To: Admission factors associated with intensive care unit readmission in critically ill oncohematological patients: a retrospective cohort study

We read the manuscript by Rodrigues et al. in the latest issue of Revista Brasileira de Terapia Intensiva Journal with great interest.(1) In this single-center, retrospective, observational cohort study, Rodrigues et al. aimed to identify risk factors associated with later readmission to the intensive care unit (ICU) among critically ill oncohematological patients by evaluating their first ICU admissions. They identified male sex, emergency surgery as the admission reason, longer length of hospital stay before ICU transfer and mechanical ventilation (MV) as independent risk factors for ICU readmissions. The hypothesis of this study was attractive because it evaluated a specific group of patients, “oncohematological patients,” who were followed in multiple critical care units and had different characteristics from the typical medical and surgical ICU populations. However, by evaluating only the admission factors of their first ICU stay, Rodrigues et al. limited their results to a narrow window. They disregarded evaluating other more important risk factors, some of which were mentioned in the manuscript’s discussion section. When evaluating readmission to ICU for a specific patient group, Gajic et al.(2) categorized the possible risk factors into 3 groups. These were (1) first ICU admission characteristics, (2) physiological characteristics, laboratory abnormalities and severity of illness at the time of ICU discharge and (3) functional status and need for nursing interventions at the time of ICU discharge. According to this categorization, Rodrigues et al.(1) should also have presented the first ICU admission characteristics of their patient group. They identified the length of hospital stay before ICU transfer as an independent predictor of readmission but did not take into consideration the length of ICU stay or the length of MV during patients’ first ICU admission. They also did not mention the type of MV applied, whether invasive MV (IMV) or noninvasive MV. Another risk factor that was excluded was ventilator-associated pneumonia development during the first ICU stay. Other relevant risk factors in oncohematological patients include immune suppression status and infections associated with this immune suppression, such as cytomegalovirus infections, pneumocystis carinii pneumonia, Aspergillus infections or invasive candidiasis. These infections are challenging to eradicate, and if not properly eradicated, can lead to readmissions in this group of patients. Second of all, physiological characteristics, laboratory abnormalities and severity of illness at the time of ICU discharge were not evaluated as risk factors of readmission by Rodrigues et al.(1) In assessing readmissions, it is important Conflicts of interest: None.