Chiara Giorni

Fenoldopam in newborn patients undergoing cardiopulmonary bypass: controlled clinical trial

We determined if low dose fenoldopam in neonates already receiving conventional diuretics improves urine output, fluid balance, acute kidney injury incidence (AKI) and time to extubation. A prospective controlled clinical trial in a pediatric cardiac intensive care unit on 40 neonates undergoing cardiac surgery with cardiopulmonary bypass, excluding simple ventricular septal defect and atrial septal defect. Fenoldopam was infused at a low dose of 0.1 microg/kg/min soon after anesthesia induction and infusion prolonged for 72 h in 20 patients. Twenty neonates with standardized perioperative therapy except fenoldopam administration served as controls. Demographic, hemodynamic, daily urine output, creatinine, creatinine clearance, serum and urinary sodium and potassium were recorded. Inotropic score (IS) was calculated as a surrogate for the degree of hemodynamic impairment. Low dose fenoldopam infusion did not show beneficial effects in renal function. The treatment did not significantly affect IS value, AKI incidence, fluid balance control, time to sternal closure, time to extubation and time to intensive care unit discharge. Low dose fenoldopam in neonates undergoing cardiac surgery with CPB did not produce effects on urine output, fluid balance and AKI incidence. Fenoldopam was well tolerated and did not negatively affect hemodynamics and vasopressor support.

Renal replacement therapy in neonates with congenital heart disease.

BACKGROUND The acute renal failure (ARF) incidence in pediatric cardiac surgery intensive care unit (ICU) ranges from 5 to 20% of patients. In particular, clinical features of neonatal ARF are mostly represented by fluid retention, anasarca and only slight creatinine increase; this is the reason why medical strategies to prevent and manage ARF have limited efficacy and early optimization of renal replacement therapy (RRT) plays a key role in the outcome of cardiopathic patients. METHODS Data on neonates admitted to our ICU were prospectively collected over a 6-month period and analysis of patients with ARF analyzed. Indications for RRT were oligoanuria (urine output less than 0.5 ml/kg/h for more than 4 h) and/or a need for additional ultrafiltration in edematous patients despite aggressive diuretic therapy. RESULTS Incidence of ARF and need for RRT were equivalent and occurred in 10% of admitted neonates. Eleven patients of 12 were treated by peritoneal dialysis (PD) as only RRT strategy. PD allowed ultrafiltration to range between 5 and 20 ml/h with a negative balance of up to 200 ml over 24 h. Creatinine clearance achieved by PD ranged from 2 to 10 ml/min/1.73 m2. We reported a 16% mortality in RRT patients. CONCLUSION PD is a safe and adequate strategy to support ARF in neonates with congenital heart disease. Fluid balance control is easily optimized by this therapy whereas solute control reaches acceptable levels.

Cerebral NIRS as a marker of superior vena cava oxygen saturation in neonates with congenital heart disease

Objectives:  To investigate the correlation between cerebral near‐infrared spectroscopy (NIRS) (rSO2c) and superior vena cava venous oxygen saturation (ScvO2) in newborn patients with congenital heart disease (CHD).

Role of vasopressin in the treatment of anaphylactic shock in a child undergoing surgery for congenital heart disease: a case report

IntroductionThe incidence of anaphylactic reactions during anesthesia is between 1:5000 and 1:25000 and it is one of the few causes of mortality directly related to general anesthesia. The most important requirements in the treatment of this clinical condition are early diagnosis and maintenance of vital organ perfusion. Epinephrine administration is generally considered as the first line treatment of anaphylactic reactions. However, recently, new pharmacological approaches have been described in the treatment of different forms of vasoplegic shock.Case presentationWe describe the case of a child who was undergoing surgery for ventricular septal defect, with an anaphylactic reaction to heparin that was refractory to epinephrine infusion and was effectively treated by low dose vasopressin infusion.ConclusionIn case of anaphylactic shock, continuous infusion of low-dose vasopressin might be considered after inadequate response to epinephrine, fluid resuscitation and corticosteroid administration.

Continuous hemofiltration dose calculation in a newborn patient with congenital heart disease and preoperative renal failure

Objective To report a case of a newborn patient with renal failure due to polycystic kidneys requiring renal replacement therapy, and total anomalous pulmonary venous return requiring major cardiosurgical intervention. Setting Pediatric cardiosurgery operatory room and pediatric cardiologic intensive care. Patient: A 6-day-old newborn child weighing 3.1 kg. Results Renal function (creatinine value and urine output) was monitored during the course of the operation and intraoperative renal replacement therapy was not initiated. Serum creatinine concentration decreased from 4.4 to 3 mg/dL at cardiopulmonary bypass (CPB) start and to 1.5 at the end of surgery: the creatinine decrease was provided by the dilutional effect of CPB priming and the infusion of fresh blood from transfusions together with an adequate filtration rate (800 m/L in about 120 minutes). After the operation, extracorporeal membrane oxygenation (ECMO) for ventricular dysfunction and continuous hemofiltration for anuria refractory to medical therapy were prescribed. The hemofiltration machine was set in parallel with the ECMO machine at a blood flow rate of 60 ml/min and a predilution replacement solution infusion of 600 ml/h (4.5 ml/min of creatinine clearance once adjusted on extracorporeal circuits; 3000 mL/m2 hemofiltration): after a single hemofiltration session lasting 96 hours, serum creatinine reached optimal steady state levels around 0.5 mg/dL on postoperative day 2 and 3. Conclusion Administration of intraoperative continuous hemofiltration is not mandatory in the case of a 3-kg newborn patient with established renal failure needing major cardiosurgery: hemodilution secondary to CPB, transfusion of hemoderivates, and optimal UF rate appear to be effective methods for achieving solute removal. If postoperative continuous hemofiltration is started, however, a “dialytic dose” of 4.5 ml/min allows an adequate creatinine clearance, quick achievement of a steady state of serum creatinine concentration and an eventual acceptable rate of inflammatory mediator removal.

Initial experience with levosimendan infusion for preoperative management of hypoplastic left heart syndrome.

Hypoplastic left heart syndrome (HLHS) in the neonatal period is characterized by pulmonary overflow coupled with systemic hypoperfusion resulting in myocardial dysfunction, multiorgan failure, and severe metabolic derangement. This condition requires that the patient be stabilized by timely medical management before surgical palliation. The safety and efficacy of levosimendan were evaluated with six neonates affected by HLHS who had clinical signs of impending pulmonary overflow/systemic hypoperfusion, defined as tachypnea ([50 breaths/min), tachycardia (heart rate, [180 beats/min), hepatomegaly, central/toe temperature gradient exceeding 10 C, and lactate levels higher than 2 mmol/l. Levosimendan is a novel inodilator agent belonging to the family of calcium sensitizer agents with documented efficacy in treating adult congestive heart failure [2]. To date, few data exist on its use for pediatric patients [4], and no data exist on its use for HLHS neonates. The institutional review board of our hospital approved the use of levosimendan for such a cohort of patients. If all predefined signs of systemic hypoperfusion remained evident for more than 4 h after initial treatment (intravenous furosemide 1 mg/kg, packed red blood cells transfusions targeting a hematocrit level higher than 45%, children warming up to a toe temperature higher than 30 C), the patients were proactively sedated and intubated. Initial ventilator settings, with an inspired oxygen fraction (FiO2) of 30%, aimed to maintain normocapnia (partial pressure of carbon dioxide in arterial gas (PaCO2), 40– 45 mmHg). A central venous catheter was placed in the superior vena cava (SVC), and levosimendan infusion at 0.1 lg/kg/min was administered as the sole inotropic agent. All patients were receiving prostaglandin E1 infusion at 0.01 lg/kg/min since birth for ductal patency. Data are expressed as mean ± standard deviation. The Mann–Whitney test was used to compare means. A p value less than 0.05 was considered significant. The mean patient age at the time of intubation was 2.2 ± 0.5 days. All the patients received a classic Norwood procedure with a Blalock-Taussig shunt after a mean levosimendan administration time of 22 ± 8 h, from start of infusion to initiation of surgery. Lactate levels decreased from 4.22 ± 2.5 to 2.1 ± 0.4 mmol/l (p \ 0.05). Base excess increased from –1.92 ± 4 to 3.5 ± 3 mmol/l (p \ 0.05). The SVC oxygen saturation/systemic saturation (a-vO2) gradient decreased from 41.4% ± 12% to 29% ± 5% (p \ 0.05). Cerebral near infrared spectroscopy (NIRS) saturation improved from 57.8% ± 15.8% to 69.2% ± 7% (p \ 0.05). Central body temperature remained constant between 36 ± 1.5 C and 36.5 ± 0.5 C, whereas peripheral temperature increased significantly from 25.5 ± 1.4 C to 30.4 ± 0.4 C (p \ 0.05). The pulmonary-to-systemic flow ratio (Qp/Qs) was calculated according to the following formula: SatO2 – SvO2/ 99 –SatO2), where SatO2 and SvO2 are the arterial and SVC oxygen saturations, respectively. The pulmonary vein oxygen saturation was assumed to be 99%. The Qp/Qs decreased from 3.8 ± 1.2 to 2.1 ± 0.34 (p \ 0.05). Heart L. Di Chiara Z. Ricci (&) C. Garisto S. Morelli C. Giorni V. Vitale S. Picardo Division of Pediatric Cardiac Anesthesia/Intensive Care Unit, Department of Pediatric Cardiology and Cardiac Surgery, Bambino Gesù Pediatric Hospital, Piazza S.Onofrio, 00100 Rome, Italy e-mail: z.ricci@libero.it

Neurally adjusted ventilatory assist and lung transplant in a child: A case report.

Objective: To report the successful application of neurally adjusted ventilatory assist to a child with cystic fibrosis who underwent single-lung transplantation. Design: Case report. Setting: Pediatric cardiac intensive care unit. Patient: A 15-yr-old male with cystic fibrosis was admitted to our pediatric cardiac intensive care unit after single-lung transplantation. The child had previously received two bowel resections at the age of 1 yr, right pneumonectomy at the age of 3 yrs, and endoscopic percutaneus gastrostomy at the age of 10 yrs. After transplant, the child failed several attempts of weaning off mechanical ventilation with pressure-support ventilation, due to infection, pneumothorax, and ventilator asynchrony that caused gastric distension and numerous episodes of nausea and vomiting. Intervention: Use of neurally adjusted ventilatory assist to avoid patient-ventilator dyssynchrony and consequent gastric distension. Conclusions: The utilization of neurally adjusted ventilatory assist allowed to limit the risk of overassistance and prevent patient-ventilator asynchrony and to successfully wean the child off mechanical ventilation after single-lung transplant.

Initial Single-Center Experience With Levosimendan Infusion for Perioperative Management of Univentricular Heart With Ductal-Dependent Systemic Circulation:

The aim of this study was to evaluate the safety and the efficacy of levosimendan, a novel calcium sensitizer agent, on postoperative hemodynamic and metabolic parameters of neonates affected by single ventricle anatomy. Twenty consecutive neonates scheduled for the Norwood procedure with Blalock Taussig shunt were prospectively enrolled. All patients received an infusion of levosimendan at 0.1 μg/kg/min commencing 24 hours before surgery, and the infusion was continued for 48 hours after surgery. No side effects (intolerance to the drug, hypotension, arrhythmias) were shown. A median inotropic score (IS) of 37 was necessary to maintain a mean arterial pressure between 45 and 50 mm Hg at intensive care unit (ICU) admission: IS was significantly reduced after 72 hours (P < .05). Brain natriuretic peptide values decreased significantly from 1210 to 459 pg/mL in 72 hours (P < .05). Median SvO2 increased significantly from 38% to 59% during the evaluated period (P < .05). Cerebral near-infrared spectroscopy values were close to 40% at ICU admission with a significant stable increase to 50% after 12 hours (P < .05). Median lactate level was 13 mmol/L at ICU admission but showed a trend to a rapid and significant decrease after 12 hours (P < .05). Median urine output was surprisingly elevated, always remaining between 5.2 and 6.2 mL/kg/h throughout the postoperative period. Survival rate was 85% at 30 days (17/20 patients) and 75% (15/20) at hospital discharge. Levosimendan infusion in a cohort of neonates with univentricular anatomy was safe and potentially beneficial on postoperative hemodynamic and metabolic parameters.

Bilateral cerebral near infrared spectroscopy monitoring during surgery for neonatal coarctation of the aorta

SIR—Neurologic complications reported as a consequence of neonatal aortic coarctation repair are rare and limited to transient paralysis of the vocal cords, Horner’s syndrome, diaphragmatic paralysis and spinal cord ischemia (1,2). Despite the possibility of the involvement of the left carotid artery in aortic cross-clamping major cerebral complications have never been reported; in fact left cerebral perfusion is generally believed to be sufficiently maintained by the right carotid and vertebral arteries through the circle of Willis. Near-infrared spectroscopy (NIRS) is a noninvasive optical technique used to monitor venous wedged brain tissue oxygenation (rSO2i) (3). We present a case of major bilateral ischemic brain injury as a complication of a neonatal coartaction repair detected by intraoperative bilateral cerebral NIRS monitoring. A 4-day old, 1.9 kg premature baby with gestational age of 35 weeks underwent surgical repair for severe isthmus coartaction and tubular hypoplasia of the trasverse aortic arch. Additional echocardiographic finding was an anomalous right subclavian artery (ARSA) originating from the descending aorta. The patient was given balanced anesthesia with i.v. fentanyl (10 lgÆkgÆh), vecuronium (0.2 mgÆkgÆh), and sevoflurane. Monitoring consisted of ECG, central venous pressure, SpO2, endtidal CO2, and cerebral saturation of both hemispheres by NIRS technology (Somanetics INVOS 5100 Inc. Troy, MI, USA). Systemic perfusion pressure was monitored via both right radial and temporal arteries (4). Basal values of cerebral saturation did not show a significant saturation difference between the hemispheres (right rSO2i 75%, left rSO2i 80%). Intraoperatively, the surgeon identified and ligated the ARSA which had previously mandated the right temporal artery cannulation. The right rSO2i dropped from 70% to 55% immediately after ARSA legation. The surgeon applied the proximal clamp very close to the right carotid artery to complete an extended end-to-end repair and clamped both the left subclavian and the left common carotid artery, while the distal clamp was applied beyond the coarctation. Unexpectedly, the left rSO2i value dropped, reaching 30% in 20 min and cerebral protection with phenobarbital (20 mgÆkg) was promptly administered. Clamping time was prolonged to 92 min because of a severe friability of the aortic tissue and progressive disruption of the end-to-end anastomosis. Low cerebral saturation lasted for the entire clamping time, but the right rSO2i remained essentially unchanged (55%). Both left and right rSO2i returned to baseline values after the release of the aortic clamps (Figure 1). After surgery the newborn was transferred to ICU with stable hemodynamics. The postoperative course was characterized by clinical seizures and coma. Brain damage was confirmed on postoperative day 10 by MRI that showed bilateral ischemic damage of the caudate, putamen and the anterior part of the mesencephalus. After 30 days the baby showed no sign of neurological recovery and was then transferred to a rehabilitation unit; Glasgow Coma Score was five. Neonatal coarctation of the aorta is defined as narrowing of the upper thoracic aorta caused by juxtaductal posterior infolding or indentation. Periductal coartation can coexist, in some cases, with various degrees of tubular hypoplasia affecting the isthmus, the distal and even the proximal aortic arch. Surgical procedure can be limited to the removal of juxtaductal narrowing and ductal tissue, or may necessitate an extended aortic arch reconstruction: in this case the proximal clamp was applied across the left subclavian, the left common carotid and part of the brachiocephalic artery. Reported neurologic complications are limited to transient paralysis of the vocal cords, Horner’s syndrome, diaphragmatic paralysis, and spinal cord ischemia, the latter being a disastrous complication resulting in paraplegia in about 0.5% of patients (1,2). To our knowledge, severe cerebral complications have never been reported as a direct consequence of aortic coarctation repair in a neonate. Despite the possibility of left carotid artery involvement in aortic Figure 1 Cerebral oxygen saturation (a) abrupt desaturation in the right hemisphere after ligation of the anomalous right subclavian artery. (b) Desaturation of the left hemisphere when the proximal cross-clamp was placed. (c) Both saturations return to baseline values after the release of the cross-clamps. 906 CORRESPONDENCE

The usefulness of near-infrared spectroscopy for detecting and monitoring status epilepticus after pediatric cardiac surgery.

HILDREN REQUIRING cardiac intensive care are at particular risk for neurologic injury, and seizures are their most common manifestations. Postoperative seizures have been reported with an incidence ranging between 4% and 15% 1-3 ; they are likely underestimated because of frequent subtle clinical expression caused by paralyzing drugs and the lack of routinely prolonged postoperative electroencephalographic (EEG) monitoring. Recently, near-infrared spectroscopy (NIRS) cerebral monitoring has been used increasingly in pediatric cardiac intensive care for assessing the adequacy of cerebral perfusion. 4,5 The authors report a case in which this technique showed its effectiveness in detecting status epilepticus (SE) occurring after cardiac surgery and characterized by the absence of any other clinical manifestations. CASE REPORT A 78-day-old female infant weighing 4 kg developed early postoperative status epilepticus after coronary stenosis correction. The patient was born with transposition of the great arteries, unique coronary ostium, and intramural coronary arteries. At the age of 19 days, an arterial switch was performed, with an uneventful postoperative course and regular discharge. At the age of 78 days, she was readmitted because of a low-cardiac-output syndrome. Echocardiographic evaluation showed severely depressed biventricular function with left ventricular ejection fraction of about 30%. Her clinical condition required immediate admission to the pediatric cardiac intensive care unit (PCICU), where she suffered from several episodes of hypotension needing mechanical ventilation and inotropic support. The patient underwent cardiac catheterization, and severe coronary ostial stenosis of both coronary arteries was diagnosed. The aortic angiography was complicated by cardiac arrest, which required prolonged cardiopulmonary resuscitation consisting of 70 minutes of external cardiac massage and 3 defibrillations at 20 J before cardiovascular function was restored. Inotropic support was administered (epinephrine, 0.3 g/kg/ min, dopamine, 10 g/kg/min, and milrinone, 1 g/kg/min), and the general condition progressively improved with reduction of metabolic acidosis, improvement of gas exchange, and hemodynamic stability. After major brain damage was excluded by cerebral evaluation, the patient underwent surgical correction of the coronary stenosis. A surgical procedure of coronary reimplantation was performed with the use of 407 minutes of cardiopulmonary bypass; after which the patient failed weaning from cardiopulmonary bypass requiring postoperative extracorporeal membrane oxygenation (ECMO). Once admitted to the PCICU, continuous hemodynamic monitoring consisted of electrocardiogram, invasive systemic arterial pressure (systolic, diastolic, and mean arterial pressure), right and left atrial pressures, transcutaneous arterial saturation (SatO2), end-tidal CO2, temperature gradient, and 2 simultaneous NIRS measurements of cerebral (central forehead probe) (cSvO2) and somatic-renal (T10 to L2 probe) (rSvO2) saturations (INVOS 300; Somatometrics, Troy, MI). Intravenous sedation was maintained with fentanyl (10 g/kg/h) and midazolam (0.1 mg/kg/h) infusions. Vecuronium bromide (0.1 mg/kg) was given when necessary. Pump flow on ECMO was maintained at 150 to 160 mL/kg/min, whereas the inspired fraction of O2 and gas flow were maintained in order to guarantee a PaCO2 of 40 mmHg and a SatO2 of 99%. Perfusion pressure was maintained stable between 40 and 50 mmHg. The baseline cSvO2 saturation was 52%, and rSvO2 was 89%. On the first postoperative day, NIRS showed significant fluctuations of cSvO2 with repetitive “spikes” despite no evidence of significant changes in blood pressure and systemic arterial saturation, suggesting the presence of seizure activity in the absence of clinical signs (Fig 1). This abnormal pattern was previously unknown by the authors’ group. The on-duty physician asked for an urgent cerebral ultrasound; it did not detect brain damage. The initial interpretation of such a pattern was hypothesized as a possible effect of the aortic cannula on cerebral blood flow; a quick examination by the on-call surgeon showed a correct cannula position. Because the patient was apparently stable with renal NIRS values at a level of about 90%, perfusion problems were excluded and a monitor artifact was suspected. The following morning an EEG was performed, as routinely prescribed by institutional protocol for ECMO patients. The EEG diagnosed the presence of SE with multifocal seizures, characterized by subcontinuous multifocal critical sequences lasting up to several minutes (Fig 2). This EEG pattern was subclinical and not associated with signs of epileptic seizures. Intravenous drugs were given in accordance with the authors’ protocol of SE treatment. Seizures persisted on the EEG despite the administration of midazolam (0.3 mg/kg), a loading dose of phenytoin (15 mg/kg), and a loading dose of phenobarbital (5 mg/kg).