influence on viral transcription, allowing for viral reactivation.
Conclusion
Zoster is a potential complication of BTX administration for chronic migraine in adults. Physician awareness can reduce the significant morbidity associated with this disease.
Introduction
Herpes zoster results from reactivation of varicella zoster virus, which lies dormant in sensory dorsal root, cranial nerve, and autonomic ganglion. It presents as a painful maculopapular or vesicular rash in a dermatomal distribution, most commonly in thoracic and cranial distributions (1). The primary risk factors are immunosuppression and advanced age. Systemic illnesses and certain malignancies also increase risk. Epidemiologic studies suggest increasing zoster incidence (2).
The most common complication is postherpetic neuralgia (PHN), which is pain persisting greater than 90–120 days after onset. PHN occurs in 10–50% of patients (2,3). Cranial nerve involvement may lead to vision changes, blindness, and stroke in herpes zoster ophthalmicus or facial palsy in Ramsay Hunt syndrome (4–6). A recent retrospective cohort study suggested herpes zoster to be a risk factor for vascular disease, including myocardial infarction and cerebrovascular disease (7). Rarely, the virus disseminates and may lead to myelitis, vasculopathy, temporal arteritis, meningitis, encephalitis, and even death (2,8,9).
Reported cases of herpes zoster following procedures.
aTime from procedure to symptoms of herpes zoster.
Case
A 72 year-old woman with a 50-year history of chronic migraine had been receiving onabotulinumtoxinA (BTX) injection for 3 years every 3 months without incident. In addition to migraine, her history was significant for herpes zoster affecting the right lower back in 1994 with subsequent post-herpetic neuralgia. In August 2013 she received BTX injections without immediate complications. The next day she developed a throbbing bioccipital headache radiating into the neck, as well as soreness wrapping around the head in a band-like fashion. This was associated with mild nausea. Two to four days after the injection she developed right-sided periorbital edema.
Examination on post-procedural day 5 revealed a 3-cm macule on the inner canthus of the right eye and mild edema of the medial right orbit. Oral antibiotics were prescribed for possible cellulitis. After a 1-week course, exam demonstrated a 5 × 4-mm scab over the superomedial right orbit and two to three healing scabs on the right side of the nose (Figure 1). A diagnosis of herpes zoster ophthalmicus was made. As a result of near resolution, treatment was not administered.
Right eyelid swelling and lesions consistent with herpes zoster in the first trigeminal distribution.
One month after the procedure, neuro-ophthalmologic examination was normal, and the patient was only reporting minor dysesthesias in the right V1 distribution. She elected not to receive further BTX administration.
Literature review
The case presented here, to the best of our knowledge, is the first published report of herpes zoster following BTX administration for chronic migraine. Our patient’s only known risk factors were her advanced age and prior history of zoster at a remote site. The BTX was administered in the dosage and distribution utilized in the PREEMPT trials (10). Our case follows several reports describing local herpes zoster outbreak following procedures (Table 1). Excluding the Dunst study (11), 24 cases were identified. The mean age of the patients was 46.5, with a range of 19–80. There were 9 men and 13 women. Excluding the Dunst study and an outlying case of zoster 2.5 years after skin grafting, the median time between procedure and symptom onset was 14.4 days, with a range of 1–60 days (12).
In the literature, only two previous cases of local zoster outbreak have been attributed to BTX injections (9). These women, 48 and 55 years old, received injections in the glabella, forehead and lateral periorbital areas for cosmesis. One patient, who had undergone at least eight prior treatments with BTX for cosmesis, presented 7 days after the injections with poorly demarcated erythema and edema with superficial erosions that were not at the exact sites of BTX injection but in a V1 distribution. The other patient first experienced paresthesias on post-procedure day 6 before progressing to the classic outbreak in a V1 distribution. Both of these patients went on to receive subsequent injections of BTX either preceded by or concomitant with prophylactic oral antivirals. They did not experience further outbreaks.
Discussion
The reason for these outbreaks following minor procedures remains elusive. There is likely to be a reporting bias in the literature, making these outbreaks potentially coincidental, although trauma-related zoster outbreaks are not uncommon (13). A 1970 case series of 100 patients with zoster included 38 local outbreaks following trauma, including two operations (13). One case-control study demonstrated a significant increased risk of zoster in the month following trauma to the outbreak site (14). Radiation and surgery have also been associated with local zoster reactivation (15,16). Interestingly, the risk of zoster remains elevated over a 2-year period following radiotherapy for breast cancer (11). Another case-control study of patients with acute facial palsy discovered eight patients who had dental surgery or orofacial surgery within the prior month. Salivary VZV DNA was detected in six and HSV-1 in two, but only 31% of controls. Three were diagnosed with Ramsay Hunt syndrome and three with zoster sine herpete. Onset of palsy occurred between five and 20 days in these patients (15).
A similar situation can be found in the literature regarding herpes simplex virus type 1 (HSV-1). Multiple dental and oral surgical procedures have been associated with reactivation of this virus, although whether any of them specifically increase the risk of HSV recurrence is not clear (17). Even with this relatively more common pathology, it remains uncertain if and to whom prophylactic antiviral therapy should be offered or if a history of recurrent herpes labialis (RHL) is a risk factor for dental procedure-related RHL. In fact, one study compared 48 patients undergoing molar extraction and 48 patients undergoing conventional restorative procedures (control group), all of whom were IgG positive for HSV-1, and showed no difference in the frequency of virus shedding between the two groups (18).
The most important observations in trying to construct a hypothesis regarding VZV reactivation following certain triggers include the lack of specificity to the environmental stressor (radiation, laser, chemical, mechanical, and thermal), the tendency for outbreaks to occur at the level of the trauma, and the fact that cases are often reported in patients without specific zoster risk factors. The consideration of whether repetitive BTX treatments over many years may contribute to reactivation is evoked by our case, as well as the patient reported by Graber, who had undergone at least eight prior treatments (9). However, a study looking at over 500 subjects aged 18–65 who received five full treatment cycles (56 weeks) of active treatment with BTX for chronic migraine did not record any cases of herpes zoster (19). Varicella virus latency in ganglionic neurons involves a characteristic profile of viral gene expression, which in turn, is epigenetically regulated. We hypothesize that the stressors in these reported cases exert a local epigenetic influence on viral transcription, which allows for viral reactivation. The details of how VZV transitions from a latent to reactivated state are not well understood, largely because of the difficulty in achieving a suitable small-animal model for study (20). In a heat stress-induced reactivation model of herpes simplex virus 1 (HSV-1), dexamethasone has been shown to accelerate the kinetics of reactivation (21). Another animal model of HSV-1 reactivation uses ultraviolet irradiation (22). Although the exact details of viral reactivation are still unknown, these in vivo models have shown that specific cytokines, such as IL-6 and tumor necrosis factor alpha, are more prominent during reactivation, and the expression of specific viral proteins, such as VP16, are required for reactivation (22,23). A local stress reaction following tissue injury from a minor procedure may be sufficient to induce VZV reactivation in a similar manner (24).
Other authors have supported local mechanisms as important for viral reactivation, a theory often cited in dental literature regarding HSV-1 (25). Another theory first postulated regarding HSV-1 reactivation suggests frequent viral release from the ganglion to form microfoci of infection in the skin, but these are usually eliminated. Procedures, such as BTX in our patient, induce physiological changes in the skin, occasionally allowing lesions to develop (26). Others have suggested a triggering hyperemia, analogous to that occurring in the kidney after administration of steroids leading to bacterial infection (13). It is generally thought that a decline in cell-mediated immunity is one factor allowing viral reactivation (27). More recently, regarding HSV-1, it has been reported that the latency associated transcript gene (the only gene highly expressed during the establishment of HSV-1 latency) is required for long-term maintenance of reactivation competent latent infections (28,29). As has been proposed for HSV-1 reactivation, it is likely, based on the wide variety of triggers leading to reactivation, that multiple pathways lead to viral reactivation from a signal transduction event that induces transcription of the latent virus (30).
Finally, an interesting recent twist in the story is that botulinum toxin is being studied as a potential treatment of post-herpetic neuralgia (31,32).
Conclusion
Although local outbreaks of shingles following procedures such as BTX may be coincidental, the frequent observation of a lack of traditional clinical risk factors suggests that these are pathophysiologically relevant. This complication is seen following diverse mechanisms of local iatrogenic injury, suggesting that there is likely nothing intrinsically pathogenic about BTX itself. We hypothesize that the stressors in these reported cases exert a local epigenetic influence on viral transcription, which allows for viral reactivation. Given the potential for untreated herpes zoster to lead to pain, disability and death; physician awareness is paramount. Physicians may counsel patients with herpes zoster risk factors (e.g. history of zoster) regarding this potential complication.
Clinical implications
Zoster reactivation may occur following minor procedures, including BTX administration. Unlike our patient, these cases tend to occur in young patients without specific risk factors. We hypothesize that the stressors in these reported cases exert a local epigenetic influence on viral transcription, allowing for viral reactivation.
Footnotes
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Conflict of interest
None declared.