SEARCH

SEARCH BY CITATION

Keywords:

  • anaesthesia;
  • endothelin-1;
  • gas exchange;
  • horse;
  • hypoxaemia;
  • MIGET ;
  • nitric oxide;
  • inline image ;
  • shunt;
  • inline image

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods Study 1
  5. Materials and methods study 2
  6. Results study 1
  7. Results study 2
  8. Discussion
  9. Conclusions
  10. Acknowledgements
  11. References

Objectives

Anaesthetized horses commonly become hypoxaemic due to ventilation/perfusion (inline image) mismatch and increased pulmonary shunt fraction (inline image). Pulse-delivered inhaled nitric oxide may improve oxygenation but may increase plasma concentration of the potent vasoconstrictor, endothelin-1 (ET-1). Objectives: Study 1) compare arterial oxygen concentration (PaO2) and saturation (SaO2), calculated inline image and ET-1 concentration; and Study 2) assess inline image matching and measured inline image in isoflurane-anaesthetized horses in left lateral recumbency receiving pulse-delivered inhaled nitric oxide (PiNO group) or inhalant gas only (C group).

Study design

Prospective research trial.

Animals

Ten Healthy adult Standardbred horses. Two horses were anaesthestized in both groups in a random cross-over design with >4 weeks between studies.

Methods

Study 1) Cardiopulmonary data including PaO2, SaO2, inline image and ET-1 concentration were measured or calculated prior to and at various points during PiNO administration in 6PiNO and 6C horses. Two-way repeated measures anova with Bonferroni significant difference test was used for data analysis with < 0.05 considered significant. Study 2) inline image matching and inline image were determined using the multiple inert gas elimination technique in 3 horses. Data were collected after 60 minutes of anaesthesia without PiNO (baseline) and 15 minutes after PiNO was pulsed during the first 30%, and then the first 60%, of inspiration. Data were descriptive only.

Results

Study 1) PaO2 and SaO2 were higher and calculated inline image was lower in the PiNO group than the C group at most time points. ET-1 was not different over time or between groups. Study 2) inline image matching and measured inline image were improved from baseline in all horses but PiNO60% provided no improvement when compared to PiNO30%.

Conclusions and Clinical Relevance

PiNO delivered in the initial portion of the inspiration effectively relieves hypoxaemia in anaesthetized horses by improving inline image matching and decreasing inline image without affecting ET-1.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods Study 1
  5. Materials and methods study 2
  6. Results study 1
  7. Results study 2
  8. Discussion
  9. Conclusions
  10. Acknowledgements
  11. References

Hypoxaemia secondary to impairment of pulmonary gas exchange is common in the anaesthetized horse (Hall et al. 1968; Nyman et al. 1988). Anaesthesia and recumbency induce significant alterations in ventilation (inline image) and blood flow (inline image) distribution in the equine lung, resulting in a inline image mismatch (McDonell 1974; Nyman & Hedenstierna 1989). The most important alteration affecting arterial oxygenation is the perfusion of unventilated dependent lung areas resulting in increased right to left vascular pulmonary shunt (inline image) and hypoxaemia (McDonell 1974; Nyman & Hedenstierna 1989; Nyman et al. 1990).

Treating, or preventing, inline image mismatch and inline image in the anaesthetized horse can be difficult (Nyman & Hedenstierna 1989; Moens et al. 1994) but the use of inhaled nitric oxide (iNO) appears to be effective (Heinonen et al. 2001, 2002; Nyman et al. 2013). Nitric oxide provides selective pulmonary vasodilation when delivered as an inhaled gas (Frostell et al. 1991, 1993). The suggested effect of iNO is selective dilation of pulmonary blood vessels in ventilated (generally nondependent) areas of the lung that receive the gas, thereby promoting redistribution of blood flow from the non-ventilated (generally dependent) regions to the ventilated regions. Although continuous delivery of iNO over the whole inspiratory phase of ventilation did not improve arterial oxygenation in anaesthetized horses (Young et al. 1999), iNO pulse-delivered (PiNO) in the early phase of inspiration did counteract hypoxaemia and inline image formation (Heinonen et al. 2000, 2001; Nyman et al. 2013) in dorsally recumbent isoflurane-anaesthetized horses. The pulse delivery of the iNO may be the key to successful utilization of the gas in anaesthetized horses and the timing of the pulse within the breath appears to be important both for optimizing perfusion distribution in the lung and for preventing rebreathing of NO, which could lead to formation of the potentially toxic by-product, nitrous dioxide (NO2), in the rebreathing circuit (Heinonen et al. 2000, 2002).

In the previous studies utilizing PiNO in anaesthetized horses (Heinonen et al. 2000, 2001; Nyman et al. 2013) there were no control groups and it is possible, albeit unlikely, that oxygenation improved with time rather than with treatment. Furthermore, the horses in all of these studies were positioned in dorsal recumbency, which, compared to positioning in lateral recumbency, is more likely to produce hypoxaemia (Nyman et al. 1988; Day et al. 1995). The degree of inline image mismatch and thus the magnitude of the response to PiNO may be insignificant in lateral recumbency. Although this would be interesting from a physiological view it would negate the need for PiNO in laterally recumbent horses.

Finally, the plasma concentration of the potent vasoconstrictor endothelin-1 (ET-1) has been measured following PiNO cessation (Grubb et al. 2008, 2013) but not during PiNO administration in the anaesthetized horse. In some species, ET-1 concentration increases during iNO administration (Chen et al. 2001; Ross et al. 2005) and abrupt cessation of iNO may allow the ET-1 effects to dominate, causing a precipitous deterioration in arterial oxygenation. Elevation of ET-1 during iNO administration could adversely impact the recovery phase of equine anaesthesia when both iNO and supplemental oxygen would be discontinued.

The objectives of this study were to:

  1. Assess the effect of PiNO on cardiopulmonary parameters, including calculated inline image and arterial oxygenation (PaO2 and SaO2), and plasma ET-1 concentration in isoflurane-anaesthetized horses placed in left lateral recumbency and to compare the data with those obtained in a control group of similarly anaesthetized horses not receiving PiNO; and
  2. Measure the effect of PiNO on inline image distribution and measured inline image in isoflurane-anaesthetized laterally recumbent horses.

Materials and methods Study 1

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods Study 1
  5. Materials and methods study 2
  6. Results study 1
  7. Results study 2
  8. Discussion
  9. Conclusions
  10. Acknowledgements
  11. References

Horses

Ten healthy Standardbred horses (eight mares and two geldings) were used for this study. The mean weight of the horses was 513 ± 37 kg (mean ± SD), and the age was 9 ± 4 years (mean ± SD). The horses were allocated such that 6 horses were in the control group (C group) that received inhalant anaesthesia but no PiNO and 6 received inhalant anaesthesia plus PiNO (PiNO group). Two horses were anaesthetized in both groups and they were randomized in a cross-over design and anaesthetized at intervals of no less than 4 weeks between studies. The study was approved by the local ethics committee for animal experiments.

Anaesthesia

Food but not water was withheld for 12 hours prior to anaesthesia. On the day of the procedure, the horses were examined and deemed healthy. After examination, as described in detail in the companion paper (Grubb et al. 2013), the horse was premedicated with acepromazine, anaesthesia induced with guaifenesin followed by thiopental, the trachea intubated, and the horse placed in lateral recumbency on a padded table. Anaesthesia was maintained with isoflurane delivered in <90% oxygen. Respiration was spontaneous.

Throughout the study, end-tidal isoflurane (Fe′ISO) concentrations of 1.5–1.7% (approximately 1.2–1.25 times the minimum alveolar concentration of isoflurane in horses) were maintained. The gas monitor was calibrated before each research period by use of a commercially prepared calibration gas. Instrumentation for data collection occurred during the first 45 minutes of anaesthesia (see below). Data collection continued for another 105 minutes, leading to a total anaesthesia time of 150 minutes. Following the last data collection, each horse was allowed to recover from anaesthesia in a padded stall and data were collected for a separate study (Grubb et al. 2013).

Instrumentation

All horses were instrumented with ECG electrodes placed for lead II analysis and measurement of heart rate. The skin over the right facial artery was clipped free of hair and aseptically prepared and a 20-gauge, 5-cm catheter was introduced percutaneously for measurement of arterial blood pressure and for collection of arterial blood samples for blood gas analysis. An area over the right jugular vein was clipped free of hair and aseptically prepared; an introducer kit was used to place a 7-F thermodilution catheter through the jugular vein and into the pulmonary artery for measurement of pulmonary arterial blood pressure and for collection of mixed venous blood samples for blood gas analysis. A pig-tail, multiport catheter was introduced by the same technique into the same jugular vein, advanced to the right ventricle, and retracted into the right atrium for injection of saline (0.9% NaCl) solution for cardiac output (inline image) determination. Catheters were positioned by use of pressure-tracing guidance and simultaneous ECG monitoring and were locked in position via a Luer-lock adapter. Systolic, diastolic, and mean arterial blood pressure values (SAP, DAP, MAP, respectively) and systolic, diastolic and mean arterial pulmonary pressure (SPAP, DPAP, MPAP, respectively) were measured by use of pressure transducers positioned at the level of the sternal manubrium, which was considered to correspond to the level of the right atrium. Cardiac output was determined by use of a thermodilution technique in which a bolus of 20 mL of cold (0 °C) saline solution was injected rapidly by hand through the pig-tailed catheter. At least 3 injections were performed, and the data were averaged at each time period. Cardiac output, arterial (systemic and pulmonary) blood pressures, heart rate (HR), FiO2, respiratory rate (fR), TV, end-tidal carbon dioxide fraction, and Fe′ISO were recorded from a standard anaesthesia monitor (AS/3- AM anaesthesia monitor; Datex-Ohmeda, Finland).

Arterial and central (mixed venous) blood samples were obtained for assessment of pH, partial pressure of oxygen and partial pressure of carbon dioxide by use of a standard electrode technique. [pHa and pHv; PaO2 and PvO2; PaCO2 and PvCO2 respectively; (ABL 500; Radiometer, Denmark)]. Arterial and mixed venous oxygen saturations were measured with a standard electrode technique and blood haemoglobin (Hb) concentrations were measured spectrophotometrically (OSM 3; Radiometer, Denmark). Samples for blood gases were stored on ice until analysis. Blood gas values were corrected for atmospheric pressure but not body temperature.

Venous blood was collected and stored for ET-1 measurement as described in detail in the companion paper (Grubb et al. 2013).

Calculated data

Shunt fraction (inline image) was calculated using the following formula:

inline image = (Cc'O2 – CaO2)/(Cc'O2 – C v̄O2), where Cc'O2, CaO2, and Cv̄O2 are oxygen content of capillary, arterial, and mixed venous blood, respectively. These values were calculated as: CxO2= (1.36 × haemoglobin concentration × SxO2) +(0.003 × PxO2).

Oxygen delivery (DO2) was calculated using: CaO2 × inline image.

Alveolar-arterial oxygen difference [(A–a)DO2] was calculated using: PAO2 – PaO2, where PAO2 is the partial pressure of oxygen in the alveoli. PAO2 was calculated using: FiO2 – (PACO2/R), where PaCO2 was used as PACO2 (the partial pressure of carbon dioxide in the alveoli) and R is the respiratory quotient (0.8). FiO2 is the fraction of oxygen in the inspired air.

Delivery of PiNO

Once completely instrumented, the horses were allowed to equilibrate under anaesthesia for 45 minutes, data were then collected and used as anaesthesia baseline data (T45). After baseline data collection, PiNO delivery commenced. The PiNO was delivered during the initial portion of each inhaled breath by use of a proprietary device that was designed at the Datex-Ohmeda Research Unit, Helsinki, Finland specifically for pulsed delivery of iNO during spontaneous breathing. The device was activated by negative pressure and delivered a volumetric dose into the endotracheal tube at the onset of inspiration in the same manner as previously described (Heinonen et al. 2000, 2002; Nyman et al. 2013). The delivery device was connected to a port located at the proximal end of the endotracheal tube. The NO was supplied in a cylinder of 2,000 ppm medical grade NO in N2 (AGA AB, Sweden). In order to determine the most-effective duration of pulse delivery for each horse, a flow sensor detecting gas flow was fitted onto the endotracheal tube to trigger the delivery of iNO during the first portion of inspiration in three separate predetermined steps of increased pulse duration. The durations were designed to correspond to approximately 30%, 45% and 60% of the total inspiratory time. The horses were allowed to equilibrate at each pulse length for 15 minutes and then data were collected before the subsequent pulse length was initiated. Pulse duration was not randomized. The pulse length that corresponded to the highest (peak) PaO2 in the individual horses was delivered for another 60 minutes resulting in a total of 105 minutes of PiNO administration.

Control horses were anaesthetized for the same duration and data were collected at the same times, but the horses did not receive PiNO.

Data collection

All data including HR; SAP, MAP and DAP; SPAP, MPAP and DPAP; FiO2; fR; TV; Hb; Fe′ISO; pHa and pHv̄; PaO2; Pv̄O2; SaO2; Sv̄O2;[A-a]DO2; and Qs/Qt were measured or calculated following 45 minutes of equilibration after commencement of isoflurane anaesthesia prior to NO administration (T45) and after each 15 minute increase in PiNO (T60, T75 and T90) and then every 15 minutes for the duration of the study (T105, T120, T135 and T150). Cardiac output was determined at T45, T105 and T135. Mixed venous blood samples were collected for ET-1 analysis at T45 and T150.

Data analysis

Data were assessed for normal distribution using the Shapiro-Wilk test. Repeated measures anova was used to compare data within and between groups at time points T45–T150. The Bonferroni significant difference test was used for post hoc comparisons. Because 2 horses were anaesthetized in each group, the data were analysed with and without these 2 horses to determine if there were differences in the data. There were no differences, so the horses were included in the groups and the data are reported in the results section. For all statistical calculations, a software package (GraphPad Prism, GraphPad Software, CA, USA) was used and a value of < 0.05 was considered significant.

Materials and methods study 2

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods Study 1
  5. Materials and methods study 2
  6. Results study 1
  7. Results study 2
  8. Discussion
  9. Conclusions
  10. Acknowledgements
  11. References

Horses, anaesthesia, and instrumentation

Three unrelated horses (3, 6 and 8 years of age; 458, 473 and 581 kgs; 1 gelding and 2 mares) were anaesthetized and instrumented as described for Study 1.

Delivery and measurement of PiNO

PiNO was delivered after collection of baseline data as described for Study 1. In addition, the dose of PiNO was measured using an AD-lite (Datex-Ohmeda, Finland) and expired NO was monitored with a chemiluminescent analyser prototype (Datex-Ohmeda, Finland) connected between the Y-piece and the point of NO administration. The analyser was calibrated with the mixture 100 μL L−1 NO in N2 (Aga AB, Lidingö, Sweden) and with room air depleted of NO with a charcoal absorber. The monitor was used to determine the end-tidal (Fe′NO) and peak expired (FpeakNO) NO fractions.

Determination of inline image and inline image

The distribution of ventilation and perfusion was estimated by a multiple inert gas elimination technique (MIGET; Wagner et al. 1974; Hedenstierna et al. 1987; Wagner 2008). Six inert gases (sulphur hexafluoride, ethane, cyclopropane, enflurane, ether and acetone), were dissolved in isotonic sodium chloride solution and infused into the jugular vein at 30 mL minute−1. After 60 minutes of infusion arterial and mixed venous blood samples were drawn, and mixed expired gas was collected from a heated mixing box connected to the expiratory limb of the large animal circle. Gas concentrations in the blood samples and exhalate were measured by a gas chromatograph (Hewlett Packard 5890 series II, GA, USA). From the blood- and mixed expired gas the inline image was calculated according to the original technique (Wagner et al. 1974). Of the data, blood flow (inline image) and standard deviation of its logarithmic distribution (log SDQ), tidal volume (inline image) and standard deviation of its logarithmic distribution (log SDV), right to left vascular shunt% of the inline image (inline image; perfusion of lung regions with inline image < 0.005), and dead space% of the VT (VD; ventilation of lung regions including apparatus dead space with inline image > 100) are presented. The residual sums of square of inline image distributions (RSS) represents the minimised sum of squared differences between the calculated and measured retention's of the six inert gases.

Data collection

Data were collected after 60 minutes of anaesthesia equilibration without PiNO (baseline). Following baseline data collection, NO delivery commenced as described above. NO was pulsed during the first 30% and 60% of inspiration, respectively and data were collected 15 minutes after equilibration at each percentage of PiNO (PiNO30% and PiNO60%). The order of delivery was not randomized.

Data analysis

Data are presented as descriptive data only. No statistical analysis was performed.

Results study 1

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods Study 1
  5. Materials and methods study 2
  6. Results study 1
  7. Results study 2
  8. Discussion
  9. Conclusions
  10. Acknowledgements
  11. References

Determination of the pulse length corresponding to peak PaO2 in individual horses

Administration of all pulse lengths of PiNO significantly increased the PaO2 at T60-T75 compared to the PaO2 at T45. In 4 horses, the optimum PaO2 was obtained at 30% of the inspiration and in 2 horses optimum PaO2 was obtained at 45% of the inspiration. Horses then continued receiving PiNO for maintenance at the percentage that resulted in the highest individual PaO2 for T90-T150.

Haemodynamic and pulmonary data

Results (mean ± SD) are presented in Table 1 and Figs 1 and 2. Significant differences (p from <0.01 to <0.0001 depending on data and time) occurred both within and between groups in data associated with pulmonary function. PaO2, SaO2, P v̄O2, S v̄O2, CaO2, C v̄O2, DO2 and (A-a)DO2 were higher and inline image were lower in the NO group than the C group at most time points. PAO2 and arterial pH were lower while PaCO2 was higher in the NO group than the C group at time points towards the end of anaesthesia but there was no difference in fR or TV.

Table 1. Cardiopulmonary data from Study 1
GroupTime: Parameter:T45T60T75T90T105T120T135T150
  1. Select cardiopulmonary data from horses anaesthetized with (PiNO) and without (C) pulse delivered iNO (Study 1). n = 6 in each group. Data are presented as mean ± SD. For abbreviations see text.

  2. a

    Significant difference between groups and.

  3. b

    Significant difference from baseline (45 minutes).

PiNOHR beats minute−136 ± 235 ± 336 ± 435 ± 236 ± 337 ± 537 ± 437 ± 5
CHR beats minute−135 ± 335 ± 236 ± 337 ± 335 ± 135 ± 335 ± 234 ± 2
PiNOfR breaths minute−13 ± 14 ± 23 ± 24 ± 23 ± 23 ± 23 ± 24 ± 2
CfR breaths minute−15 ± 25 ± 24 ± 25 ± 25 ± 25 ± 25 ± 35 ± 3
PiNO

PaCO2 kPa

(mmHg)

8.6 ± 0.56

(63 ± 4)

10.5 ± 1.6b

(79 ± 12b)

9.6 ± 0.4b

(72 ± 3b)

11.1 ± 1.5a

(83 ± 11a)

11.9 ± 1.5ab

(90 ± 11ab)

11.8 ± 1.2ab

(88 ± 9ab)

11.7 ± 1.5ab

(88 ± 11ab)

12.2 ± 1.5ab

(92 ± 11ab)

C

PaCO2 kPa

(mmHg)

7.8 ± 1.0

(59 ± 8)

8.3 ± 1.3

(62 ± 10)

8.3 ± 1.3

(62 ± 10)

8.6 ± 1.5

(64 ± 11)

8.6 ± 1.5

(64 ± 11)

9.1 ± 1.9b

(68 ± 14b)

9.0 ± 1.8b

(68 ± 14b)

9.2 ± 2.1b

(69 ± 16b)

PiNOpHa7.24 ± 0.03a7.22 ± 0.05a7.20 ± 0.05a7.19 ± 0.05ab7.17 ± 0.05ab7.18 ± 0.04ab7.1 8 ± 0.05ab7.16 ± 0.04ab
CpHa7.32 ± 0.037.30 ± 0.047.30 ± 0.057.29 ± 0.057.28 ± 0.057.27 ± 0.067.27 ± 0.067.27 ± 0.07
PiNOQt Litres minute−126.2 ± 4.626.2 ± 4.2     26.6 ± 2.1
CQt Litres minute−127.6 ± 4.626.3 ± 3.4     25.2 ± 2.5
PiNOMPAP mmHg27 ± 926 ± 725 ± 825 ± 724 ± 824 ± 825 ± 924 ± 9
CMPAP mmHg27 ± 423 ± 824 ± 926 ± 9b27 ± 8b28 ± 7b27 ± 10b27 ± 9b
PiNOMAP mmHg66 ± 878 ± 5b80 ± 4b80 ± 4b77 ± 7b79 ± 10b78 ± 10b79 ± 14b
CMAP mmHg70 ± 1172 ± 1077 ± 1180 ± 882 ± 882 ± 683 ± 682 ± 5
PiNOET-1 Picomol mL−16.62 ± 1.89      7.82 ± 1.16
CET-1 Picomol mL−16.71 ± 2.04      7.38 ± 2.92
image

Figure 1. Select respiratory data from isoflurane-anaesthetized horses with (PiNO) and without (C) pulse delivered iNO. Data are presented as mean ± SD. See text for abbreviation. * = significant difference between groups. + = significant difference from baseline (45 minutes).

Download figure to PowerPoint

image

Figure 2. inline image in horses anaesthetized with (PiNO) and without (C) pulse delivered iNO. Data are presented as mean ± SD. * = significant difference between groups + = significant difference from baseline (45 minutes).

Download figure to PowerPoint

HR, SAP, MAP, DAP, SPAP, MPAP, DPAP, fR, inline image and ET-1 concentrations were not different between groups. Hb ranged from 9.0 to 11.5 throughout the studies and was not different over time or among groups.

Results study 2

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods Study 1
  5. Materials and methods study 2
  6. Results study 1
  7. Results study 2
  8. Discussion
  9. Conclusions
  10. Acknowledgements
  11. References

Descriptive data (mean ± SD) are presented in Table 2 and Fig. 3. The NO delivery was 6.5 ± 3.8 and 12.8 ± 7.4 μ mol minute−1 during the PiNO30% and PiNO60%, respectively. The exhaled peak NO fractions was 0.5 ± 0.2 and 2.6 ± 2.0 ppm and end-tidal NO was 0.2 ± 0.1 and 0.3 ± 0.1 ppm, in the PiNO30% and PiNO60%, respectively.

Table 2. Ventilation-perfusion distribution and cardiorespiratory data from Study 2
Horse123
  1. Cardiopulmonary, inline image and inline image data from 3 horses before (baseline) and during PiNO pulsed into the first 30% and 60% of the inspiratory time (Study 2). Data are descriptive only and are presented as mean ± SD. See text for abbreviations. See Fig. 3 for graphic presentation of ventilation/perfusion data.

Duration of PiNO0%30%60%0%30%60%0%30%60%
inline image (%)433027443538302533

PaO2 (kPa)

(mmHg)

11.8

87

27.8

209

28.0

210

15.4

115

27.6

207

27.2

204

28.8

216

35.7

268

27.8

209

logSDQ0.680.620.650.850.760.640.840.730.68
logSDV0.630.620.620.830.760.670.690.660.61
VD/VT (%)555253495853696871
HR (beats minute−1)393837423940414237
inline image (litres minute−1)29.233.934.442.337.136.021.525.431.4
MAP (mmHg)849492888988717073
MPAP (mmHg)182631222424322830
fR (breaths minute−1)3.73.33.12.02.02.24.44.44.7
VT (litres minute−1)5.75.65.67.27.27.14.74.95.0
RSS2.001.820.531.411.181.193.895.341.69
image

Figure 3. Ventilation-perfusion distribution in three horses (Study 2). The ventilation-perfusion distribution (inline image) presented as a log scale and the ventilation and perfusion in L minute−1 as open and closed circles, respectively. Development of a large right to left vascular shunt (inline image = 0) and an increased distribution of ventilation-perfusion were evident during isoflurane anaesthesia (Base line, left panels). Pulsed inhalation of nitric oxide (PiNO) during the first 30% or 60% of the breath are presented as PiNO30% (middle panels) and PiNO60% (right panels). PiNO30% resulted in a clear reduction of the shunt but no further improvement occurred at PiNO60%. See Table 1 for corresponding data.

Download figure to PowerPoint

When compared to baseline, PaO2 increased while inline image and logSDQ decreased in all horses with PiNO30%. In horses 1 and 2, PiNO60% caused minimal changes from PiNO30%. In horse 3, PiNO60% caused deterioration in PaO2 compared to PiNO30% and returned towards baseline values.

Compared to baseline, inline image decreased in all three horses during PiNO30% and was lower in two horses and higher in one horse during PiNO60%. Log SDQ decreased and log SDV decreased or did not change during PiNO compared to baseline and VD/VT changed minimally throughout the study. The RSS represents the weighted residual sum of squares of the model fits. Model fitting is performed using a weighted least squares approach, where values of model parameters were found so as to minimize the difference between measured (m) and model predicted (p) values of excretion and retention.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods Study 1
  5. Materials and methods study 2
  6. Results study 1
  7. Results study 2
  8. Discussion
  9. Conclusions
  10. Acknowledgements
  11. References

In Study 1, the administration of PiNO pulsed into the first 30% or 45% of the inspiration resulted in a significant and sustained improvement in pulmonary function as evidenced by an increase in PaO2 and SaO2 and a decrease in calculated inline image in PiNO horses as compared to C horses. In Study 2, PaO2 was higher, measured inline image was lower and inline image distributions were more closely matched in all horses receiving PiNO in the first 30% of inspiration than values at baseline. No further improvement was measured when PiNO was distributed during the first 60% of the inspiration. The data from the two phases of this study demonstrate that it is possible to improve arterial oxygenation with PiNO in laterally recumbent isoflurane-anaesthetized horses. The marked improvement in arterial oxygenation during PiNO was due to an evident reduction of inline image. Further, log SDQ decreased during PiNO, indicating more efficient matching of inline image. Although the PiNO horses had a lower inline image than the C horses from T60 to T150, all horses in this study had a large inline image, which continued to increase in the C horses. The large inline image was unexpected since the horses were in lateral recumbency and not expected to develop a shunt fraction of the same magnitude as that produced when in dorsal recumbency. However, the horses in our study were overweight and not exercised and actually may be representative of many of the clinical cases that are anaesthetized.

The reduction in shunt with concomitant improvement in arterial oxygenation during iNO has been suggested to be a result of redistribution of blood flow from low to high inline image regions (Putensen et al. 1995; Hopkins et al. 1997; Heinonen et al. 2000, 2001, 2002). This is also most likely the mechanism in the horse, although this is still unproven and somewhat in question since success of this therapy would require overcoming gravitational effects caused by the large vertical lung gradient in an adult horse. Although normal pulmonary blood flow distribution in the horse is only partially gravity-dependent, the fact that the shunt occurs in the dependent lung means that blood redistributed from the shunt would have to more against gravity (Dobson et al. 1985; Bernard et al. 1996; Hlastala et al. 1996; Erickson et al. 1999; Harmegnies et al. 2002). However, there is evidence that this does indeed occur as blood flow was shunted to the nondependent lung in anaesthetized laterally-recumbent horses (Stolk 1982).

The advantage of PiNO when compared to continuous iNO is that the PiNO delivered in the first part of the inspiration is likely to selectively reach the well-ventilated alveoli that actively participate in gas exchange, whereas continuous iNO could also reach alveoli that only open occasionally and thereby only inconsistently participate in gas exchange (Heinonen et al. 2000, 2001, 2002). Selective redistribution within the ventilated lung area receiving PiNO may well explain the therapeutic effect observed in this study compared to the lack of effect with continuous distribution of iNO that occurred in a previous study (Young et al. 1999). This may also explain why, in the inert gas portion of our study (Study 2), the inline image was decreased and inline image matching and PaO2 were increased in all horses with PiNO delivered into the first 30% of the inhalation but did not change (horses 1 and 2) or trended back towards baseline (horse 3) when PiNO delivery was increased to the first 60% of the inspiration.

In addition to the enhanced effectiveness of pulse-delivered when compared to continuously-delivered iNO, PiNO may also improve the safety of iNO therapy. When iNO is delivered continuously or for a large percentage of the inspiratory time, NO can accumulate in the breathing circuit and be re-breathed in uncontrollable doses. In the circuit, the NO reacts with oxygen forming the toxic gas NO2. This gas acts mainly as an irritant affecting the respiratory mucosa but extremely high-dose exposure may result in pulmonary oedema and diffuse lung injury (Elsayed 1994). Although we did not measure NO in the breathing circuit in phase 1 of the study, pulse administration of PiNO during the first 30% of inspiration during phase 2 resulted in an average end-tidal NO of 0.2 ppm, which was negligible and unlikely to result in NO2 formation (Nyman et al. 2013).

In our study, the PaCO2 was significantly higher and pHa significantly lower in the NO horses as compared to the C horses, and this is similar to previously reported results (Nyman et al. 2013) Although the respiratory rates and tidal volume were not statistically different between the two groups, they appear to have been clinically different, leading to the differences in PaCO2 and pHa. The high PaCO2 also affected the (A-a)DO2 since CO2 is part of the formula. Had the CO2 been equal in both horses, the (A-a)DO2 would likely have been even lower in the PiNO horses. Because we did not want to affect the ability to determine the impact of PiNO, we did not ventilate the horses but hypercarbic horses anaesthetized in the clinical setting would be ventilated.

Intriguingly, the PiNO horses did not have lower PAP as would have been expected when considering the mechanism of action of iNO. However, the same result occurred in earlier studies from our laboratory (Heinonen et al. 2001, 2002; Nyman et al. 2013). Although no significant changes in overall PAP were detected, the lung is a very heterogeneous organ when it comes to inline image matching and the inability to detect a global change does not exclude the fact that regional differences might occur. Increased pulmonary arterial pressure often occurs secondary to hypoxaemia because of hypoxic pulmonary vasoconstriction (HPV). In HPV, pulmonary arteries constrict in order to divert blood from atelectatic lung areas to well-oxygenated alveoli (Nagendran et al. 2006). Although conflicting evidence exists, this vasoconstriction may be due to an increase in the potent vasoconstrictor, ET-1 (Allen et al. 1993; Wang et al. 1995). In our study, ET-1 was not different within or between groups in spite of the fact that the PaO2 in the C group was significantly lower than that of the PiNO group at all time points other than baseline. This could be due to the fact that ET-1 may not be responsible for producing HPV in all species (Douglas et al. 1993; Wong et al. 1993), that horses don't have a profound HPV response (Elliott et al. 1991) or that HPV is blunted by inhalant anaesthetics (Karzai et al. 1998; Middelveld & Alving 2002). Finally, no changes in SAP, MAP, DAP, HR or inline image occurred during the study, emphasizing the fact that iNO has a local pulmonary effect but does not affect the systemic circulation.

There are two further potential limitations of our study. The first is that the horses in our study were normoxic and would not have required PiNO. Our research, as with most research, started with healthy research animals to determine the effect of a treatment. Based on the success of PiNO, the next phase of study will be hypoxemic horses. The second limitation is that the impact of intra-operative hypoxaemia in horses is unknown. Although many horses recover without incident, hypoxaemia associated with anaesthesia has been directly implicated in postanaesthetic cerebral necrosis (McKay et al. 2002), hepatic insult in combination with halothane (Whitehair 1996), decreased skeletal muscle oxygenation (Steffey et al. 1992; Whitehair 1996; Portier et al. 2009), increased serum lactate (Portier et al. 2009) and lactic acidemia (Taylor 1999). Poor muscle perfusion caused by hypotension is directly implicated in anaesthesia-related myopathies and neuropathies (Grandy et al. 1987; Lindsay et al. 1990; Richey et al. 1990; Young & Taylor 1993; Duke et al. 2006), although the damage in this case is probably due to decreased oxygen delivery and ischemia rather than to the mere presence of hypotension. This is evidenced by the fact that lactate is often elevated, indicating anaerobic metabolism and the likelihood of local hypoxia (Lindsay et al. 1990). More research is needed to determine the impact of intraoperative hypoxaemia in horses.

Conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods Study 1
  5. Materials and methods study 2
  6. Results study 1
  7. Results study 2
  8. Discussion
  9. Conclusions
  10. Acknowledgements
  11. References

In conclusion, PiNO pulsed into the first 30% of the inspiration improves oxygenation due to reduced inline image and improved inline image matching in laterally recumbent isoflurane anaesthetized horses without affecting ET-1.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods Study 1
  5. Materials and methods study 2
  6. Results study 1
  7. Results study 2
  8. Discussion
  9. Conclusions
  10. Acknowledgements
  11. References

This study was supported by grants from Swedish-Norwegian Foundation for Equine Research. The authors are grateful to Anneli Rydén and Eva-Maria Hedin for skillful technical assistance.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods Study 1
  5. Materials and methods study 2
  6. Results study 1
  7. Results study 2
  8. Discussion
  9. Conclusions
  10. Acknowledgements
  11. References