When Uncommon Uncovers: Mucosal Tuberculosis in a Medical Tourist From Burundi


Moti Grupper, MD, Infectious Disease Unit, Bnai Zion Medical Center, 47 Golomb St, P.O. Box 9490, Haifa 31048, Israel. E-mail: moti.grupper@b-zion.org.il


Tuberculosis confined to the mucus membranes is a rare presentation in the era of effective chemotherapy. We describe a case of mucosal tuberculosis in a “medical tourist” from Burundi that went undiagnosed for 6 years. Starting as conjunctivitis, the disease has spread to involve the nose and larynx as well. The clinical, pathophysiological, and epidemiological aspects are discussed.

Mucosal tuberculosis is a rare presentation of a common disease. Mucosal tuberculosis may affect the conjunctivae, the nasal mucosa, the pharynx, the larynx, and the trachea.1–14 Some of these presentations carried a grave prognosis in the prechemotherapy era, as tuberculous laryngitis was responsible for more than one third of deaths.1 We present a unique case of mucosal tuberculosis from Africa, causing destruction of the eyelids, nose, and nasopharynx that had been undiagnosed for 6 years.

Case Report

A 26-year-old student from Burundi arrived at our facility for diagnostic purposes (self-referred medical tourist). He presented with slowly progressive, bilateral, lower palpebral inflammation (Figure 1); this was accompanied by dyspnea and weight loss of 15 kg over the last year. The lesions began 6 years ago in the lower left eyelid, progressed within a few months to the right, and subsequently spread to the left nostril and pharynx. Multiple blood tests, palpebral biopsy, and antibiotics (topical and oral) in Burundi were non-yielding.

Figure 1.

Palpebral and nasal lesions, with ulcerations and granulation tissue.

Fiberoptic nasopharyngoscopy at our hospital revealed: “granulomatous” lesions of the left nostril and the right nasal septum, extensive epiglotic/arytenoid destruction, and narrowed airway (diameter = 4 mm). Blood count showed mild anemia (hemoglobin = 11.4 g%), but chemistry, serum immunoglobulins, and complement were normal. Serologies for antineutrophilic cytoplasmic antibodies, HIV, Brucella, Borrelia burgdorferi, and syphilis were all negative. Palpebral culture grew methicillin sensitive Staphylococcus aureus. Computerized tomography of the chest was remarkable for bilateral, upper lobe, and nodular lesions.

The patient was scheduled for palpebral biopsy which demonstrated acute on chronic inflammation including non-necrotizing granulomata. However, specific stains for Mycobacteria, fungi, and other organisms were negative. Polymerase chain reaction (PCR) assays using pan-bacterial, pan-fungal, pan-mycobacterial, and Chlamydial primers were negative, except for S aureus that was not considered to be the culprit pathogen. Culture, serology, and PCR for Leishmania were negative as well. Four weeks after the biopsy, cultures of the palpebral biopsy yielded growth of a Mycobacterium tuberculosis complex organism on Löwenstein–Jensen medium. The strain was identified at the national tuberculosis laboratory as M tuberculosis, sensitive to all first line antituberculous agents. Subsequently, a tuberculin skin test (5 TU) yielded an induration of 16 × 25 mm. Mycobacterial cultures of sputum or gastric apirates were not obtained because of technical issues. The patient was started on a four-drug regimen (isoniazide, rifampicin, pyrazinamide, and ethambutol) and flew back to Burundi.

Within 2 weeks after initiation of the antituberculous therapy the palpebral and nasal lesions started to dry, and he was capable of swallowing solid food more freely. After 2.5 months of treatment he had gained 14 kg and his mood was reportedly much improved.


Aside from the satisfaction of diagnosing a chronic, treatable disease, this case raises several important features. Firstly, the disease process included widespread involvement confined to the mucosal membranes. Scattered reports of either nasal, conjunctival, nasopharyngeal, pharyngeal, or laryngeal tuberculosis can be found,2–14 all emphasizing that these are uncommon and hard to diagnose presentations, even in endemic countries. To our knowledge there are no other case reports describing the simultaneous involvement of all these mucous sites.

The combination of mucosal lesions, macroscopic appearance of ulcerations with granulation tissue, histology of non-caseating granulomata with absent acid-fast bacilli, positive mycobacterial culture, and positive PPD is most consistent with the diagnosis of lupus vulgaris, one of the paucibacillary forms of cutaneous tuberculosis.15 The pathophysiological basis for the current process distribution is not completely clear. One possible explanation would be a primary, simultaneous exogenous inoculation of tubercule bacilli into both the respiratory tract and the mucosal surfaces of the eyes and nose. Likewise, a sequential autoinoculation may have occurred. Namely, infection of one eyelid first, then the other, followed by the nose and the larynx. On the other hand, autoinoculation by contaminated lung/laryngeal secretions from post primary tuberculosis may be responsible. Hematogenous spread from an endogenous site had also been emphasized as a possible mechanism in cases of lupus vulgaris of the face.1 The complex interaction of mycobacteria with M cells (specialized cells which are part of the mucosa-associated lymphoid tissue), resulting in endocytosis of the first, has a probable major role regarding tropism to mucous membranes.16

This case highlights important public health aspects. The patient, who had laryngeal and probably pulmonary involvement with tuberculosis (although unproven microbiologically), had considerable air travel with a notoriously communicable disease. The possible transmission of infectious diseases, particularly tuberculosis, by international flights, has been widely addressed, including by WHO guidelines.17 Notably, most passengers arriving by commercial air flights are not screened for tuberculosis in any country.17 Consequently, the key to limiting these problematic scenarios is the suspicion or diagnosis of the communicable disease before departure. It is clear that the probability of achieving this goal in developing countries is quite low. Burundi, for example, is one of the poorest countries in the world, with only one physician per 44,000 people18; it is thus not surprising that this case went undetected for a long time.

The healthcare marketplace is globalizing, and medical tourism is increasingly recognized; however, emphasis is mainly given to the trend of traveling from developed to less developed countries for receiving medical care (eg, travel to India for transplantation).19 Our case illustrates that the road to the tropics is a two-way road and attention should also be given to air travelers who are “medical tourists” from developing countries. As it seems intuitive that these passengers have a higher likelihood of carrying a communicable disease, screening this specific group should be considered by public health ministries and port authorities.

In conclusion, we presented a unique case of mucosal tuberculosis with both diagnostic and public health challenges. Clinicians should be vigilant to rare presentations of common diseases. [Note: Ten months after the growth of mycobacteria at the local laboratory, workup carried out at the Infectious Diseases Pathology Branch of the CDC was positive for the 16S rRNA gene of M tuberculosis complex (paraffin embedded sections).]

Declaration of Interests

The authors state that they have no conflicts of interest.