Anaphylactic shock is a rare but potentially lethal complication during anaesthesia even in previously apparently healthy individuals.1,2 It is characterised by systemic vasodilatation that results in circulatory collapse and low tissue perfusion.3,4 In addition, as demonstrated by our group, anaphylactic shock may decrease specifically cerebral perfusion, beyond that which would be expected from the level of arterial hypotension, thus resulting in brain ischaemia and hypoxia that could contribute to unexpectedly high morbidity and mortality even in individuals without major comorbidities and adequately resuscitated.5
Epinephrine (EPI) is recommended in most current guidelines as the first-line treatment of anaphylactic shock.6,7 Nevertheless, EPI may fail to restore adequate organ perfusion rapidly, or it can be completely ineffective (EPI resistance).8–10
Several recent clinical cases have reported that arginine vasopressin (AVP) administration has potential benefits in cases of anaphylactic shock refractory to EPI.11–13 Although these observations were reported following sequential administration of EPI and AVP, it has been suggested that AVP might represent a new first-line therapeutic option to treat anaphylactic shock.14,15 However, no controlled trials of treatment in humans are available and recommendations for the use of AVP are based on experts’ opinions. In addition, there are limited experimental data in models of anaphylactic shock regarding the effects of AVP on regional circulation of vital organs, especially the brain. Given the severe decrease of brain oxygenation in anaphylactic shock, beyond what would be expected from arterial hypotension per se, investigations comparing the effects of EPI vs. AVP on cerebral oxygenation during anaphylactic shock are necessary. We consider that brain oxygenation in anaphylactic shock could be of crucial importance and may help to explain the mechanism of death in many patients. Human autopsy studies do not show a specific signature (e.g. pulmonary oedema or myocardial ischaemia) of anaphylactic shock16; therefore, it is possible that brain ischaemia/hypoxia may be a mechanism that could explain death in the absence of other measurable consequences on vital organs such as the heart or the lungs. The aim of our study was to compare resuscitation with EPI vs. AVP on brain perfusion and oxygenation in a rat model of lethal anaphylactic shock.
Animals and sensitisation protocol
All animal procedures and care were performed in accordance with the European Communities Council Directive of 24 November 1986 (86/609/EEC). Our study was also conducted according to the official recommendations of the French Ministry of Agriculture (Paris, France) and the recommendations of the Helsinki Declaration. The protocol was approved (AL/105/112/13) by the Ethical Committee for Animal Experiments of Strasbourg University (16 January 2013, Chairperson Dr Fabielle Angel). This study was conducted in compliance with the ARRIVE statement.17
Ten-week-old brown Norway male rats weighing 250 to 300 g (Janvier, Le Genest-St-Isle, France) were used for these experiments. They were kept under standard conditions (temperature 21 ± 1°C; light from 6 a.m. to 6 p.m.) and given a standardised diet (A04; UAR, Villemoisson-sur-Orge, France) and water (Aqua-clear; Culligan, Northbrook, Illinois, USA) ad libitum. Rats were sensitised by subcutaneous administration of grade VI chicken egg albumin (Ovalbumin, 1 mg; Sigma-Aldrich, Saint-Quentin Fallavier, France) and aluminium hydroxide (OHA1, 4 mg; Merc Euro Lab, Briare Le Canal, France) diluted in 1 ml of 0.9% saline solution at days 0, 4 and 14, similar to previous studies.18–20
Surgical procedure and measurement of haemodynamic variables
The surgical procedure was performed on day 21 following the initial sensitisation. Anaesthesia was induced using 3% isoflurane. The trachea was intubated, and the lungs were mechanically ventilated with 100% oxygen using a Harvard Rodent respirator model 683 (Harvard Apparatus, Cambridge, Massachusetts, USA). Maintenance of anaesthesia and analgesia was ensured with 1% isoflurane, and a subcutaneous dose of 40 μg kg−1 of sufentanil. Rectal temperature was maintained at 37 ± 0.5°C by intermittent warming with a heating pad. A fluid-filled polyethylene catheter (internal diameter, 0.58 mm; outer diameter, 0.96 mm; Biotrol Diagnostic, Chennevières Les Louvres, France) was inserted into the right femoral vein for fluid maintenance (10 ml kg−1 h−1 of 0.9% saline solution) and administration of drugs. Mean arterial pressure (MAP) was continuously monitored with a fluid-filled polyethylene catheter (internal diameter, 0.58 mm; outer diameter, 0.96 mm; Biotrol Diagnostic, Chennevières Les Louvres, France) using a strain gauge pressure transducer (DA-100; Biopac Systems, Northborough, Massachusetts, USA) inserted into the right femoral artery. Carotid artery blood flow (CaBF) was measured using peri-vascular ultrasonic flow probes (RB) (Transonic System Inc, Ithaca, New York, USA) placed on the right carotid artery, as previously described.5
Following this preparation, animals were placed in the prone position and secured in a stereotaxic frame. A hole was drilled in the skull 4.6 mm anterior to and 2 mm lateral to lambda. A flexible Clark-type polarographic oxygen electrode (diameter 0.5 to 0.6 mm, length 1 mm) connected to a computer-supported Licox system (GMS, Mielkendorf, Germany) was then inserted into the right hippocampus to a depth of 3 mm for cerebral oxygen partial (PtiO2) pressure monitoring. Cerebral cortical blood flow (CBF) was monitored using a PeriFlux PF 5010 laser Doppler monitor (Perimed AB, Stockholm, Sweden) with a laser Doppler needle probe (reference 402, diameter of 0.45 mm, fibre separation of 0.14 mm) inserted through a second hole (2 mm rostral to the oxygen electrode) onto the surface of the right cerebral cortex.
After a 20-min stabilisation period, anaphylactic shock was induced in 20 rats by the intravenous injection of 1 mg ovalbumin (OVA) diluted in 1 ml of 0.9% saline – T0. Animals were randomly assigned (sealed envelopes) to four different groups and received the following: vehicle (1 ml 0.9% saline solution) only in the control group (CON); no-treatment in the OVA group; EPI (Adrénaline; Aguettant, Lyon, France): two boluses of EPI 2.5 μg were injected, one at 3 min (T3) and the other at 5 min (T5) after shock induction (T0), respectively; continuous infusion of EPI (10 μg kg−1 min−1) was initiated immediately after the last bolus injection (EPI); arginine vasopressin [(Arginine8)-Vasopressin Grade Solution APP; Sigma-Aldrich]: the bolus dose of AVP 0.03 IU was injected at T3 followed by continuous infusion of AVP (0.08 IU min−1) (AVP). The perfusion rates for the continuous infusions were adjusted to reach and maintain a MAP value of 75 mmHg in the EPI and AVP groups. When the MAP goal was reached and stable over a 10-min period, drug infusion was progressively decreased, as previously described.20
In all animals, MAP, CaBF, CBF and hippocampal PtiO2 values were continuously monitored. Values were recorded prior to shock induction (T0) and then at regular intervals after OVA: 1 min (T1), 2.5 min (T2.5), 5 min (T5), 7.5 min (T7.5) and 10 min (T10), then every 5 min until the end of the experiment (T60). Arterial blood gases were measured prior to shock induction (T0) and at the end of the experiment (T15 for the OVA group and T60 for the other groups) using an automated blood gas analyser (Radiometer ABL 725, Copenhagen, Denmark).
All animals were eventually euthanised by an overdose of thiopentone sodium.
Results are expressed as mean ± SD. All data were analysed with Statview 5.0 software (Deltasoft, Meylan, France). Haemodynamic variables were analysed by two-way repeated measures analysis of variance or by the non-parametric Kruskal–Wallis test if the normality test failed. If a time-group interaction was identified, the differences among groups were tested specifically for the time periods that were physio-pathologically relevant (e.g. shock induction, following onset of therapy). For each group, the linear regression equation (y = ax + b), where ‘a’ is the regression coefficient (i.e. the slope of the regression line) and ‘b’ is y-intercept, and r2 were determined: r2 is the squared correlation coefficient, that is, the proportion of the total variation of the dependent variable accounted for by the independent variable.
Biochemical variables were analysed by one-way analysis of variance. When a significant difference was detected by the analysis of variance, post hoc analysis was performed with the Bonferroni test for pair-wise comparisons, corrected for the number of comparisons. Significance was assumed when P < 0.05.
Twenty-seven OVA-sensitised brown Norway rats were studied and randomly allocated to the CON group (263.0 ± 4.5 g, n = 7 rats), OVA group (260.4 ± 5.4 g, n = 7 rats), EPI group (260.3 ± 3.0 g, n = 6 rats) and AVP group (264.0 ± 6.9 g, n = 7 rats). In the four groups, baseline values of all measured variables did not differ during the stabilisation period.
Time course profiles for MAP are presented in Fig. 1. OVA challenge resulted in a profound decrease of MAP in the OVA, EPI and AVP groups. In both EPI and AVP groups, a transient increase in MAP was observed resulting from the combined effect of 1 ml volume expansion and the vasopressor injection. Subsequently, MAP continued to decrease slowly reaching a minimal value of 37 ± 3.0 mmHg in the EPI group and 37 ± 2.6 mmHg in the AVP group. After this point, MAP values progressively increased in both groups. There were no significant differences between the EPI and the AVP groups.
The effects of EPI and AVP on the changes in CaBF values are presented in Fig. 2. In groups receiving treatment with EPI or AVP, CaBF decreased to 16 ± 10 and 11 ± 9% of baseline values, respectively, within 10 min. A progressive and significant increase in CaBF was observed after T15 reaching 44 ± 23.7 and 32 ± 19.3% of baseline values at T60 in the EPI and AVP groups, respectively. Although higher CaBF values were achieved in the EPI group, the difference was not statistically significant.
The effects of EPI and AVP on CBF values are presented in Fig. 3; CBF values decreased gradually until T15 following anaphylactic shock induction, then increased progressively to 47.7 ± 13.8 and 63.5 ± 28.5% of baseline values at T60 in the EPI and AVP groups, respectively. The difference was not statistically significant.
The effects of EPI and AVP on PtiO2 values are presented in Fig. 4. After an initial decrease, PtiO2 values were partially restored in both treatment groups, but with a significantly different time profile: PtiO2 values were significantly higher in EPI group at T15, T20 and T25 (P < 0.05). In the group treated with EPI, the PtiO2 decrease was limited, reaching a minimal value of 25.0 ± 2.2 mmHg at T20, and remained higher than the ischaemic threshold of 20 mmHg throughout the entire study period. In contrast, in the AVP group, after the induction of anaphylactic shock, a sharp decrease was observed, with PtiO2 values falling below the cerebral ischaemic threshold at T10. The lowest PtiO2 values (4.9 ± 2.0 mmHg) were observed at T20 (P < 0.05 vs. the EPI group and the CON group), and then they increased progressively, reaching 20.8 ± 2.0 mmHg at the end of the study period.
Comparing CaBF and MAP, the EPI and AVP regression coefficients were not significantly different (Fig. 5a). In EPI and AVP groups, the mean linear regression equations were, respectively, y = 0.81x − 14.2 (r2 = 0.24) and y = 1x − 27.8 (r2 = 0.58). The relationship between CBF and MAP is shown in Fig. 5b. In EPI group, the mean linear regression equation was y = 0.51x + 13.2 (r2 = 0.18) and in AVP group, the regression coefficient (0.96) was significantly higher (P < 0.05): y = 0.96x − 4.8, r2 = 0.63). The relationship between MAP and PtiO2 is shown in Fig. 5c. The linear regression equation in the AVP group (y = 0.22x + 7.31, r2 = 0.5) had significantly higher regression coefficient (0.22, P < 0.05) compared with EPI group (y = 0.1x + 20, r2 = 0.12). The higher r2 values are evidence of a greater dependence of CaBF, CBF and PtiO2 on MAP.
Blood gas results are presented in Table 1. In the absence of treatment, anaphylactic shock resulted not only in a profound decrease in both PaCO2 and PaO2 values but also in a dramatic increase in lactate levels. EPI and AVP administration limited this increase in lactate levels and prevented the decrease in PaCO2 and PaO2 values.
The main findings of our study were as follows: in the absence of treatment, rats in the OVA group died within 15 min after induction of anaphylactic shock, whereas administration of EPI or AVP increased MAP, CaBF, CBF and PtiO2, and extended survival time; EPI, as compared with AVP, resulted in less pressure-dependence for CaBF, CBF and PtiO2; despite similar macro-haemodynamic and cerebral haemodynamic profiles, treatment with EPI was associated with significantly higher hippocampal PtiO2 values compared with AVP.
Pharmacological elevation of blood pressure is a therapeutic goal in patients with severe anaphylactic shock in order to prevent cerebral ischaemia. Whereas maintenance of cerebral blood flow during arterial hypotension is essential for homeostasis and the prevention of irreversible brain ischaemia damage, dissociation between MAP values and local circulation, including the brain, may exist during anaphylactic shock before and after the onset of treatment.5,18 In our study, we focused on the effects of EPI and AVP on cerebral blood flow for comparable MAP and CaBF values.
A key reason for studying the effects of AVP on the cerebral perfusion was the suggestion that AVP might be a potential therapeutic drug in catecholamine-refractory anaphylactic shock.11–13 The rationale is that, in clinical settings, catecholamine α1-adrenergic receptors may become desensitised or down-regulated, especially in case of systemic acidosis,21–23 limiting the vasopressor effects of EPI. Because AVP acts through vascular V1 receptors, it could restore vascular tone even if α1-adrenergic receptors are down-regulated. Moreover, in previous work, we showed that AVP and EPI had comparable effects on mean arterial pressure.19 However, the effects of AVP on cerebral perfusion in anaphylactic shock were still unknown.
We monitored PtiO2 in the hippocampus because this brain region is known to be vulnerable to cerebral ischaemia and hypoxia. As expected, in our study, EPI and AVP partially restored systemic haemodynamic variables and global cerebral blood flow (both treatments resulted in comparable MAP, CaBF and CBF values). There were two major differences between AVP and EPI. First, for the AVP group, the slope of the MAP vs. PtiO2 regression line (see Fig. 5c) suggests more pressure dependence in the AVP group compared with the EPI group. These results are in favour of the use of EPI to improve brain oxygenation. The second is that AVP was associated with significantly lower hippocampal PtiO2 values compared with EPI.
The overall effects of EPI and AVP on cerebral perfusion and oxygenation in this model of anaphylactic shock are probably a consequence of their systemic and cerebral effects. EPI, by activating vascular β1-adrenergic and β2-adrenergic receptors,24 dilates the cerebral micro-vasculature and improves cerebral perfusion.25–27 These mechanisms could explain our finding that EPI improved CBF of the cortex and the PtiO2 of the hippocampus.
On the contrary, the precise mechanisms explaining the vascular effects of AVP remain unclear. The effects of vasopressin vary according to the individual's physiological state and the dose used.28–31 For instance, AVP has been shown to vasodilate large cerebral arteries and to vasoconstrict small cerebral arterioles. These effects, specifically observed in rats, could explain our results. During the first 10 min of resuscitation, treatment with AVP was associated with significantly lower PtiO2 values in the AVP compared with the EPI group despite similar CaBF and CBF values. PtiO2 values are determined by the product of CBF and cerebral arteriovenous oxygen tension difference (AVTO2).32 As CBF was comparable between groups, the difference was probably due to a difference in AVTO2, resulting from reduced oxygen diffusion in the AVP group. This supports the hypothesis of an AVP-induced vasoconstriction in small vessels in some areas of the brain combined with normal or hyper-perfusion in other areas and thus overall CBF is maintained. Such heterogeneity, which has been previously suggested as a mechanism of brain hypoxia, even with normal or high CBF after acute ischaemic stroke,33,34 may also be present in our model of anaphylactic shock.
Finally, because arterial PO2 values were not repeatedly monitored throughout the study, we cannot exclude a possible increased oxygen extraction in peripheral organs combined with the lower cardiac output which might have led to lower PaO2 values at some time points, giving rise to lower PtiO2 values.
Some limitations of this study should be taken into account: animals were ventilated using 100% oxygen and we observed significant changes in PaCO2 that might have played a role in the alterations in cerebral perfusion we observed; neither blood–brain barrier disruption nor cerebral metabolism was investigated and these might have played a role on the observed effects of EPI on CBF; isoflurane administration per se might have influenced cerebral perfusion, although its effects on vasopressor modulation of cerebral perfusion by EPI or vasopressin have been shown to be similar35; the mechanisms and mediators of the immune reaction during anaphylactic shock may be different among species and extrapolation to humans should be done with caution.
We have demonstrated in a rat model of anaphylactic shock that EPI is superior to AVP in preserving PtiO2 in the hippocampus, one of the most vulnerable areas to ischaemia in the early phase of anaphylactic shock.
Acknowledgements relating to this article
Assistance with the study: the authors are indebted to Chantal Montémont for technical assistance.
Financial support and sponsorship: this work was supported by the Groupe Choc, Contrat AVENIR INSERM, Faculté de Médecine, Université de Lorraine, Vandoeuvre-les-Nancy, France and Laboratoire EA 3072: «Mitochondries, stress oxydant et protection musculaire», Institut de Physiologie, Faculté de Médecine, Université de Strasbourg.
Conflicts of interest: none.
1. Dong SW, Mertes PM, Petitpain N, et al. GERAP. Hypersensitivity reactions during anesthesia. Results from the ninth French survey (2005-2007). Minerva Anestesiol
2. Lienhart A, Auroy Y, Pequignot F, et al. Survey of anesthesia-related mortality in France. Anesthesiology
3. Lee JK, Vadas P. Anaphylaxis: mechanisms and management. Clin Exp Allergy
4. Kemp SF, Lockey RF. Anaphylaxis: a review of causes and mechanisms. J Allergy Clin Immunol
5. Davidson J, Zheng F, Tajima K, et al. Anaphylactic shock decreases cerebral blood flow more than what would be expected from severe arterial hypotension. Shock
6. Kemp SF, Lockey RF, Simons FE. World Allergy Organization ad hoc Committee on Epinephrine in Anaphylaxis. Epinephrine: the drug of choice for anaphylaxis. A statement of the World Allergy Organization. Allergy
7. Sheikh A, Shehata YA, Brown SG, Simons FE. Adrenaline for the treatment of anaphylaxis: cochrane systematic review. Allergy
8. Chengot T, Goncalves J, Marzo K. Back from irreversibility: use of percutaneous cardiopulmonary bypass for treatment of shock from refractory anaphylaxis during coronary intervention. J Invasive Cardiol
9. Di Chiara L, Stazi GV, Ricci Z, et al. Role of vasopressin in the treatment of anaphylactic shock in a child undergoing surgery for congenital heart disease: a case report. J Med Case Rep
10. Schummer C, Wirsing M, Schummer W. The pivotal role of vasopressin in refractory anaphylactic shock. Anesth Analg
11. Hussain AM, Yousuf B, Khan MA, et al. Vasopressin for the management of catecholamine-resistant anaphylactic shock. Singapore Med J
12. Kill C, Wranze E, Wulf H. Successful treatment of severe anaphylactic shock with vasopressin. Two case reports. Int Arch Allergy Immunol
13. Meng L, Williams EL. Case report: treatment of rocuronium-induced anaphylactic shock with vasopressin. Can J Anaesth
14. Schummer W, Schummer C. Vasopression and suspected anaphylactic reactions associated with anaesthesia. Anaesthesia
15. Schummer W, Schummer C, Wippermann J, Fuchs J. Anaphylactic shock: is vasopressin the drug of choice? Anesthesiology
16. Shen Y, Li L, Grant J, et al. Anaphylactic deaths in Maryland (United States) and Shanghai (China): a review of forensic autopsy cases from 2004 to 2006. Forensic Sci Int
17. NCRRGW Group. Animal research: reporting in vivo experiments: the ARRIVE guidelines. J Physiol
18. Dewachter P, Jouan-Hureaux V, Franck P, et al. Anaphylactic shock: a form of distributive shock without inhibition of oxygen consumption. Anesthesiology
19. Dewachter P, Jouan-Hureaux V, Lartaud I, et al. Comparison of arginine vasopressin, terlipressin, or epinephrine to correct hypotension in a model of anaphylactic shock in anesthetized brown Norway rats. Anesthesiology
20. Dewachter P, Raeth-Fries I, Jouan-Hureaux V, et al. A comparison of epinephrine only, arginine vasopressin only, and epinephrine followed by arginine vasopressin on the survival rate in a rat model of anaphylactic shock. Anesthesiology
21. Heck DA, Bylund DB. Mechanism of down-regulation of alpha-2 adrenergic receptor subtypes. J Pharmacol Exp Ther
22. Mitra JK, Roy J, Sengupta S. Vasopressin: its current role in anesthetic practice. Indian J Crit Care Med
23. Overgaard CB, Dzavik V. Inotropes and vasopressors: review of physiology and clinical use in cardiovascular disease. Circulation
24. Hu T, Beattie WS, Mazer CD, et al. Treatment with a highly selective beta(1) antagonist causes dose-dependent impairment of cerebral perfusion after hemodilution in rats. Anesth Analg
25. Edvinsson L, Lacombe P, Owman C, et al. Quantitative changes in regional cerebral blood flow of rats induced by alpha- and beta-adrenergic stimulants. Acta Physiol Scand
26. Vincent R. Drugs in modern resuscitation. Br J Anaesth
27. Hare GM, Worrall JM, Baker AJ, et al. Beta2 adrenergic antagonist inhibits cerebral cortical oxygen delivery after severe haemodilution in rats. Br J Anaesth
28. Holmes CL, Patel BM, Russell JA, Walley KR. Physiology of vasopressin relevant to management of septic shock. Chest
29. Holmes CL, Landry DW, Granton JT. Science review: vasopressin and the cardiovascular system. Part 1: receptor physiology. Crit Care
30. Holmes CL, Landry DW, Granton JT. Science review: vasopressin and the cardiovascular system. Part 2: clinical physiology. Crit Care
31. Bein B, Cavus E, Dorges V, et al. Arginine vasopressin reduces cerebral oxygenation and cerebral blood volume during intact circulation in swine: a near infrared spectroscopy study. Eur J Anaesthesiol
32. Rosenthal G, Hemphill JC 3rd, Sorani M, et al. Brain tissue oxygen tension is more indicative of oxygen diffusion than oxygen delivery and metabolism in patients with traumatic brain injury. Crit Care Med
33. Itoh Y, Suzuki N. Control of brain capillary blood flow. J Cereb Blood Flow Metab
34. Ostergaard L, Jespersen SN, Mouridsen K, et al. The role of the cerebral capillaries in acute ischemic stroke: the extended penumbra model. J Cereb Blood Flow Metab
35. Bruins B, Kilbaugh TJ, Margulies SS, Friess SH. The anesthetic effects on vasopressor modulation of cerebral blood flow in an immature swine model. Anesth Analg