Effect of different pulses of nitric oxide on venous admixture in the anaesthetized horse

E. Heinonen1,3, G. Nyman2, P. Meriläinen3 and M. Högman*,1

1Department of Medical Cell Biology, Section of Integrative Physiology, Uppsala University, Box 571, SE-75123 Uppsala, Sweden. 2Department of Large Animal Clinical Sciences, Faculty of Veterinary Medicine, Swedish University of Agricultural Sciences, Uppsala, Sweden. 3Datex-Ohmeda Research Department, Helsinki, Finland*Corresponding author

Accepted for publication: November 9, 2001


    Abstract
 Top
 Abstract
 Introduction
 Material and methods
 Results
 Discussion
 References
 
Background. Dependent atelectatic lung areas open towards the end of inspiration when the lung opening pressure increases, and recollapse during expiration. We hypothesized that inhaled nitric oxide (NO) counteracts hypoxic vasoconstriction in these collapsing lung areas, resulting in increased pulmonary shunt perfusion.

Methods. We administered NO as a pulse and varied the pulse timing during inspiration in equine anaesthesia, where atelectasis develops regularly. Six spontaneously breathing standard breed trotters were studied under isoflurane anaesthesia in lateral recumbency. NO pulsed into the first 30% of inspiration (group NOp1) was assumed to affect open lung areas. To cover more open lung areas NO was then pulsed into the first 60% of inspiration (group NOp2). In a third group, administration between 50 and 80% of inspiration was aimed at the intermittently opening lung areas (group NOp3).

Results. With NOp1, venous admixture decreased by 8 (2)% (mean (SEM), P=0.045) and with NOp2 by 10 (1)% (P=0.01). With NOp3, venous admixture reduction was insignificant.

Conclusions. Pulsed administration of NO in early inspiration is optimal in reducing right to left vascular shunt in atelectatic equine lung. This reduction is positively correlated with the magnitude of the initial shunt. With administration in early inspiration, NO is mostly taken up by the lung. This prevents NO accumulation and NO2 formation in rebreathing circuits. These findings may be important in humans when atelectasis occurs increasingly with overweight and age during anaesthesia, but also in postoperative intensive care and in ARDS.

Br J Anaesth 2002; 88: 394–8

Keywords: pharmacology, nitric oxide; lung, atelectasis; anaesthesia, equine; horse


    Introduction
 Top
 Abstract
 Introduction
 Material and methods
 Results
 Discussion
 References
 
Nitric oxide (NO) delivered by inhalation induces vasodilatation in the lung.1 2 This effect favours perfusion in ventilated lung areas, equating ventilation and perfusion distribution. The improved matching of ventilation and perfusion results in an increase in pulmonary artery oxygen partial pressure (PaO2).3 In humans with ventilation– perfusion mismatching, i.e. chronic obstructive pulmonary disease and adult respiratory distress syndrome (ARDS), inhalation of NO sometimes has no effect,4 5 or even causes a decline in oxygenation.6 These conflicting results are explained by NO inhalation occurring in lung compartments of low or no ventilation, counteracting hypoxic vasoconstriction, and increasing shunt perfusion.6 7 Alveolar recruitment enhances the effect of NO therapy in ARDS patients where one-third of the lung may be aerated, but dependent lung areas are consolidated and atelectatic.810 The boundary zone between the normal and altered areas where the lung opens and closes intermittently forms an area of low ventilation. As the inspired gas flow favours the open lung areas, the low ventilation areas are only reached towards the end of inspiration.9 We hypothesize that NO, because of its high alveolar diffusion capacity, induces vasodilatation in the boundary zone where the ventilation is low, possibly increasing ventilation–perfusion mismatch and impairing gas exchange.11

In horses, atelectasis and ventilation–perfusion mismatch develop regularly during anaesthesia.12 We therefore used equine anaesthesia to study the effect of NO delivered to various lung areas, in reducing shunt.


    Material and methods
 Top
 Abstract
 Introduction
 Material and methods
 Results
 Discussion
 References
 
Six standard bred trotters with a mean (SEM) [range] age of 5 (1) [2–8] yr and weight of 513 (28) kg were examined. The local Ethics Committee for Animal Experiments approved the study.

Procedure and NO delivery
One hour after induction of anaesthesia, a period used for catheterization and to achieve stable anaesthesia, the first baseline readings were recorded and then NO was administered during the first 30% of inspiration (NOp1). Fifteen minutes after cessation of NOp1, a further baseline was recorded and NO was administered during the first 60% of inspiration (NOp2). After completion of NOp2, there was another 15-min resting period, followed by a new baseline recording. NO was then administered during 50–80% of inspiration (NOp3) to reach poorly ventilated areas of the lung. The response was recorded after 5 min of NO delivery in all delivery modes. As NOp3 involved a risk of a severe increase in shunt and hypoxaemia, no randomization of the order of administration of NO was performed.

NO was injected with a pulsed delivery device developed by Datex-Ohmeda Research Department, Helsinki, Finland. The device generates a NO pulse by regulating the flow of the NO-containing gas mixture for the required period. For the flow regulation, the device has a proportionally adjustable valve and a dose flow sensor. The NO pulse was injected during inspiration (NOp1, NOp2, or NOp3). A D-liteTM (Datex-Ohmeda) type flow sensor, dimensioned appropriately for the monitoring of horse ventilation, was used to synchronize the dose with breathing. The pulse was injected at the beginning of the tracheal tube to allow selective administration to different parts of the lungs. The NO supply was 2000 µl litre–1 NO in N2 (AGA AB, Lidingö, Sweden).

Anaesthesia
During the12 h preceding anaesthesia, the horses were allowed to drink only water. Acepromazine 0.05 mg kg–1 was given intramuscularly approximately 30 min before induction of anaesthesia. An i.v. infusion of 7.5% guaifenesin was given until the horse became ataxic. Then anaesthesia was induced by the i.v. injection of thiopental 5 mg kg–1. The trachea was intubated and the horses were positioned in lateral recumbency, and connected to a large-animal breathing circuit. Anaesthesia was maintained with spontaneous breathing of isoflurane in oxygen with a fresh gas flow of 5–6 litre min–1. The end-tidal isoflurane fraction was 1.3–1.7%, adjusted for stable anaesthesia.

A thermodilution catheter (7F, Swan-Ganz) was inserted with an introducer kit (Arrow Int. Inc., Reading, PA, USA) through a right jugular vein to the pulmonary artery. This catheter was used for pulmonary artery pressure (PAP) and cardiac output (Q·t) measurement, and mixed venous blood sampling. A pigtail, multi-hole catheter (Cook Europe A/S, Söborg, Denmark) was inserted with a similar technique into the same jugular vein, advanced to the right ventricle, and retracted into the right atrium. This catheter was used as an injection port for Q·t measurement. A catheter to measure mean arterial pressure (MAP) and to sample arterial blood was introduced percutaneously into the facial artery (Insyte-W, 18GA, Becton-Dickson, Ohmeda, Helsingborg, Sweden). The catheters were positioned under pressure-tracing guidance with simultaneous ECG monitoring and locked in position with a Luer-lock adaptor.

Monitoring
Expired NO was monitored with a chemiluminesence analyser prototype (Datex-Ohmeda) connected between the Y-piece and the point of NO administration. The analyser was calibrated with the mixture 100 µl litre–1 NO in N2 (AGA AB) and with room air depleted of NO with a charcoal absorber. The signal rise time of the analyser was 200 ms, which allowed analysis of the expired breath pattern. The monitor was used to determine the end-tidal and peak expired NO fractions.

MAP and PAP were measured by pressure transducers positioned at the level of the sternal manubrium, which was considered to correspond to the level of the right atrium. These pressures, and also the inspired (FIO2) and expired (FEO2) gas oxygen fractions, respiratory rate (RR), tidal volume (VT), end-tidal carbon dioxide and oxygen fractions, and isoflurane fractions, were recorded on an AS/3 AMTM (Datex-Ohmeda) anaesthesia monitor.

Cardiac output was measured with the thermodilution technique (Cardiac Output Computer Model 9520 A, Edwards laboratory, Santa Ana, CA, USA); 20 ml 0°C 0.9% saline was injected into the right atrium through the pigtail catheter.

Arterial and central venous blood was obtained and analysed with a standard electrode technique (ABL 5, Radiometer, Copenhagen, Denmark). The arterial and venous (PO2) oxygen partial pressures were measured at standard electrode temperature (37°C) and the oxygen saturations (SaO2 and SvO2, respectively) were calculated from the blood samples using the human P50 value 3.57 kPa, which is close to the equine value of 3.41 kPa.

To estimate the pulmonary shunt, venous admixture was calculated from the blood gas values using the method described by Berggren.13

Statistical analysis
Repeated measurement ANOVA was used to compare data within the group on different parts of the study. The Tukey honest significant difference test was used for post hoc comparisons and probability values were calculated. For all statistical calculations, the Statistica/w 5.0 software package (StatSoft Inc., Tulsa, OK, USA) was used. Results are given as mean values (SEM). In the analysis, the probability P<0.05 was considered as significant.


    Results
 Top
 Abstract
 Introduction
 Material and methods
 Results
 Discussion
 References
 
Administration of NO did not affect ventilation, Q·t, or PAP, irrespective of the NO delivery period (Table 1). As a result of the longer pulse duration, the amount of NO delivered in NOp2 was twice that delivered in NOp1 and NOp3 to keep the NO fraction constant in the affected zones (Table 1). The NO fraction inspired during the pulse was calculated from the dose, respiration rate, NO pulse duration, and average inspiration flow (Table 1). There were no significant differences in the baseline values of venous admixture and PaO2 in NOp1, NOp2, and NOp3. The venous admixture was 27 (4), 29 (4), and 30 (3)%, respectively. The PaO2 was 27 (7), 22 (5), and 20 (4) kPa, respectively. With NOp1, the venous admixture decreased (P=0.045) and PaO2 increased (P=0.033). With NOp2, the venous admixture also decreased (P=0.01) and PaO2 increased (P=0.021). Despite the difference in the delivered amounts of NO in NOp1 and NOp2 (P=0.011), the effects of NO were equal. NOp3 did not improve venous admixture nor PaO2. The changes in venous admixture and PaO2 induced by the different delivery modes are presented in Figure 1.


View this table:
[in this window]
[in a new window]
 
Table 1 Tidal volume (VT), respiration rate (RR), cardiac output (Q·t), mean pulmonary artery pressure (PAP), NO delivery rate, and fraction in the inhaled gas when administering NO during the first 30% (NOp1), during the first 60% (NOp2), and during the period between 50 and 80% of inspiration (NOp3). The NO fraction is calculated from the NO delivery, pulse duration, and ventilation
 


View larger version (9K):
[in this window]
[in a new window]
 
Fig 1 The change in venous admixture (Qs/Qt) and PaO2 when NO is delivered as a pulse in different parts of inspiration. Administration of NO during the first 30% (NOp1), during the first 60% (NOp2), and during the period between 50 and 80% of inspiration (NOp3), respectively. (A) NOp1, (B) NOp2, and (C) NOp3.

 
When NO was delivered in early inspiration, the degree of reduction in venous admixture depended on its baseline value (r=0.94, P=0.006). The correlation for NOp1 is presented in Figure 2. Regression analysis shows that with this NO therapy, venous admixture could not be reduced when the baseline venous admixture was below 14%. However, baseline venous admixture above 14% could be reduced by 55%.



View larger version (13K):
[in this window]
[in a new window]
 
Fig 2 Correlation between the reduction in venous admixture and the venous admixture before NO delivery in the NOp1 mode (delivery during the first 30% of inspiration). The solid line is the regression line (r=0.94) and the dashed curves are 95% confidence range.

 
The expired peak NO fraction was higher during NOp2 (P=0.047) and NOp3 (P<0.001) than during NOp1. The end-tidal NO fraction was equally low in all NO delivery modes (P=0.13, Fig. 3). Because of the large NO expiration during NOp3, the fresh gas flow had to be increased to the minute ventilation to prevent rebreathing of NO from the circuit.



View larger version (8K):
[in this window]
[in a new window]
 
Fig 3 The exhaled NO peak (closed bar) and end-tidal NO (open bar) fractions in the different delivery modes. (For explanation of NOp1, NOp2, and NOp3, see Fig. 1.)

 

    Discussion
 Top
 Abstract
 Introduction
 Material and methods
 Results
 Discussion
 References
 
The large degree of venous admixture (27–30%) and low PaO2 (20–27 kPa) in relation to the FIO2 (1.0) indicate the presence of atelectasis in the horses in this study.12 NO had a different effect on venous admixture when delivered in different phases of inspiration. Delivered during early inspiration, NO resulted in a reduction of venous admixture and improved arterial oxygenation. The venous admixture of the horses did not change significantly when NO was administered from 50 to 80% of inspiration, supporting the study hypothesis.

When NO was delivered in early inspiration, the decrease in venous admixture was positively correlated with its baseline value. A similar correlation was observed in our previous study of five horses in dorsal recumbency, where NO was only delivered during the first half of inspiration in three different procedures.14 The venous admixture was reduced by 14% from a baseline of 37% when 9 µmol min–1 of NO was delivered during spontaneous breathing (reduction predicted from the line of identity of Fig. 2 is 13%); a reduction of 10% from a baseline of 32% occurred when 17 µmol min–1 of NO was delivered during spontaneous breathing (predicted value 10%) and a reduction of 7% from a baseline of 26% occurred when 30 µmol min–1 of NO was delivered during mechanical ventilation (predicted value 7%).14 In another study of seven horses given NO at a constant 10 p.p.m. inspired concentration, the reduction in venous admixture fell below the line of identity with a baseline value of 21% and reduction of 1% (predicted value 4%).15 This is comparable with the NO delivery given in NOp3 with corresponding figures of 30% and 5%, respectively (predicted value 9%).

The reduction in venous admixture in human ARDS patients when NO is administered during the whole of inspiration is independent of the initial level,4 estimations derived from that data suggest reduction of only 5% from 40% (14% predicted from Fig. 2). The smaller reduction in venous admixture when NO is delivered during the whole of inspiration may be explained by NO affecting the border zone, dilating the vessels affected by local hypoxic vasoconstriction, which counteracts any reduction in shunt caused by NO in the ventilated lung areas.

The proposed mechanism for the decrease in venous admixture with NO inhalation is vasodilatation as a result of decreased pulmonary vascular resistance. Even though the venous admixture was reduced with NO delivery during early inspiration in the present study, NO did not alter PAP or Q·t. This surprising result may be because of vasoconstriction occurring in the zones not receiving NO. Perfusion redistribution without a change in PAP or Q·t has also been observed in humans when NO is administered to one lung only.16 This was associated with a significant increase in the pulmonary vascular resistance in the lung not receiving NO. Down-regulation of endogenous NO production in both lungs during unilateral administration of NO is proposed as one explanation of this finding. Another explanation involves vessel elasticity. Distension in regions receiving NO would allow the vessels to constrict elsewhere, because Q·t remains unaltered, thus preserving PAP.

The alveolar NO uptake determined experimentally in humans is 95–100%.17 In simulation experiments, the uptake of NO administered in the first third of inspiration in pigs is 80–90%.18 From the mean of the exhaled peak- and end-tidal fractions (Fig. 3), and the ventilation and NO delivery data in Table 1, the uptake in the NOp1 mode can be estimated at 92%, suggesting that NOp1 is delivered successfully to the alveolar region. When NO is delivered at a constant inspired concentration, the uptake rate decreases to 5619 or 50–60%,18 because of the expiration of NO from the anatomical dead space. In a rebreathing circuit, the expired NO is recycled leading to administration of an uncontrollable dose. In this circuit, the NO also reacts with oxygen forming NO2. Hence in rebreathing circuits, NO should be delivered only in short pulses synchronous with early inspiration.

During human anaesthesia, the ventilation–perfusion ratio is high in the uppermost regions and low at the base of the lungs, which are subject to atelectasis.20 Anaesthesia also induces at least a 70% increase in closing volume.21 This increase is positively correlated with age and body mass index.21 22 In morbidly obese patients, an observed 50% reduction in functional residual capacity compared with the pre-anaesthesia value has been found to be closely related to the development of atelectasis.23 24 During general anaesthesia with mechanical ventilation, 75% of the impairment of arterial oxygenation may be explained by atelectasis and airway closure.25 Common measures to compensate for the impairment are the use of a higher FIO2 or increasing the positive end-expiratory pressure (PEEP). The development of atelectasis correlates with a high FIO2, and PEEP may increase the shunt, leading to deteriorating oxygenation.25 26 Pulsed NO may provide a new method for maintenance of arterial oxygenation during anaesthesia, which could even reduce atelectasis formation by allowing a reduction in FIO2.

In conclusion, to reduce pulmonary shunt, NO delivered in pulses in early inspiration is the optimal method. With this type of administration, the degree of improvement correlates positively with the initial magnitude of the shunt. In rebreathing systems, NO should be delivered only in pulses to permit accurate dosing and to avoid NO2 formation in the breathing circuit.


    Acknowledgement
 
The authors wish to thank Anna Edner, Karin Thulin and Eva-Maria Hedin for laboratory assistance.


    References
 Top
 Abstract
 Introduction
 Material and methods
 Results
 Discussion
 References
 
1 Frostell C, Fratacco MD, Wain JC, Jones R, Zapol WM. Inhaled nitric oxide. A selective pulmonary vasodilator reversing hypoxic pulmonary vasoconstriction. Circulation 1991; 83: 2038–47[Abstract]

2 Pepke Zaba J, Higenbottam TW, Dinh Xuan AT, Stone D, Wallwork J. Inhaled nitric oxide as a cause of selective pulmonary vasodilatation in pulmonary hypertension. Lancet 1991; 338: 1173–4[ISI][Medline]

3 Rossaint R, Falke KJ, López F, Slama K, Pison U, Zapol WM. Inhaled nitric oxide for the adult respiratory distress syndrome. N Engl J Med 1993; 328: 399–405[Abstract/Free Full Text]

4 Rossaint R, Gerlach H, Schmidt-Ruhnke H, et al. Efficacy of inhaled nitric oxide in patients with severe ARDS. Chest 1995; 107: 1107–15[Abstract/Free Full Text]

5 Brett SJ, Hansell DM, Evans TW. Clinical correlates in acute lung injury response to inhaled nitric oxide. Chest 1998; 114: 1397–404[Abstract/Free Full Text]

6 Barberà JA, Roger N, Roca J, Rovira I, Higenbottam TW. Worsening of pulmonary gas exchange with nitric oxide inhalation in chronic obstructive pulmonary disease. Lancet 1996; 347: 436–40[ISI][Medline]

7 Gerlach H, Roissant R, Pappert D, Falke KJ. Time-course and dose-response of nitric oxide inhalation for systemic oxygenation and pulmonary hypertension in patients with adult respiratory distress syndrome. Eur J Clin Invest 1993; 23: 499–502[ISI][Medline]

8 Johannigman JA, Davis K, Campbell RS, Luchette FA, Frame SB, Branson RD. Positive end-expiratory pressure and response to inhaled nitric oxide: changing nonresponders to responders. Surgery 2000; 127: 390–4[ISI][Medline]

9 Tobin MJ. Culmination of an era in research on the acute respiratory distress syndrome. N Engl J Med 2000; 342: 1360–1[Free Full Text]

10 Ware LB, Matthay MA. The acute respiratory distress syndrome. N Engl J Med 2000; 342: 1334–49[Free Full Text]

11 Borland CD, Higenbottam TW. A simultaneous single breath measurement of pulmonary diffusing capacity with nitric oxide and carbon monoxide. Eur Respir J 1989; 2: 56–63[Abstract]

12 Nyman G, Funkquist B, Kvart C, et al. Atelectasis causes gas exchange impairment in the anaesthetised horse. Equine Vet J 1990; 22: 317–24[ISI][Medline]

13 Berggren SM. The oxygen deficit of arterial blood caused by non-ventilating part of the lungs. Acta Physiol Scand 1942; 4 (Suppl XI): 1–92

14 Heinonen E, Nyman G, Meriläinen P, Hedenstierna G, Högman M. Pulsed delivery of nitric oxide counteracts hypoxaemia in the anaesthetised horse. Vet Anesth Analg 2001; 28: 3–11

15 Young LE, Marlin DJ, McMurphy RM, Walsh K, Dixon PM. Effects of inhaled nitric oxide 10 p.p.m. in spontaneously breathing horses anaesthetized with halothane. Br J Anaesth 1999; 83: 321–24[Abstract/Free Full Text]

16 Hambraeus-Jonzon K, Bindslev L, Frostell C, Hedenstierna G. Individual lung blood flow during unilateral hypoxia: effects of inhaled nitric oxide. Eur Respir J 1998; 11: 565–70[Abstract/Free Full Text]

17 Nathorst Westfelt U, Lundin S, Stenqvist O. Uptake of nitric oxide in acute lung injury. Acta Anaesthesiol Scand 1997; 41: 818–23[ISI][Medline]

18 Heinonen E, Högman M, Meriläinen P. Theoretical and experimental comparison of constant inspired concentration and pulsed delivery in NO therapy. Int Care Med 2000; 26: 1116–23[ISI][Medline]

19 Nathorst Westfelt U, Benthin G, Lundin S, Stenqvist O, Wennmalm Å. Conversion of inhaled nitric oxide to nitrate in man. Br J Pharmacolol 1995; 114: 1621–4[Abstract]

20 Tokics L. Radiospirometry V/Q. Acta Anaesthesiol Scand 1991; 35 (Suppl 95): 97–101[Medline]

21 Rothen HU, Sporre B, Engberg G, Wegenius G, Hedenstierna G. Airway closure, atelectasis and gas exchange during general anaesthesia. Br J Anaesth 1998; 81: 681–6[Abstract/Free Full Text]

22 Strandberg A, Tokics L, Brismar B, Lundquist H, Hedensttierna G. Constitutional factors promoting development of atelectasis during anaesthesia. Acta Anaesthesiol Scand 1987; 31: 21–4[ISI][Medline]

23 Damia G, Mascheroni D, Croci M, Tarenzi L. Perioperative changes in functional residual capacity in morbidly obese patients. Br J Anaesth 1988; 60: 574–8[Abstract]

24 Pelosi P, Ravagnan I, Giurati G, et al. Positive end-expiratory pressure improves respiratory function in obese patients but not in normal subjects during anesthesia and paralysis. Anesthesiology 1999; 91: 1221–31[ISI][Medline]

25 Rothen HU, Sporre B, Engberg B, et al. Prevention of atelectasis during general anaesthesia. Lancet 1995; 345: 1387–91[ISI][Medline]

26 Tokics L, Hedenstierna G, Strandberg A, Brismar B, Lunquist H. Lung collapse and gas exchange during general anaesthesia: effects of spontaneous breathing, muscle paralysis, and positive end-expiratory pressure. Anesthesiology 1987; 66: 157–67[ISI][Medline]