There have been conflicting reports of false positive galactomannan assay results in patients with systemic cryptococcosis. We sought to determine the frequency of GM positivity in patients with pulmonary cryptococcosis and confirm the source of this cross-reactivity in vitro. We conducted a retrospective study to elucidate the rate of galactomannan (GM) false positivity and cause in a cohort of 29 patients with pulmonary cryptococcal disease. The production of GM cross-reacting substances by clinical isolates and laboratory isolates of C. neoformans was tested in vitro. The mean serum GM index (Platelia Aspergillus) in patients with pulmonary cryptococcosis was 1.06, with 16 (55.2%) of patients having values above the positive cutoff value of 0.5. GM index values significantly decreased after treatment of cryptococcosis. There was no significant correlation between galactomannan and cryptococcal glucuronoxylomannan antigen (Eiken Latex test) results. Culture supernatants from clinical isolates and wild-type C. neoformans did not react in the GM assay; however, growth in the presence of 6% sodium chloride induced the production of cross-reacting GM antigens in culture supernatants from clinical isolates, wild type and a glucuronoxylomannan-deficient mutant of C. neoformans, but not in culture supernatants from a galactoxylomannan-deficient strain. Our results support the cross-reactivity of cryptococcal galactoxylomannan with the serum GM assay in vitro and in patients with pulmonary cryptococcal infection.