Protocols to Improve the Care of Critically Ill Pediatric and Adult Patients

In addition to the celebrated technological and pharmaceutical advances in intensive care, an unassuming type of research testing the use of protocols in care has yielded some of the most important improvements in patient mortality and morbidity, and in the costs of critical illness. The basic concept is that routine clinical care is enhanced when interdisciplinary teams of health professionals use evidence-based protocols to complement their clinical judgment.¹ Perhaps the greatest value obtained from the use of protocols is that of reducing unnecessary variations in intensive care practice. Positive randomized trials now support the role of protocols for critical care interventions such as lung-protective ventilation,² weaning from mechanical ventilation,^{3 - 8} transfusion of blood products,⁹ sepsis resuscitation,¹⁰ glycemic control,¹¹ and sedation and analgesia.^{12 - 13}

The benefit of protocols has been shown most consistently in approaches for weaning from mechanical ventilation. A recent review of the literature found that 4 randomized trials and 11 additional controlled studies have shown favorable outcomes such as a reduced duration of mechanical ventilation and an increased rate of successful extubation in critically ill adults.¹⁴ Similar findings have been reported among critically ill children randomly assigned to protocol-directed vs physician-directed weaning.¹⁵ Among the larger trials of weaning from mechanical ventilation (N≥300),^{5 - 7 ,15} these consistent effects were both statistically significant and clinically important.

The implications of protocols for improving the care of critically ill patients appear substantial: reduced time receiving mechanical ventilation, shorter lengths of intensive care unit (ICU) and hospital stay; and fewer nosocomial infections. One study showed that the cost of ICU care was reduced by more than $5000 per patient among those treated with a weaning protocol incorporating daily spontaneous breathing trials driven by respiratory care and nursing members of the ICU team, compared with those weaned without a protocol.⁵ Furthermore, duration of mechanical ventilation is an independent contributor to mortality.¹⁶ The solid foundation of evidence in this area has prompted many multidisciplinary ICUs to incorporate weaning protocols into clinical practice.

In this issue of THE JOURNAL, Randolph and colleagues¹⁷ describe a second randomized trial of weaning protocols in critically ill children. The primary goals of this multicenter trial from the Pediatric Acute Lung Injury and Sepsis Investigators (PALISI) Network were to compare the use of weaning protocols with standard care and to compare a pressure-support weaning protocol with another protocol using volume support. A secondary goal was to study the relationship between weaning outcomes and sedation. Strengths of this trial include concealed randomization, a representative population of critically ill children from multiple pediatric ICUs, the inclusion of a "standard care" group, complete follow-up, and an intention-to-treat analysis. After the first 182 children were randomized, the study was stopped due to lack of apparent differences among the groups with respect to any weaning outcomes. In the face of these unexpectedly negative findings, the investigators examined the overall trial data and found that increased sedative use during the first 24 hours of weaning (the only time during which these data were collected) was an important predictor of weaning duration (P<.001) and weaning failure (P = .04).

The authors concluded that the weaning protocol did not significantly decrease the relatively short duration of ventilation in infants and children, but that sedative use was the major factor affecting duration of weaning in this study. Conceivably, participating institutions may already be sufficiently well-resourced and evidence-based in their weaning practices such that weaning protocols have less to offer than in other centers. There are other alternative and likely explanations for these findings. As the authors suggest, perhaps the most relevant issue related to the protocol is that specific weaning modes and weaning criteria may be less important than incorporating routine scheduled assessments (ie, a daily spontaneous breathing trial) to assess patients' readiness for weaning and for extubation.⁸ In addition, the sample size, diverse patient ages (ranging from neonates to adolescents), and limited adherence to protocols (66% adhered fully) might have reduced the ability to detect a true benefit associated with one or both of the experimental weaning strategies.

The study results relating sedation use to weaning outcomes underscore the importance of recognizing when a patient is ready to wean. The impact of excessive sedation on weaning in adults is clearly established. Brook et al¹³ found that a nursing-implemented protocol for targeted sedation reduced sedation use, duration of ventilation, need for tracheostomy and duration of ICU stay by nearly 2 days (P = .01) among patients with respiratory failure. Kress et al¹² found that daily awakening of mechanically ventilated patients through interruption of sedation reduced the duration of ventilation and ICU stay by 3.5 days (P = .02). In addition, these investigators found that patients receiving daily interruption of sedative administration had fewer cranial computed tomography scans performed to evaluate persistent cognitive deficits.

The next major advance in the area of weaning from mechanical ventilation may arise from clinical trials of interventions focused on interactions between the lungs and the brain. Cognitive impairment has been associated with increased rates of ventilator-associated pneumonia and failed extubation.^{18 - 20} Recent data on the prevalence of delirium in mechanically ventilated patients, derived from newly validated delirium assessment instruments for intubated patients,^{21 - 23} have shown that its incidence may be as high as 82%. By altering the daily patterns and dosages of psychoactive drug delivery via goal-directed sedation protocols, in combination with weaning protocols, significant improvements may be achieved in the outcomes of pediatric and adult patients receiving mechanical ventilation, including their long-term cognitive recovery.

Protocol use is not a synonym for "cookbook medicine." Protocols for weaning from ventilator support should not be so rigid that the safety and comfort of patients assume secondary importance. Nor should detailed protocols be expected to transfer smoothly from one ICU to another, or from a research venue to general practice. The time, organization, and administrative resources required to successfully implement and sustain use of weaning protocols should not be underestimated.²⁴ Moreover, weaning protocols should be customized to each setting, incorporating 2 patient-centered principles: (1) minimize sedation to facilitate safe awakening, and (2) routinely assess ability to breathe spontaneously and to breathe without support.

In summary, several protocols have been shown to improve the practice of critical care for children and adults, and have translated into improved patient outcomes. Findings from further pediatric trials may highlight important differences in therapeutic responses between critically ill children and adults. Further research may reveal that some of the relationships among sedation, weaning protocols, and patient outcomes differ in important ways between children and adults. Multicenter, collaborative trials in pediatric ICUs are key to advancing knowledge in this field. Thus, the PALISI Network Trial of weaning protocols in pediatric ICUs represents an important milestone for the care of the sickest children and infants.