The endotracheal route has been suggested by the European Committee for Resuscitation (1) and the American Heart Association (2) as an alternative method for the administration of adrenaline, the most important medication used in advanced cardiac life support, if IV access cannot be attained (1–4). The European Heart Association presently recommends that when the endotracheal route is used, the adrenaline dosage should be two to three times larger than for the IV route (i.e., 0.02–0.03 mg/kg). However, the optimal dose of endotracheal adrenaline for augmenting aortic diastolic blood pressure (BP) is unknown. Although most of the previous investigators who studied endotracheal adrenaline administration in animal models have reported favorable results, the initial decrease of BP associated with endotracheal adrenaline was generally ignored (5–14). In a previous study, we speculated that it was the β-adrenergic–mediated effect of adrenaline that accounts for this deleterious decrease in BP (15). Gonzalez et al. (16) showed that the hemodynamic effects of adrenaline are influenced by the dose and the rate of administration. When given in larger doses (infusion rates of 0.2–1 μg · kg−1 · min−1), adrenaline was a vigorous vasoconstrictor that caused an increase in diastolic BP and improved coronary blood flow. Moreover, the β-adrenergic receptors responded to smaller doses of adrenaline than the α-adrenergic receptors (16,17). Thus, in contrast to the importance of α-adrenergic stimulation during resuscitation, β2 stimulation may be counterproductive by causing peripheral vasodilatation, decreasing aortic pressure and reducing myocardial perfusion pressure (16,17). This may have been the explanation for clinical studies that suggested that the standard IV dose administered endotracheally was less than optimal, often ineffective, and even possibly deleterious (8,18). In this study, we approached the problem by comparing the hemodynamic effects of endotracheal adrenaline using three different regimens in five dogs: endotracheal adrenaline with pretreatment with propranolol, endotracheal adrenaline without this nonselective β-blocker, and endotracheal administration of 10 mL of normal saline (sham). A prospective, randomized, cross-over trial was designed to determine the pharmacodynamic variables of heart rate, BP, and oxygenation after endotracheal adrenaline administration (0.05 mg/kg).
The study was approved by the Animal Care and Users Committee of the Tel Aviv University, and the animals were cared for in accordance with national and institutional guidelines.
Five adult mongrel dogs of both sexes, weighing 10.5–21.0 kg, were anesthetized with IV sodium pentobarbital (25 mg/kg) and intubated orally with a low-pressure cuffed endotracheal tube. The animals were ventilated (fraction of inspired oxygen = 0.21) at a respiratory rate of 20 to 24 breaths/min to maintain a PAco2 of 25–45 mm Hg and a PAo2 of >90 mm Hg. Both femoral arteries were percutaneously cannulated for the measurement of arterial BP, and for blood gas sampling. Heart rate (standard lead 2) and arterial BP were continuously monitored with a polygraph recorder.
To achieve steady state conditions, 20 min was allowed for stabilization after the induction, endotracheal intubation, and catheter insertions. Baseline blood samples for arterial blood gases were drawn after which adrenaline hydrochloride (1:1000), 0.05 mg/kg (Teva Pharmaceuticals Industries, Petach Tikva, Israel) diluted with 10 mL of normal saline was administered into the endotracheal tube of each dog. After adrenaline administration, five forced manual ventilations were delivered with an Ambu bag (Laerdal, Stavanger, Norway). Serum samples for arterial blood gases were then collected before and at 0.25, 0.5, 0.75, 1, 2, 3, 4, 5, 10, 15, 20, 25, 30, and 60 min after the drug administration. The following protocol was repeated three times: 1) administration of 10 mL of normal saline endotracheally (sham), 2) pretreatment by IV nonselective β-blocker (propranolol 0.115 mg/kg) 10 min before the endotracheal adrenaline and after hemodynamic stabilization (note: four preliminary dose adjustment experiments had been performed earlier to test which dose of propranolol would produce a 10% decrease in heart rate), and 3) administration of endotracheal adrenaline without pretreatment of a β-blocker. The order in which each dog underwent each regimen was randomized, and there was at least a one-week interval between trials.
Data, expressed as mean ± sem, were analyzed by using BMDP Statistical Software (SPSS, Chicago, IL). The statistical analysis included analysis of variance with repeated measures. P < 0.05 was considered significant.
The hemodynamic changes after endotracheal adrenaline administration with and without pretreatment with propranolol and those that occurred after saline administration are illustrated in Figures 1–3. The endotracheal administration of adrenaline produced a significant decrease in systolic, diastolic, and mean arterial BP in all the dogs compared with the endotracheal saline administration (P < 0.03). The maximum decrease of mean BP occurred at the first 2 min after the administration and lasted for 10 min.
The changes in heart rate that were recorded after the administration of endotracheal adrenaline with and without propranolol pretreatment are shown in Figure 3. A significant tachycardia developed immediately after the endotracheal adrenaline administration; it reached a maximal level at 30 s after the injection and lasted until the end of the trial (P < 0.05). The endotracheal adrenaline given after propranolol pretreatment caused no significant changes in heart rate, and there were no significant differences in the arterial blood gases between the study groups (Table 1).
In this study, we investigated the phenomenon of an initial decrease in arterial BP associated with endotracheal adrenaline, an event that had been observed by others but which had not been subjected to in-depth study.
The endotracheal administration of adrenaline 0.05 mg/kg in a dog model produced an early significant decrease in systolic, diastolic, and mean arterial BP that paralleled a long-lasting significant increase in heart rate. These effects were abolished when the dogs were pretreated with a nonspecific β-blocker (propranolol). Although most of the previous investigators who studied endotracheal adrenaline administration in animal models reported favorable results, this initial decrease of arterial BP associated with endotracheal adrenaline was generally ignored (5–7). Naganobu et al. (13) compared the effects of distilled water and normal saline as diluents for the endobronchial administration of adrenaline in anesthetized dogs and found that the initial hemodynamic response to endotracheal adrenaline diluted with saline was a decrease of arterial BP. In our dogs to which saline-diluted adrenaline was delivered endobronchially, the mean arterial BP decreased after the adrenaline administration. Roberts et al. (10) compared the effect of endotracheal versus IV adrenaline administration and found that both routes produced favorable effects on heart rate and BP, but they failed to comment on the initial decrease in BP after the endotracheal adrenaline administration. Mazkereth et al. (11) studied plasma adrenaline concentrations after peripheral bronchial versus endotracheal adrenaline delivery in a nonarrest canine model: the mean arterial BP decreased after the adrenaline administration in both groups. This was considered the result of using a small dose of adrenaline. Similarly, in another animal study comparing different volumes of dilution, we showed that endotracheal adrenaline was associated with a decrease in arterial BP (9). In a normotensive and hypotensive shock model, Orlowski et al. (12) demonstrated that endotracheal adrenaline at doses of 0.01–0.04 mg/kg did not produce a consistent increase in BP. Again, the authors disregarded the early decrease in both systolic and diastolic arterial pressures that lasted for 20 minutes. Similar studies in swine showed that endotracheal adrenaline (0.01 mg/kg) produced no significant increase in plasma adrenaline or arterial BP compared with controls (19).
Several clinical studies have suggested that the standard adrenaline dose administered endotracheally was less than optimal and often ineffective. In comparing endotracheal and peripheral IV adrenaline administered to five patients in asystolic cardiac arrest, Quinton et al. (18) found that 10 mL of 1:10000 adrenaline, which is approximately 0.01 mg/kg, was ineffective. McCrirrick and Monk (20) examined the hemodynamic effects of endotracheal adrenaline over a range of doses to a maximum of 10 μg/kg and suggested that these doses are likely to be ineffective. McCrirrick and Kestin (8) compared systolic BP responses to IV and endotracheal (0.5 μg/kg diluted to 5 mL) adrenaline administration in 12 patients undergoing elective hip replacement. They reported that the endotracheal administration of adrenaline failed to produce a clinically important increase in systolic arterial BP in any patient, and that arterial BP decreased by 10 mm Hg or more after endotracheal adrenaline administration in 4 patients. Similarly, poor absorption of adrenaline was noted in adults with asystolic arrest receiving 2.5 mg of 1:10,000 adrenaline by the endotracheal tube (21). In a recent study, Raymondos et al. (14) investigated the effect of endotracheal adrenaline in 34 patients receiving implantable cardioverter defibrillators under general anesthesia and demonstrated an increase in mean arterial BP. However, the authors did not comment on certain features that were shown in the figures, one of them being the decrease in mean arterial BPs in the initial phase after the adrenaline administration. Furthermore, some of the described patients (nonspecified) were on β-blockers that may have abolished the β-mediated decrease of arterial BP.
Why did β-blockers abolish the decrease in arterial BP? Adrenaline is the final product of catecholamine synthesis stimulating both α- and β-adrenergic receptors in a dose-dependent fashion. With larger doses (0.2–1 μg · kg−1 · min−1), it is a vigorous vasoconstrictor (17). This is associated with arteriolar vasoconstriction, causing increased systemic vascular resistance and, thus, increased aortic pressure leading to improved coronary perfusion pressure and myocardial blood flow (21). The increase in aortic diastolic pressure associated with adrenaline during cardiopulmonary resuscitation (CPR) is critical for maintaining coronary perfusion pressure and myocardial blood flow and is the key to its effectiveness in successful resuscitation. With small doses (0.05–0.1 μg · kg−1 · min−1), adrenaline predominantly affects β1 and β2 receptors, resulting in positive inotropic and chronotropic effects (17). However, in contrast to the importance of α stimulation during resuscitation, β2 stimulation may be counterproductive by causing peripheral vasodilatation, lowering aortic pressure, and reducing myocardial perfusion pressure. Furthermore, in humans undergoing CPR, cardiac output decreases to <30% of normal values (22). This may limit blood flow and the transport of drugs from the alveoli to the central circulation, thereby further decreasing adrenaline blood concentration to the β-adrenergic dose range, and this may account for the failure of the endotracheal administration of adrenaline during CPR, a feature that was noted in recent clinical reports. These effects were abolished in the current study when the dogs were pretreated with a nonspecific β-blocker (propranolol).
Several limitations of the present study must be considered. The physiological conditions during cardiac arrest differ from those in our earlier study (22). In addition, there is an extremely large individual variation in response to catecholamines administered during CPR (23). Thus, our results may not necessarily apply to other hemodynamic scenarios, such as cardiac arrest or ventricular fibrillation models. However, as suggested by previous studies, there are also indications for endobronchial drug administration with (still) stable circulation (9,13,24). Furthermore, it is likely that for any given dose, tracheal absorption of adrenaline will be less effective during cardiac arrest than it was in our study, thus causing predominantly β-mediated effects (8). A second limitation of the study is that dogs were anesthetized with pentobarbital. However, after the induction, an interval of 20 minutes was allowed for stabilization after endotracheal intubation and catheter insertion to achieve steady state conditions. Another limitation is that the concentration of adrenaline in the blood was not measured, and we could not eliminate an interference of a β-blockade in decreasing the blood concentrations of adrenaline.
Despite these limitations, the results of our study, taken together with previous animal and human data, point to a potential hazard of small doses of endotracheal adrenaline. Thus, we suggest that 0.05 mg/kg adrenaline is associated with predominantly β-mediated effects, causing peripheral vasodilatation unopposed by α receptor-mediated vasoconstriction. We propose that the standard dose of endotracheal adrenaline that is recommended by the European Heart Association and the UK Resuscitation Council may be too small, and that this approach can exert deleterious effects. In addition, while some studies in animals and humans suggest larger IV doses of adrenaline than currently recommended may improve resuscitation rates (25), none has addressed the endotracheal equivalent of large-dose IV adrenaline. It is likely that larger doses of endotracheal adrenaline are needed to produce a beneficial effect, a possibility that should be further investigated. We suggest that the incorporation of larger endotracheal doses as first-line therapy in combination with the measurement of adrenaline plasma concentrations warrants additional studies.
We thank Esther Eshkol for editorial assistance, and P. Lilos for the statistical support and data analyses.
1. European Resuscitation Council and Resuscitation Council (UK). Guidelines for advanced cardiac life support. Resuscitation 2000; 46: 155–62.
2. Guidelines for cardiopulmonary resuscitation and emergency cardiovascular care: international consensus on science agents to optimize cardiac output and blood pressure. Circulation 2000; 102 (suppl I): 135.
3. Goldberg A. Cardiopulmonary arrest. N Engl J Med 1974; 290: 381–5.
4. Redding JS, Pearson JW. Evaluation of drugs for cardiac resuscitation. Anesthesiology 1963; 24: 203–7.
5. Redding JS, Asuncion FS, Pearson JW. Effective routes of drug administration during cardiac arrest. Anesth Analg 1967; 46: 253–8.
6. Greenberg MI, Roberts RJ, Krusz JC, Baskin SI. Endotracheal adrenaline in a canine anaphylactic shock model. J Am Coll Emerg Phys 1979; 8: 500–3.
7. Ralston SH, Tacker WA, Showen L. Endotracheal versus intravenous adrenaline during electromechanical dissociation with CPR in dogs. Ann Emerg Med 1985; 14: 1044–8.
8. McCrirrck A, Kestin I. Haemodynamic effects of tracheal compared with intravenous adrenaline. Lancet 1992; 340: 868–70.
9. Paret G, Vaknin Z, Ezra D, et al. Adrenaline pharmacokinetics and pharmacodynamics following endotracheal administration in dogs: the role of volume of diluent. Resuscitation 1997; 35: 77–82.
10. Roberts JR, Greenberg MI, Knaub MA, Baskin SI. Comparison of the pharmacological effects of adrenaline administered by the intravenous and endotracheal routes. J Am Coll Emerg Phys 1978; 7: 260–4.
11. Mazkereth R, Paret G, Ezra D, et al. Adrenaline blood concentrations after peripheral bronchial versus endotracheal administration of adrenaline in dogs. Crit Care Med 1992; 20: 1582–7.
12. Orlowski JP, Gallagher JM, Porembka DT. Endotracheal adrenaline is unreliable. Resuscitation 1990; 19: 103–13.
13. Naganobu K, Hasabe Y, Uchiyama Y, et al. A comparison of distilled water and normal saline as diluents for endobronchial administration of adrenaline in the dog. Anesth Analg 2000; 91: 317–21.
14. Raymondos K, Panning B, Leuwer M, et al. Absorption and hemodynamic effects of airway administration of adrenaline in patients with severe cardiac disease. Ann Intern Med 2000; 132: 800–3.
15. Paret G, Vaknin Z, Ezra D, et al. Endotracheal adrenaline: should the recommended dose be reconsidered? J Prehosp Care 1998; 2: 1–5.
16. Gonzalez ER, Ornato JP, Garnett AR, et al. Dose-dependent vasopressor response to adrenaline during CPR in human beings. Ann Emerg Med 1989; 18: 920–6.
17. Keeley SR, Bohn DJ. The use of inotropic afterload-reducing agents in neonates. Clin Perinatol 1988; 15: 467–78.
18. Quinton DN, O’Byrne G, Aitkenhead AR. Comparison of endotracheal and peripheral intravenous adrenaline in cardiac arrest: is the endotracheal route reliable? Lancet 1987; i: 828–9.
19. Crespo SG, Schoffstall JM, Fuhs LR, Spivey WH. Comparison of two doses of endotracheal adrenaline in a cardiac arrest model. Ann Emerg Med 1991; 20: 230–4.
20. McCrirrick A, Monk CR. Comparison of i.v. and intratracheal administration of adrenaline. Br J Anaesth 1994; 72: 529–32.
21. Schuttler J, Hornchen U, Bremer F. Pharmacokinetic and comparative efficacy of adrenaline during out-of-hospital CPR. Anesthesiology 1991; 75: 287.
22. Del Guercio LRM, Feins NR, Cohn JD, et al. Comparison of blood flow during external and internal cardiac massage in man. Circulation 1965; 31 (suppl): 171–80.
23. Kern KB, Elchisak MA, Sanders AB, et al. Plasma catecholamines and resuscitation from prolonged cardiac arrest. Crit Care Med 1989; 17: 786–91.
24. Hahnel JH, Lindner KH, Schurmann C, et al. Endobronchial drug administration: does deep endobronchial delivery have advantages in comparison with simple injection through the endotracheal tube? Resuscitation 1990; 20: 193–202.
25. Goetting M, Paradis N. High-dose adrenaline in refractory pediatric cardiac arrest. Crit Care Med 1989; 17: 1258–62.