The authors reply
Our study (2) was primarily focused on the relationship between measures of respiratory muscle weakness and extubation outcomes and was not specifically looking for risk factors for respiratory muscle weakness. Nevertheless, a crucial next step would be to identify the independent risk factors for respiratory muscle weakness in children and develop strategies to prevent it. Mechanisms implicated in acquired respiratory muscle weakness (i.e., ventilator-induced diaphragm dysfunction) relate to the underlying disease status of the patient, the severity of inflammation, the use of therapies like neuromuscular blockade and corticosteroids, the degree of protein catabolism, the nutritional status of the patient, and the degree of diaphragm contraction during mechanical ventilation (3–9). However, the adaptations of malnutrition on diaphragmatic function are unclear and may be disparate based on whether the malnutrition is acute or chronic. In rodent models of acute malnutrition, there is loss of cross-sectional area but preserved diaphragm contractility and improved fatigue resistance with restoration of cross-sectional area with refeeding. In contrast, in human cystic fibrosis patients with chronic malnutrition, there is a clear association between lower body mass index (BMI) (or weight for height for children < 2 yr) z score and decreased diaphragmatic strength (10–12). As such, the nutritional status of the patient upon initiation of mechanical ventilation, as well as acquired nutritional deficiencies or electrolyte abnormalities during the course of mechanical ventilation, may be important in the pathogenesis or preservation of diaphragmatic function. However, the development of ventilator-induced diaphragm dysfunction is likely multifactorial, and the effects of individual risk factors may be additive or multiplicative.
Unfortunately, we did not specifically track electrolytes in our study (2), so we do not have phosphorous levels. However, on secondary analysis, we found that children who had weight for height z score (for children < 2) or BMI z score (for children > 2), two SDs above or below normal (i.e. chronic malnutrition or obesity) had a nonsignificant trend for higher rates of reintubation (p = 0.07) (Table 1). However, this relationship did not appear to be from low aPiMax, as a higher proportion of obese or malnourished children maintained aPiMax greater than 30 cm H2O (the cut point we found to be most associated with extubation failure). Although cardiac surgery patients may have more swings in their electrolytes, patients undergoing cardiac surgery were not at higher risk for reintubation (p = 0.8), although there was a trend for these patients to have lower aPiMax (p = 0.06). However, because there is a larger proportion of neonates undergoing cardiac surgery, after controlling for age, this relationship was no longer present (p = 0.18).
Accumulating data indicate that greater than 50% of critically ill adults who are on mechanical ventilation greater than 72 hours have thinning of the diaphragm (based on ultrasound) within the first few days of mechanical ventilation (5), and there is a dose-response relationship between diaphragm atrophy and increasing ventilator driving pressure (5).