Conventional gas ventilation, liquid-assisted high-frequency oscillatory ventilation, and tidal liquid ventilation in surfactant-treated preterm lambs.

This study was designed to compare the efficacy and potential protective or injurious effects of tidal liquid ventilation (TLV), liquid-assisted high-frequency oscillatory ventilation (LA-HFOV), and high PEEP conventional mechanical ventilation (CMV) in neonatal respiratory distress syndrome. Preterm lambs (124-126 days gestation), prophylactically treated with natural surfactant, were allocated to one of the treatment modalities or to an untreated fetal control group (F), euthanised after tracheal ligation. LA-HFOV animals received an intratracheal loading dose of 5 mL x kg(-1) followed by a continuous intrapulmonary instillation of 12 mL x kg(-1);h(-1) FC-75 perfluorocarbon liquid. The ventilation strategies aimed at keeping clinically appropriate arterial blood gases for a study period of 5 hours. A histological lung injury score was calculated and semiquantitative morphometry was performed on lung tissue fixed by vascular perfusion. The alveolar-arterial pressure difference for O2 was significantly lower throughout the study in TLV compared to CMV lambs; at 1, 2, and 5 hours, oxygenation was better in TLV when compared to LA-HFOV. Total lung injury scores in TLV lambs were significantly lower than in either CMV or LA-HFOV animals, but higher when compared to F. CMV and LA-HFOV induced an excess of collapsed and overdistended alveoli, whereas in TLV alveolar expansion was normally distributed around predominantly normal alveoli. CMV and LA-HFOV, but not TLV, were associated with an excess of dilated airways. Thus, in the ovine neonatal RDS model, TLV compared favourably to either gas ventilation strategy by its more uniform ventilation, reduced lung injury, and improved gas exchange.

Observations on the onset of torsade de pointes arrhythmias in the acquired long QT syndrome.

OBJECTIVE: Premature ectopic beats may create a specific sequence of events (e.g. short-long-short) preceding Torsade de Pointes arrhythmias (TdP) in the long QT syndrome. The relevance of this sequence for the initiation of TdP is not clear. In our dog model of TdP, interventricular dispersion (DeltaAPD=left-right ventricular monophasic action potential duration: APD) is associated with TdP, therefore we tested the hypothesis that the ectopic beats contributes to DeltaAPD. METHODS: In 17 anaesthetized dogs with chronic AV-block, which showed spontaneous TdP after class III medication, APD was analyzed to 1. quantitate the alterations due to (multiple) ectopic beats on the left and right APD (measured with endocardial catheters) and 2. compare the DeltaAPD prior to the occurrence of premature beats (steady state) in dogs with non-sudden onset of TdP (n=10) and sudden onset TdP (n=7). Three phases were distinguished: phase 1: steady state beats prior to ectopic beats, phase II: the beat(s) belonging to the dynamic phase, and phase III: the beat causing TdP. Because the coupling interval of premature beats in this condition often falls within the APD, the DeltaAPD(50) was validated as an alternative for the previously applied DeltaAPD(100) (r=0.51, P<0.01). RESULTS: In steady state (phase I) DeltaAPD(50) is longer in the sudden onset TdP (130+/-35 ms) as in the non-sudden onset TdP (65+/-40 ms). In the non-sudden TdP group the dynamic phase II contribute to the heterogeneity in APD, i.e. LV-APD increases more than RV-APD leading to a DeltaAPD(50) increase to 130+/-100 ms (P<0.01) just preceding TdP (phase III). CONCLUSION: The synergism between ectopic beats (short-long-short sequence) and DeltaAPD create the circumstances for TdP initiation.

A feedback controller for the maintenance of FRC during tidal liquid ventilation: theory, implementation, and testing.

The necessity of controlling functional residual capacity (FRC) during tidal liquid ventilation has been recognized since the first description of this respiratory support technique by Kylstra et al in 1962. We developed a microcomputer feedback system that adjusts the inspired tidal volume (VT,I) of a liquid ventilator based on the end-expiratory quasi-static alveolar pressure (PA,EE), in order to maintain a stable FRC. The system consists of three subunits: (1) a tracheal pressure catheter to estimate breath by breath FRC changes, derived from PA,EE changes, and (2) a roller pump interfaced with (3) a personal computer in which a closed-loop control is implemented. The regulator sets the actual PA,EE against the corresponding desired value. Any discrepancy is offset by changes in VT,I and the required change in pump velocity is communicated to the roller pump. The size of any change in pump velocity is determined to both the observed and target or desired PA,EE (i.e., the error) and the (calibration) pressure-volume curve. To evaluate the efficacy of the controller, a set of laboratory bench tests were conducted under steady state and transient conditions. Closed-loop control was effective in keeping FRC and PA,EE near the desired level, with an acceptable oscillatory behaviour. The feedback controller successfully compensated for transient disturbances of PFC liquid balance. The steady state stability was confirmed during a five hour period of liquid ventilation in five preterm lambs.

The dynamic behavior of the diastolic slope of monophasic action potential can be related to the occurrence and maintenance of delayed afterdepolarization dependent arrhythmias.

We have described the value of the diastolic slope of the MAP recording at the end of a pacing train as a qualifying marker for the induction of delayed afterdepolarization (DAD) dependent arrhythmias. In the present study (1) the behavior of the slope at different time points during a pacing train was quantified and related to the arrhythmogenic outcome (group A) and (2) termination of DAD dependent VT was related to changes in the slope steepness (group B). In dogs with chronic complete AV block, a MAP was recorded during (1) ventricular pacing, before and after ouabain administration (group A) and (2) 6 spontaneous and 6 lidocaine induced VT terminations (group B). During control (group A), the slope at the end of pacing train was 5 +/- 3 m V/s (mean +/- SD), independent of the pacing duration. During ouabain, this increased to 20 +/- 15 mV/s (P < 0.05), varying with the duration of pacing. The slope was steeper after pacing for 4 seconds, compared to 20 seconds (26 +/- 12 mV/s vs 16 +/- 13 mV/s, P < 0.05) which corresponded with more frequent VT induction. In spontaneously terminating VTs (group B), CL increased from 353 +/- 54 ms at the start to 434 +/- 78 ms (P < 0.05) before VT termination. This corresponded with a decreasing steepness of the slope from 19 +/- 10 mV/s to 6 +/- 5 mV/s (P < 0.05). In lidocaine induced VT termination, the CL and the steepness of the slope showed an identical behavior. There is a dynamic variation in the steepness of the diastolic slope during pacing, which depends on the duration of pacing and predicts arrhythmogenic outcome. Furthermore, a decrease in steepness of the slope during DAD dependent VT can be used to predict VT termination.