Reply to Rokx et al

We thank Dr Rokx and colleagues for their interesting comments on our brief report recently published in Clinical Infectious Diseases and titled “Hemophagocytic lymphohistiocytosis associated with Bartonella henselae infection in an HIVinfected patient” [1]. Positive diagnosis of disseminated Bartonella henselae infection was based on the association of liver biopsy data (including peliosis hepatis and positive Warthin-Starry [WS] staining) andstrongly positive B. henselae serology (immunoglubulin G = 1/8192). The B. henselae serology used in our laboratory has more than 95% specificity for titers ≥1/128 [2]. A frozen sample was not available for molecular diagnostic confirmation. No other pathogen cross-reacting with WS staining was isolated (negative serology for syphilis and Lyme disease, negative stool and liver cultures for microsporidiosis). Second, other causes of secondary hemophagocytic lymphohistiocytosis (HLH) were ruled out: