Detection and quantitative analysis of patient-ventilator interactions in ventilated infants by deep learning networks.

BACKGROUND: The study of patient-ventilator interactions (PVI) in mechanically ventilated neonates is limited by the lack of unified PVI definitions and tools to perform large scale analyses. METHODS: An observational study was conducted in 23 babies randomly selected from 170 neonates who were ventilated with SIPPV-VG, SIMV-VG or PSV-VG mode for at least 12 h. 500 breaths were randomly selected and manually annotated from each recording to train convolutional neural network (CNN) models for PVI classification. RESULTS: The average asynchrony index (AI) over all recordings was 52.5%. The most frequently occurring PVIs included expiratory work (median: 28.4%, interquartile range: 23.2-40.2%), late cycling (7.6%, 2.8-10.2%), failed triggering (4.6%, 1.2-6.2%) and late triggering (4.4%, 2.8-7.4%). Approximately 25% of breaths with a PVI had two or more PVIs occurring simultaneously. Binary CNN classifiers were developed for PVIs affecting ≥1% of all breaths (n = 7) and they achieved F1 scores of >0.9 on the test set except for early triggering where it was 0.809. CONCLUSIONS: PVIs occur frequently in neonates undergoing conventional mechanical ventilation with a significant proportion of breaths containing multiple PVIs. We have developed computational models for seven different PVIs to facilitate automated detection and further evaluation of their clinical significance in neonates. IMPACT: The study of patient-ventilator interactions (PVI) in mechanically ventilated neonates is limited by the lack of unified PVI definitions and tools to perform large scale analyses. By adapting a recent taxonomy of PVI definitions in adults, we have manually annotated neonatal ventilator waveforms to determine prevalence and co-occurrence of neonatal PVIs. We have also developed binary deep learning classifiers for common PVIs to facilitate their automatic detection and quantification.

Particular genomic and virulence traits associated with preterm infant-derived toxigenic Clostridium perfringens strains.

Clostridium perfringens is an anaerobic toxin-producing bacterium associated with intestinal diseases, particularly in neonatal humans and animals. Infant gut microbiome studies have recently indicated a link between C. perfringens and the preterm infant disease necrotizing enterocolitis (NEC), with specific NEC cases associated with overabundant C. perfringens termed C. perfringens-associated NEC (CPA-NEC). In the present study, we carried out whole-genome sequencing of 272 C. perfringens isolates from 70 infants across 5 hospitals in the United Kingdom. In this retrospective analysis, we performed in-depth genomic analyses (virulence profiling, strain tracking and plasmid analysis) and experimentally characterized pathogenic traits of 31 strains, including 4 from CPA-NEC patients. We found that the gene encoding toxin perfringolysin O, pfoA, was largely deficient in a human-derived hypovirulent lineage, as well as certain colonization factors, in contrast to typical pfoA-encoding virulent lineages. We determined that infant-associated pfoA+ strains caused significantly more cellular damage than pfoA- strains in vitro, and further confirmed this virulence trait in vivo using an oral-challenge C57BL/6 murine model. These findings suggest both the importance of pfoA+ C. perfringens as a gut pathogen in preterm infants and areas for further investigation, including potential intervention and therapeutic strategies.

Comparison of flexible nasogastric tube and semi-rigid catheter during less invasive surfactant administration.

BACKGROUND: Various flexible and semi-rigid catheter techniques have been reported for surfactant delivery during less invasive surfactant administration (LISA) in preterm infants. Data on the effect of catheter selection on procedural success rates and adverse events are limited. Our objective was to compare the rates of success and adverse events of LISA performed with nasogastric tube and semi-rigid catheter. METHODS: This was a post-hoc analysis of data from a quality improvement project. LISA was performed according to the standardized local protocol. Baseline characteristics, data on performance of LISA, degree of difficulty in laryngoscopy and vital parameters after the initiation of LISA were collected and outcomes were compared between groups. RESULTS: Fifty-six infants were included (21 with nasogastric tube, and 35 with semi-rigid catheter). Procedure success rate (defined as a single LISA attempt resulting in intratracheal administration of the planned dose of surfactant), incidence of adverse events, heart rate and oxygen saturation values and outcomes did not differ significantly between the two groups. When using a nasogastric tube for LISA, a significantly higher fraction of inspired oxygen was needed in the 3rd (0.62 vs. 0.48, P=0.024), 4th (0.61 vs. 0.37, P<0.001) and 5th minute (0.48 vs. 0.37, P=0.001) to maintain normal oxygen saturations. CONCLUSIONS: Use of the semi-rigid catheter was associated with better oxygenation during and shortly after the procedure. Our results may help neonatal units to develop local guidelines.

Low inflating pressures during neonatal tidal volume targeted ventilation: occurrence and significance.

OBJECTIVES: We investigated the inflating pressures (Pinfl, the difference between peak inspiratory pressure and positive end-expiratory pressure) in infants receiving volume targeted ventilation. METHODS: Data were collected and analysed from 195 infants. Median Pinfl was determined before each blood gas (n = 3425). Ventilator parameters and blood gases were compared between periods when Pinfl was <5 mbar and periods when it was higher. RESULTS: 1-hour periods when median Pinfl was <5 mbar occurred in 30% of the babies and were associated with similar tidal volumes and minutes ventilation as periods with higher Pinfl. Babies triggered more ventilator inflations, had more spontaneous breaths and lower oxygen requirement when Pinfl was low. There was no difference in blood gases when Pinfl was <5 mbar or when it was higher. CONCLUSIONS: Episodes of low inflating pressure occur frequently in babies receiving volume targeted ventilation, but they do not lead to changes in blood gases.

Synchronized intermittent mandatory ventilation with volume guarantee and pressure support in neonates: Detailed analysis of ventilator parameters.

OBJECTIVE: To analyse the relationship between peak inflating pressure, expired tidal volume, respiratory rate, and inspiratory time of volume-guaranteed ventilator inflations and pressure-supported spontaneous breaths during synchronized intermittent positive pressure mode with volume guarantee and pressure support (SIMV-VG-PS) in neonates. METHODS: Ventilator parameters were downloaded every second from 16 babies ventilated with SIMV-VG-PS mode using Dräger Babylog VN500 ventilators over 137 days. Transcutaneous carbon dioxide (tcCO2 ) data were also collected. Data were computationally analysed using Python. The average of each ventilator parameter was determined during each minute separately for ventilator inflations and for spontaneous breaths. These values were compared and their effect on tcCO2 levels was also analysed. RESULTS: The relationship between the peak inflating pressure of the volume guaranteed inflations (PIPVG ) and pressure-supported spontaneous breaths (PIPPS ) was highly variable. The PIPPS /PIPVG ratio differed significantly from the value (0.66) targeted by clinicians (group median: 0.80, range: 0.50-1.00). PIPPS frequently exceeded PIPVG . When PIPPS /PIPVG was >0.66, the expired tidal volume and the rate of the pressure-supported spontaneous breaths were also significantly (p < 0.0001) higher, but there was no difference in tcCO2 levels. The flow-cycled spontaneous breaths had significantly shorter inspiratory times than ventilator inflations. CONCLUSIONS: During SIMV-VG-PS it is difficult to ensure a pressure support level proportionate to the inflating pressure of ventilator inflations and to achieve the stability of tidal volumes.

Acceleration during neonatal transport and its impact on mechanical ventilation.

OBJECTIVE: During interhospital transfer, critically ill neonates frequently require mechanical ventilation and are exposed to physical forces related to movement of the ambulance. In an observational study, we investigated acceleration during emergency transfers and if they result from changes in ambulance speed and direction or from vibration due to road conditions. We also studied how these forces impact on performance of the fabian+nCPAP evolution neonatal ventilator and on patient-ventilator interactions. METHODS: We downloaded ventilator parameters at 125 Hz and acceleration data at 100 Hz sampling rates, respectively, during the emergency transfer of 109 infants. Study subjects included term, preterm and extremely preterm infants. We computationally analysed the magnitude, direction and frequency of ambulance acceleration. We also analysed maintenance and variability of ventilator parameters and the shape of pressure-volume loops. RESULTS: While acceleration was <1 m/s2 most of the time, most babies were occasionally exposed to accelerations>5 m/s2. Vibration was responsible for most of the acceleration, rather than speed change or vehicle turning. There was no significant difference between periods of high or low vibration in ventilation parameters, their variability and how well targeted parameters were kept close to their target. Speed change or vehicle turning did not affect ventilator parameters or performance. However, during periods of intense vibration, pressure-volume ventilator loops became significantly more irregular. CONCLUSIONS: Infants are exposed to significant acceleration and vibration during emergency transport. While these forces do not interfere with overall maintenance of ventilator parameters, they make the pressure-volume loops more irregular.

Incidence, predictors of success and outcome of LISA in very preterm infants.

OBJECTIVES: The aim of this study was to examine the success rate of less invasive surfactant administration (LISA), to identify early predictive factors for the outcome of LISA, and to compare neonatal outcomes between the LISA failure group and the group of infants who were successfully treated with LISA. DESIGN: A retrospective cohort study. PATIENTS: Infants born at less than 33 weeks of gestation (n = 158) and treated with LISA for respiratory distress syndrome. RESULTS: LISA was successful in 86 cases (54.4%); 72 preterm infants (45.6%) needed additional surfactant therapy and/or mechanical ventilation in the first 72 h (LISA failure). In a multivariate logistic regression analysis, six independent predictors of LISA success were identified: core temperature at the time of admission (adjusted odds ratio (OR): 3.56), dose of poractant alfa (<200 mg/kg; adjusted OR: 0.254), elevated C-reactive protein (>10 mg/L) at 24 h of life (adjusted OR: 0.28), highest respiratory severity score (RSS) during the first hour of life or at the time of LISA (adjusted OR: 0.463), maternal age (adjusted OR: 0.923), and birth weight (adjusted OR: 1.003). The receiver operating curve created by using the identified factors indicates good predictive power with an area under the curve of 0.85. LISA failure was associated with a substantially higher risk of complications. CONCLUSION: LISA success can be predicted by variables available before the intervention. Failure of LISA is relatively frequent event in very preterm infants and is associated with adverse outcomes. Prevention of hypothermia during early stabilization and appropriate dosing of surfactant may increase LISA success rates and improve patient outcome.

Volume-Targeted Ventilation.

Volume-targeted ventilation (VTV) has been increasingly used in neonatology. In systematic reviews, VTV has been shown to reduce the risk of neonatal morbidities and improve long-term outcomes. It is adaptive ventilation using complex computer algorithms to deliver ventilator inflations with expired tidal volumes close to a target set by clinicians. Significant endotracheal tube leak and patient-ventilator interactions may complicate VTV and make ventilator parameters and waveforms difficult to interpret. In this article, we review the rationale for using VTV and the evidence supporting its use and provide practical advice for clinicians ventilating newborn infants.

Cerebral oxygenation in preterm infants during maternal singing combined with skin-to-skin care.

BACKGROUND: Our aim was to investigate the effect of music therapy in combination with skin-to-skin care (SSC) on regional cerebral oxygenation (rSO2) measured with near-infrared spectroscopy (NIRS) in premature infants and to study physiological stability during the interventions. METHODS: This was a prospective single-center observational cohort study conducted in a tertiary neonatal intensive care unit. The study consisted of four phases: (1) baseline measurements in an incubator for 30 min; (2) quiet SSC for 30 min (SSC-Pre); (3) SSC with live maternal singing accompanied by live guitar music for 20 min (SSC-Music); (4) final quiet SSC for another 30 min (SSC-Post). RESULTS: The primary outcome measure of mean rSO2 for the 31 preterm infants analyzed showed a significant increase from baseline during SSC-Music (76.87% vs 77.74%, p = 0.04) and SSC-Post (76.87% vs 78.0%, p = 0.03) phases. There were no significant changes observed in heart rate (HR), peripheral oxygen saturation (SpO2), and cerebral fractional tissue oxygen extraction (cFTOE). The coefficient of variation (CV) of rSO2 and SpO2 decreased during each intervention phase. CONCLUSION: Combining music therapy with SSC appears to be safe in preterm neonates. The impact of the small increase in rSO2 and reduced variability of SpO2 and rSO2 warrants further investigation. IMPACT: Music therapy combined with skin-to-skin care (SSC) is safe in clinically stable premature infants and could be encouraged as part of developmental care. This is the first report where near-infrared spectroscopy (NIRS) was used to detect the simultaneous effect of music therapy and SSC on cerebral rSO2 in preterm infants. Music therapy with SSC caused a modest increase in rSO2 and decreased the coefficient of variation of rSO2 and peripheral oxygen saturation (SpO2), which suggest short-term benefits for preterm infants.

Computational analysis of neonatal ventilator waveforms and loops.

BACKGROUND: Modern neonatal ventilators allow the downloading of their data with a high sampling rate. We wanted to develop an algorithm that automatically recognises and characterises ventilator inflations from ventilator pressure and flow data. METHODS: We downloaded airway pressure and flow data with 100 Hz sampling rate from Dräger Babylog VN500 ventilators ventilating critically ill infants. We developed an open source Python package, Ventiliser, that includes a rule-based algorithm to automatically discretise ventilator data into a sequence of flow and pressure states and to recognise ventilator inflations and an information gain approach to identify inflation phases (inspiration, expiration) and sub-phases (pressure rise, pressure plateau, inspiratory hold etc.). RESULTS: Ventiliser runs on a personal computer and analyses 24 h of ventilation in 2 min. With longer recordings, the processing time increases linearly. It generates a table reporting indices of each breath and its sub-phases. Ventiliser also allows visualisation of individual inflations as waveforms or loops. Ventiliser identified >97% of ventilator inflations and their sub-phases in an out-of-sample validation of manually annotated data. We also present detailed quantitative analysis and comparison of two 1-hour-long ventilation periods. CONCLUSIONS: Ventiliser can analyse ventilation patterns and ventilator-patient interactions over long periods of mechanical ventilation. IMPACT: We have developed a computational method to recognize and analyse ventilator inflations from raw data downloaded from ventilators of preterm and critically ill infants. There have been no previous reports on the computational analysis of neonatal ventilator data. We have made our program, Ventiliser, freely available. Clinicians and researchers can use Ventiliser to analyse ventilator inflations, waveforms and loops over long periods. Ventiliser can also be used to study ventilator-patient interactions.

Volume guarantee ventilation in neonates treated with hypothermia for hypoxic-ischemic encephalopathy during interhospital transport.

OBJECTIVE: We investigated if volume guarantee (VG) ventilation in babies with hypoxic-ischemic encephalopathy (HIE) during interhospital transport decreases tidal volumes and prevents hypocapnia. STUDY DESIGN: We computationally collected and analyzed ventilator data of babies ventilated with synchronized intermittent mandatory ventilation (SIMV) with VG (n = 28) or without VG (n = 8). RESULT: The expiratory tidal volume of ventilator inflations was lower with SIMV-VG (median [IQR]: 4.9 [4.6-5.3] mL/kg) than with SIMV only (median [IQR]: 7.1 [5.3-8.0] mL/kg, p = 0.01). Babies receiving SIMV-VG had lower peak inflating pressures (median: 10.7 cmH2O, versus 17.5 cmH2O, p = 0.01). There was no significant difference in minute ventilation or in pCO2. Babies with strong spontaneous breathing had a mean PIP < 10 cmH2O but this did not result in adverse events or worsening of acidosis. CONCLUSIONS: The use of VG ventilation in babies with HIE reduces tidal volumes and frequently results in very low inflating pressures without affecting pCO2.

Fifteen-minute consultation: How to interpret and manage ventilator alarms in the neonatal intensive care unit.

Modern neonatal intensive care units use a large number of monitoring and therapeutic devices. Most of them have alarms with varying degree of standardisation. Mechanical ventilator alarms alert clinicians about technical problems with equipment, acute deterioration of the patient, changes in his or her clinical condition or in respiratory mechanics. However, frequent ventilator alarms interfere with developmental care and they may lead to alarm fatigue, missed alarms and clinical incidents. In this article, we discuss the most important ventilator alarms and their clinical significance. We also provide advice how to respond to ventilator alarms and how to set alarm limits.

Effect of pressure rise time on ventilator parameters and gas exchange during neonatal ventilation.

BACKGROUND: Pressure rise time (PRT), also known as slope time to the peak inflating pressure can be set on some modern neonatal ventilators. On other ventilators, PRT is determined by the set circuit flow. Changing slope time can affect mean airway pressure (MAP), oxygenation, and carbon dioxide elimination. Our aim was to investigate the effect of PRT on ventilator parameters and gas exchange during volume-guaranteed ventilation. METHODS: In a crossover study, 12 infants weighing greater than 2 kg were ventilated using a Dräger Babylog VN500 ventilator with synchronized intermittent positive pressure ventilation with volume guarantee (SIPPV-VG) and pressure support ventilation with volume guarantee (PSV-VG). During both modes PRTs between 0.08 and 0.40 seconds were used in 15-minute epochs. Data from the ventilator and patient monitors were downloaded with 1- and 100-Hz sampling rate and analyzed using the Python computer language. RESULTS: During PSV-VG, longer PRTs were associated with longer inspiratory time (P < .0001) and with lower peak inflating pressure (PIP; P = .003), but the MAP was similar. During SIPPV-VG the PIP was not significantly different; however, MAP was lower with longer PRT (P = .001). With a short PRT (0.08 seconds), the PIP was higher during PSV-VG than during SIPPV-VG (19.8 vs 16.5 mbar; P = .042). There were no significant differences in tidal volume delivery, respiratory rate, minute volume, oxygen saturations, or end-tidal CO2 with different PRTs in either mode. CONCLUSIONS: During SIPPV-VG or PSV-VG, using short or long PRTs affects some ventilation parameters but does not significantly change oxygenation or carbon dioxide elimination.

Volume-targeted ventilation with a Fabian ventilator: maintenance of tidal volumes and blood CO2.

OBJECTIVE: To analyse the performance of the Fabian +NCPAP evolution ventilator during volume guarantee (VG) ventilation in neonates at maintaining the target tidal volume and what tidal and minute volumes are required to maintain normocapnia. METHODS: Clinical and ventilator data were collected and analysed from 83 infants receiving VG ventilation during interhospital transfer. Sedation was used in 26 cases. Ventilator data were downloaded with a sampling rate of 0.5 Hz. Data were analysed using the Python computer language and its data analysis packages. RESULTS: ~107 hours of ventilator data were analysed, consisting of ~194 000 data points. The median absolute difference between the actual expiratory tidal volume (VTe) of the ventilator inflations and the target tidal volume (VTset) was 0.29 mL/kg (IQR: 0.11-0.79 mL/kg). Overall, VTe was within 1 mL/kg of VTset in 80% of inflations. VTe decreased progressively below the target when the endotracheal tube leak exceeded 50%. When leak was below 50%, VTe was below VTset by >1 mL/kg in less than 12% of inflations even in babies weighing less than 1000 g. Both VTe (r=-0.34, p=0.0022) and minute volume (r=-0.22, p=0.0567) showed a weak inverse correlation with capillary partial pressure of carbon dioxide (Pco2) values. Only 50% of normocapnic blood gases were associated with tidal volumes between 4 and 6 mL/kg. CONCLUSIONS: The Fabian ventilator delivers volume-targeted ventilation with high accuracy if endotracheal tube leakage is not excessive and the maximum allowed inflating pressure does not limit inflations. There is only weak inverse correlation between tidal or minute volumes and Pco2.

Neonatal encephalopathy therapy optimization for better neuroprotection with inhalation of CO2: the HENRIC feasibility and safety trial.

BACKGROUND: There is an association between hypocapnia and adverse neurodevelopmental outcome in infants with neonatal encephalopathy (NE). Our aim was to test the safety and feasibility of 5% CO2 and 95% air inhalation to correct hypocapnia in mechanically ventilated infants with NE undergoing therapeutic hypothermia. METHODS: Ten infants were assigned to this open-label, single-center trial. The gas mixture of 5% CO2 and 95% air was administered through patient circuits if the temperature-corrected PCO2 ≤40 mm Hg. The CO2 inhalation was continued for 12 h or was stopped earlier if the base deficit (BD) level decreased <5 mmol/L. Follow-up was performed using Bayley Scales of Infant Development II. RESULTS: The patients spent a median 95.1% (range 44.6-98.5%) of time in the desired PCO2 range (40-60 mm Hg) during the inhalation. All PCO2 values were >40 mm Hg, the lower value of the target range. Regression modeling revealed that BD and lactate had a tendency to decrease during the intervention (by 0.61 and 0.55 mmol/L/h, respectively), whereas pH remained stable. The rate of moderate disabilities and normal outcome was 50%. CONCLUSIONS: Our results suggest that inhaled 5% CO2 administration is a feasible and safe intervention for correcting hypocapnia.

Volume Guaranteed Ventilation During Neonatal Transport.

OBJECTIVES: To compare tidal volumes, inflating pressures and other ventilator variables of infants receiving synchronized intermitted mandatory ventilation with volume guarantee during emergency neonatal transport with those of infants receiving synchronized intermitted mandatory ventilation without volume guarantee. DESIGN: Retrospective observational study. SETTING: A regional neonatal emergency transport service. PATIENTS: We enrolled 77 infants undergoing emergency neonatal transfer. Forty-five infants were ventilated with synchronized intermittent mandatory ventilation with volume guarantee and 32 with synchronized intermitted mandatory ventilation without volume guarantee. INTERVENTIONS: Infants received synchronized intermitted mandatory ventilation with or without volume guarantee during interhospital emergency neonatal transport using a Fabian + nCPAP evolution neonatal ventilator (Software Version: 4.0.1; Acutronic Medical Instruments, Hirzel, Switzerland). MEASUREMENTS AND MAIN RESULTS: We downloaded detailed ventilator data with 0.5 Hz sampling rate. We analyzed data with the Python computer language and its data science packages. The mean expiratory tidal volume of inflations was lower and less variable in infants ventilated with volume guarantee than in babies ventilated without volume guarantee (group median 4.8 vs 6.0 mL/kg; p = 0.001). Babies ventilated with synchronized intermittent mandatory ventilation with volume guarantee had on average lower and more variable peak inflating pressures than babies ventilated without volume guarantee (group median 15.5 vs 19.5 cm H2O;p = 0.0004). With volume guarantee, a lower proportion of the total minute ventilation was attributed to ventilator inflations rather than to spontaneous breaths between inflations (group median 66% vs 83%; p = 0.02). With volume guarantee, babies had fewer inflations with tidal volumes greater than 6 mL/kg and greater than 8 mL/kg (group medians 3% vs 44% and 0% vs 7%, respectively; p = 0.0001). The larger tidal volumes in the non-volume guarantee group were not associated with significant hypocapnia except in one case. CONCLUSIONS: During neonatal transport, synchronized intermittent mandatory ventilation with volume guarantee ventilation reduced the occurrence of excessive tidal volumes, but it was associated with larger contribution of spontaneous breaths to minute ventilation compared with synchronized intermitted mandatory ventilation without volume guarantee.

Flow cytometry for near-patient testing in premature neonates reveals variation in platelet function: a novel approach to guide platelet transfusion.

BACKGROUND: Neonatal haemorrhaging is often co-observed with thrombocytopenia; however, no evidence of a causal relationship with low platelet count has been reported. Regardless, the administration of a platelet transfusion is often based upon this parameter. Accurate measurement of platelet function in small volumes of adult blood samples by flow cytometry is well established and we propose that the use of the same technology could provide complementary information to guide the administration of platelet transfusions in premature neonates. METHODS: In 28 neonates born at 27-41 weeks gestation, platelet function after stimulation agonists was measured using fibrinogen binding and P-selectin expression (a marker of degranulation). RESULTS: Platelets of neonates with gestation of ≤36 weeks (n = 20) showed reduced fibrinogen binding and degranulation with ADP, and reduced degranulation with CRP-XL. Degranulation Scores of 7837 ± 5548, 22,408 ± 5301 and 53,131 ± 12,102 (mean ± SEM) identified significant differences between three groups: <29, 29-36 and >36 weeks gestation). Fibrinogen binding and degranulation responses to ADP were significantly reduced in suspected septic neonates (n = 6) and the Fibrinogen Binding scores clearly separated the septic and healthy group (88.2 ± 10.3 vs 38.6 ± 12.2, P = 0.03). CONCLUSIONS: Flow cytometric measurement of platelet function identified clinically different neonatal groups and may eventually contribute to assessment of neonates requiring platelet transfusion.

Analysis of peak inflating pressure and inflating pressure limit during neonatal volume guaranteed ventilation.

OBJECTIVES: To report how peak inflating pressure (PIP), tidal volume (VT) and low-tidal volume alarms are affected by maximum allowed inflating pressure (Pmax) during volume guarantee (VG) ventilation. STUDY DESIGN: Ventilation data were analysed with 1 Hz sampling rate from 25 neonates receiving synchronised intermittent positive pressure ventilation with VG for >12 h. RESULTS: The difference between Pmax and PIP (Pdiff) ranged between 5 and 20 mbar (median = 11 mbar) despite a protocol to keep Pmax 5 mbar above the "working PIP". Pmax was reached in 5.2% of inflations. Computational modelling demonstrated that had Pdiff been kept at 5 mbar more consistently, >10% of inflations would have reached Pmax. The frequency of low-tidal volume alarms showed inverse correlation with Pdiff. CONCLUSIONS: It is difficult to implement a simple Pmax strategy due to variability of PIP. Setting Pmax close to the "working PIP" limits VT delivery and triggers frequent alarms.

High-frequency oscillatory ventilation with volume guarantee: a single-centre experience.

OBJECTIVE: High-frequency oscillatory ventilation (HFOV) is widely used in neonatology. The Dräger Babylog VN500 ventilator offers volume-guaranteed HFOV (HFOV-VG) mode when the high-frequency tidal volume (VThf) to be delivered can be set. We investigated how HFOV-VG maintains VThf in the short and longer term and how it affects other ventilator parameters and blood gases. METHODS: We downloaded ~3.2 million seconds (36.7 days) of ventilator data from 17 infants ventilated using HFOV-VG during clinical care with 1 Hz sampling rate. To process and analyse the data, we used the Python computer language. RESULTS: Overall, the median VThf was 1.93 mL/kg (IQR 1.64-2.45 mL/kg). The difference between set and delivered tidal volume was <0.2 mL/kg for 83% of time. In the individual recordings, the median VThf ranged between 1.44 and 3.31 mL/kg. During HFOV-VG, the VThf varied from 1 second to another, but it was very close to the target value when averaged over 5 min periods. After weight correction, the VThf or the diffusion coefficient of carbon dioxide (DCO2) showed weak inverse correlation with partial pressure of CO2(pCO2) (for VThf, r=-0.162, 95% CI -0.282 to -0.037, p=0.01). Uncorrected values showed no correlation. Of the 53 blood gas measurements taken when VThf was >2.5 mL/kg, there were only six (11%) with a pCO2 >8 kPa. CONCLUSIONS: During HFOV-VG, the tidal volume of oscillations varies in the short term but is maintained very close to the target over the longer term. VThf or DCO2 have poor correlation with CO2 levels but a volume of >2.5 mL/kg VThf is rarely needed.