Correspondence

Tuberculous Bronchiolitis

N Engl J Med 1996; 334:1748-1749June 27, 1996

Article

To the Editor:

In his discussion of Case 6-1996, which involved Mycobacterium avium bronchiolitis, Murphy (Feb. 22 issue)1 mentions that he is not aware of a reported case of M. avium complex infection in a patient without predisposing factors, chronic lung disease, or immunodeficiency. We have reported the case of a 32-year-old woman with disseminated M. avium infection who was not infected with the human immunodeficiency virus (HIV) and who did not have primary or acquired immunodeficiency.2

We believe that the patient discussed by Murphy should be evaluated for an underlying disease, such as hairy-cell leukemia or Hodgkin's disease, because M. avium infection may antedate the clinical onset of neoplasia.3 Moreover, mild splenomegaly was noted on physical examination. We also recommend an immunologic evaluation, because opportunistic infections have been associated with idiopathic CD4+ lymphocytopenia.4 We also suggest that lung-biopsy specimens be stained with periodic acid–Schiff to detect alveolar proteinosis, which may be associated with pulmonary M. avium infection.5

Louis-Jean Couderc, M.D.
Isabelle Caubarrère, M.D.
Hôpital Foch, 92150 Suresnes, France

5 References

1. 1

   Case Records of the Massachusetts General Hospital (Case 6-1996). N Engl J Med 1996;334:521-526
   Full Text | Web of Science | Medline

2. 2

   Couderc LJ, Epardeau B, Dazza MC, et al. Disseminated mycobacterium avium intracellulare infection without predisposing conditions. Lancet 1992;340:731-731
   CrossRef | Web of Science | Medline

3. 3

   Di Bella NJ, Buchanan BD, Koontz CH. Disseminated atypical tuberculosis -- antedating the clinical onset of neoplasia. Cancer 1977;40:1276-1279
   CrossRef | Web of Science | Medline

4. 4

   Smith DK, Neal JJ, Holmberg SD, Centers for Disease Control Idiopathic CD4+ T-Lymphocytopenia Task Force. Unexplained opportunistic infections and CD4+ T-lymphocytopenia without HIV infection -- an investigation of cases in the United States. N Engl J Med 1993;328:373-379
   Full Text | Web of Science | Medline

5. 5

   Witty LA, Tapson VF, Piantadosi CA. Isolation of mycobacteria in patients with pulmonary alveolar proteinosis. Medicine (Baltimore) 1994;73:103-109
   Web of Science | Medline

To the Editor:

We wonder whether the pathological evidence in Case 6-1996 is sufficient to establish the diagnosis of M. avium bronchiolitis in the patient, described as a 40-year-old male social worker with the recent onset of dyspnea and bilateral, diffuse nodular densities on the chest film; two negative HIV tests and two negative tuberculin skin tests; prior residence in New York, Massachusetts, California, and Cuba; and most notably, a history of recent treatment with antibiotics, including a quinolone, and a slow response to unspecified therapy for M. avium infection. The species of the small number of beaded acid-fast organisms detected in the lung-biopsy specimen was not established, and the ultimate isolation of M. avium from the specimen does not prove it was the sole or primary pathogen.

We first recommend that formalin-preserved lung tissue be examined by polymerase chain reaction (PCR) for M. tuberculosis. If this test is negative, we then recommend that the lung tissue be stained by an immunoperoxidase technique with an antibody to M. bovis, strain bacille Calmette–Guérin, which is more sensitive than conventional acid-fast stains for identifying mycobacteria in tissues.1 If additional mycobacteria are revealed by this technique, their morphologic features may suggest their identity (M. avium, for example, is usually larger than M. tuberculosis), but the preferred technique for determining the species of these mycobacteria is PCR. If PCR for M. tuberculosis and immunoperoxidase staining with an antibody to M. bovis are both negative, we recommend silver methenamine and Warthin–Starry (or Steiner) stains to detect fungi and uncommon bacteria that may escape detection on gram's staining of tissue. Negative results of all these proposed examinations would eliminate our doubt that M. avium can cause the clinical and pathological entity described in the case report.

Elizabeth L. Wiley, M.D.
Stephen D. Nightingale, M.D.
4703 Broad Brook Dr., Bethesda, MD 20814

1 References

1. 1

   Wiley EL, Mulhollan TJ, Beck B, Tyndall JA, Freeman RG. Polyclonal antibodies raised against Bacillus Calmette-Guerin, Mycobacterium duvalii and Mycobacterium paratuberculosis used to detect mycobacteria in tissue with the use of immunohistochemical techniques. Am J Clin Pathol 1990;94:307-312
   Web of Science | Medline

Author/Editor Response

Dr. Eugene J. Mark, associate editor of the Case Records of the Massachusetts General Hospital, replies:

To the Editor: Drs. Wiley and Nightingale suggest many possible causes of the bronchiolitis. The determination of causation in this case was based on the best available evidence. Mycobacteria were seen in tissue, and mycobacteria grew in culture. PCR is a powerful tool for identifying small numbers of organisms, but a recent review,1 which assessed mycobacterial DNA in tissue, used culture as the confirmatory test, and in this case, the culture did grow M. avium and did not grow M. tuberculosis. The case is peculiar for M. avium infection, but it would also be unusual for M. tuberculosis infection.

Having spent hundreds of hours looking for acid-fast mycobacteria under the high-power lens, I hope that I will never be called on to distinguish M. tuberculosis from M. avium by the size of an organism in tissue. Fungi other than histoplasma do not require special stains for visualization in tissue, and silver stains showed no organisms. The attending physician treated the patient for M. avium infection with ethambutol, clarithromycin, and clofazimine, and he is now markedly improved.

Drs. Couderc and Caubarrère describe disseminated M. avium infection without predisposing conditions and question the immune status of this patient. No immune deficit has become apparent. Pulmonary alveolar proteinosis is a diagnosis made with the hematoxylin–eosin stain. It can be corroborated with the periodic acid–Schiff stain. No alveolar proteinosis was present in this case.

Eugene J. Mark, M.D.
Massachusetts General Hospital, Boston, MA 02114

1 References

1. 1

   Richter E, Greinert U, Kirsten D, et al. Assessment of mycobacterial DNA in cells and tissues of mycobacterial and sarcoid lesions. Am J Respir Crit Care Med 1996;153:375-380
   Web of Science | Medline