Approach to the Diagnosis of Sarcoidosis

LettersApproach to the Diagnosis of Sarcoidosis

To the Editor: We read with interest Dr. DeRemee's reviewof sarcoidosis, which was published in the February 1995issue of the Mayo Clinic Proceedings (pages 177 to 181).We agree with Dr. DeRemee about tissue confirmation fordiagnosis of stage I sarcoidosis; however, we think it isreasonable to consider not only the benefits but also the risksand costs of the various diagnostic techniques.'

Lung abnormalities are almost universal in sarcoidosis,independent of the sarcoidosis stage. Transbronchial biopsyis the most adequate initial test for stage I sarcoidosis.' Fourof the most important current studies that assessedtransbronchial biopsy of sarcoidosis showed that this proce­dure had a sensitivity of 66% for the diagnosis of stage Isarcoidosis.'? Moreover, a work by Wang and colleagues'on the value of flexible transbronchial needle aspirationshowed a sensitivity of 83% for the diagnosis of stage Isarcoidosis. Inasmuch as computed tomography (CT) of thechest is more sensitive than is chest roentgenography, CT ishelpful for localizing affected areas for biopsy of chest x­ray-determined stage I sarcoidosis.' Thus, the sensitivity ofthe transbronchial tests could be even greater when CT of thechest is performed.

Severe complications are uncommon with transbronchialbiopsy.t-' The most frequent problems are pneumothorax,which has been reported in less than 3% of cases, and insig­nificant bleeding, which has been described in 1.7%.2 Thereported complication rate associated with mediastinoscopyvaries. Sarrazin and Dyon" reported a mediastinoscopy com­plication rate of 1.5 to 3.8%. Recently, mediastinoscopy wasassessed as an outpatient procedure; complications occurredin 1.37% of patients, and significant complications werenoted in I % who required hospitalization.' Finally,transbronchial biopsy-induced mortality was described asbeing as low as 0 to 0.12% in comparison with 0 to 0.5% formediastinoscopy.v' The 95% sensitivity of mediastinoscopyfor stage I sarcoidosis':' and its low complication rate arehandicapped, however, by the cost and requirement of gen­eral anesthetic.

On the basis of these comparisons, if a biopsy specimenby transbronchial tests is negative for granuloma and ifclinical examination and thorax CT performed a few weekslater do not add any diagnostic clue, mediastinoscopy shouldbe considered for the diagnosis of stage I sarcoidosis.

Jose Bordon, M.D.Jose Blanco, M.D.Mar Mosteiro, M.D.Adela Acuna, M.D.Hospital Xeral of VigoVigo, Spain

REFERENCES1. KirtlandSH, WinterbauerRH. Pulmonarysarcoidosis. Semin

Respir Med 1993; 14:344-3522. Shure D. Transbronchial biopsy and needle aspiration. Chest

1989; 95:1130-11383. Wang KP, Fuenning C, Johns CJ, Terry PB. Flexible

trans bronchial needle aspiration for the diagnosis ofsarcoidosis. Ann Otol RhinolLaryngol 1989;98:298-300

4. SarrazinR, DyonJF. Mediastinoscopy. Rev MalRespir 1992;9:99-110

5. CybulskyII, BennettWF. Mediastinoscopy as a routineoutpa­tient procedure. Ann ThoracSurg 1994;58:176-178

In response: My response to Dr. Borden and associatesconcerns only the relative merits of mediastinoscopy versustransbronchial lung biopsy in the diagnosis of stage Isarcoidosis. My colleagues and I have not assessed theWang technique of needle aspiration of mediastinal lymphnodes for pathologic study pertinent to sarcoidosis, andhence I cannot make a valid analysis of this procedure. Isuspect, however, that the sensitivity rate of 83% is the resultof Dr. Wang's expertise and special interest in this area,which probably cannot be duplicated in many institutions.

Both mediastinoscopy and trans bronchial lung biopsyhave very low associated risks of morbidity and mortality.In light of the young population in whom these proceduresare used in sarcoidosis, they must be essentially equal inthese regards. Hence, in selecting these procedures, the mainconsiderations are cost; patient discomfort, both physicaland mental; and final accuracy of diagnosis. In regard to costat Mayo, mediastinoscopy is approximately two times that ofthe bronchoscopic approach. Thus, if one assigned two unitsof cost to mediastinoscopy and one unit to bronchoscopy andconsidered 100 patients with stage I sarcoidosis apparent onchest roentgenography, one could make the following calcu­lations. With the bronchoscopic approach, the sensitivity is66%, and thus the cost will be 100 units for 66 diagnoses;however, 34 patients will need to undergo another proce­dure, most likely mediastinoscopy, incurring an additional68 units to the already expended 100 units for bronchoscopy.For the 100 patients, the total units will be 168. If the patientinitially undergoes mediastinoscopy, which is virtually100% sensitive, the total cost would be 200 units, a differ­ence of 32 units in favor of the bronchoscopic approach;however, if the sensitivity of the bronchoscopic approach is50%, which it is in many institutions, no difference existsbetween the two approaches.

In former times, I was accustomed to using thebronchoscopic route, but having been frequently frustratedbecause of a negative result and having to explain to thepatient the need for yet another procedure, and in light of thesuccess rate and low morbidity associated withmediastinoscopy, I changed my practice. To some extent,this approach is tempered by the nature of my practice,

Mayo Clin Proc 1995; 70:1019-1020 1019 © 1995 Mayo Foundation for Medical Education and Research

For personal use, Mass reproduce only with permission from Mayo Clinic Proceedings,