Abstract
BACKGROUND: We describe the risk indicators for oral mucositis (OM) in paediatric oncology patients hospitalised in the Institut Gustave Roussy (Villejuif-Paris) and treated with alkylant chemotherapy with autologous peripheral blood progenitor cells. METHODS: The sample was selected using PIGAS software. Three groups of subjects received different chemotherapy regimens: A. Melphalan, B. Busulfan and C. other alkylant protocols. The degree of mucositis was recorded by CTC version 2.0 (Common Toxicity Criteria). Descriptive statistics were performed. The association between mucositis and risk indicator variables was tested using a chi2 test. The association between case status and covariates was tested using unconditional logistic regression analysis. RESULTS: Of the 337 children enrolled, 241 showed mucositis (group 1) and 96 did not show mucositis (group 2) during alkylant chemotherapy. There was a higher prevalence of male patients in both groups. The three different chemotherapy regimen groups are correlated with the appearance of oral mucositis (chi2 = 22.42, p < 0.01). Weight loss was higher in group 1 (chi2 = 6.31, p = 0.01). The duration of aplasia was lower in the Busulfan protocol (7.5 days) than in the Melphalan group (9.3 days) or the other regimens (8.6 days). The use of Bufulfan was directly associated with case status (presence of oral mucositis): odds ratio [OR] = 2.1 and confidence interval [95%CI] = 1.3-3.0. Also, occurrences of germinal tumours and secondary bacterial infections were directly linked with case status: [OR] = 1.4 and 1.8, confidence interval [95%CI] = 1.2 - 1.7 and 1.1 - 2.5, respectively. CONCLUSION: The presence of OM was associated with the three different chemotherapy regimens considered; in particularly patients treated with Busulfan had the highest prevalence.