Rate of Complications in Immunocompromised Patients and Unexpectedly High Proportion of Zygomycetes in Computed Tomography-Guided Percutaneous Lung Biopsy Specimens

To the Editor—We read with great interest the article by Lass-Flörl et al. [1] about the utility of examination of CTguided lung biopsy specimens for diagnosis of invasive fungal infection in immunocompromised patients. For greater appreciation and to generalize the article’s findings, some issues need additional clarification. First, CT-guided transthoracic needle biopsy of the lung has become a common procedure to elucidate the nature of pulmonary nodules [2]. The diagnostic performance of this procedure for tumors depends on the size, location, and radiologic and histologic characteristics of the lesion, and the procedure is reported to have high sensitivity and specificity for malignancy [3]. At present, studies of the diagnostic accuracy of this approach are scarce in the context of invasive pulmonary mycosis. The technique is reported to have a considerable rate of complications in several reports, and we have some concern regarding potentially severe complications in the studied population. Although pneumothorax is reported to occur in 8%–60% of patients, it seldom requires drainage (1%–13% of patients) among persons who present with pulmonary nodules [4– 6]. Use of a thoracic tube to treat pneumothorax in immunocompromised patients may constitute an important risk factor for infection. Major hemorrhage, which is considered to be a very rare adverse event following CT-guided needle biopsy, may also constitute a relevant hazard in invasive pulmonary aspergillosis. Patients with hematologic malignancies and invasive pulmonary mycosis have an increased risk of developing bleeding complications due to both thrombocytopenia and hypervascularization of the lesion. Therefore, it would be of great interest to have additional detailed data about the rate of complications in the described immunocompromised population. Second, thrombocytopenia is common in patients with hematologic malignancies. The described technique is contraindicated for patients with platelet counts !50,000 platelets/mL. We are concerned that there was no information about platelet counts and transfusion. In our opinion, Lass-Flörl et al. [1] should have provided the percentage of patients with hematologic malignancies for whom this diagnostic approach would have been contraindicated and who required platelet transfusion. Third, the authors documented an unexpectedly high proportion of patients with Zygomycetes infection (13 [21.3%] of 61 patients). This finding is of great concern and could have major consequences in clinical practice, because drugs that are commonly used for these patients (e.g., voriconazole and echinocandins) are not effective against these fungi. Therefore, we asked ourselves whether the high rates of Zygomycetes infection observed by the authors could be associated with local epidemiologic features, prophylactic regimens, and/or greater diagnostic accuracy of their approach. The generally observed increase in the rate of zygomycosis seems to be associated with the type of prophylaxis administered [7, 8], as reflected by the relationship between the use of voriconazole prophylaxis and the increased frequency of zygomycosis documented in several recent reports [9, 10]. Additional information on local, specific fungal epidemiology, the prophylactic regimens used for solid-organ transplant recipients, and hematologic malignancies in the studied population should have been provided to clarify the high rate of Zygomycetes infection.