Neural network-based detection of esophageal intubation in anesthetized patients

OBJECTIVE

To test whether a neural network-based method could differentiate between tracheal and esophageal intubation in anesthetized patients by recognizing breathing circuit pressure and flow waveform patterns.

METHODS

Tracheal tubes were placed in the trachea and in the esophagus of adult patients undergoing elective operations. After ensuring for proper oxygenation, ventilator settings were changed to 5 ml/kg tidal volume (VT) and 15 cpm and circuit pressure and flow were recorded for 15 seconds. Then, the breathing circuit was switched to the tube placed in the esophagus, and signals were recorded for an additional 15-second period. During off-line analysis, individual waveforms were separated. Tracheal breaths were labeled with a score of 1 while esophageal "breaths" were labeled with -1. A neural network was defined to learn to associate waveforms to their corresponding scores. Data from 54% of the patients were used to train the neural network. Data from the remaining subjects were used for testing.

RESULTS

Forty-six patients were studied. Neural network training was achieved with 100 tracheal and 94 esophageal waveforms from 25 patients. Neural network performance was tested on 84 tracheal and 76 esophageal waveforms from 21 subjects. The neural network assigned scores of 0.99 +/- 0.05 (mean +/- SD) to tracheal waveforms and -0.99 +/- 0.03 to esophageal waveforms. The difference between mean esophageal and tracheal scores was -1.99 with a 99.999% confidence range of -2.01 to -1.96. Any arbitrary cutoff threshold, ranging between -0.76 and 0.7, separated tracheal and esophageal score regions, yielding no false positive or negative results.

CONCLUSION

A neural network differentiated consistently tracheal from esophageal intubation when the ventilation test mode was used. The ventilation mode employed is feasible in most adult patients undergoing elective procedures under general anesthesia. Further research is required to train neural networks to recognize esophageal intubation in different age groups and when different ventilation modes are applied.

Effect of crystalloid and colloid preloading on uteroplacental and maternal haemodynamic state during spinal anaesthesia for caesarean section

We have studied the effects of crystalloid 1 litre (lactated Ringer's) or colloid 0.5 litre (hydroxyethyl starch) preloading in 26 healthy parturients undergoing elective Caesarean section under spinal anaesthesia. Maternal placental uterine artery circulation was measured using a pulsed colour Doppler technique with simultaneous measurement of maternal haemodynamics. A high incidence of maternal hypotension was observed during spinal anaesthesia in the crystalloid group (62%) but the incidence was lower in the colloid group (38%). Central venous pressure was increased significantly in both groups after preload but decreased shortly after induction of spinal anaesthesia to baseline values. The mean pulsatility index (PI) in the uterine arteries did not change during preload or spinal block. A surprising finding was the widespread variation and some high values for the uterine artery PI after spinal anaesthesia. These individual increases in PI were transient and always returned to baseline values within 2 min. These results suggest that preloading with either solution is ineffective in preventing maternal hypotension and that changes in maternal heart rate, systolic arterial pressure and central venous pressure during spinal anaesthesia were not associated with rapid individual increases in uteroplacental vascular resistance. These changes seemed not to have any major effect, however, on the clinical condition of the newborn, as assessed by Apgar scores and umbilical artery pH values.

Uterine and fetal hemodynamics and fetal cardiac function after atenolol and pindolol infusion. A randomized study

OBJECTIVE

To evaluate the short-term effects of intravenously given atenolol and pindolol on utero- and umbilicoplacental vascular impedance, fetal hemodynamics and cardiac function in patients suffering from pregnancy-induced hypertension.

STUDY DESIGN

A total of 24 women were randomized to receive atenolol or pindolol infusion. By using pulsed color Doppler techniques, uterine, placental arcuate, umbilical fetal middle cerebral and renal arteries were examined before, at the end and 30 min after the end of infusion. Pulsatility indices (PI) were calculated to assess vascular impedance. Fetal myocardial function was evaluated by using pulsed Doppler and M-mode echocardiography. Peak systolic velocities from the ascending aorta and pulmonary trunk, and also inner diameters and fractional shortenings of both ventricles were measured.

RESULTS

Both drugs significantly decreased maternal blood pressure. Immediately after the infusion, maternal heart rate was significantly decreased in both groups; but the decrease was clearer and lasted longer in the atenolol group. Pindolol caused no changes in utero- or umbilicoplacental vascular impedance, while atenolol increased it in the nonplacental uterine artery. After atenolol infusion, PI in the umbilical artery was higher than after pindolol. Pindolol had no effects on fetal hemodynamics, while atenolol decreased PI value in the fetal renal artery. Peak systolic velocity in the pulmonary trunk was decreased after atenolol. Pindolol did not affect the fetal cardiac function. In subgroups with originally increased utero- or umbilicoplacental vascular impedance, the responses in uterine and umbilical vascular impedance and in fetal hemodynamics and cardiac function after atenolol and pindolol were different compared to whole groups.

CONCLUSION

Differently acting antihypertensive agents seem to affect differently uteroplacental vascular impedance. Atenolol may have some direct effects on fetal hemodynamics and cardiac function. According to our results, pindolol seems to be more preferable in the treatment of pregnancy-induced hypertension than atenolol.

The effects of epidural ropivacaine and bupivacaine for cesarean section on uteroplacental and fetal circulation

BACKGROUND

Ropivacaine is a new long-acting amide local anesthetic that has been shown in animal studies to have less dysrhythmogenic and cardiotoxic potential than bupivacaine. The intravenous administration of ropivacaine has not been associated with any detrimental effects on uterine blood flow in pregnant ewes. The purpose of this randomized, double-blind study was to examine the effects of epidural ropivacaine for cesarean section on blood flow velocity waveforms in uteroplacental and fetal arteries with color Doppler ultrasound and to assess whether the block modified fetal myocardial function.

METHODS

Healthy parturient women with singleton, uncomplicated pregnancies at term received 115-140 mg 0.5% ropivacaine (n = 11) or 0.5% bupivacaine (n = 10) in incremental epidural doses. The first ultrasound measurement was performed before injection of the study drug. Pulsatility indexes (PI) were derived for the blood flow velocity waveforms of the maternal placental and nonplacental uterine arteries; the placental arcuate artery; and the fetal umbilical, middle cerebral, and renal arteries. The fetal heart then was examined by echocardiography. The PI of the maternal uterine arteries and the fetal umbilical artery were measured 5 min after the injection of the local anesthetic. When sensory analgesia had reached the T6-T4 level, the ultrasound measurement was repeated with the same methods and targets as in the baseline measurement.

RESULTS

Both drugs provided adequate surgical anesthesia for cesarean section. In the bupivacaine group, the PI values for the maternal placental and nonplacental uterine arteries increased significantly 5 min after the main dose (P = 0.01, P = 0.002) and when sensory analgesia had reached the T6-T4 level (P = 0.004, P = 0.01) as compared with the baseline measurement. Simultaneously, the PI in the fetal middle cerebral artery decreased significantly (P = 0.02). The PI for the maternal uterine artery increased significantly (P = 0.01) after ropivacaine administration but only on the nonplacental side and not until sensory analgesia had reached the T6-T4 level. No effect on the Doppler indexes obtained from the umbilical artery was observed in either group. There were no significant differences relative to baseline values in any fetal myocardial measurement or in any ultrasound measurement between the groups. Neither drug had any detrimental effect on Apgar scores or umbilical cord acid-base status. None of the neonates' conditions was markedly depressed according to neurobehavioral testing.

CONCLUSIONS

Within this small study, epidural 0.5% ropivacaine for cesarean section did not compromise the utero-placental circulation in healthy parturient women with uncomplicated pregnancies. It provided surgical anesthesia that was equally effective as that provided by 0.5% bupivacaine.

Fetal cardiac function and ductus arteriosus during indomethacin and sulindac therapy for threatened preterm labor: a randomized study

OBJECTIVE

The purpose of this randomized study was to evaluate the effects of indomethacin (10 patients) and sulindac (10 patients) given for 4-days on the fetal cardiac function and ductus arteriosus in pregnancies complicated by threatened premature labor between 28 and 32 gestational weeks.

STUDY DESIGN

By use of pulsed color Doppler techniques the pulsatility index in fetal ductus arteriosus was calculated. Peak systolic velocities in the fetal ascending aorta and pulmonary trunk were also measured. By M-mode echocardiography both ventricular inner end-diastolic and end-systolic diameters were measured and ventricular fractional shortenings were calculated. Tricuspid valve regurgitation was evaluated by pulsed and color Doppler techniques. The ultrasonic examinations were made before and 4, 24, 48, and 72 hours after the start and 24 hours after the end of medication.

RESULTS

Indomethacin significantly decreased the mean pulsatility index in fetal ductus arteriosus 4 hours after the beginning of medication. This decrease became greater later during medication, and it was associated with a significant increase in both ventricular inner end-diastolic diameters and with a significant decrease in right ventricular fractional shortening. The mean pulsatility index in the fetal ductus arteriosus increased to control values at 24 hours after the end of medication. Sulindac significantly decreased the mean pulsatility index in fetal ductus arteriosus only 24 hours after the beginning of medication. All other mean pulsatility index values did not differ from control values. Other cardiac parameters remained unchanged during sulindac treatment.

CONCLUSION

Indomethacin has a significant reversible constrictive effect on the fetal ductus arteriosus that is associated with secondary changes, especially in the right ventricle. Sulindac seems to have only a mild and transient constrictive effect on the fetal ductus arteriosus.

Interfacing between spontaneous breathing and mechanical ventilation affects ventilation-perfusion distributions in experimental bronchoconstriction

The effect of interfacing between spontaneous and mechanical ventilation on ventilation-perfusion (VA/Q) distributions was determined during pressure-support ventilation (PSV) and in the presence and absence of spontaneous breathing during biphasic positive airway pressure (BIPAP) in 10 pigs with methacholine-induced bronchoconstriction. Whereas BIPAP without spontaneous breathing provides full and PSV breath-to-breath synchronized ventilatory support, BIPAP allows unrestricted spontaneous breathing throughout the mechanical cycle. Compared with BIPAP with and without spontaneous breathing, PSV effected an increase in ventilatory rate (p < 0.05) and a higher minute ventilation (VE) (p < 0.05). Spontaneous breathing during BIPAP accounted for 15 +/- 1% of the VE and increased cardiac output (CO) from 4.5 +/- 0.2 to 5.3 +/- 0.2 L/min (p < 0.05), Pao2 from 55 +/- 3 to 80 +/- 4 mm Hg (p < 0.05), and oxygen delivery (DO2) from 442 +/- 39 to 630 +/- 43 ml/min (p < 0.05). PSV did not increase CO, Pao2, and DO2. Spontaneous breathing did not affect oxygen consumption. During BIPAP spontaneous breathing accounted for a 15 +/- 2% decrease (p < 0.05) in blood flow to shunt units and a 16 +/- 2% increase (p < 0.05) in the perfusion of normal VA/Q units. Perfusion of shunt and normal VA/Q units was similar during PSV and BIPAP without spontaneous breathing. Dead space ventilation decreased with spontaneous breathing during BIPAP by 12% compared with PSV (p < 0.05). Dispersion of ventilation distribution was lowest during BIPAP. Uncoupling of spontaneous and mechanical ventilation during BIPAP improved gas exchange by allowing better VA/Q matching during experimental bronchoconstriction.(ABSTRACT TRUNCATED AT 250 WORDS)

Improvement in VA/Q distributions during inhalation of nitric oxide in pigs with methacholine-induced bronchoconstriction

Effects of nitric oxide (NO) and aerosolized terbutaline inhalation on ventilation-perfusion (VA/Q) distributions were determined during methacholine-induced bronchoconstriction in nine mechanically ventilated pigs. Animals inhaled, in random order, zero, 20, and 80 parts per million (ppm) of NO or aerosolized terbutaline. Inhalation of either 20 ppm NO or terbutaline produced similar reduction in pulmonary resistance and increase in lung compliance. Bronchodilation was most pronounced during inhalation of 80 ppm NO. NO inhalation increased PaO2 from 65 +/- 4 to 90 +/- 5 (20 ppm NO) and 104 +/- 6 mm Hg (80 ppm NO) (p < 0.05), and oxygen delivery (DO2) from 484 +/- 49 to 565 +/- 25 (20 ppm NO) and 619 +/- 43 ml/kg/min (80 ppm NO) (p < 0.05) compared with control. Aerosolized terbutaline did not increase PaO2 and DO2. NO inhalation accounted for a decrease in blood flow to shunt units (20 ppm NO: 14 +/- 1%, 80 ppm NO: 19 +/- 2%; p < 0.05) and an increase in the perfusion of normal VA/Q units (20 ppm NO: 12 +/- 1%, 80 ppm NO: 18 +/- 1%; p < 0.05). Perfusion of shunt and normal VA/Q units was similar in the absence of NO inhalation with and without aerosolized terbutaline. Pulmonary vascular resistance decreased from 510 +/- 55 to 332 +/- 22 dyn.s/cm5 with 20 ppm NO (p < 0.05) and to 329 +/- 41 dyn.s/cm5 with 80 ppm NO (p < 0.05) but did not change with terbutaline.(ABSTRACT TRUNCATED AT 250 WORDS)

Method of delivering constant nitric oxide concentrations during full and partial ventilatory support

OBJECTIVE

The objective of our study was to evaluate the precision and safety of administering nitric oxide (NO) during full and partial ventilatory support.

METHODS

NO was administered either using a microprocessor-controlled servo ventilator, substituting an NO-N2 mixture for the ventilator's usual air supply or by adding an NO-N2 mixture with a constant flow at the proximal end of the tracheal tube. NO, nitrogen dioxide (NO2), and nitrous and nitric acid (HNOx) was quantified selectively with a sequential, selective, hollow tube preconcentration and chemiluminescence analysis in a respiratory system model during various modes of full and partial ventilatory support.

RESULTS

The servo valve system of the ventilator provided accurate NO concentrations during full and partial ventilatory support. Interaction of spontaneous and mechanical ventilation during partial ventilatory support resulted in irregular inspiratory flow patterns and a difference of 3.6% to 44.1% between the desired and measured inspiratory NO concentrations when NO was administered at a constant flow to the proximal end of the tracheal tube. NO2 was not detected. Small amounts of 0.6 to 0.8 ppm HNOx were detected when 80 ppm NO was administered in a humid gas mixture of 37 degrees C in the presence of 90% oxygen.

CONCLUSIONS

NO can be administered accurately without formation of NO2 during full and partial ventilatory support with the electronically controlled valve system of the ventilator. Formation of HNOx is a potential problem at high NO and O2 concentrations in the presence of moisture.

Uteroplacental and fetal haemodynamics and cardiac function of the fetus and newborn after crystalloid and colloid preloading for extradural caesarean section anaesthesia

We have studied the effects of randomized preloading with either a crystalloid (lactated Ringer's) 15 ml kg-1 or colloid (hydroxyethyl starch) 7.5 ml kg-1 solution in 20 parturients undergoing elective Caesarean section under extradural anaesthesia, on blood flow in maternal placental and non-placental uterine and placental arcuate arteries and in fetal umbilical, renal and middle cerebral arteries, using a pulsed colour Doppler technique. Simultaneously, fetal and neonatal myocardial function were investigated by pulsed Doppler and M-mode echocardiography. We found no changes in maternal or fetal blood velocity waveform indices after crystalloid preloading, but the pulsatility index of the maternal non-placental uterine artery increased significantly after colloid preloading. Fetal heart rate decreased after preloading with crystalloid solution. There were no differences in fetal or neonatal myocardial function between the groups, and the outcome of the newborn infants were uneventful in all cases. These results suggest that preloading with either a crystalloid or colloid solution may lead to different uterine and fetal haemodynamics but these solutions had only minimal effects on fetal and neonatal myocardial performance and no effect on the clinical condition of newborns in uncomplicated pregnancies.

Oxygen consumption following pediatric cardiac surgery

Metabolic responses during recovery from cardiac operations for various congenital heart defects were studied in 30 mechanically ventilated pediatric patients in two groups: infants 1 year or less (group I) and children more than 1 year old (group II). Oxygen consumption (VO2) and carbon dioxide production (VCO2) were measured using a pediatric metabolic monitor intermittently after induction of anesthesia, after skin closure, 2 to 4 hours postoperatively, and on the first postoperative morning in the pediatric intensive care unit. Energy expenditure and respiratory quotient were determined from respiratory gas measurements. Rectal and skin temperatures and hemodynamic variables were recorded at the same time. VO2 increased during rewarming 2 to 4 hours after the operation by 12 +/- 15% in group I and by 24 +/- 19% in group II, while rectal temperature increased by 2.0 +/- 1.2 degrees C and 1.8 +/- 1.4 degrees C, respectively. No further increase in VO2 occurred until the first postoperative morning. A hypermetabolic response was not seen in all cases despite marked thermal changes. High-dose fentanyl anesthesia partly explains the low responses. On the other hand, low cardiac output may also compromise oxygen supply. Sixty-three percent of infants were treated for cardiac failure before surgery and 75% needed inotropic support immediately after the operation. Low central venous oxyhemoglobin saturation values (ScvO2 < 60%) were observed during rewarming, indicating an increase in oxygen extraction secondary to an increased oxygen demand in the brain during recovery from anesthesia, and a low cardiac output or delayed restoration of cerebral blood flow after CPB and deep hypothermia.

Continuous positive airway pressure modulates effect of inhaled nitric oxide on the ventilation-perfusion distributions in canine lung injury

OBJECTIVES

The present study was designed to evaluate if continuous positive airway pressure (CPAP) augments the effect of nitric oxide (NO) inhalation on matching between ventilation and perfusion (VA/Q) during acute lung injury.

DESIGN

Prospective, randomized study.

SETTING

A research laboratory at a university medical center.

SUBJECTS

Ten anesthetized mongrel dogs with oleic acid-induced lung injury.

INTERVENTIONS

Zero or 40 parts per million of NO in the inspiratory gas, with and without 10 cm H2O CPAP in random order.

MEASUREMENTS AND MAIN RESULTS

Gas exchange was assessed by estimating the VA/Q distributions using the multiple inert gas elimination technique. Application of CPAP decreased blood flow to shunt units by 26 +/- 2 percent (mean +/- SD) and increased the fraction of cardiac output to normal VA/Q units (VA/Q ratio of 0.1 to 10) by 26 +/- 2 percent (p < 0.05). Inhalation of NO during CPAP accounted for a further 10 +/- 2 percent decrease in the blood flow to shunt units and an 8 +/- 2 percent increase in the fraction of the cardiac output to normal VA/Q units (p < 0.05). Inhalation of NO alone had no significant effect on the VA/Q distributions. Inhalation of NO decreased mean transmural pulmonary artery pressure (Ppatm) both without (Ppatm from 30 +/- 2 to 23 +/- 2 mm Hg; PVR from 323 +/- 44 to 228 +/- 43 dynes.s .cm-5; p < 0.05) and with CPAP (Ppatm from 25 +/- 2 to 20 +/- 2 mm Hg; PVR from 255 +/- 30 to 173 +/- 31 dynes.s.cm-5; p < 0.05).

CONCLUSIONS

Although pulmonary vascular resistance can be lowered with NO inhalation alone, recruitment of gas exchange units with CPAP is necessary to produce a beneficial effect of NO inhalation on VA/Q matching and oxygenation. When recruitment of gas exchange units with CPAP brings gaseous NO in contact with enough pulmonary blood vessels, NO-induced vasodilation will augment VA/Q matching by a steal mechanism.

Effect of interfacing between spontaneous breathing and mechanical cycles on the ventilation-perfusion distribution in canine lung injury

BACKGROUND

Improved matching between ventilation and perfusion (VA/Q) has been proposed to be a major advantage of partial ventilatory support compared with controlled mechanical ventilation. This study was designed to determine whether a difference in gas exchange exists between partial ventilatory support techniques that allow unsupported spontaneous breathing to occur during any phase of the mechanical ventilatory cycle and those that provide mechanical support for each spontaneous inspiratory effort.

METHODS

Ten anesthetized dogs with oleic acid-induced lung injury received, in random order, pressure-support ventilation (PSV) and airway pressure-release ventilation (APRV) with and without spontaneous breathing using equivalent airway pressure limits. Gas exchange was assessed by conventional blood gas analysis and by estimating the VA/Q distributions using the multiple inert-gas elimination technique.

RESULTS

During APRV, spontaneous breathing accounted for 10 +/- 1% of the total expiratory minute ventilation. Breath-to-breath ventilatory support with PSV resulted in the highest total expiratory minute ventilation (P < 0.05). During spontaneous breathing with APRV, cardiac output increased from 3.9 +/- 0.3 to 4.6 +/- 0.41.min-1 (P < 0.05), arterial oxygen tension from 75 +/- 3 to 107 +/- 8 mmHg (P < 0.05), and oxygen delivery from 567 +/- 47 to 719 +/- 73 ml.kg.min-1 (P < 0.05). PSV did not increase cardiac output, arterial oxygen tension, and oxygen delivery. Spontaneous breathing did not increase oxygen consumption. During APRV spontaneous breathing accounted for a 13 +/- 2% decrease (P < 0.05) in blood flow to shunt units (VA/Q < 0.005) and a 14 +/- 2% increase (P < 0.05) in the perfusion of normal VA/Q units (0.1 < VA/Q < 10). Pulmonary blood flow distribution to shunt and normal VA/Q units was similar during PSV and APRV without spontaneous breathing. Dead space (VA/Q > 100) ventilation decreased by 6% during APRV with spontaneous breathing compared with PSV (P < 0.05).

CONCLUSIONS

Spontaneous breathing superimposed on mechanical ventilation contributes to improved VA/Q matching and increased systemic blood flow. Apparently, the spontaneous contribution to a mechanically assisted breath during PSV is not sufficient to counteract the VA/Q maldistribution of positive pressure lung insufflation during acute lung injury.

Effect of endogenous and inhaled nitric oxide on the ventilation-perfusion relationships in oleic-acid lung injury

Previous investigations have shown that the ventilation-perfusion (VA/Q) mismatch caused by acute lung injury can be alleviated either by inducing vasodilation in ventilated lung units with inhaled nitric oxide (NO) or by inhibiting the synthesis of endogenous NO, which opposes hypoxic pulmonary vasoconstriction. To determine the effects of a combination of these interventions, we evaluated cardiopulmonary function and VA/Q distributions in 10 dogs with oleic acid-induced lung injury. Each animal received, in random order, zero or 40 ppm of NO in inspiratory gas, with and without intravenous infusion of NG-monomethyl-L-arginine (L-NMMA) (5 mg/kg/h). The multiple inert-gas elimination technique was used to estimate VA/Q distributions. Systemic L-NMMA administration alone did not affect VA/Q inequality and gas exchange, but increased pulmonary and systemic vascular resistance. Inhaled NO improved gas exchange by redistributing blood flow from shunt units to lung units with a nearly ideal VA/Q ratio, without affecting pulmonary or systemic vascular resistance. Improved VA/Q matching and gas exchange was most pronounced when NO was inhaled in the presence of systemic L-NMMA. Inhalation of NO reversed the pulmonary but not the systemic vasoconstriction caused by L-NMMA. These results suggest that endogenous NO release is not limited to hypoxic lung regions in animals with oleic acid-induced lung injury. Inhaled NO reversed L-NMMA-induced pulmonary vasoconstriction and improved VA/Q matching by selectively dilating the pulmonary vasculature in ventilated lung units.

Ventilation-perfusion distributions during mechanical ventilation with superimposed spontaneous breathing in canine lung injury

Biphasic positive airway pressure (BIPAP) allows unrestricted spontaneous breathing throughout mechanical ventilation. Effects of spontaneous breathing during BIPAP on pulmonary gas exchange were studied on a randomized basis in 12 dogs with oleic acid-induced lung injury using the multiple inert gas elimination technique. Spontaneous breathing during BIPAP, accounting for 10% of minute ventilation (VE), increased PaO2 from 61 +/- 2 to 78 +/- 3 mm Hg (mean +/- SE) (p < 0.01), cardiac output from 4.2 +/- 0.3 to 4.6 +/- 0.3 L/min (p < 0.05), and oxygen delivery from 537 +/- 51 to 716 +/- 58 ml/kg/min (p < 0.05), whereas oxygen consumption and total VE remained unchanged. Improved pulmonary gas exchange caused by better ventilation/perfusion (VA/Q) matching was indicated by a 17 +/- 3% decrease (p < 0.01) in blood flow to shunt units (VA/Q < 0.005), a 15 +/- 3% increase (p < 0.05) in perfusion of normal VA/Q units (0.1 < VA/Q < 10), and a 6 +/- 3% reduction in ventilation of dead space (VA/Q > 100) areas (p < 0.05). Spontaneous breaths superimposed on mechanical ventilation may convert shunt VA/Q units to normal by increased ventilation of poorly or nonventilated units and/or increase blood flow to previously minimal or nonperfused areas.

Blood pressure after cardiopulmonary bypass: which technique is accurate?

To evaluate the accuracy with which a patient's aortic blood pressure can be estimated upon separating from cardiopulmonary bypass (CPB), simultaneously recorded radial artery pressure, oscillometric brachial artery pressure, pressure in the CPB circuit, and the surgeon's estimate of blood pressure by aortic palpation were compared to directly measured aortic root pressure. After obtaining institutional approval and written informed consent, 20 patients requiring CPB for cardiac operations were studied. General anesthesia was induced and maintained with fentanyl, vecuronium, and enflurane. Blood pressure measurements were made before CPB and repeated 2, 5, and 10 minutes after discontinuation of CPB. Radial artery systolic pressure before CPB and radial artery mean pressure 10 minutes after CPB were different from the aortic root pressures (P < 0.05). Although the other radial artery pressures and the surgeon's estimate of systolic aortic pressure were statistically similar to the aortic root pressures, the range of differences was clinically significant. The oscillometric technique and CPB line were poor estimates of aortic root pressure. Of the techniques used to estimate aortic blood pressure, including radial arterial, oscillometric, aortic line of the CPB circuit, and digital palpation, the radial arterial was the best, and the aortic line from the CPB machine and palpation by the surgeon were the worst. When a clinician is unsure of the blood pressure during separation from CPB, direct measurement of central aortic blood pressure is advised.

Continuous positive airway pressure by mask in patients after coronary surgery

Thirty patients who underwent coronary artery bypass grafting were randomized to receive 30% oxygen by mask either with an ambient airway pressure or with 7.4 mmHg (1 kPa) continuous positive airway pressure (CPAP) for 8 h after extubation. Arterial blood oxygen tension (PaO2) decreased remarkably in the control group after extubation (from 19.2 +/- 5.3 kPa to 12.4 +/- 2.7 kPa) but less in the CPAP group (from 16.4 +/- 3.3 kPa to 14.0 +/- 2.1 kPa). On the second postoperative morning PaO2 was equally low in both groups (control: 8.4 +/- 1.5 kPa, CPAP: 8.9 +/- 1.9 kPa). Atelectatic areas were seen with similar frequency in both groups, 17% (whole material) on the first and 50% on the second postoperative morning. Atelectasis was more common in patients with internal thoracic artery grafting and/or pleural drainage. In conclusion, CPAP therapy was well tolerated, and minimized the decrease in PaO2 after extubation, but could not prevent the poor oxygenation or the late development of atelectatic areas on the second postoperative day.

Neural network-based detection of esophageal intubation

To improve the accuracy of early detection of inadvertent esophageal intubation, we designed, trained, and tested a neural network-based computer system to detect the mechanical differences between lung and esophagogastric ventilation. Ten 25 to 30-kg anesthetized swine were sequentially ventilated with tidal volumes of 9, 12, and 15 mL/kg, using tubes placed in the trachea and in the esophagus, while flow and pressure waveforms were collected for 9-10 breaths. Gas remaining in the stomach was aspirated after each period of gastric ventilation. A computer program identified each mechanical inspiration, extracted the first 37 flow and pressure data points from each record, and normalized them to an equal amplitude. A back-propagation single-hidden-layer neural network was trained to identify the origin of flow and pressure waveforms as tracheal or esophageal. Ten different training and testing groups were assembled. In each group, data from nine subjects were used for training and data from the remaining subjects were used for testing. A total of 291 esophageal and 300 tracheal flow and pressure waveforms were analyzed by the network. The network identified esophageal intubation correctly during the first five breaths of all esophageal recordings. In one subject, the network identified the eighth esophageal breath as tracheal and could not identify three breaths. All tracheal intubations were identified correctly. Flow and pressure "signatures" of pulmonary and gastric ventilation are easily learned by a neural network. Therefore, neural-network recognition of esophageal intubation from flow and pressure signals is possible, and the development of an on-line detector for tracheal tube misplacement seems feasible.

Does a paracervical block with bupivacaine change vascular resistance in uterine and umbilical arteries?

Paracervical block during labor following normal, uncomplicated pregnancy is occasionally accompanied by fetal bradycardia. To evaluate whether a paracervical block with bupivacaine causes changes in the vascular resistance of uteroplacental and umbilicoplacental blood flow, a total of 12 singleton, uncomplicated pregnancies in active labor at the end of pregnancy were included to this study. By using pulsed color Doppler ultrasound techniques the pulsatility indices were measured from both uterine and umbilical arteries before, one minute and 20 minutes after a paracervical block with bupivacaine. Pulsatility indices of both uterine and umbilical arteries remained unchanged throughout the study period, as did fetal and maternal heart rates. In two cases fetal bradycardia developed, causing a marked increase in the vascular resistance of the umbilical artery one minute after the paracervical block. In the uterine arteries there was only a minor increase. When the fetal bradycardia ceased the pulsatility indices returned to the levels at the beginning of the study. Paracervical block with bupivacaine in normal pregnancies without signs of chronic or acute fetal distress does not change the vascular resistance in the uterine or umbilical arteries. If fetal bradycardia develops, it seems to be due to the direct effect of bupivacaine on the fetus, mainly on the umbilical vessels.

Catheter-induced pulmonary artery haemorrhage resulting from a pneumothorax

We present a case of pulmonary artery perforation in a patient who developed a pneumothorax after cardiac surgery. In the process of inserting a chest tube the patient became tachypnoeic, and developed haemoptysis. The trachea was intubated, and right bronchial intubation was performed with persistent bleeding. The pulmonary artery catheter was gently withdrawn and the balloon inflated, with cessation of bleeding. The patient was taken to the operating room, a bronchial blocker was placed in the right lower lobe bronchi, and the pulmonary artery catheter was removed. The bronchial blocker was removed the following day with no bleeding. The aetiology of perforation was secondary to the pneumothorax, which caused a shift of the mediastinum to the right, elevated pulmonary artery pressures, and the distal migration of the catheter through the pulmonary artery. It is recommended that treatment include tracheal intubation, inflation of the pulmonary artery catheter balloon, and the placement of a right lower lobe bronchial blocker.

Effect of labetalol infusion on uterine and fetal hemodynamics and fetal cardiac function

The effect of intravenously-administered labetalol (1 mg/kg) on uterine and fetal hemodynamics and fetal cardiac function was studied in 13 women with pregnancy-induced hypertension. Maternal mean blood pressure had decreased significantly by the end of the labetalol infusion and 30 min later. The pulsatility indices reflecting peripheral vascular resistance did not change in the main uterine, placental arcuate, umbilical, and fetal middle cerebral and renal arteries, nor did the parameters reflecting fetal cardiac function. A subgroup of seven patients with a more pronounced decrease in maternal blood pressure and three cases with original signs of fetal asphyxia did not react hemodynamically in a different manner from the group as a whole with respect to the circulatory effects of labetalol. These findings obtained by Doppler and colour Doppler methods in as comprehensive manner as possible seem to demonstrate a wide tolerance of maternal and fetal hemodynamics to the moderate decrease in maternal blood pressure achieved by labetalol infusion under short-term conditions in cases of pregnancy-induced hypertension.

Uteroplacental and fetal circulation during extradural bupivacaine-adrenaline and bupivacaine for caesarean section in hypertensive pregnancies with chronic fetal asphyxia

We have studied the effects of an extradural block during Caesarean section using either bupivacaine plain or with adrenaline 85-100 micrograms on blood velocity waveforms of maternal uterine and placental arcuate arteries and fetal umbilical, renal and middle cerebral arteries, in 20 hypertensive parturients with chronic fetal asphyxia. Fetal myocardial function was investigated at the same time by M-mode echocardiography. Extradural anaesthesia resulted in a significant decrease in maternal mean systolic and diastolic arterial pressures in both groups, but this was more marked after plain bupivacaine. There were no significant differences in any of the Doppler recordings relative to baseline values after plain bupivacaine, but after bupivacaine with adrenaline there were significantly increased blood flow velocity indices for the maternal uterine and placental arcuate arteries and significantly decreased indices in the fetal renal and middle cerebral arteries. Neonatal outcome as evaluated by Apgar scores and acid-base values in the umbilical cord were similar in the two groups. The results suggest that adrenaline added to the solution of bupivacaine increased vascular resistance in the uteroplacental circulation, indicating impaired blood flow.

Aggressive direct treatment of a fetus with supraventricular tachycardia and hydrops fetalis

We report a case, in which direct fetal therapy by amiodarone injected into the umbilical vein during the last trimester of pregnancy was used for the treatment of fetal supraventricular tachyarrhythmia in the presence of severe hydrops fetalis. Eight injections were needed due to the recurrence of supraventricular tachycardia 1-9 days after the initial normalization of fetal tachycardia after each puncture. A severe fetal hydrops was maintained despite the achievement of sinus rhythm, and thus two ascites draining procedures were performed during the last 3 days before delivery to expand the fetal lungs. A normorhythmic hydropic baby was born by Cesarean section at 34 weeks + 6 days. Only mild respiratory difficulties occurred after birth. Her electrocardiogram suggests a Wolff-Parkinson-White syndrome but the overall recovery has been uneventful.

Adaptation of pressure support ventilation to increasing ventilatory demand during experimental airway obstruction and acute lung injury

OBJECTIVE

To estimate the changes in the relative amount of ventilatory assistance offered by inspiratory pressure support during changing ventilatory demand with external airway obstruction and with oleic acid-induced acute lung injury.

DESIGN

Prospective, controlled, crossover study.

SETTING

Experimental laboratory in a university anesthesiology department.

SUBJECTS

Eight dogs anesthetized with pentobarbital.

INTERVENTIONS

An external resistor was placed in the breathing circuit to produce increased resistance to breathing. Acute lung injury was produced with oleic acid. Ventilatory demand was increased by increasing the CO2 concentration in inspired gas to produce an increase of 20 torr (2.7 kPa) in end-tidal CO2 tension. During airway obstruction, pressure support was adjusted to reduce the inspiratory decrease in intrathoracic pressure to the level present during unobstructed breathing. During acute lung injury, pressure support was applied to reduce the maximum negative deflection of intrathoracic pressure by 50%.

MEASUREMENTS AND MAIN RESULTS

The ventilator effort was estimated by calculating the pressure-time integral of proximal airway pressure; the spontaneous ventilatory effort was estimated in a similar manner from esophageal pressure. The pressure support averaged 10 +/- 3 cm H2O during airway obstruction and 7 +/- 2 cm H2O during lung injury. The CO2 challenge effected an average increase in mean minute ventilation of 78% during airway obstruction (p < .001) and 120% during acute lung injury (p < .01). The augmentation of minute ventilation was accomplished by increasing the ventilatory rate and the transpulmonary pressure while inspiratory time remained unchanged. The pressure-time integrals measured using both airway (p < .001) and esophageal pressure (p < .01) increased significantly during each CO2 challenge, reflecting an increase in the contribution of both the ventilator and the animal to the required breathing effort. Significant decreases in the relative magnitude of the ventilator effort both during airway obstruction (p < .05) and during lung injury (p < .01) indicated that the increase in the spontaneous effort was predominant over the increase in mechanical ventilatory support.

CONCLUSIONS

A ventilatory rate-dependent adaptation of pressure support to increased ventilatory demand occurs in an experimental setting both during airway obstruction and lung injury. The results of this study confirm an advantage of breath-to-breath inspiratory pressure support over techniques designed to supply a predetermined mechanical minute volume.