Etiology, clinical manifestations and prevention of perianal infection in patients with hematological malignancies

Introduction . Perianal infection (PI) in patients with hematological malignancies is characterized by a wide spectrum of pathogens and a variety of clinical manifestations and mechanisms of development of the infectious process. Aim : to study the pathogenetic mechanisms of PI development in patients with hematological malignancies and to develop prevention tactics. Materials and methods . The prospective study included 132 patients with hematological malignancies who had episodes of PI. The infectious process in the pararectal fi ber was registered based on the data of clinical examination or magnetic resonance imaging. Sources of infection and microbiologic results were studied. Results : Two main mechanisms of PI development were revealed, the frequency of which was statistically significantly different depending on the presence of neutropenia (p < 0.0001, odds ratio (OR) = 24.42 (confidence interval (CI) 95% [9.82–60.74]). In PI episodes that developed against the background of neutropenia, the predominant mechanism of infection was the penetration of microorganisms through broken tissue barriers (75 %): anal fissures were the entry gate of infection in 62.9 % of episodes; perianal ulcers and skin erosions — in 12.1 %. The alternative route of infection (cryptogladular mechanism) was registered in the majority of patients without neutropenia (66.7 %) and was represented by the involvement of crypts of the anal canal (28.6 %) or the presence of pararectal fistulas (38.1 %). Clinical manifestations of PI were associated with leukocyte count (p < 0.0001) and mechanism of infection (p < 0.0001). The median leukocyte count in patients with abscesses (2.12×10 9 /L) was statistically significantly higher than in infiltrates (0.57×10 9 /L) and necrosis (0.74×10 9 /L). The main source of infection in infiltrates was anal fissures (70.4 %), while in abscesses the main sources of infection were crypts of the anal canal (39 %) and pararectal fistulas (36 %). The main causative agents of PI were Gram-negative bacteria ( Escherichia coli (43 %), Klebsiella spp. (15 %), Pseudomonas aeruginosa (4.4 %)), and Enterococci (12.5 %). P. aeruginosa was isolated more often in necrosis (22 %) than in other clinical forms of PI (3–5 %) (p = 0.0033), while the frequency of detection of other bacteria was independent of the clinical manifestations of PI. PI was a microbiologically proven source of sepsis in 9.5 % of PI episodes. The probability of PI-associated bloodstream infections was highest at 5 days and was significantly higher in patients with neutropenia (10 % vs 2 %) (p = 0.0044). Conclusion : Different mechanisms of pararectal cell infection should be taken into account when forming a strategy for the prevention of PI.