Burning mouth syndrome associated with varicella zoster virus

Abstract

We present two cases of burning mouth syndrome (BMS)—of 8-month duration in a 61-year-old woman and of 2-year duration in a 63-year-old woman—both associated with increased levels of antivaricella zoster virus (VZV) IgM antibodies in serum and with pain that improved with antiviral treatment. Combined with our previous finding of BMS due to herpes simplex virus type 1 (HSV-1) infection, we recommend evaluation of patients with BMS not only for VZV or HSV-1 DNA in the saliva, but also for serum anti-VZV and anti-HSV-1 IgM antibodies. Both infections are treatable with oral antiviral agents.

Background

Burning mouth syndrome (BMS) is characterised by chronic, orofacial burning pain for which no dental or medical cause is found. Pain persists and is often intractable despite treatment with antidepressants, analgesics, hormones, α-lipoic acid and anticonvulsants.¹ We recently described a case of BMS in a 65-year-old woman that was due to herpes simplex virus type 1 (HSV-1), virologically verified by the presence of HSV-1 DNA in her saliva, and based on the disappearance of pain and viral DNA after treatment with oral valacyclovir.² We now present two cases of BMS produced by another alphaherpesvirus, varicella zoster virus (VZV); both cases responded to oral antiviral therapy, although the duration of treatment was longer than that needed for BMS caused by HSV-1.

Case presentation

Case 1

A healthy 61-year-old woman developed sudden burning pain on both sides of her tongue, hard palate and buccal mucosa anteriorly. Pain developed within 1 month of routine dental cleaning and persisted for 8 months. Pain was 4–5/10 in the morning and rapidly increased to 10/10 as the day progressed. Concurrently, she developed chronic bilateral occipital headaches that radiated to the left side of her head, along with intermittent shooting pain in the left V2 distribution. She had been immunised with zoster vaccine 10 months earlier. She noted a dry mouth 4 months before the onset of mouth pain. There was no history of herpes labialis or genitalis. She did not use tobacco, alcohol or recreational drugs. Multiple dental and medical evaluations were negative. On examination, sensation was normal on the face and tongue, and in the mouth. There was no loss of smell or taste, no tongue atrophy and no weakness. Several bilateral, non-tender, flat erythematous papules were seen on the posterior palate.

Routine blood count, liver and renal function tests were normal. Thyroid-stimulating hormone (TSH), free T3, free T4 and thyroglobulin antibody were normal; thyroid peroxidase antibody was elevated at 345 U/mL (normal <60 U/mL). ACE, erythrocyte sedimentation rate (ESR), C reactive protein (CRP) and vitamin D were normal. Serum contained neither anti-HSV-1 nor HSV-2 IgG/IgM antibodies. Serum anti-VZV IgG antibody was present and anti-VZV IgM antibody was elevated at 1.88 (normal <0.90). Neither VZV, HSV-1 nor HSV-2 DNA was detected in saliva.

Case 2

A few months after extensive dental work, a healthy 63-year-old woman developed persistent burning 10/10 mouth pain reaching a 2-year duration over the palate, gums, lips and tip of her tongue bilaterally. At times, her lips felt swollen and her left ear was sensitive to the touch. She had no changes in taste or smell. Multiple dental and medical evaluations were negative. Her history was remarkable for vitamin D deficiency and recurrent sinus infections. She had been immunised with zoster vaccine 1 year before the onset of mouth pain. There was no history of herpes labialis or genitalis and she did not use tobacco, alcohol or recreational drugs. Examination was normal, particularly sensation on the face and tongue, and in the mouth. There was no loss of taste, tongue atrophy, weakness or oral/mucosal lesions.

Routine blood count, liver and renal function tests were normal. TSH, ESR and CRP were normal. Serum contained neither anti-HSV-1 nor anti-HSV-2 IgG/IgM antibodies. Serum anti-VZV IgG antibody was present and anti-VZV IgM antibody was elevated at 1.02 (normal range <0.90 ISR). Saliva contained neither VZV, HSV-1 nor HSV-2 DNA.

Differential diagnosis

Other causes of mouth pain to consider are: (1) xerostomia caused by age-related reductions in oestrogen and progesterone, and other endocrine abnormalities; (2) infections including Candida species and coliforms such as Enterobacter and Klebsiella; (3) allergic reactions to dietary antigens and dental metals; (4) autoimmune diseases including Sjögren's syndrome and systemic lupus erythematosus; (5) nutritional deficiencies including vitamin B₁, B₂, B₆, B₁₂, folic acid and zinc; (6) drugs, particularly ACE inhibitors and angiotensin receptor blockers, levodopa and topirimate; and (7) compressive lesions of the trigeminal nerve or ill-fitting dentures.²

Outcome and follow-up

Case 1

The patient was treated with oral valacyclovir, 1 g three times daily for 2 months, after which she reported only minimal pain compared to chronic severe daily pain before treatment. Over the next 5 months, multiple attempts to lower the valacyclovir to 1 g daily or discontinue treatment led to recurrent severe pain. At a 1 g three times daily dose of valacyclovir, the patient has continued to improve.

Case 2

The patient was treated with oral valacyclovir, 1 g three times daily for 3 months, after which she reported less mouth pain and improved sleep. Anti-VZV IgM antibody remained elevated at 1.15. She continued the same dose of valacyclovir for another 3 months, after which the pain was further reduced by 60%. The valacyclovir was later reduced to 1 g daily. After 8 months, the patient is pain-free 3–4 days per week and otherwise experiences only mild pain.

Discussion

The two patients with BMS described had elevated levels of serum anti-VZV IgM antibodies and favourable responses to treatment with antiviral agents. Previously, we described a 65-year-old woman with a 6-month history of BMS whose saliva contained 3.4×10⁸ copies of HSV-1 DNA per mL, and in whom antiviral treatment led to complete resolution of pain and disappearance of HSV-1 DNA in saliva.³ While neither of our patients here had VZV DNA or HSV-1 DNA in their saliva, both had serum anti-VZV IgG antibody, an indication of past exposure to VZV. The diagnostic value of anti-VZV IgG antibody is useful when it is increased fourfold or more over time, but monitoring antibody levels over time is not always practical. In contrast, both of our patients had elevated levels of anti-VZV IgM antibodies, which usually develop 7–10 days after acute infection, increase to maximum levels after 2–3 weeks and decline to undetectable levels after about 3 months. While the presence of virus-specific IgM antibody suggests current or recent primary infection, IgM-specific antibody responses are not restricted to primary infection and can arise from reactivation of latent infections (eg, zoster or recurrent HSV infection). Importantly, antiviral IgM antibody can persist if infection becomes chronic.⁴ Overall, the detection of anti-VZV IgG antibody indicates past exposure to VZV typically encountered during varicella (chickenpox), while the elevated anti-VZV IgM antibody in both of our patients with BMS most likely reflects reactivation of latent VZV.

Alphaherpesviruses as a cause of chronic, orofacial pain is not surprising since VZV is latent in >90% and HSV in >70% of human trigeminal ganglionic neurons.^5–9 Virus reactivation from trigeminal ganglia can cause pain in the face and mouth with or without the associated rash of herpes zoster or herpes labialis. A remarkable case of VZV trigeminal ganglionitis without rash involved a 39-year-old immunocompetent man who had three episodes of left maxillary-distribution zoster with pain that resolved after treatment with antiviral agents; 2 years later, left maxillary-distribution pain recurred without zoster rash and persisted for 3 months until he died of severe hypertensive cardiomyopathy.¹⁰ Immunohistochemistry revealed VZV antigen but not HSV antigen in trigeminal nerve roots and in the adventitia and wall of a meningeal artery, with corresponding inflammation in the ganglia and trigeminal nerve roots. Another case of chronic VZV ganglionitis without rash involved a 45-year-old immunocompetent woman with right facial numbness and pain in the maxillary division of the right trigeminal nerve, of 13-month duration; brain imaging revealed a homogeneously enhancing mass in the right Gasserian fossa, and examination of the excised ganglionic mass revealed inflammation and VZV antigen but not HSV antigen in the ganglion and trigeminal nerve.¹¹ Overall, chronic productive virus infection caused the prolonged maxillary-distribution pain without rash in these two immunocompetent patients.

Although VZV reactivation in both of our patients cannot be attributed to a specific event, it should be noted that they had undergone dental procedures preceding the onset of mouth pain. While trauma does not precede VZV reactivation as often as in HSV reactivation, there is a report of two patients in whom surgery-induced contralateral homologous dermatomal distribution zoster within 3–4 weeks.¹² Both of our patients also received zoster vaccine within a year of the onset of BMS, but it seems unlikely that immunisation contributed to their condition since none of more than 19 000 adults who received zoster vaccine in the Shingles Prevention Study developed BMS.¹³

Finally, the International Headache Society describes BMS as an intraoral burning sensation for which no dental or medical cause can be found, with the following diagnostic criteria: (1) daily pain in the mouth persisting for most of the day; (2) oral mucosa of normal appearance; and (3) absence of local and systemic disease. Both of our patients satisfied these clinical criteria for BMS, and, on further evaluation, had anti-VZV IgM antibody in their serum and responded favourably to antiviral agents, although over a longer period than our earlier patient with BMS caused by HSV-1. Clinicians should recognise that the absence of improvement within the first 2 weeks of antiviral therapy in patients with confirmed VZV infection in BMS implies the need for prolonged antiviral treatment. Importantly, given these cases and an earlier report that alphaherpesviruses can cause BMS,³ the updated International Headache Society classification of primary BMS should include exclusion of alphaherpesvirus infection.

Overall, we recommend that evaluation of patients with suspected BMS should include: (1) assessment of serum anti-VZV, anti-HSV-1 and anti-HSV-2 IgM antibody; and (2) PCR analysis of saliva by PCR or an oral swab for the presence of VZV, HSV-1 and HSV-2 DNA. Diagnosis of alphaherpesviruses as the cause of BMS is essential since these infections are treatable with antiviral agents.

Learning points.

• Varicella zoster virus (VZV), as well as herpes simplex virus type 1 (HSV-1), can cause burning mouth syndrome, with or without rash.

• Diagnosis can be confirmed by either the presence of elevated serum anti-VZV or anti-HSV-11 IgM antibodies, or positive PCR for VZV or HSV-1 DNA in saliva or oral swabs.

• Prolonged antiviral treatment may be required to alleviate the pain of VZV-associated burning mouth syndrome.