Mechanical Ventilation for Children With Hypoxemic Respiratory Failure After Stem Cell Transplantation: Quo Vadis?*
Over the last 40 years, adults and children with severe life-threatening hypoxemic respiratory failure who receive mechanical ventilation in ICUs continue to have improved short- and long-term survival with mortality decreased from over 60% to under 30%. In 1981, Lyrene and Truog (1) described 60% mortality in children with adult respiratory distress syndrome (ARDS); two of six survivors had severe neurologic problems. A recent systematic review of children with ARDS showed a pooled mortality of 34% and noted that the mortality had not altered over the last 20 years; children with comorbidities such as sepsis, immunosuppression or cancer did worse, as did children with indirect lung injury. Importantly, there were substantial regional and geographical differences (2). Similarly, a Brazilian study showed a 43% mortality in children with ARDS and unsurprisingly, children with only single organ (respiratory) failure, did better with a 23% mortality. Also, children with extra pulmonary (indirect) ARDS did worse with 64% mortality, when compared with direct ARDS with 35% mortality (3). Many factors including actual cause of ARDS, severity of ARDS, and nature of comorbidities, systems of healthcare and attitudes of healthcare professionals also affect the outcome of children with ARDS. However, until recently two fundamental problems existed with all clinical reports on the outcome of children with ARDS, namely inconsistency over the definition of pediatric ARDS (PARDS) and the lack of clear stratification methods to measure the severity of disease. This has now been addressed with the publication of consensus recommendations from the Pediatric Acute Lung Injury Consensus Conference (PALICC) (4). Using an oxygenation index (OI), mild, moderate, and severe PARDS were defined as an OI of four to less than eight, eight to less than 16, and 16 or greater, respectively. The OI will be familiar to most neonatologists and pediatric intensivists as it was originally developed in the 1980s to quantify the severity of hypoxemic lung disease as a criterion for prediction of death, and hence the use of an experimental therapy called “extracorporeal membrane oxygenation” (ECMO) in those sickest patients who were considered “likely to die.”
The outcome of children with hematologic malignancy admitted to ICU has always varied depending on the reason for admission. In general, children are usually admitted for hypoxemic respiratory failure, sepsis/septic shock, encephalopathy, or recovery after a procedure that may require short-term ventilation or close observation. In 1988, in a study of the outcome of children with hematologic malignancy admitted to ICU, the mortality for these four groups was 75%, 75%, 48%, and 7%, respectively (5). The authors stated that ICU treatment should be improved, as patients with severe cardiorespiratory failure, at that stage, did not appear to benefit from ICU treatment. A group, from the Cleveland clinic, challenged this view (6) and stated that they had had (in the previous 13 mo) only three of 11 children with sepsis or interstitial pneumonitis die, and hence, the previous authors view was overly pessimistic. The initial authors restated that in their group of patients, 18 of 21 children with interstitial pneumonitis died and 18 of 20 children with septic shock and positive blood cultures died (7). Unfortunately, no information about the severity of illness, admission criteria, and treatments received by patients or the philosophy and attitudes about quality of life and hence ongoing care were given by the Cleveland group, thus rendering comparison of the outcomes very difficult.
Since then, there are numerous reports on the outcome of children with hematologic malignancy and hematopoietic stem cell transplant (HSCT) following ICU admission.