Eur Respir J 1990, 3, 450-455

Comparison of pulmonary and systemic effects of adenosine triphosphate in chronic obstructive pulmonary

disease - ATP: a pulmonary controlled vasoregulator?

S.J.M. Gaba, C. Pretaut

Compared pulmonary and systemic effects of adenosine triphosphate in chronic obstructive pultfWIUJry disease - ATP: a pulmonary con/rolled vasoregulator? SJM. Gaba. C. Pre[aut. ABSTRACT: During stepwlse Incremental Intravenous adenosine triphosphate (ATP) Infusion, systemic and pulmonary vascular effects were compared In 10 patients with stable chronic obstructive pulmonary disease (COPD). Pulmonary vasodilation was: 1) predominant and maximal as early as the lowest dose Infusion (0.1 j.lmOI·kg·1·mln·1) with pulmonary arterial mean pressure (lfpa) (-16%; p<O.Ol) and pulmonary vascular resistance (PVR) decreases (·28%; p<0.005) and simultaneous Increasing 6 PVR/6SVR ratio; 2) associated with worsening hypoxaemia (·14%; p<0.001), but also with Increasing alveolar-~rterlal oxygen pressure differ· ence (ll(A·n)o

1) and venous admixture (Qs/Qt) suggesting some Inhibition

or hypoxic pulmonary vasoconstriction. Systemic vasodllatlon wa.~: 1) clearly dose-dependent, but only reached slgnlflcant level at 0.2 ).l.mol·kg·1·mln·1

with systemic arterial mean pressure (Psa) (-12.5%; pdl.OS) and l>')'Stemlc vascular resistance (SVR) decreases (·30%: pc:O.Ol); 2) associated with arterial carbon dioxide tenslon(Paco

1) decrease (·12.5 %; p<:0.005) and re­

curring uncontrollable hyperpnoea, suggesting a ventllatory stimulatory effect of ATP In man. In patients wltlt stable COPD, ATP Infusion has dual acute haemodynamic effects depending on the dose-level. The predominant pulmonary vnsodllallng effect occurs as early as the lowest dose-levels without any furUter Increase of pulmonary vasodilation. This contrasts with the dose-related systemic vasodJiatlon effect. Such a dual haemody­namic effect Is an Indirect Indication of in 1•ivo lung metabolism of ATP. Eur Respir J., 1990, 3, 450-455.

Service d'Exptoration de la Fonction Respiratoire, Hopital Aiguelongue, Avenue du Major Flandre, Mootpellier, 34059 cedex, France.

Correspondence: S.J.M. Gaba, Service d'Exploration de la Fonction Respiratoire, Hopital Aiguelongue, Avenue du Major Flandre, Montpellier, 34059 cedex, France.

Keywords: Adenosine triphosphate; controlled hypotension; hypoxic pulmonary vasoconstriction; pulmonary circulation; systemic vessels; vasodilation.

Received: April, 1988; accepted after revision November 15, 1989.

Supported in part by a grant from Ayerst Laboratories, Montrouge, France.

Adenosine triphosphate (ATP) is the high energy intra-cellular natural compound. Many extra-cellular effects have also been described, particularly on vessels [1, 2].

described for A TP on pulmonary endothelial cells in culture [6].

In man, A TP-induced systemic hypotension has been observed in annesthetized palients during surgery [3, 4]. However, haemodynam ic data have never been analysed in conscious man during A TP-induced systemic vasodi­lation.

Recently, we have described ATP-induccd pulmon::try vasodilation during low dose-level infusion (0.05 and 0.1 J.Unol·kg·•·m in·1

) in chronically hypoxaem ic patients [5) with chronic obstructive pulmonary disease (COPD). ln this study we did noL observe systemic vasodilation; however, the dose levels we used were four times lower than that used in anaesthetized patients (3]. These different results may indicate different A TP reactivity levels in pulmonary and systemic circulation, which in turn could be accounted for by lung metabolism of ATP. Indeed, a saturable metabolic pathway has previously been

The aim of this s tudy was to detennine the dose-effect relationship of A TP on pulmonary and systemic circula­tion in order to ascertain whether regional differences exist. Such differences could illustrate indirectly an in vivo metabolism of circulating A TP by the pulmonary vascular bed.

Methods

Patients

Ten patients (10 men), 52- 78 yrs of age, with stable moderate to severe COPD, were studied (table 1). Selection criteria [7] included clinical data, chest roentgenogram, electrocardiogram (ECG) and pulmonary function testing; the minimal amount of bronchial obstruction required was percentage forced expiratory

PULMONARY VS SYSTEMIC EFFECTS OF ATP IN COPD 451

volume in one second/forced vital capacity (FEV 1/FVC %) lower than 60. Patients with reversible airway ob­struction were excluded. None of the patients had any other lung disease or a history of thromboembolism or left heart disease.

Treatment with methylxanthines or corticoids was continued; on the other hand, patients using vasodilating drugs (calc ium-blockers, beta-adrenergic drugs or nitrates), those requiring continuous oxygen therapy, and almitrine-treatcd patients were excluded from the study.

Preliminary informed consent was obtained from each patient.

Methods

Heart rate (HR) and ECG were continuously monitored. A Swan-Ganz flow-directed thermodilution catheter (Edwards laboratories, Santa-Anna, CA) was inserted through the right internal jugular vein into the pulmonary artery under continuous pressure wave monitoring. Pressures were measured using a Statham P 23 DB pressure transducer (Stalham instrument, Hato Rey, PR). The zero reference was situated at mid-chest, wil.h l.he subject Jying supine. Pressures were recorded using a thermal-writing recorder (Philips EM 110) during two spontaneous breathing cycles for each assessment Mean pulmonary artery pressure (Ppa), mean pulmonary wedge pressure (Ppaw) and mean right atrium pressure (Pra) were recorded. Cardiac output (Qt) was assessed in triplicate by the thermodilution method using a cardiac index computer (SP 1435; Gould Instruments, Cleveland, OH). Iced (4°C) 9% saline serum was used for this. A teflon catheter was inserted percutaneously into the brachial artery. Mean systemic artery blood pressure (Psa) was recorded using a Statham P 23 DB pressure transducer and the same thermal writing recorder. Arterial and mixed venous blood gases were assessed immediately, using standard electrodes (pH-blood gas analyser 168, Coming Medical Instruments, Medfield, MA).

Calculated data

Cardiac index (Cl) was computed as Cl = Qt/body area (/·min·1·m·Z); stroke index (SI) was calculated as SI = CI/HR (ml ·m·Z·per pulse); pulmonary vascular res is ta nce (PVR) was c al c ul a te d a s P VR = 80·f>pa - Ppaw/QL (dyne·cm·5·s·1) and systemic vascular resistance (SVR) was calculated as SVR = 80·Psa - Pra/ Qt (dyne-cm·5-s·1). P uJmonary and systemic vasodilating effects were compared using their variation ratio (t:.P VR/ t:.SVR), calculated as the ratio of variation of PVR (%) from baseline level to the variation of SVR (%) during the same dose-level of A TP infusion. An increase of this ratio indicates a predominant effect on systemic circula­tion.

Gasometric data allowed us to calculate arterial (Cao2)

and mixed venous (Cvo2) oxygen content as Cxo2 (ml·l-1

) = (1.39-Hb·Sxo2 + 0.003·Pxo2)·10, where Hb in

the haemoglobulin level and Sxo2 and Pxo2 the arterial and mixed venous oxygen saturation and tension, respec­tively. The indexed oxygen consumption was calculated as Vo2 (ml·min·1·m·2) = Cao2·CJ.l0. The alveolar-arterial oxygen pressure gradient (P(A-a)o1) was caJculated as P(A-a)o

2 (kPa) = Plo

2- PAcoJR-Pao2, where Pto

2 was the

inspiratory pressure of 0 2, PAc02 was the aJveolar pres­sure of co2 (assumed to be equal to the arterial Pco1 (Paco~). R was the respiratory exchange ratio, (assumed to be 0.8) and Pao2 .w~ the arteriaJ pressure of Or The venous admixture (Qs/Qt) was calculated on ambient air as Qs/Qt (%) = Cc'o2 - CaojCc'o2 - CVo2 (%) where Cc'o

2 was the capillary oxygen content, estimated using

the calculated alveolar oxygen pressure (PA02), corre­sponding saturation being determined using KELMAN's subroutine [8].

Table 1. -Anthropometric and lung function data of the 10 patients

Age Height Weight Hb FVC yr cm kg g•l' 1 % pred

mean 61 so 10

167 4.1

70.6 155 13.2 10

68 15.1

FEVI % pred

39 16.7

FVC: forced vital capacity; FEV1: forced expiratory volume in

one second; Reference values ore those of the Commission des Communautes Europeennes [9].

Protocol

Subjects rested at least 30 min after the insertion of catheters; no premedication was used and the protocol started when HR and Psa levels were stable. A TP was infused intravenously (iv) using STRIADYNE (Ayerst Laboratories, Montrouge, France) in incremental dose-levels of 0.05, 0.1, 0.15, 0.2 and 0.25 IJlllOl-kg·1·min·l, each infusion lasting for 20 min. The solvent (chlorbutol), an alcoholised compound, was infused at an average total dose of 52±6 ml. Data were serially collected before and during the infusion, between the 15th and 20th min for each infusion.

Statistical analysis

Data variations during incremental A TP infusion were assessed by ANOV A. When significant, variations on data were compared for each dose-level to baseline, by using the t-test for paired data. The Bonferroni method (9] for correcting simultaneous comparison procedure was used (critical Z-value at p=0.005).

Results

All data are summarised in table 2. During the whole procedure, significant variations were found in HR and CI, Psa. an~ SVR, P pa and PVR, Pao2, Paco1 ,

P(A-a)o2 and Qs/Ql. Some were observed beginning with the firs t dose (0.05 j.lmol·kg·1·min·1) (fig. 1). The SVR variations only reached significant level with the

452 S.J.M. GABA, C. PREFAUT

Table 2. - Effects of ATP infusion using stepwise incremental dose-levels

TO 0.05 0.1 0.15 0.2 0.25 ANOVA

Pao2 kPa 8.4±0.3 7.5±0.3 7.2±0.3 7.4±0.3 7.5±0.4 7.3±0.4 0.001 Paco2 kPa 5.6±0.3 5.4±0.4 5.5±0.4 5.2±0.4 5.0±0.4 4.9±0.4 0.001 PYo2 kPa 4.7±0.2 4.6±0.2 4.5±0.2 4.7±0.2 4.8±0.3 4.7±0.3 NS HR c·min·1 81.5±3.7 81.9±3.7 82.9±3.4 84.6±3.3 87.6±3.3 90.2±3.7 0.025 Psa mmHg 95.3±3.3 92±3.6 92.2±3.5 87.4±3.6 83.4±3.5 81.2±3.5 0.05 !J>a mmHg 20.1±1.9 18.3±1.9 16.9±1.5 17.8±1.6 18±1.7 17.9±1.5 0.05 !Yaw mmHg 6.2±1.6 5.9±1.5 5.6±1.5 6.0±1.3 6.2±1.6 5.8±1.6 NS Pra mmHg 4.2±1.2 4±1.8 3.7±1.5 4.3±1.9 3.9±1.6 4.4±2.1 NS Cl /·min·1·m·2 3.39±0.86 3.59±1.04 3.77±1.07 3.88±0.98 4.19±1.04 4.21±1.16 0.025 SI 41.9±2.7 44.0±3.3 45.3±3.5 45.4±3 47.8±3.2 46.8±3.6 NS PVR D·cm·5·s·1 189±7.6 159±6.9 137±5.7 138±5.9 126±5.6 132±5.4 0.001 SVR D·cm·5·s·1 1297±17 1183±17 1128±16 1003±13 904±13 891±15 0.001 t\PVR/.ASVR 1.81 2.11 1.19 1.09 0.96 J;>(A-.a)o2 kPa 4.7±0.2 5.9±0.3 6.0±0.2 6.0±0.3 6.2±0.2 6.3±0.3 0.001 Qs/Qt% 34.4±2.1 42.9±2 43±2.6 43.5±2.4 44.8±2.7 44.5±2.5 0.01

Values are mean±si!M; TO: baseline data; 0.05, 0.1, 0.15, 0.2 and 0.25: dose-level as j.lmol·kg·1·min·1; Pao2

: arterial oxygen te_!!sion; Paco2: arterial carbon dioxide ~nsion; PVo

2: mixed venous 0

2 pressure; 'P'sa:__!Jiean syst~ic ~ery

pressure; Ppa: mean pulmonary artery pressure; Ppaw: mean pulmonary artery wedge pressure; Pra: mean nght atnum pressure; Cl: cardiac index; SI: stroke index; PVR: pulmonary vascular resistance; SVR: systemic vascular resis~an~e; t\PVR!.ASVR: PVR over SVR relative variation ratio; P(A-a)o

2: alveolar-arterial oxygen pressure gradient; Qs/Qt:

venous admixture; NS: not significant; ANOV A: analysis of variance.

A%

+30

+20

+10

100%

-10

-20

0

< 20 m in >

1

~mol·kg-1

2 3 4 5

P(A-a)oll (pc0.001)

Qs/Cu (pc0.001) Cl (pc0.05)

Fig. 1. -Pulmonary data variations during stepwise incremental ATP infusion. Variations are expressed as percent of baseline level (mean±sEM). For definitions of significant values; see table 2. p: ANOVA significance level; when significant, Bonferroni corrected paired t-test over TO level {*: p<0.005, u: p<O.OOl).

fourth dose-le~el (0.20 Jlmol·kg·1 ·min· 1); three variables: HR, Psa, and Paco

2 reached significant

variations only with the fifth dose-level (0.25 )..l.mol·kg·1·min-1

) (fig. 2). During this study, other haemodynamic and gasometric variables showed no changes.

The APVR/.::lSVR variation ratio reached its highest value: 211% (fig. 3), with the second dose-level (0.1 Jlmol·kg·1 ·min·1) indicating an initial predominant pulmonary vascular effect. This ratio then decreased

regularly towards its lowest value: 90%, reached with the highest dose-level (0.25 Jlmol·kg·1-min·1),

indicating a subsequent predominant systemic vascular effect.

Occasional side-effects occurred during high dose-level A TP infusion in 7 out of 10 patients: nausea, flush or hyperpnoea; all remained moderate and transient and it was never necessary to stop the procedure. However, such side-effects reoccurred periodically at 1-2 min intervals until the end of the infusion.

PULMONARY VS SYSTEMIC EFFECTS OF ATP IN COPD 453

A%

+20

+10

100%

-10

-20

0

0 20 mln O 3 4

Cl tpc0.05)

* . H.R. (pc0.05)

Pacoa (pc0.001)

p-* (pc0.05)

I SVR * (pc0.001)

5

Fig. 2.- Systemic data variations during stepwise incremental ATP infusion. Variations are expressed as percent of baseline level (mean±sBM). For definitions of significant values, see table 2. p: ANOVA signifcance level; when significant, paired !-test over TO level (*: p<O.OOS, ••: p<O.OOI) .

fl. PV R K

fl.SVR%

•

....... SYSTEMIC VD side-effects •••

rso~m~2~~~0Q·~----·~ ..... ,................... -@ Ci> @ @ @ @ ATP

j.!JnOI·kg·1

Fig. 3.- Variations ratio of pulmonary (PVR) and systemic (SVR) vascular resistance during stepwise incremental ATP infusion. APVR/ASVR: variations ratio of PVR (%)on SVR (%); VD: vasodilation.

Discussion

In patients with stable COPD during incremental A TP infusion, our results show that; 1) low dose-levels induced predominant pulmonary vasodilation, with pPa and PVR decrease and an increase of LlPVR/~SVR; 2) the highest dose-level induced predominant systemic vaso­dilation, with Psa and S VR decrease, a drop of LlPVR/.1SVR ratio to its lowest level, and intermittent side-effects.

Many compounds have previously been described as pulmonary vasodilators but have shown, in fact, only unspecific vasodilating effects, acting simultaneously on pulmonary and systemic vessels with predominant systemic effects [10] . Some of these drugs, like

hydralazine, may induce dramatic deleterious effects such as sustained systemic hypotension or a worsening of hypoxaemia making them difficult to use in the management of pulmonary hypertension in COPD [11]. The theoretically ideal pulmonary vasodilating compound would induce predominant pulmonary vasodilation [12]. Thus far, this pulmonary predominant activity has been described only during short term infusion of urapidil, a post-synaptic alpha

1-blocker [13], or prostaglandin E

1 [14],

but has not been recorded during low rate infusion of prostacyclin [15], isoproterenol [16) and acetylcholine [17] in man.

In the present study, we describe a natural short-lived compound leading to a predominant pulmonary vasodilating effect during low dose-level infusion

454 S.J.M. OABA, C. P~FAUT

(maximal during ATP infusion using 0 . 1 J.l.mOl·kg·1·min·1). The effect of ATP on pulmonary vessels was not dose-dependent and remained unchanged at higher dose-levels. However, at these higher levels, there was a shift in the predominance of vasodilation from the' pulmonary system to the systemic circulation. All these data suggest a selective pulmonary vasodilating effect only during low dose ATP infusion. This predomi­nant pulmonary effect disappears at a certain dose level and may be explained by some saturable metabolic pathway in pulmonary vessels, which acts like a metabolic filter on circulating infused ATP. Such a saturable pathway could be the endothelium specific uptake of A TP (6], endothelium endocytosis using caveolae [18]. or the ATP dephosphorylation induced by endothelium ectonucleotidases [19]. Although in vivo pulmonary vascular uptake of A TP was described as early as 1950 in dogs [20], in vivo studies remain incapable of assessing the respective roles of the above-mentioned mechanisms in the function of the pulmonary endothelium as a metabolic filter of A TP.

In our study systemic hypotension was associated with a corresponding increase of Cl and HR without any change of SI. This fact, previously observed in conscious dogs with intact baroreflex [21], is suggestive of a reactive adrenergic state which probably interferes with the vasodilating effect of ATP [22]. On the contrary, in ATP-induced hypotension during anaesthesia in animals and in man no change or a slight decrease of heart rate has usually been recorded [23]. This result could be accounted for by decreased baroreflex control during anaesthesia. It has, however, been shown that baroreflex control is still effective during this state [24]. It must be taken into consideration that during anaesthesia A TP dose­levels were nearly fourfold higher than in our study and could have raised bradycardia by direct cardiac effect [25]. Further studies are necessary to better understand these mechanisms; however, very high dose-levels of A TP remain impossible to use in conscious subjects as side effects occur.

Although there are only a few reports about the hypotensive effect of ATP in man [23], systemic vasodilating and hypotensive effects of adenosine have been evaluated during human surgery [26]. Adenosine vascular effects, however, cannot be extrapolated to ATP because these compounds act on different receptors {P1 vs P2) and they use distinct mechanisms [25] for a relax­ing effect, with A TP being the only one that induces an endothelium-dependent relaxation.

During A TP-induced pulmonary vasodilation, and in the same manner, Pao2 decreased and W?S a;-;sociated with significant worsening of P(A-a)o

2 and Qs/Qt. During the

two lowest infusion levels these results, associated wiLh the lack of variations in other data such as Cl, Ppaw, Psa and Paco2, were highly suggestive of a worsening of the ventilation-perfusion relationship, as previously described during other pulmonary vasodilating drug trials [27). During ATP infusion, as inhibition of hypoxic pulmo­nary vasoconstriction has previously been recorded in dogs [2~) and could at least partly explain a wors~ning of the V A/Q relationship by increasing the low V A/Q

lung units and therefore inducing a decrease of Pao1 in the patients. Results suggest that drug-induced pulmonary vasodilation on chronically hypoxaemic patients with stable COPD could be safe if the pulmo­nary vasodilation (i.e. pulmonary driving pressure drop) is associated with cardiac output increase. This phenome­non would limit the hypoxaemia worsening and could also increase the oxygen-delivery to peripheral tissues.

The significant decrease of Paco1 during systemic vasodilation induced by the A TP infusion was an interesting observation. During pulmonary vasodilation trials in COPD patients with nifedipine or nitrendipine, Paco

2 did not vary [27, 29]; on the contrary acute drug­

induced pulmonary vasodilation with hydralazine [30] and isoproterenol [16] induced similar decreases of Paco1 in these patients. Hydralazine infusion induced a ventilatory drive increase that could explain, at least partly, the decrease of Paco2 [31]. During our study, many patients complained of dyspnoea with periodic difficulty of breathing and a peculiar recurring sensation of uncon­trollable hyperventilation. Similar clinical side-effects were previously noticed during A TP infusion in conscious man [32] but these data have never been analysed. More specific studies are necessary to confirm the hypothetical ventilatory stimulant effect of ATP.

In this study of chronically hypoxaemic patients, A TP infusion induced dual haemodynamic effects: 1) pulmonary vasodilation was not dose-dependent but predominant during low dose-levels, with slightly wors­ening hypoxaemia, possibly due to partial inhibition of hypoxic pulmonary vasoconstriction; 2) on the contrary, systemic vasodilation was dose-dependent and became predominant only at the highest dose levels. Such a dual haemodynamic effect is an indirect argument in favour of a saturable metabolic pathway of pulmonary circulation for ATP. For these highest dose levels clinical side-effects were noticed which suggest a vemi­latory stimulant effect of ATP.

Haemodynamic and ventilatory effects of A TP and other purinergic compounds in health and disease call for further studies. Physiological or therapeutic implications remain, so far, premature.

Acknowledgements: The writers thank R. Joly for her technical and S . Bemabeu for his photographic assistance.

References

1. Su C. - Extracellular functions of nucleotides in heart and blood vessels. Ann Rev Physiol, 1985, 47, 665-676. 2. Somlyo AP, Somlyo AV. - Vascular smooth muscle ll. Pharmacology of normal and hypertensive vessels. Pharmacol Rev, 1970, 22, 249-253. 3. Fukunaga AF, Sodeyama 0, Matsuzaki Y, Ikeda K, Sato K. - Hemodynamic and metabolic changes of ATP-induced hypotension during surgery (abstract). Anesthesiology, 1983, 59, A12. 4. Numajiri Y, Mokuji T, Kagami K.- ATP induced hypo­tensive anesthesia during prosthetic hip surgery. J Clin A11esthesiol, 1979, 3, 279-284.

PULMONARY VS SYSTEMIC EFFECTS OF ATP IN COPD 455

5. Caba SJM, Bourgouin-Karaouni D, Dujols P, Michel FB, Prefaut Ch. - Effects of Adenosine Triphosphate on pulmo­nary circulation in chronic obstructive pulmonary disease. Am Rev Respir Dis, 1986, 134, 1140-1144. 6. Dieterle Y, Ody C, Ehrensberger A, Stalder H, Junod AF. - Metabolism and uptake of adenosine triphosphate and adeno­sine by porcine aortic and pulmonary endothelial cells and fibroblasts in culture. Circ Res, 1978, 42, 869-876. 7. American Thoracic Society. - Chronic bronchitis, asthma and pulmonary emphysema. Statement by the Committee on Diagnosis Standards for Non-tuberculosis Respiratory Diseases. Am Rev Respir Dis, 1962, 85, 762-768. 8. Kelman CR. - Digital computer subroutine for the conversion of oxygen tension into saturation. J Appl Physiol, 1966, 21, 375-376. 9. Wallenstein S, Zucker CL, Fleiss JL. - Some statistical methods useful in circulation research. Circ Res, 1980, 47, 1-9. 10. Rubin LJ. - Vasodilators and pulmonary hypertension: where do we go from here? Am Rev Respir Dis, 1987, 135, 287. 11. Lupi-Herrera E, Seoane M, Verdejo J. - Hemodynamic effect of hydralazine in advanced, stable chronic obstructive pulmonary disease with cor pulmonale. Chest, 1984, 85, 156-163. 12. Hyman AL. KadowiiZ PJ.- Vasodilator therapy for pul­monary hypertensive disorders. Chest, 1984, 85, 145-147. 13. Adnot S, Defouilloy C, Brun-Buisson C, Abrouk F. Piquet J, Lemaire F. - Hemodynarnic effects of Urapidil in patients with pulmonary hypertension. Am Rev Respir Dis, 1987, 135, 288-293. 14. Naeijet R, Melot C, Mols P, Hallemans R.- Reduction in pulmonary hypertension by Prostaglandin E

1 in decompensated

chronic obstructive pulmonary disease. Am Rev Respir Dis, 1982, 125, 1-5. 15. Szczeklik J, Szczeklik A, Nizankowski R. -Haemodynamic changes induced by prostacyclin in man. Br Heart J, 1980, 44, 254-258. 16. Lockhart A, Lissac J, Salmon D, Zappacosta C, Benismail M. - Effects of isoproterenol on the pulmonary circulation in obstructive airways disease. Clin Sci, 1967, 32, 177-187. 17. Charms BL. - Primary pulmonary hypertension. Effect of unilateral pulmonary artery occlusion and infusion of acetylcholine. Am J Cardiol, 1961, 8, 94-99. 18. Smith U, Ryan JW. - An electron microscopic study of the vascular endothelium as a site for bradykinin and adeno­sine-5'-triphosphate inactivation in rat lung. Adv E:xp Med Bioi, 1970, 8, 249-261. 19. Cordon JL. - Extracellular ATP: effects, sources and fate. Biochem J, 1986, 233, 309-319. 20. Binet L, Bumestein M. - Poumon et action vasculaire de !'adenosine triphosphate (A.T.P.). Presse Med, 1950, 68, 1201-1203. 21. Pagani M, Vatner SF. Braunwald E. - Hemodynarnic effects of intravenous sodium nitroprusside in the conscious dog. Circulation, 1978, 57, 144-151. 22. Vcrhacghe RH, Lorcnz RR, McGrath MA, Shepherd IT, Vanhoulte PM. - Metabolic modulation of neurotransmitter release - adenosine, adenine, nucleotides, potassium, hypcros­molarity, and hydrogen ion. Fed Proc, 1978, 37, 208-211. 23. Kien ND, White DA, Reitan JA, Eisele JH. -Cardiovascular function during controlled hypertension induced by Adenosine Triphosphate or sodium nitroprusside in the anesthetized dog. Anesth Analg, 1987, 66, 103-110. 24. Fukunaga AF, Flacke WE, Bloor BC. - Hypotensive effects of Adenosine and Adenosine triphosphate compared with sodium nitroprusside. Aneslh Analg, 1982, 61, 273-278.

25. Bumstock G, Kennedy C. - Purinergic receptors in the cardiovascular system. Progress in Pharmacology, 1986, 6 (2), 112-132. 26. Sollevi A, Lagerkranser M, !rested TL, Gordon E, Lindquist C. - Controlled hypotension with Adenosine in cere­bral aneurysm surgery. Anesthesiology, 1984, 61, 400-405. 27. Melot C, Hallemans R, Naeije R, Mols P, Lejeune P. -Deleterious effect of nifedipine on pulmonary gas exchange in chronic obstructive pulmonary disease. Am Rev Respir Dis, 1984, 130, 612-616. 28. Benumof JL, Fukunaga AF, Trousdale FR. - A TP inhib­its hypoxic pulmonary vasoconstriction. Anesthesiology, 1982, 57, A474. 29. Rubin LJ, Moser K.- Long-term effects of nitrendipine on hemodynamics and oxygen transport in patients with cor pulmonale. Chest, 1986, 89, 141-145. 30. Keller CA, Shepard JW, Chun SD, Dolan GF, Vasquez P, Vu-Dinh Minh. - Effect of hydralazine on hemodynamics, ventilation, and gas exchange in patients with chronic obstruc­tive pulmonary disease and pulmonary hypertension. Am Rev Respir Dis, 1984, 130, 606-611. 31. Corriveau ML, Shepard JW, Dolan OF. -Effects of hy­dralasine on mouth occlusion pressure and ventilatory re­sponse to hypercapnia in patients with chronic obstructive pulmonary disease and pulmonary hypertension. Am Rev Respir Dis, 1987, 135, 118-122. 32. Davies DF, Cropper AL. Schroeder HA. - Circulatory and respiratory effects of Adenosine Triphosphate in man. Circulation, 1951, ID. 543-550.

Ejfets pulmanaires et systemiques compares du triphosphate d' adenosine dans les bronchopneumopathies chroniques obstructives. L'ATP: un vasoregulateur a controle pulmanaire? S. Gaba, C. Prefaut. RESUME: Nous avons compare les effets vasculaires pulmonaires et systemiques de 1' A TP administre en perfusion par voie intra-veineuse, ~ dose croissante, chez 10 patients atteints de BPCO stable. La vasodilation pulmonaire obtenue etait 1) preferentielle et maximale des les plus faibles doses (0.1 ~M·kg· 1·min·1) comme en temoigncnt la diminution de la P'ap (-16%; p<0.01). des RVP (-28%; p<0.005) et !'augmenta­tion simultanee du rapport des variations des resistances pulmonaires et systemique 6RVP/6RVS. 2) associee a une aggravation de l'hypoxemie (-14%; p<;0.001) mais aussi a une augmentation de P(A-a)o, et de Qs/Qt, suggerant une inhibition de la vasoconstriction pulmonaire hypoxique. La vasodilation systemique etait par contre: 1) dependante de la dose, ne devenant significative qu'a partir de la dose de 0.2 ~-tM·kg·1 ·min·1 pour la Pas (-12.5%; p<0.05) et les RVS (-30%; p<0.01). 2) associee ~des effets secondaires survenant de f~on cyclique; 3) associce a une diminution de Paco, (-12.5%; p<0.005) et a un besoin incoercible d'hyperventiler, suggerant un effet stimulant ventilatoire de I' ATP chez l'homme. Chez des patients atteints de BPCO stable, la pcrfusion d' A TP possede une dualite d'action seton la dose administree: un effet vasodilatateur pulmonaire prCferentiel est induit par les faibles doses sans etre dose-dependant contraircment a l'effet vasodilatateur systemique. Cette dissociation entre effet vasodilatateur pulmonaire et systemique en fonction de la dose administree, pourrait traduire in vivo le role mctabolique de l'endothelium pulmonaire chez l'homme vis a vis de l'ATP. Eur Respir J., 1990, 3, 450-455.