Mark J. Heulitt

Single-institution experience with interhospital extracorporeal membrane oxygenation transport: A descriptive study.

Objective: Patients with refractory cardiopulmonary failure may benefit from extracorporeal membrane oxygenation, but extracorporeal membrane oxygenation is not available in all medical centers. We report our institutions nearly 20-yr experience with interhospital extracorporeal membrane oxygenation transport. Design: Retrospective review. Setting: Quaternary care childrens hospital. Patients: All patients undergoing interhospital extracorporeal membrane oxygenation transport by the Arkansas Childrens Hospital extracorporeal membrane oxygenation team. Interventions: Data (age, weight, diagnosis, extracorporeal membrane oxygenation course, hospital course, mode of transport, and outcome) were obtained and compared with the most recent Extracorporeal Life Support Organization Registry report. Results: Interhospital extracorporeal membrane oxygenation transport was provided to 112 patients from 1990 to 2008. Eight were transferred between outside facilities (TAXI group); 104 were transported to our hospital (RETURN group). Transport was by helicopter (75%), ground (12.5%), and fixed wing (12.5%). No patient died during transport. Indications for extracorporeal membrane oxygenation in RETURN patients were cardiac failure in 46% (48 of 104), neonatal respiratory failure in 34% (35 of 104), and other respiratory failure in 20% (21 of 104). Overall survival from extracorporeal membrane oxygenation for the RETURN group was 71% (74 of 104); overall survival to discharge was 58% (61 of 104). Patients with cardiac failure had a 46% (22 of 48) rate of survival to discharge. Neonates with respiratory failure had an 80% (28 of 35) rate of survival to discharge. Other patients with respiratory failure had a 62% (13 of 21) rate of survival to discharge. None of these survival rates were statistically different from survival rates for in-house extracorporeal membrane oxygenation patients or for survival rates reported in the international Extracorporeal Life Support Organization Registry (p > .1 for all comparisons). Conclusions: Outcomes of patients transported by an experienced extracorporeal membrane oxygenation team to a busy extracorporeal membrane oxygenation center are very comparable to outcomes of nontransported extracorporeal membrane oxygenation patients as reported in the Extracorporeal Life Support Organization registry. As has been previously reported, interhospital extracorporeal membrane oxygenation transport is feasible and can be accomplished safely. Other experienced extracorporeal membrane oxygenation centers may want to consider developing interhospital extracorporeal membrane oxygenation transport capabilities to better serve patients in different geographic regions.

Reliability of displayed tidal volume in infants and children during dual-controlled ventilation

Objective: Previous studies have shown a significant difference between ventilator-measured tidal volume and actual-delivered tidal volume. However, these studies used external methods for measurement of compression volume. Our objective was to determine whether tidal volume could be accurately measured at the expiratory valve of a conventional ventilator using internal computer software to compensate for circuit compliance with a dual control mode of ventilation. Design: Clinical study during an 8-month period. Setting: Pediatric intensive care unit. Patients: All patients admitted to the pediatric intensive care unit during the enrollment period who were mechanically ventilated using the Servo I (Maquet, Bridgewater, NJ) were eligible for this study. Interventions: Patients were ventilated using a dual-control mode of ventilatory support and either an infant or adult circuit (with and without circuit compensation). Measurements and Main Results: Tidal volume measured at the endotracheal tube using a pneumotachometer was compared with ventilator-displayed tidal volume. Sixty-eight patients were studied between September 2004 and April 2005. Age range was 2 days to 18 yrs (median, 23 mos) and weight range was 2.3 kg to 103 kg (median, 14.5 kg) with 41 male patients (60%). We found ventilator-displayed tidal volume, without circuit compensation, generally overestimates true-delivered tidal volume and, with circuit compensation, generally underestimates true-delivered tidal volume. However, agreement between tidal volume measured at the patients airway and that measured with and without compensation for circuit compliance was good. The error in both cases, without and with circuit compensation, is relatively greater in infants and small children. Conclusions: There is an underestimation of delivered tidal volume when compensating for circuit volume loss measured at the ventilator. There is no improvement in measured tidal volume using circuit compensation in small infants and children.

Pharmacokinetics and Pharmacodynamics of Famotidine in Children

The pharmacokinetics and pharmacodynamics of intravenous famotidine were studied in 12 children (1.1–12.9 years of age; mean weight ± standard deviation = 27.6 ± 21.2 kg) who were given the drug for prophylactic management of stress ulceration. After a 0.5‐mg/kg infusion of famotidine, timed blood (n = 10) and urine (n = 6) samples and repeated evaluations of intragastric pH (n = 13) were obtained from each subject. Pharmacokinetic parameters were determined from curve fitting of serum concentration data. The mean (± SD) maximum serum concentration (Cmax) was 527.6 ± 281.2 ng/mL, the elimination half‐life (t1/2) was 3.2 ± 3.0 hours, and the apparent steady‐state volume of distribution (Vdss) was 2.4 ± 1.7 L/kg. Plasma clearance (Cl) and renal clearance (ClR) were 0.70 ± 0.34 L/hr/kg and 0.43 ± 0.24 L/hr/kg, respectively. Over 24 hours, 73.0 ± 27.3% of the dose was excreted unchanged in the urine (Fel). Pharmacodynamic analysis of gastric pH data using the sigmoid Emax model predicted that 50% of the maximal effect of famotidine (EC50) occurs at a serum concentration of 26.0 ± 13.2 ng/mL. Children who did not have an initial intragastric pH ≤4 did not have a significant response in pH after receiving famotidine. Although Vdss and Cl were higher in these children than those seen in adults, statistically significant relationships between these parameters and age were not observed in the study population. The pharmacodynamics and pharmacokinetics of famotidine in children older than one year of age appear to be similar to those noted in adults.

Infections during extracorporeal life support

Little data exist on the type of infections patients acquire during extracorporeal life support. Through a retrospective analysis of 109 patients who underwent 115 episodes of venoarterial extracorporeal life support, it was determined that nosocomial infections developed in 18 patients (16%). Patients with nosocomial infections were supported for longer periods of time (230 versus 140 hours; P < .05) and were more likely to have an open chest (P = .02) than those who did not have infectious complications. Blood-borne infections occurred most often while patients were cannulated for extracorporeal life support, with urinary tract and wound infections more commonly occurring after decannulation. Fungal organisms were isolated in 50% of nosocomial infections. Patients with blood or wound fungal infections had a higher case-fatality rate than those patients with bacterial complications (P = .03). Because it is unlikely that the duration of extracorporeal life support can be shortened significantly, the authors recommend an increased level of awareness of nosocomial infections in patients on prolonged extracorporeal life support. Further research is needed to assess the effects of antifungal prophylaxis or immune modulation to prevent nosocomial infections.

Pharmacokinetics and Pharmacodynamics of Ranitidine in Neonates Treated with Extracorporeal Membrane Oxygenation

The pharmacokinetics and pharmacodynamics of ranitidine were studied in 13 term neonates with stable renal and hepatic function who were treated with extracorporeal membrane oxygenation (ECMO). Ranitidine was initially administered as a single 2 mg/kg dose over 10 minutes and intragastric pH was monitored to determine response. Within 90 minutes after administration of ranitidine, intragastric pH for all of the patients whose initial reading was ≤ 4 had increased to > 5. Intragastric pH remained < 4 for a minimum of 15 hours. Mean ± 1 standard deviation elimination half‐life was 6.61 ± 2.75 hours, and 41.5 ± 22.2% of the single dose was eliminated in urine within 24 hours. Total plasma clearance of ranitidine correlated well with estimated glomerular filtration rate. Twenty‐four hours after the initial dose, a continuous infusion of ranitidine (2 mg/kg/24 hr) was started and continued for 72 hours or until ECMO was discontinued. Eleven patients completed 48 hours of continuous infusion and seven completed all 72 hours. Plasma clearance and elimination half‐life were determined from steady‐state plasma ranitidine concentrations 24, 46, and 72 hours after the start of the infusion. There were no significant differences in clearance between these intervals. These data suggest that for term neonates with stable renal and hepatic function, ranitidine does not need to be administered more frequently than every 12 hours. A continuous infusion of 2 mg/kg/24 hours maintained intragastric pH above 4 in more than 90% of our patients, and in our opinion is the preferred method for delivering ranitidine to term neonates undergoing ECMO who require H2 antagonists. Response to therapy should be monitored by repeated measurement of gastric pH and the dose should be adjusted accordingly.

Reliability of measured tidal volume in mechanically ventilated young pigs with normal lungs

ObjectiveThis study examined whether volumes can be accurately measured at the expiratory valve of a conventional ventilator using pressure support ventilation and positive end expiratory pressure with software compensation for circuit compliance available in the Servo ί ventilator.Design and settingComparison of two methods for measuring tidal volume in an animal laboratory.SubjectsTwenty healthy, intubated, sedated, spontaneously breathing pigs.InterventionsVolume was measured in ten neonatal-sized and ten pediatric-sized pigs ventilated with the Servo ί ventilator using pressure support ventilation and positive end expiratory pressure with and without circuit compliance compensation. We compared volume measured at the airway opening by pneumotachography to volume measured at the expiratory valve of a conventional ventilator.Measurements and resultsThe use of circuit compliance compensation significantly improved the agreement between the two volume methods in neonatal-sized piglets (concordance correlation coefficient: with circuit compliance compensation, 0.97; without, 0.87, p=0.002). In pediatric-sized pigs there was improvement in agreement between the two measurement methods due to circuit compliance compensation (concordance correlation coefficient with circuit compliance compensation, 0.97; without, 0.88, p=0.027). With circuit compliance compensation off there was positive bias: mean difference (bias) 2.97±0.12 in neonatal-sized and 3.75±0.38 in pediatric-sized pigs.ConclusionsOur results show that volume can be accurately measured at the expiratory valve of a conventional ventilator in neonatal- and pediatric-sized animals.

Neurally triggered breaths have reduced response time, work of breathing, and asynchrony compared with pneumatically triggered breaths in a recovering animal model of lung injury

Objective: Our objective was to compare response time, pressure time product as a reflection of work of breathing, and incidence and type of asynchrony in neurally vs. pneumatically triggered breaths in a spontaneously breathing animal model with resolving lung injury. Design: Prospective animal study. Setting: Experimental laboratory. Subjects: Male Yorkshire pigs. Interventions: Intubated, sedated pigs were ventilated using neurally adjusted ventilatory assist and pressure support ventilation with healthy and sick/recruited lungs. After injury, the lung was recruited using a computer-driven protocol. Respiratory mechanics were determined using a forced oscillation technique, and airway flow and pressure waveforms were acquired using a pneumotachograph. Measurements and Main Results: Waveforms were analyzed for trigger delay, pressure time product, and asynchrony. Trigger delay was defined as the time interval (ms) from initiation of a breath to the beginning of ventilator pressurization. Pressure time product was measured as the area of the pressure curve for animal effort (area A) and ventilator response (area B). Asynchrony was classified according to triggering problems, adequacy of flow delivery, and adequate breath termination. Mean values were compared using the Wilcoxon signed-ranks test (p < .05). Trigger delay (ms) was less in neurally triggered breaths (pressure support ventilation healthy 104 ± 27 vs. neurally adjusted ventilatory assist healthy 72 ± 30, pressure support ventilation sick/recruited 77 ± 18 vs. neurally adjusted ventilatory assist sick/recruited 38 ± 18, p < .01). Pressure time product areas A and B were decreased for neurally triggered breaths compared with pressure support ventilation in both healthy and recruited animals (p ⩽ .02). Overall, the percentage of asynchrony was less for neurally adjusted ventilatory assist breaths in the recruited animals (pressure support ventilation 27% and neurally adjusted ventilatory assist 6%). Conclusions: Neurally triggered breaths have reduced asynchrony, trigger delay, and pressure time product, which may indicate reduced work of breathing associated with less effort to trigger the ventilator and faster response to effort. Further study is required to demonstrate if these differences will lead to decreased days of ventilation and less use of sedation in patients.

Validation of a noninvasive blood pressure monitoring device in normotensive and hypertensive pediatric intensive care patients

Objective. To evaluate the performance and to define limitations of a noninvasive blood pressure monitoring device in the critically ill pediatric population. Method. Patients were included in the study if they were admitted to the Pediatric Intensive Care Unit, were between the ages of 1 month and 18 years with wrist circumferences of ≥ 10 cm, and had an indwelling arterial line. Patients were excluded if their systolic blood pressure differed by ≥ 7.5% between their upper extremities. The measurements were collected simultaneously with those from an arterial line by a computer interfaced with the noninvasive blood pressure monitoring system and the patient’s monitor. Heart rates were calculated from the recorded pulse waveforms of the arterial lines. Comparison analyses were performed via bias and precision plots of the blood pressure and heart rate data in addition to calculation of Pearson’s correlation coefficients and concordance correlation coefficients. As a nonparametric method of comparison, the proportion of measurements that differed by greater than 10% was calculated. Results. Blood pressures and heart rates of 20 patients between the ages of 12 months and 17 years were monitored by a noninvasive blood pressure monitor for 30 min per patient. This data collection resulted in 2015 data points for each blood pressure and heart rate for comparison of methods. Concordance correlation coefficients were the following: systolic blood pressure, 0.93; diastolic blood pressure, 0.93; mean blood pressure, 0.94; and heart rate, 0.85. Conclusions. The noninvasive blood pressure monitor is capable of producing an accurate blood pressure measurement every 12–15 heartbeats in addition to providing a pulse waveform and digital display of the heart rate. Our study showed good agreement between the methods in the normotensive and hypertensive critically ill pediatric population with a wrist circumference limitation defined at ≥ 11 cm.

Mechanical Ventilation in the Pediatric Cardiac Intensive Care Unit The Essentials

Ventilating a child or newborn in the postoperative course after repair of congenital heart disease requires a solid basic understanding of respiratory system mechanics (pressure–volume relationship of the respiratory system and the concept of its time constants) and cardiopulmonary physiology. Furthermore, careful attention has to be paid to avoid damaging the lungs by potentially injurious mechanical ventilation. Optimizing ventilator settings during controlled and assisted ventilation, allowing as early as possible for spontaneous ventilation by still assisting mechanically the patient’s respiratory efforts are important features for lung protection, for minimizing potential hemodynamic side effects of positive pressure ventilation, and for early weaning from mechanical ventilation. In the search for being less invasive, the use of noninvasive ventilation in the cardiac intensive care setting is rapidly increasing despite still lacking evidence of its theoretical superiority and requires good knowledge of specific techniques and equipment available for this approach in this setting. This review will address many of these aspects and highlight the essentials to be known when ventilating a child in the Cardiac Intensive Care Unit (CICU).

Comparison of total resistive work of breathing in two generations of ventilators in an animal model

Spontaneous breathing through an endotracheal tube and ventilator circuit is associated with an increased work of breathing (WOB). Recently, pediatric ventilators have introduced improved features to optimize patient‐ventilator interactions. We performed an experiment utilizing an animal model to compare total resistive WOB of two widely used ventilators, the Siemens Servo Ventilator 300 (SV300) with patient‐optimized features, such as flow‐triggering and rapid response time, and the Siemens 900C (S900C) without those features. A total of 120 experiments of 10 minutes duration each were performed in 6 anesthetized, intubated lambs. In each experiment, the animal was randomized to either pressure support ventilation (PSV) of 5 cm H2O, or continuous positive airway pressure (CPAP) with 0 cmH2O end expiratory pressure (ZEEP) while supported by the SV300 or the S900C. Each animal was used as its own control. WOB was measured with a Bicore monitoring device as WOB of the animal (WOBp), WOB of the ventilator (WOBv), and the pressure time product (PTP) for each breath during the experiment. Oxygen consumption (VO2) of the animal was measured using breath‐by‐breath gas analysis with a customized metabolic monitoring system. A Wilcoxon signed rank sum test was used for analysis. All comparisons between the ventilators for both CPAP and PSV showed a statistically significant difference (p < 0.001). WOBp was reduced by 47% during pressure support ventilation (PSV) and by 47% during CPAP when the SV300 was used compared to the S900C. We conclude that WOB is significantly lower in animals ventilated with the SV300 than with the S900C ventilator, and we speculate that ventilators with the features of the SV300 may offer advantages in ventilating pediatric patients. Pediatr Pulmonol. 1996; 22:58–66.