Effects of arginine vasopressin on oxygenation and haemodynamics during one-lung ventilation in an animal model.
In a case of arterial hypotension during one-lung ventilation, haemodynamic support may be required to maintain adequate mean arterial pressure. Arginine vasopressin, a potent systemic vasoconstrictor with limited effects on the pulmonary artery pressure, has not been studied in this setting. Twelve female pigs were anaesthetised and ventilated and arterial, central venous and pulmonary artery catheters were inserted. A left-sided double lumen tube was placed via tracheostomy and one-lung ventilation was initiated. The animals were in the left lateral position, with the left lung ventilated and right lung collapsed. Respiratory and haemodynamic values were recorded before and during a continuous infusion of arginine vasopressin sufficient to double the mean arterial pressure. The arginine vasopressin caused a decrease in cardiac output (3.8[+ or -]1.1 vs. 2.7[+ or -]0.7 l/min, P <0.001) and mixed-venous oxygen tension (39.1[+ or -]5.8 vs. 34.4[+ or -]5 mmHg, P=0.003). Pulmonary artery pressure was unchanged (24[+ or -]2 vs. 24[+ or -]3 mmHg, P=0.682). There was no effect of the arginine vasopressin on arterial oxygen tension (226[+ or -]106 vs. 231[+ or -]118 mmHg, P=0.745). However, there was a significant decrease in shunt fraction (28.3[+ or -]6.2 vs. 24.3[+ or -]7.8%, P=0.043) and a significant proportional increase in perfusion of the ventilated lung (78.8[+ or -]9.5 vs. 85.5[+ or -]7.9%, P=0.036). In our animal model of one-lung ventilation, doubling mean arterial pressure by infusion of arginine vasopressin significantly affected global haemodynamics, but had no influence on systemic arterial oxygen tension.
Key Words: arginine vasopressin, one-lung ventilation, shunt, systemic oxygenation
One-lung ventilation (OLV) is frequently required during thoracic surgery. During OLV, oxygenation depends on the distribution of blood flow between the ventilated and the non-ventilated lung. Different factors, such as hypoxic pulmonary vasoconstriction (HPV), gravity-dependent effects of positioning of the patient and effects of reduced or increased cardiac output may alter lung perfusion during OLV (1,2). Following initiation of OLV, HPV in the non-ventilated lung, by redistribution of pulmonary blood flow away from areas with poor aeration (i.e. the non-ventilated lung) to well aerated areas (i.e. the ventilated lung), minimises the increase in pulmonary shunt fraction and ameliorates the decrease in arterial oxygen tension ([P.sub.a][O.sub.2]).
Hypotension during thoracic surgery may require treatment with vasoactive drugs such as catecholamines. While catecholamines may be used to correct hypotension, in the situation of OLV they may also exhibit direct and indirect effects on pulmonary perfusion, such as pulmonary vasodilation, vasoconstriction and changes in cardiac output, which may interfere with HPV (3,4), resulting in subsequent changes in oxygenation (4).
Arginine vasopressin (AVP), a potent systemic vasoconstrictor is reported to exhibit minimal effects on the pulmonary circulation (5) and hence could be an attractive alternative drug to correct arterial hypotension during OLV (e.g. for a patient undergoing OLV in septic shock due to empyema (6)).
However, to our knowledge, no data are available on the effects of AVP in the specific situation of OLV Accordingly, to investigate the effects of AVP on oxygenation and haemodynamics during OLV, we developed an instrumented porcine animal model in which AVP was titrated to achieve a marked, predefined increase in baseline mean arterial pressure.
MATERIALS AND METHODS
The study was approved by the local Animal Protection Committee and by the governmental Animal Care Office (Thuringer Landesamt fur Lebensmittelsicherheit and Verbraucherschutz, Weimar, Germany).
After overnight fasting with free access to water, 12 healthy female pigs (German landrace, bred by Charles River Laboratories, Sulzfeld, Germany, 25 to 35 kg) were premedicated with ketamine (500 mg intramuscularly) to allow placement of an intravenous cannula in an auricular vein and to initiate pulse oximetry and continuous ECG monitoring. General anaesthesia was then induced with propofol (2 to 3 mg/kg) and rocuronium (0.9 to 1.2 mg/kg). The trachea was orally intubated with a 7.5 to 8.5 ID endotracheal tube (Safety Flex[TM], Mallinckrodt, Athlone, Ireland) and mechanical ventilation adjusted to maintain end-tidal C[O.sub.2] tension at approximately 33 to 37 mmHg during subsequent surgical preparation. Anaesthesia was maintained with a 60:40% mixture of nitrous oxide ([N.sub.2]O) and oxygen ([O.sub.2]), and continuous infusion of propofol (20 to 35 mg/kg/h), remifentanil (10 to 20 [micro]kg/kg/h) and pancuronium (0.1 to 0.2 mg/kg/h). During preparation animals received 500 ml of hydroxyethylstarch, followed by 10 ml/kg of body-warm balanced electrolyte solutions.
Using sterile technique, a femoral arterial catheter was placed and advanced 20 to 25 cm to be positioned in the abdominal aorta for haemodynamic monitoring and arterial blood gas sampling. A flow-directed thermodilution pulmonary artery catheter (PAC) (Edwards Lifesciences, Irvine, California, U.S.A.) was inserted through an 8.5 Fr introducer sheath (Arrow International, Reading, U.S.A.) into the right external jugular vein. Using pressure recording, the tip of the PAC was positioned just beyond the pulmonary valve to ensure placement in the main pulmonary artery (i.e. the catheter was not advanced to the wedge-position). The PAC was connected to a cardiac output (CO) device (COLD-Z021[TM], Pulsion Medical Systems, Munich, Germany). Cardiac output measurements were performed in triplicate by central venous injection of 10 ml cold saline (1 to 5[degrees]C) and averaged for each time-point. A central venous catheter was placed in the right internal jugular vein. Body temperature was maintained by covering animals with a forced-air warming blanket (Warmtouch[R], Mallinckrodt, El Paso, Texas, U.S.A.) and was continuously monitored by the thermistor of the PAC. The temperature was kept between 36.5 and 37.5[degrees]C.
Tracheostomy was performed and the orotracheal tube was replaced under fibreoptic control by a left-sided, specially designed 39 Fr double-lumen tube (DLT) (Mallinckrodt, Athlone, Ireland). This DLT ensured that the right upper bronchus of the pigs (originating from the trachea) could also be ventilated or accessed through the tracheal limb of the tube. Animals were placed in the left lateral decubitus position and kept in this position throughout the experiment. An 8.0 ID endotracheal tube was passed through a small (approximately 5 cm) right-sided thoracotomy into the right pleural space. Ventilation to the right lung was discontinued and lung collapse was verified by fibreoptic control (thoracoscopy) of the right pleural space via the tube.
After the experimental preparation was completed, remifentanil and [N.sub.2]O were discontinued. Anaesthesia was provided by continuous intravenous infusion of propofol without changing the infusion rate throughout the experiment.
Mechanical ventilation (Evita 2, Drager, Lubeck, Germany) was provided with a Fi[O.sub.2] of 1.0 in a pressure control mode at 25 cm[H.sub.2]O driving pressure and 5 cm[H.sub.2]O end-expiratory pressure. Ventilation pressures were kept constant during the experiment for each animal. Respiratory rate was adjusted to maintain end-tidal C[O.sub.2] at between 33 to 37 mmHg.
After a stabilisation time with stable cardiorespiratory variables for at least 30 minutes (no more than 10% variation of the baseline at the beginning of the 30 minutes), respiratory and haemodynamic parameters including mean arterial pressure (MAP), heart rate, end-tidal carbon dioxide concentration, [O.sub.2] saturation via pulse oximetry (AS/3, Datex, Helsinki, Finland) and CO were recorded. Arterial blood and mixed venous blood samples were analysed by an automated blood gas analyser (ABL 625, Radiometer, Copenhagen, Denmark).
After baseline measurements, a continuous infusion of AVP was started to achieve a 100% increase in MAP compared to baseline values (e.g. if the baseline MAP was 50 mmHg AVP infusion was titrated to achieve an MAP of 100 mmHg). After maintaining the desired mean arterial pressure ([+ or -]10%) without changing the infusion rate of AVP for at least 30 minutes, haemodynamic values were recorded and an arterial and mixed-venous blood gas analysis performed.
Colour-coded microspheres for intravenous injection were available for the last six animals of the series, giving us the opportunity to assess left and right lung perfusion before and during AVP-infusion. Application and methodological consideration of microsphere measurements in pigs have been presented and discussed in detail elsewhere (7).
Briefly, differently coloured microspheres were injected into the central venous line at baseline and during AVP-infusion. Microspheres were flow dependently trapped in the pulmonary microvasculature. Lungs were harvested at the end of the experiment and microspheres were chemically extracted from lung tissue (7). Right and left lung perfusion in each study phase was calculated according to the amount of microspheres of each colour detected by spectrophotometry in right and left lung samples.
During the study, lung separation and correct DLT placement was verified by continuous dual capnography (i.e. capnography of the dependent and the non-dependent lung), fibreoptic bronchoscopy (at the end of study periods) and thoracoscopy of the right hemithorax (i.e. verification of right lung collapse) at the end of the baseline measurements and after the AVP phase.
At the end of the experiment the animals were sacrificed under deep anaesthesia with potassium chloride. The lungs were removed and examined macroscopically to rule out gross pathology (e.g. pleural lesions, pneumonia). The lungs of the last six animals were further processed for the determination of lung perfusion.
Shunt fraction, blood oxygen content and oxygen extraction ratio were calculated using different equations. (Refer to Appendix for detailed mathematical equations.)
Statistical analysis was performed using the computing software "Statistical Package for the Social Sciences" SPSS (14th version, SPSS Inc., Chicago, IL, U.S.A.). After verifying normal distribution of the data (Kolmogorov-Smirnov-Test), a paired t-test was used to compare the changes before and after AVP infusion. Data are presented as mean [+ or -] SD. The level of statistical significance was set at P <0.05.
All animals finished the study protocol. The mean infusion rate of AVP to achieve the predefined increase in mean arterial pressure was 0.23[+ or -]0.07 U/kg/h.
AVP caused a significant decrease in heart rate, stroke volume and cardiac output, a significant increase in central venous pressure and systemic vascular resistance. The pulmonary arterial pressure was unchanged during AVP infusion (Table 1). The decrease in cardiac output and the increase in central venous pressure were seen in every single animal.
The [P.sub.a][O.sub.2] and [P.sub.a]C[O.sub.2] remained unchanged, whereas oxygen extraction rate ([O.sub.2] ER) increased. Minute ventilation, mixed-venous oxygen tension (Pv[O.sub.2]) and shunt fraction significantly decreased. Proportional perfusion of the dependent left lung increased significantly (Table 2). This effect was seen in each of the six animals where lung perfusion was assessed.
To our knowledge, this is the first study investigating the effects of AVP during one-lung ventilation. It demonstrated that during OLV, infusion of AVP caused major haemodynamic alterations but did not impair arterial oxygen tension.
The decrease in cardiac output (29%) and heart rate (13%) seen in this study during the infusion of AVP is consistent with the effects of AVP reported from other clinical conditions like septic shock or post cardiac arrest (8,9). Most likely the decrease in Pv[O.sub.2] (12%) is secondary to the decrease in CO, which (under conditions of unchanged arterial oxygen content) would result in decreased oxygen delivery and hence (under conditions of unchanged oxygen demand) result in an increased tissue oxygen extraction ratio as shown in our investigation.
In the situation of OLV, alterations in CO may affect oxygenation. Russell et al showed that increasing CO with inotropic agents like adrenaline or dobutamine significantly impairs the oxygenation due to increased shunt (4-10). In contrast, a decrease in CO leads to a preferential increase of perfusion of the ventilated lung (3), which in theory (because of a reduced shunt fraction) could result in an increase in [P.sub.a][O.sub.2]. Interestingly, in our study with the AVP-induced decrease in CO, we did not see an increase in [P.sub.a][O.sub.2].
During OLV, oxygenation is influenced by more than one factor. In our study, the mixed venous oxygen tension (Pv[O.sub.2]) significantly decreased during the infusion of AVE If at a given intrapulmonary shunt the Pv[O.sub.2] decreases, this may ultimately, by mixing of the mixed venous (shunted) blood with oxygenated blood, lead to a decrease in systemic oxygenation (11). On the other hand, a decrease in Pv[O.sub.2] as seen in our study, may enhance HPV, increase perfusion of the dependent lung (12) and hence, in theory, reduce shunt. In our study, we observed a decrease in Qs/Qt with the infusion of AVP, suggesting that this mechanism may have played a role.
We postulate that in our study, the interplay of effects potentially leading to a decrease in oxygenation and effects improving perfusion of the ventilated lung (and thereby reducing Qs/Qt) resulted in a net unchanged [P.sub.a][O.sub.2] with the AVP-induced decrease in CO.
The infusion of AVP did not affect PAP in our animal model. In one study in rats, it was shown that AVP is capable of inducing pulmonary vasodilatation during hypoxic pulmonary vasoconstriction (5). In our study we did not measure HPV directly. Further, due to the position of the tip of the PAC, we were not able to measure pulmonary artery occlusion pressure and hence unable to calculate pulmonary vascular resistance (PVR). However, given the marked decrease in CO during infusion of AVP in the presence of similar PAP before and during the infusion of AVP, one could assume that the PVR increased during the study period. In addition, an increase in PVR may increase the workload of the right ventricle. Our finding of an overall increase in CVP with the infusion of AVP is consistent with this assumption.
A high PVR can increase the fraction of CO delivered to the collapsed nondependent lung (2). However our data on lung perfusion prior to and during infusion of AVP indicate that this effect did not play a major part in our study. The increase of perfusion of the dependent lung is more consistent with the decrease in CO (3) and this may explain the reduced overall shunt fraction.
Our study has limitations. First, our model of OLV, while reflecting the clinically relevant aspects of lateral positioning and unilateral lung collapse, lacks surgical manipulations to the non-dependent lung, a factor which has been shown to affect pulmonary haemodynamics (13). Second, our data were obtained in healthy mammals, while many patients undergoing thoracic surgery preoperatively present with significant cardiac and pulmonary co-morbidities, which may affect the response to OLV Third, while the amount of AVP used in this study was comparable to doses used in other porcine studies investigating effects of AVP (14), the target of doubling MAP by AVP infusion was an arbitrary one. Because with AVP, due to its relatively long half-life, a subtle increase in MAP is more difficult to achieve and control than by infusion of catecholamines, in this first study aiming to identify general effects of AVP during OLV, we decided to provoke a rather pronounced increase in MAP. However, from our data it remains unclear if lesser alterations in MAP would similarly affect global haemodynamics and oxygenation. Hence, further studies, more closely reflecting the scenario of correcting intraoperative hypotension during OLV, are required to assess the potential value of AVP as compared to other vasoactive agents in this situation.
Lastly, the microsphere technique used to assess proportional lung perfusion was only available during the later part of the study and hence was only applied in six of 12 animals. However, in these animals effects of AVP on proportional lung perfusion were uniform. Further, haemodynamic effects of AVP in these six animals were not different from the effects seen in the other six animals not receiving microspheres. We therefore assume that our data on lung perfusion are reliable.
In conclusion, in this animal model of OLV, infusion of AVP caused a major decrease in CO and mixed venous oxygen saturation but did not alter arterial oxygenation. Most likely the interplay of indirect effects on HPV and an increased 02 ER, resulting in a lower Pv[O.sub.2], led to a balance between the potentially positive and potentially negative effects on oxygenation of AVP, resulting in a net unchanged arterial oxygen content.
Calculations and equations
The shunt fraction was estimated using the equation:
Qs / Qt = (Cc[O.sub.2] - Ca[O.sub.2]) / (Cc[O.sub.2] - Cv[O.sub.2]),
where Qs represents shunt flow, Qt total flow, Ca[O.sub.2]=arterial, Cv[O.sub.2]=mixed venous and Cc[O.sub.2]=capillary oxygen content.
The arterial oxygen content was calculated using the equation:
Ca[O.sub.2] = hb x [S.sub.a][O.sub.2] x 1.34 + [P.sub.a][O.sub.2] x 0.0031,
where hb=haemoglobin concentration, [S.sub.a][O.sub.2]=arterial oxygen saturation and [P.sub.a][O.sub.2]=arterial oxygen tension. For calculation of the mixed venous oxygen content the mixed-venous saturation and the mixed venous oxygen tension were used as appropriate. For calculation of Cc[O.sub.2], capillary oxygen saturation was assumed to be 100% and capillary oxygen tension was assumed to equal alveolar oxygen tension ([P.sub.A][O.sub.2]), which was calculated according to the alveolar gas equation.
The alveolar gas equation was used as followed:
[P.sub.A][O.sub.2] = ([P.sub.b] - [P.sub.SVP]) x Fi[O.sub.2] - [P.sub.a]C[O.sub.2] / RQ,
where [P.sub.b]=barometric pressure (760 mmHg at sea level), [P.sub.SVP]=pressure of saturated water vapour (47 mmHg at 37[degrees]C), Fi[O.sub.2]=fraction of inspired oxygen, which was set at 1.0. [P.sub.a]C[O.sub.2]=arterial carbon dioxide tension and RQ=respiratory quotient (estimated to be 0.8).
To obtain the oxygen extraction ratio ([O.sub.2]ER), oxygen delivery (D02) and oxygen consumption (V02) were calculated according to the following equations:
1) D[O.sub.2] = CO x Ca[O.sub.2]
2) V[O.sub.2] = CO x (Ca[O.sub.2] - Cv[O.sub.2])
Then the following equation was used to calculate [O.sub.2]ER:
3) [O.sub.2]ER = V[O.sub.2] / D[O.sub.2] x 100
Accepted for publication on November 13, 2007.
(1.) Bardoczky GI, Szegedi LL, d'Hollander AA, Moures JM, de Francquen P, Yernault JC. Two-lung and one-lung ventilation in patients with chronic obstructive pulmonary disease: the effects of position and F(IO)2. Anesth Analg 2000; 90:35-41.
(2.) Malmkvist G, Fletcher R, Nordstrom L, Werner O. Effects of lung surgery and one-lung ventilation on pulmonary artery pressure, venous admixture and immediate postoperative lung function. Br J Anaesth 1989; 63:696-701.
(3.) Eisenkraft JB. Effects of anaesthetics on pulmonary circulation. Br J Anaesth 1990; 65:63-78.
(4.) Russell WJ, James ME The effects on increasing cardiac output with adrenaline or isoprenaline on arterial haemoglobin oxygen saturation and shunt during one-lung ventilation. Anaesth Intensive Care 2000; 28:636-641.
(5.) Walker BR, Haynes J, Wang HL, Voelkel NF. Vasopressin-induced pulmonary vasodilation in rats. Am J Physiol 1989; 257:H415-422.
(6.) Dellinger RP, Carlet JM, Masur H, Gerlach H, Calandra T, Cohen J et al. Surviving Sepsis Campaign guidelines for management of severe sepsis and septic shock. Crit Care Med 2004; 32:858-873.
(7.) Walter B, Bauer R, Gaser E, Zwiener U. Validation of the multiple colored microsphere technique for regional blood flow measurements in newborn piglets. Basic Res Cardiol 1997; 92:191-200.
(8.) Klinzing S, Simon M, Reinhart K, Bredle DL, Maier-Hellmann A. High-dose vasopressin is not superior to norepinephrine in septic shock. Crit Care Med 2003; 31:2646-2650.
(9.) Kern KB, Heidenreich JH, Higdon TA, Berg RA, Hilwig RW, Sanders AB et al. Effect of vasopressin on postresuscitation ventricular function: unknown consequences of the recent Guidelines 2000 for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Crit Care Med 2004; 32 (9 Suppl):S393-397.
(10.) Russell WJ, James ME The effects on arterial haemoglobin oxygen saturation and on shunt of increasing cardiac output with dopamine or dobutamine during one-lung ventilation. Anaesth Intensive Care 2004; 32:644-648.
(11.) Slinger P, Scott WA. Arterial oxygenation during one-lung ventilation: a comparison of enflurane and isoflurane. Anesthesiology 1995; 82:940-946.
(12.) Domino KB, Wetstein L, Glasser SA, Lindgren L, Marshall C, Harken A et al. Influence of mixed venous oxygen tension (Pvo2) on blood flow to atelectatic lung. Anesthesiology 1983; 59:428-434.
(13.) Ribas J, Jimenez MJ, Barbera JA, Roca J, Gomar C, Canalis E et al. Gas exchange and pulmonary hemodynamics during lung resection in patients at increased risk: relationship with preoperative exercise testing. Chest 2001; 120:852-859.
(14.) Knotzer H, Pajk W, Maier S, Ladurner R, Kleinsasser A, Wenzel V et al. Arginine vasopressin reduces intestinal oxygen supply and mucosal tissue oxygen tension. Am J Physiol Heart Circ Physiol 2005; 289:1-1168-173.
L. HUTER *, K. SCHWARZKOPF ([dagger]), N. P. PREUSSLER ([double dagger]), E. GASER ([dagger]), R. BAUER ([sections]), H. SCHUBERT **, T. SCHREIBER ([double dagger])
Department of Anaesthesiology and Intensive Care Medicine, Center for Molecular Biomedicine and Institute for Experimental Animals, University of Jena, Jena, Germany
* M.D., Staff Anesthesiologist, Department of Anesthesiology and Intensive Care Medicine. ([dagger]) M.D., Staff, Department of Anaesthesiology and Intensive Care Medicine. ([double dagger]) M.D., D.E.S.A., Staff Anaesthesiologist, Department of Anaesthesiology and Intensive Care Medicine. ([sections]) M.D., Staff, Center for Molecular Biomedicine. ** D.V.M., Head, Institute for Experimental Animals.
Address for reprints: Dr L. Huer, Klinik fur Anasthesiologie and Intensivtherapie, Klinikum der Friedrich-Schiller-Universitat Jena, Erlanger Allee 101, 07747 Jena, Germany.
TABLE 1 Haemodynamic variables at baseline and during arginine vasopressin (A VP) infusion (n=12 animals). Values are given as mean [+ or -] SD. The given P values are for the paired t-test. Baseline During AVP P value HR (bpm) 87[+ or -]16 76[+ or -]10 -0.001 MAP (mm Hg) 52[+ or -]13 99[+ or -]20 -0.001 PAP (mm Hg) 24[+ or -]2 24[+ or -]3 0.682 CVP (mmHg) 9.3[+ or -]3.3 10.5[+ or -]3.2 0.001 CO (l/min) 3.8[+ or -]1.1 2.7[+ or -]0.7 -0.001 SV (ml) 43[+ or -]8 36[+ or -]8 0.001 SVR (dyn/s/[cm.sup.5]) 951[+ or -]278 2825[+ or -]1261 <0.001 HR=heart rate, MAP=mean arterial pressure, PAP=mean pulmonary arterial pressure, CVP=central venous pressure, CO=cardiac output, SV=stroke volume, SVR=systemic vascular resistance. TABLE 2 Respiratory and oxygenation variables at baseline and during arginine vasopressin (A VP) infusion (n=12 animals, except for left lung perfusion which was assessed in only six animals) Values are given as mean [+ or -] SD. The given P values are for the paired t-test. Baseline During AVP P value [P.sub.a][O.sub.2] (mm Hg) 226[+ or -]106 231[+ or -]118 0.745 [P.sub.a]C[O.sub.2] (mmHg) 39.5[+ or -]2.7 39.2[+ or -]5.3 0.821 Qs/Qt (%) 28.3[+ or -]6.2 24.3[+ or -]7.8 0.043 Left lung perfusion (%) 78.8[+ or -]9.5 85.5[+ or -]7.9 0.036 [O.sub.2] ER (%) 31.6[+ or -]6.1 40.2[+ or -]6.3 0.001 PV[O.sub.2] (mm Hg) 39.1[+ or -]5.8 34.4[+ or -]5 0.003 MV (l/min) 6.8[+ or -]1.2 6.5[+ or -]1.6 0.028 [P.sub.a][O.sub.2]=arterial oxygen tension, [P.sub.a]C[O.sub.2]=arterial carbon dioxide tension,Qs/Qt =calculated shuntfraction, left lungperfusion=perfusion of the dependent left lung, [O.sub.2] ER=oxygen extraction ratio, PV[O.sub.2]=mixed-venous oxygen tension, MV=minute ventilation.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Original Papers|
|Author:||Huter, L.; Schwarzkopf, K.; Preussler, N.P.; Gaser, E.; Bauer, R.; Schubert, H.; Schreiber, T.|
|Publication:||Anaesthesia and Intensive Care|
|Article Type:||Clinical report|
|Date:||Mar 1, 2008|
|Previous Article:||Treatment of recurrent chest pain in a heart transplant recipient using spinal cord stimulation.|
|Next Article:||Anaesthetic implications of hyperekplexia--'startle disease'.|