Validation of the CAS neonatal NIRS system by monitoring vv-ECMO patients: preliminary results.

The CAS neonatal NIRS system determines absolute regional brain tissue oxygen saturation (SnO2) and brain true venous oxygen saturation (SnvO2) non-invasively. Since NIRS-interrogated tissue contains both arterial and venous blood from arterioles, venules, and capillaries, SnO2 is a mixed oxygen saturation parameter, having values between arterial oxygen saturation (SaO2) and cerebral venous oxygen saturation (SvO2). To determine a reference for SnO2, the relative contribution of SvO2 to SaO2 drawn from a brain venous site vs. systemic SaO2 is approximately 70:30 (SvO2:SaO2). If the relationship of the relative average contribution of SvO2 and SaO2 is known and does not change to a large degree, then NIRS true venous oxygen saturation, SnvO2, can be determined non-invasively using SnO2 along with SaO2 from a pulse oximeter.

Systemic hypertension associated with venovenous extracorporeal membrane oxygenation for pediatric respiratory failure.

BACKGROUND/PURPOSE: Arterial hypertension (HTN) is common in neonates on venoarterial (VA) extracorporeal membrane oxygenation (ECMO), but HTN in pediatric venovenous (VV) ECMO has not been well described. The authors noted HTN in their VV ECMO experience and hypothesized that HTN was associated with fluid status, steroid use, and renal insufficiency. METHODS: Records of 50 patients receiving VV ECMO for respiratory failure were reviewed. HTN was defined as systolic blood pressure greater than 95th percentile for age for > or =1 hour, unresponsive to sedation/analgesia. Hypertensive index (HI) is defined as total hypertensive hours per total ECMO hours. Fluid status was estimated by a fluid index (FI = total fluid balance during ECMO per ECMO hours per weight). RESULTS: Forty-seven of 50 patients (94%) had HTN. Median HI was 0.21 (range, 0.01 to 1.0). Thirteen patients had renal insufficiency, 39 received steroids, and 23 received continuous venovenous hemofiltration (CVVH). There was no association between HI and FI, steroid use, or renal insufficiency. Thirty-three patients were treated for HTN, often requiring multiple agents. Bleeding complicated the course of 18 patients, and HI was significantly higher in those patients (P = .03). HI was not different between survivors (37 of 39 with HTN) and nonsurvivors (10 of 11 with HTN). CONCLUSIONS: Hypertension is a common complication associated with VV ECMO with unclear etiology. HTN was frequently difficult to control. This study emphasizes the need for the development of treatment protocols to decrease the incidence, severity, and associated morbidity. Improved insight into the etiology of HTN associated with pediatric VV ECMO, including evaluation of the renin-angiotensin system, would help guide therapy.

Fluid overload before continuous hemofiltration and survival in critically ill children: a retrospective analysis.

OBJECTIVE: Continuous venovenous hemofiltration (CVVH) is used for renal replacement and fluid management in critically ill children. A previous small study suggested that survival was associated with less percent fluid overload (%FO) in the intensive care unit (ICU) before hemofiltration. We reviewed our experience with a large series of pediatric CVVH patients to evaluate factors associated with outcome. DESIGN: Retrospective chart review. SETTING: Tertiary children's hospital. PATIENTS: CVVH pediatric ICU patients from November 1997 to January 2003. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: %FO was defined as total fluid input minus output (up to 7 days before CVVH for both hospital stay and ICU stay) divided by body weight. One hundred thirteen patients received CVVH; 69 survived (61%). Multiple organ dysfunction syndrome (MODS) was present in 103 patients; 59 survived (57%). Median patient age was 9.6 yrs (25th, 75th percentile: 2.5, 14.3). Median %FO was significantly lower in survivors vs. nonsurvivors for all patients (7.8% [2.0, 16.7] vs. 15.1% [4.9, 25.9]; p =.02] and in patients with > or =3-organ MODS (9.2% [5.1,16.7] vs. 15.5% [8.3, 28.6]; p =.01). The Pediatric Risk of Mortality Score III at CVVH initiation also was associated with survival in these groups, but by multivariate analysis, %FO was independently associated with survival in patients with > or =3-organ MODS (p =.01). CONCLUSIONS: Survival in critically ill children receiving CVVH in this large series was higher than in previous reports. CVVH survival may be associated with less %FO in patients with > or =3-organ MODS. Prospective studies are necessary to determine whether earlier use of CVVH to control fluid overload in critically ill children can improve survival.

Primary use of the venovenous approach for extracorporeal membrane oxygenation in pediatric acute respiratory failure.

OBJECTIVES: To describe a single center's experience with the primary use of venovenous cannulation for supporting pediatric acute respiratory failure patients with extracorporeal membrane oxygenation (ECMO). DESIGN: Retrospective chart review of all patients receiving extracorporeal life support at a single institution. SETTING: Pediatric intensive care unit at a tertiary care children's hospital. PATIENTS: Eighty-two patients between the ages of 2 wks and 18 yrs with severe acute respiratory failure. INTERVENTIONS: ECMO for acute respiratory failure. MEASUREMENTS AND MAIN RESULTS: From January 1991 until April 2002, 82 pediatric patients with acute respiratory failure were cannulated for ECMO support. Median duration of ventilation before ECMO was 5 days (range, 1-17 days). Sixty-eight of these patients (82%) initially were placed on venovenous ECMO. Fourteen patients were initiated and remained on venoarterial support, including six in whom venovenous cannulae could not be placed. One patient was converted from venovenous to venoarterial support due to inadequate oxygenation. Venoarterial patients had significantly greater alveolar-arterial oxygen gradients and lower PaO(2)/FIO(2) ratios than venovenous patients (p <.03). Fifty-five of 81 venovenous patients received additional drainage cannulae (46 of 55 with an internal jugular cephalad catheter). Thirty-five percent of venovenous patients and 36% of venoarterial patients required at least one vasopressor infusion at time of cannulation (p = nonsignificant); vasopressor dependence decreased over the course of ECMO in both groups. Median duration on venovenous ECMO for acute hypoxemic respiratory failure was 218 hrs (range, 24-921). Venovenous ECMO survivors remained cannulated for significantly shorter time than nonsurvivors did (median, 212 vs. 350 hrs; p =.04). Sixty-three of 82 ECMO (77%) patients survived to discharge-56 of 68 venovenous ECMO (81%) and nine of 14 venoarterial ECMO (64%). CONCLUSIONS: Venovenous ECMO can effectively provide adequate oxygenation for pediatric patients with severe acute respiratory failure receiving ECMO support. Additional cannulae placed at the initiation of venovenous ECMO could be beneficial in achieving flow rates necessary for adequate oxygenation and lung rest.