\s=b\The incidence of pneumonia was evaluated in 844 children undergoing initial treatment for acute lymphoblastic leukemia (ALL). A total of 310 episodes occurred in 239 patients followed up for five to 36 months after diagnosis. The peak incidences occurred in the periods 0 to 20 days and 40 to 80 days after the start of antileukemic therapy. Bacterial pneumonias occurred primarily during the first 20 days after diagnosis of ALL. No episode of Pneumocystis carinii pneumonia was noted before 40 days, and the majority of instances occurred 50 to 120 days after diagnosis. In 80% of all episodes, a specific causative organism was not detected. The incidence of P carinii pneumonia was greater in patients receiving intrathecal methotrexate as part of CNS prophylaxis than in those receiving only CNS irradiation. Pneumonia is a frequent event during the therapy of ALL in childhood.(Am J Dis Child 134: [28][29][30][31][32][33][34] 1980) Pul monary infections frequently occur during the therapy of acute leukemia and other malignancies of the hemopoietic system, often result¬ ing in interruption of intensive chem¬ otherapy programs and/or in death.17 According to the literature, pneumo¬ nia accounts for 28% to 43% of all fatal and nonfatal infections in these patients.'Ii Bacterial and fungal orga¬ nisms comprise more than 90% of all the documented etiologic agents, al¬ though viruses and protozoans are recovered more frequently than from normal age-matched patient popula¬ tions'3 and are the most common causes of death due to pneumonia in some series.6 In addition, the direct pulmonary effects of radiation and specific chemotherapeutic agents, such as methotrexate, have been increasingly implicated in these pro¬ cesses as they are incorporated into standard treatment regimens for acute leukemia.818Previous studies of pulmonary dis¬ ease in acute leukemia have consisted of investigations that stress severe or fatal episodes in relatively small patient populations.1 •-•e·7·-" To deter¬ mine the frequency and characteris¬ tics of pulmonary disease during the therapy of childhood acute lympho¬ blastic leukemia (ALL) in a large representative population and to iden¬ tify factors contributing to the devel¬ opment of pulmonary complications, an evaluation was conducted by the Childrens Cancer Study Group (CCSG) as an integral part of an ongo¬ ing treatment program for childhood ALL.
SUBJECTS AND METHODS
Patient PopulationAll patients aged birth to 16 years entered on CCSG protocols 101 and 143 were eligible for this study. Three standard phases of therapy are identified for both protocols: induction, intensification, and maintenance. Protocol CCSG-101 (Fig 1) was designed, in part, to determine the most effective and least toxic means of controlling extramedullary disease in ALL by randomizing patients who attain bonemarrow remission status to one of four intensification regimens, as follow: regi¬ men 1, craniospinal irradiation 2,400 rads and 1,200 rads to an "extended" field that includes total abdomen and pelvis (the t...