Postoperative respiratory failure (RF) after pediatric liver transplantation (LT) is associated to high rate of morbidity and mortality due to this procedure. Use of noninvasive ventilation (NIV) for management of RF in postoperative period is debated.
Evaluate perioperative risk factors of RF treated with CPAP in pediatric patients undergone LT.
Single center retrospective experience. Population was divided in two group based on the need of NIV after extubation. Preoperative, intraoperative and postoperative variables were collected to understand determinants of RF treated by CPAP by multivariate logistic model.
In this retrospective cohort, 172 pediatric patients undergone to LT. Ninety four patients (94/172) developed RF after extubation and were treated with helmet CPAP. No differences were found between the two groups for type of graft, donor, surgical complications, infection rate. Patients treated with CPAP received more blood transfusion, and abdominal mesh wall was inserted. At multivariate analysis, biliary atresia (OR=3.6, 95% CI 1.3-9.8, p=.01), PELD>22 (OR=1.03; 95% CI 1.0-1.07, p=.02) and use of vasopressors during intraoperative period (OR=2.9; 95% CI 1.4-6.0, p=.005) expose pediatric liver recipient to development of postoperative RF treated with CPAP. Weight > 10 kg (OR=0.9, 95% CI 0.93-0.99, p=.03) could reduce the risk of postoperative RF. In the NIV group rate of intubation for NIV failure was 7.4% (7/94).
Biliary atresia, PELD>22 and use of intraoperative inotropes/vasopressors expose pediatric liver recipients to a greater odds of developing postoperative RF. CPAP application in early post-transplant period may be a good choice to treat postoperative RF after major abdominal surgery.
Establishing the ideal PEEP (PEEPid) is a challenge; the balance between collapse and over distension is difficult to assess. Electrical Impedance Tomography (EIT) can be used to evaluate the regional compliance changes at different PEEP levels, measuring areas of collapse and over distention.
Evaluate the PEEPid in patients with bronchiolitis.
Patients with bronchiolitis ventilated on PICU were included. We used a Pulmovista 500 (Draeger) electrical impedance tomography monitor and a 16 electrode belt; a decremental PEEP (14/12/10/8/6/4 cmH2O) was applied over 10 minutes. Data was collected prior and 2, 4, and 6 hours after the trial. PEEPid was defined as the point of best compliance.
38 PEEP trials were done in 27 children with no adverse events. Mean age was 8.5 months (1-27) and weight 6.6 kg (12.5-4). Initial mean satO2 was 94% with a mean FiO2 of 0.45 and a PEEP of 7.9 (5-10) cmH2O. PEEPid was more broadly distributed: 6/37 patients were at the PEEPid, with 14 patients starting at below and 10 above the PEEPid.There was no association between PEEPid and age. Patients varying from PEEPid had increasing compliance loss with increasing distance from PEEPid+ or – 2cmH2O meant a compliance loss of 14.2 or 12.9%, with over 30% for variations of 6cm H2O.
PEEP trials are feasible and practicable in children with bronchiolitis. PEEP levels prior to PEEP trials were different to PEEPid in 31/37 patients. Population predictions of PEEPid cannot be made. Compliance loss increases rapidly the further from the PEEPid the patient is set.
In recent years, ventilatory support in severe bronchiolitis has changed, with increased use of non-invasive ventilation (NIV) and the implementation of high-flow nasal cannula oxygen therapy (HFNC).
Analyse the impact of HFNC on patients with bronchiolitis: admission to PICU, length of stay (LOS) in the PICU, LOS in the hospital, describing changes observed between the periods 2010-2011 and 2016-2018.
Retrospective, observational, descriptive and comparative study of patients younger than one year that were admitted to the PICU with bronchiolitis. Demographic and clinical data collected include the ventilatory support received: HFNC, NIV or conventional mechanical ventilation (CMV), and the use of HFNC after CMV or NIV. LOS in both the PICU and the hospital was recorded. Statistical significance was set at p <0.05.
248 patients were included; 53 from 2010/2011 and 195 from 2016/2018. No differences were found between clinical variables at admission, nor in the LOS in PICU. The use of HFNC before PICU admission was significantly greater in 2016/2018. There was a 3% reduction in PICU admissions between the two periods, and the need for CMV decreased around 20% in the second period. Hospital’s LOS was significantly shorter for patients who received pre-admission HFNC 17.8 versus 11 days (p= 0.0005). HFNC for weaning did not reduce the PICU’s LOS 7 versus 7.5 days (p= 0.7).
The use of HFNC does not significantly reduce PICU admission, nor LOS in the PICU. However, a significant reduction in hospital LOS is observed for patients receiving pre-admission HFNC.
Pediatric acute respiratory distress syndrome (PARDS) is associated with high mortality in children (33% in severe PARDS). Lung protective ventilation strategies include low tidal volume and high-frequency ventilation, permissive hypercapnia and hypoxemia. Many institutions utilize high frequency oscillation ventilation at high mean airway pressures (MAP). An alternative strategy uses pressure control (PC) mode on a conventional ventilator with a high frequency, low tidal volume, high MAP and low FiO2.
To demonstrate that high frequency, low stretch ventilation with higher MAP is safe in patients with severe PARDS.
This is a single center, retrospective review of patients admitted between 2011 and 2018. Patients met PALICC (Pediatric Acute Lung Injury Consensus Conference) severe PARDS definition. Ventilation followed these guidelines: (1) avoidance of oxygen toxicity (FiO2 <60%) (2) MAP titrated with PEEP to maintain PaO2 > 55 mmHg and SpO2 >88% (3) tidal volume 2-6 ml/kg (4) ventilator rates of 30 to 100/minute. Demographic, management and outcome data were collected and analyzed.
In this cohort of patients (N=35), 46% were male with a mean age of 10 years and mean oxygenation index of 39 (median 30, range 16.3-125). Ventilation data (cm H2O): mean PEEP 21 (range 14-38), mean MAP 31 (range 23-47), mean driving pressure 22. Mortality was 11% and air leak requiring chest tube occurred in 6%.
PALICC recommends lower ventilator pressures. In this retrospective study higher airway pressures using a conventional ventilator with lung protective strategy resulted in low mortality and morbidity.
Bronchopulmonary dysplasia (BPD) is the chronic lung disorder of preterm infants characterized by multifactorial etiology. Despite the biggest role of genetic susceptibility, several studies were conducted to search for contributing factors.
To reveal the antenatal and postnatal factors linked with BPD in a single-center cohort of very preterm infants.
We carried out this retrospective study in a single-center cohort between January 2014 and December 2018. Preterm infants, < 32 gestational weeks of age and < 1500 g with the diagnosis of moderate/severe BPD were included. All ante-, peri and neonatal data possibly regarding BPD were recorded.
Data of 626 infants were analyzed. Mean gestational ages and birth weight of the whole cohort were 28±1.4 weeks and 1084±225 grams respectively. Of those ninety seven (15.4%) were diagnosed as moderate/severe BPD. Infants with BPD had significantly lower gestational age and birth weight as compared to non-BPD group (27±1.5 weeks and 933±201 grams respectively-p<0.05). All factors significant for moderate/severe BPD in the univariate analysis were included in the logistic regression model. Advanced resuscitation in the delivery room (OR 2.64, Cl [1.57-4.4], low gestational age (OR 0.80, Cl [0.67-0.95]), hemodynamically significant patent ductus arteriosus (OR 1.78, Cl [1.05-3.03]) and late full oral feeding (OR 1.05, Cl [1.02-1.08]) were linked with a higher rate of moderate/severe BPD.
We demonstrated that day of full oral feeding is associated with BPD. Despite the knowledge about the oral feeding problems in preterm infants with BPD, it is unclear whether early oral feeding attempts improve respiratory adverse outcomes.
Postnatal growth failure is frequently encountered in infants with bronchopulmonary dysplasia (BPD).
To investigate the predictive role of early weekly postnatal weight gain on BPD in preterm infants.
We performed this retrospective study in a single-center cohort between January 2014 and January 2018. Preterm infants, < 32 gestational weeks of age and < 1500 g with the diagnosis of moderate/severe BPD were included. We assessed the weight gain of the preterm infants weekly during the first 4 weeks of life, then at sixth and eighth weeks. Data about feeding characteristics and weight gain regarding BPD were recorded. Univariate and multivariate logistic regression analyses were performed including all confounding factors.
Data of 474 infants were analyzed. Mean gestational ages and birth weight of the whole cohort were 28.6±1.5 weeks and 1135±227 grams respectively. Of those, eighty-one (17%) were diagnosed as moderate/severe BPD. Infants with BPD had significantly lower gestational age and birth weight as compared to non-BPD group (27±1.5 weeks and 950±207 grams respectively-p<0.05). In the univariate analysis, postnatal weekly weight gain during the first 4 weeks of life was less than non-BPD group (p<0.05). In the logistic regression analysis; lower weight gain in the fourth week of life was associated with higher rate of moderate/severe BPD ((OR 0.89, Cl [0.83-0.95]).
Despite the knowledge of unfavourable effect of poor weight gain on bronchopulmonary dysplasia, there is limited data about exact timing. We demonstrated that poor weight gain during the fourth week of life was an independent risk factor for BPD.
Predicting which extremely preterm infants (EPI, <28 weeks’) will be successfully extubated to non-invasive ventilation is difficult. We aimed to compare the characteristics and outcomes of EPI who had extubation success and failure.
To understand the differences in infants who are unsuccessfully extubated from invasive mechanical ventilation compared to their successfully extubated counterparts.
Dual centre retrospective audit of the first extubation of EPI conducted between 2016-2017. Extubation failure was defined as reintubation within 7 days.
one hundred of 198 (50.51%) EPI required reintubation. There were important differences in the characteristics of infants in both groups. Infants requiring reintubation had poorer outcomes than those who did not (table). In a multivariable regression model GA remained the only significant predictor of extubation success (area under a receiver operating characteristic curve = 0.73).
| Demographic | Success (n=98) | Failure (n=100) | P |
| Gestational age in weeks (mean, SD) | 26.5 (1.1) | 25.6 (0.9) | <0.001 |
| Birth-weight, in grams (mean, SD) | 901 (179) | 771 (157) | 0.001 |
| Pre-extubation MAP (mmHg) | 7.6 (1.6) | 8.6 (1.5) | 0.003 |
| Pre-extubation achieved VT (mL/kg)• | 4.2(3.3-4.8) | 3.7 (2.9-4.2) | 0.023 |
| Outcomes | |||
| Use of postnatal corticosteroids | 4(9%) | 16(37%) | 0.001 |
| BPD at 36 weeks PMA | 18(40%) | 33(80%) | 0.008* |
| Supplemental O2 (days) in survivor | 76.5 (28.6) | 123.3 (41.9) | 0.001 |
| Hospitalization (days) in survivors | 88.9 (29.1) | 130.9 (40.4) | 0.001 |
Lower GA and birthweight and greater ventilator support are associated with an increased risk of extubation failure. Failure is associated with longer durations of supplemental oxygen and hospitalisation and an increased risk of BPD.
Bronchiolitis remains a major cause of respiratory failure, with RSV infection being the most frequent cause.
To characterize the cases of acute bronchiolitis admitted to a Pediatric Intensive Care Unit (PICU).
Retrospective study of patients admitted to our PICU with the diagnosis of acute bronchiolitis, from 1-10-2016 to 30-09-2018.
26 patients were admitted with the diagnosis of bronchiolitis; 16 were males and the median age was 3 months [1-24]. The median duration of hospitalization was 6 days [2-44] and the main cause of hospitalization was acute respiratory failure (24/26). Thirteen patients had comorbidities, the most common being prematurity (7/13) and congenital heart disease (5/13). Fifteen patients required noninvasive ventilation (NIV, mean duration of 4 days) and 10 required invasive mechanical ventilation (IMV, mean duration of 6 days). One patient required extracorporeal membrane oxygenation. The most frequently isolated virus was RSV (15/20), followed by Metapneumovirus (3/20). Seventeen patients had bacterial infection; the most frequent agents were Staphylococcus aureus(11/17) and Haemophilus influenzae(10/17). RSV infection was associated with a median duration of hospitalization of more than 7 days (p <0.05), but not with superior requirement of NIV or IMV. No statistically significant differences were found between the groups of patients with and without comorbidities, regarding the duration of hospitalization and the need for NIV or IMV. No deaths were recorded.
Although NIV is increasingly used in bronquiolitis, 40% of our patients required IMV. We also highlight the fact that half of the children had comorbidities, which is according to the literature.
Children with significant comorbidities may have a greater ventilation requirement during a PICU stay. With its associated risks, ways to avoid the intubation, such as HFNC, are being attempted.
To determine if the use of HFNC in children with significant comorbidities reduces their duration of ventilation or time spent in PICU.
Notes of all children admitted to a regional PICU during a 12 month period requiring invasive ventilation were examined.
Of 178 children, 107 had pre-existing comorbidities (60.1%).
There were no significant differences in age (p=0.3), weight (p=0.06) or PIM3 score (p= 0.32) between those with comorbidities and those without. Both groups had similar lengths of post-extubation PICU stay (p=0.27).
There was a tendency for higher mortality in the children with comorbidity (8.4%) than in those without (1.4%), p=0.053.
Use of HFNC prior to intubation was similar in both groups although there was substantially greater use of HFNC post extubation in those with comorbidity (38.2%) than in those without (17.1%), p=0.004.
Children with comorbidities had fewer hours free of ventilation at 30 days – median (IQR) 638.5 (550.7-696.5) vs 666 (624-702), p=0.047 and fewer PICU free days at 30 days – median (IQR) 24.2 (17.2-28) vs 26.6 (23.8-28.2), p=0.024.
Critically ill children with pre-existing illness have a longer duration of respiratory support, more often receiving HFNC following extubation. The length of stay post extubation is similar in those with or without comorbidity. Further studies are needed to determine whether HFNC is of clinical benefit in reducing re-intubation or other adverse complications.
Despite the lack of evidence of HFOV in ARDS in adult trials, this strategy is still a rescue therapy in children. Over the last three years, of the 2000 children Nottingham University Hospitals has ventilated , 2% were oscillated.
We aimed to identify the indications, demographics, and outcomes of patients on HFOV.
Retrospective study of children ventilated on HFOV between January 2015-March 2018; demographics, diagnosis, indications and clinical data, adjuvants therapies and outcomes were analysed
49 patients were identified; 16 excluded. The mean age was 2.2 +/- 3.5 years [DP(&YP1] (3 days- 18 years) with a mean LOS of 17.9 days. 70% of the patient had ARDS or viral pneumonitis and 60% had comorbidities (prematurity13%, neurodevelopment 13%, respiratory 11%, congenital heart disease 7%, haemato-oncology 7%).Before HFOV children were on Conventional Mechanical Ventilation for a mean of 47 hours; Oxygenation Index was 27.8 (7.3 -67.4); FiO2 85% (45-100), PaO2 9.1Kpa (3.6-19.1), PCO2 11.4Kpa (7.5-20), MAP 18 (26-12). The OI had a significant improvement after 12 (21.0+/- 14.1) and 24 (10.5 +/- 5.2) hours. The pH (7.27 +/- 0.1) and the pCO2 (8.9+/- 3.1) had a significant improvement 2 hours after HFOV. Nitric Oxide was used in 40% of the children; 5 patients were accepted for ECMO and 5 died (1 on ECMO).
Most patients presented with severe respiratory failure, mainly ARDS, have multiple comorbidities, with a survival rate of 93%. Oscillation was effective with improvement of the oxygenation and ventilation in this group of patients.
Pulmonary Hypertension (PH) occurs in 30-45% of infants with moderate to severe Broncho-pulmonary Dysplasia (BPD) Farrow and Steinhorn 2014). Sildenafil is a potent inhibitor of 5phosphodiesterase acting on cGMP (cyclic guanosine monophosphate), leading to vasodilation of pulmonary vessels, therefore, hypothetically effective in infants with BPD-PH.
A clinical question: In preterm infant (Patient) with BPD-PH could sildenafil (Intervention) as compared to no intervention (Control) help reduce length of invasive ventilation (Outcome)?
Literature search using PICO format yielded studies which were reviewed.
Three trials in Cochrane systematic review (Shah PS and Ohlsson A 2011) showed improved oxygenation index with decreased mortality in term/near term infants, however long term effects weren't explored.
Backes et al (2016) in retrospective study reported inter-hospital variation, with sildenafil reserved for sicker infants (< 29 weeks) with severe disease. There were inconsistencies in diagnosis of PH (echocardiography/ cardiac catheterisation), variation in doses, duration and starting age of treatment.
Wardle et al (2015) in retrospective study demonstrated that in 14 preterm infants (24-28 weeks), systolic PAP (Pulmonary Artery Pressure) was reduced as compared to 5 infants without its use. The study was however small with selection bias.
Use of sildenafil has been investigated in small low powered data in this age group. It is unlicensed in infants < 1 year of age with warning against its use in higher doses. There is no consensus in the dose, duration nor weaning strategy. The evidence found for its use in preterm infant with BPD-PH is limited and further trials should be awaited.
Acute respiratory failure is a major cause of pediatric intensive care unit (PICU) admission in pediatric cancer and hematopoietic stem cell transplant (HSCT) patients. The use of high-flow nasal cannula oxygen therapy (HFNC) is growing as an alternative to standard oxygen. However, its use in patients treated for malignancies including HSCT, is controversial.
The aim of this study was to assess outcomes of pediatric cancer and HSCT patients (including non-malignant indications) with acute hypoxemic respiratory failure treated with high-flow nasal oxygen on the ward.
In this retrospective cohort study, pediatric cancer and (non)-cancer HSCT patients with acute hypoxemic respiratory failure treated with HFNC were included. Among 39 patients included in the study, 53 episodes of HFNC treatment were analyzed. Of these episodes, 18 (34%) failed and patients required subsequently PICU admission. A significant median higher CRP (175 (range (72-308) versus 80 (13.5-187.8) mg/dL, p = 0.006) and higher Bedside Pediatric Early Warning Score (PEWS) 1-4 hours after initiation of HFNC (10.1 ± 0.8 versus 7.1 ± 0.4, p=0.001) was found in the failure group compared to the non-failure group. Among the 18 patients admitted to PICU, 14 (78%) needed intubation. Five (28 %) patients died during their PICU admission. None of the patients died outside the PICU.
In this study, one third of the pediatric cancer and HSCT patients receiving HFNC on the ward, eventually required PICU admission. A significant higher CRP and a higher Bedside PEWS 1-4 hours after initiation of HFNC were associated with HFNC treatment failure.
Bronchopulmonary dysplasia (BPD) is one of the most fatal respiratory complication of preterm infants. Anti-inflammatory therapy with corticosteroid is the most effective treatment to prevent BPD. Intratracheal instillation of corticosteroid as new route of drug delivery has been studied to improve the local effect and to decrease the systemic side effect of steroid.
We analyzed the effect of early intratracheal instillation of budesonide with surfactant to prevent BPD in very low birth weight preterm infants.
Very low birth weight infants diagnosed with severe respiratory distress syndrome were enrolled. Severe respiratory distress syndrome was diagnosed as chest radiography and requiring mechanical ventilator with fractional inspired oxygen (FiO2) of > 0.5. Treatment group were given budesonide and surfactant and control group were given surfactant.
There were no differences in gestational age and body weight between two groups. Five vs 4 infants needed 2nd dose of surfactant in treatment group vs control group. One infant vs 4 infants were dead in treatment group vs in control group. Survival without BPD were 5 vs 3 in treatment group vs control group. Mild BPD and moderate BPD were 2, 2 vs 1, 2 in treatment group vs in control group. Invasive ventilation duration was 8 days vs 15.9 days in treatment group vs control group. Non-invasive ventilation such as SNIPPV or NCPAP duration was 4.7 days vs 10.7 days in treatment group vs control group.
Intratracheal administration budesonide with surfactant shortly after birth may prevent BPD and improve neonatal outcome.
Pulmonary hypoplasia is common in the perinatal period and a significant cause of death in newborn infants. Definitive diagnosis is by histopathology. Clinical diagnosis and management is often challenging. There are no studies describing outcomes of the neonates with clinically suspected pulmonary hypoplasia.
Outcomes of preterm and term infants with clinical diagnosis of pulmonary hypoplasia
Retrospective observational study over a period of 10 years from January 2009 to December 2018. Data was collected from Badger database (Neonatal patient record database) on gestational age, birth weight, ventilator support, use of nitric oxide, CLD ( Chronic Lung Disease), mortality and two year neurodevelopmental outcome
Total number of infants identified with clinical diagnosis of pulmonary hypoplasia: 27; with mortality of 19/26 (73%), survival of 7/26 (26%) 1 outcome unknown. The proportion of infants with associated anomalies and survival: anhydramnios/oligohydramnios due to PROM (Prolonged Rupture of Membranes) (11/5), skeletal dysplasia (2/0), renal anomalies (7/1), multiple congenital anomalies (5/0), postnatal diagnosis of Congenital Diaphragmatic Hernia (1/1). Among the survivors 3/7 had CLD. Two year neurodevelopmental outcome showed, 2 with no or mild delay, 1 with moderate delay and none had severe delay.
Anhydramnios/ Oligohydramnios secondary to PROM: n- 11
| Survivors (n-5) | Non-survivors (n-6) | |
| Mean GA (weeks) | 29+6 | 26+5 |
| Mean BW (gms) | 1484grams | 757grams |
| Nitric oxide | 2 | 3 |
In non-surgical tertiary NICU, PROM leading to anhydramnios/oligohydramnios is the most common cause for pulmonary hypoplasia. Mortality remains very high in infants born with pulmonary hypoplasia. Aggressive treatment for pulmonary hypoplasia secondary to PROM and anhydramnios/oligohydramnios is required to improve the outcomes.