- Email: [email protected]
Role of sodium in contrast medium-induced polymorphic ventricular tachycardia: Results in a rabbit model of lengthened QT interval
Role of Sodium in Contrast Medium-induced Polymorphic Ventricular Tachycardia: Results in a Rabbit Model of Lengthened QT Interval Jean-Marc Idle, PharmD, Christine Bourgoin-Balut, BS, Thierry Lefevre, MD, Philippe Zamia, PhD Val~rie Goulas, BS, Sophie Galliard, MD, Corinne Mathias, BS, Bruno Bonnemain, PharmD
Rationale and Objectives. The authors: (a) Compared the proarrhythmic effects of i0xaglate (!52 mmol/L sodium) and iohexol (no sodium) in a rabbit model and (b) as;sessed the effect of adding 150 mmot/L:sodium to iso:tonic iohexo!. : : M a t e r i a l s and Metlaods; Either ioxagtate:(320 mg of iodine:per milliliter) or iohexot (350 m g o f iodine per mitli:liter) was selectively injected into the right Coronary aritery (t.5 mL over 30 seconds) of 10 rabbits, some of which also received the %,adrenergic receptor agonist :::methoxamine.To validate the model;: the class IlI antiar:::rhythmic agent clofilium Was injected intravenously during methoxamine infusion. Frontal electrocardiography was performed continuously todetect polymorphic ventricular tachycardia (PVT): In a second study; the authors '::i assessed the frequency of arrhyttmaiaS:after hnjection of :::isotonic iohexot Solution (145 mg of iodine per miIlili: ter), either atone or with i50 mxr~olFLsodium. : : :Results. Methoxamine :significantly iengthened the QT, Q T c , and RR:intervalS (P < .05). The use of clofilium :: alone induced no PVT, whereas five of:eight methoxa: ~ne-infused rabbits developed PVT: after: Ciofilium injection (P = .03). Both contrast media prolonged the repolarization period. IohexoI alone induced a higher fiequency of PVT than did ioxaglate a!one (P = .0006). Methoxamine infusion did not Potentiate the frequency of p\,r-f in the ioxagiate-injected rabbits. The addition of Sodium to isotonic iohexol prevented: the occurrence of :PVT (P = :.0006). :: :.... :: :: :::Conclusion, Although ioxagiate prolonged :the repolari ization period, itdid not Causea higher frequency of ar::: :rhythmia When injected in association with: methoxa: mine. Iohexol, which contains no Sodium, induced a t'Agh :frequencyof arrhythrnia. The addition of a physiologic Concentration of sodium to isotonic iohexoI can prevent :::i:ventricuiax arrhythmias: :i ::: : :Key Words: Angiographyi e;ntrast :media; contrast media, comparative studies; heart, an'hythrrfia; iohexoI; ? ;loxagtate. i::i:: : :::: :::::i: : :::: :i
Iodinated contrast medium-induced arrhythmia may occur during coronary angiography or percutaneous transluminal coronary angioplasty (1,2). The most serious arrhythmia that may occur during such procedures is ventricular fibrillation, that is, chaotic asynchronous fractionated activity of the heart (3). Ventricular fibrillation generally occurs as a consequence of the inherent properties of the contrast medium, the patient's clinical status, and operator technique (4,5). The contrast medium-induced component of such arrhythmia is still unclear. Sodium concentration is generally considered to be an important contrast medium-related parameter (4). Pedersen et al (6) found a marked lengthening in monophasic action potential duration when ioxaglate, an ionic low-osmolar contrast medium containing 152 mmol/L sodium, was injected into the anterior descending branch of the left coronary artery during a simulated wedged-catheter situation in dogs. They found that ioxaglate caused a high frequency of ventricular fibrillation in their particular model. They suggested that excessive lengthening of ventricular repolarization induced by this contrast medium may be arrhythmogenic, as seen in patients with long QT syndrome and after using some class III antiarrhythmic drugs (6,7). They hypothesized that such excessive lengthening of ventricular repolarization (and the consecutive proarrhythmic risk) may be due to the high sodium concentration of the ioxaglate solution. To investigate this hypothesis, we compared the ef-
Acad Radio11998; 5:435-443 1From the Departments of Pharmacology (J.MI., C.B,B., V,G., C.M,), Biostatistics (P.Z.), and Clinical Research (S.G.), Laboratoire Guerbet, BP 50400, F-95943 Roissy-Charles-de-Gaulle Cedex, Aulnay-sous-Bois, France; and the Department of Interventional Cardiology, Institut Cardiovasculaire Paris-Sud, Massy, France (T.L.). Received June 2, 1997; revision requested August 1; revision received August 18; acc e p t e d September 20. Address reprint requests to J.M.I. © AUR, 1998
435
fects of ioxaglate with those of a nonionic contrast medium, iohexol (which contains no sodium), on a recent and validated rabbit model of arrhythmia. This model, which consists of animal sensitization and QT interval prolongation with methoxamine, revealed that some drugs, such as class III antiarrhythmic drugs or cisapride, are associated with polymorphic ventricular tachycardia (PVT), which results from prolonged cardiac repolarization (8-10). If the hypothesis of Pedersen et al (6) is correct, ioxaglate should induce a higher frequency of arrhythmia when associated with methoxamine than when injected alone. The same authors also found that adding 150 mmol/L sodium to a nonionic contrast medium led to ventricular fibrillation (6). To investigate this eventuality, we evaluated the electrocardiographic consequences of the addition of 150 mmol/L sodium to iohexol. To avoid any possible interference with osmolality, we used an isotonic iohexol solution.
Animal Preparation Seventy-seven male New Zealand white rabbits (Elevage Cunicole de Lantages, Sainte Savine, France; mean weight + standard error of the mean, 3.0 kg + 0.1) were anesthetized by means of intramuscular injection of 5.0 mg/kg midazolam (Hypnovel; Roche, Neuilly-surSeine, France) and 50 mg/kg ketamine (Imalgene; RhtneMtrieux, Lyon, France), and a tracheostomy was performed. From the femoral artery, a catheter (16 gauge; Vygon, Ecouen, France) was advanced into the aorta for blood pressure recordings. The marginal vein of each ear was catheterized to allow infusion of methoxamine and the validation agent when necessary. The left common carotid artery was isolated surgically and catheterized for right coronary injection via a preshaped 3-F polyethylene catheter (Guerbet Biomtdical, Louvres, France). A few seconds before the injection, the catheter was advanced selectively into the right coronary artery. The catheter was positioned and the injection performed under fluoroscopic control (Stenoscop; GE Medical Systems, Issy-les-Moulineaux, France). Immediately after the injection, the catheter was withdrawn to the aortic arch.
Electrophysiologic Measurements Frontal I and II and precordial V 3 leads were recorded at electrocardiography with an AT-3 recorder (Schiller,
436
Baar, Switzerland). Monitoring was continuous. The QT interval was corrected for heart rate changes by using the following equation (11): QTc = QT x ~](1,000/RR interval), where QTc, QT, and RR intervals are measured in milliseconds. PVT was defined as a series of rapid repetitive polymorphic QRS complexes with characteristic continuous undulation around the isoelectric line usually associated with a pause or slowing of heart rate and was considered to have occurred if four or more closely coupled repetitive extrasystoles with this particular pattern were observed (12,13).
Hemodynamic Measurements Aortic blood pressure was measured with a Spectramed Statham P 23 XL transducer (Gould Electronique, Ballainvilliers, France) and was continuously recorded and displayed on a Gould ES 1000 polygraph and a Gould V 1000 monitor, respectively. Continuous data acquisition and treatment were carried out by means of HEM software (Notocord, Croissy-sur-Seine, France). All experiments were carried out in compliance with the European Economic Community directive (86/609/ EEC) on animal welfare.
Test Solutions Sodium meglumine ioxaglate (Hexabrix; Laboratoire Guerbet, Aulnay-sous-Bois, France) and iohexol (Omnipaque; Nycomed, Oslo, Norway) were tested in their commercially available forms (ioxaglate: 320 mg of iodine per milliliter, osmolality at 37°C = 600 mOsm/kg; iohexol: 350 mg of iodine per milliliter, osmolality at 37°C = 880 mOsm/kg). In a second series of experiments, iohexol was diluted in distilled water to reach isotonicity (osmolality at 37°C = 308 mOsm/kg; iodine concentration, 145 mg of iodine per milliliter) and, in another test, 150 mmol/L sodium was added to this solution (osmolality at 37°C = 635 mOsm/kg).
Experimental Protocol After surgery, animals were allowed to equilibrate for 10 minutes before baseline measurements were obtained and the study was started. Time points of PVT occurrence and the duration of the arrhythmic episode were measured for all experiments. Validation of the experimental modeL--Rabbits (eight per group) were sensitized according to the technique used by Carlsson et al (8), which involves continuous intravenous infusion (10 gg/kg/min over 30 minutes) of methoxamine (Sigma, Saint Quentin-Fallavier, France)
vol 5, No 6, June 1998
I!! ii,¸.If¸ I
CONTRAST MEDIUM-INDUCED POLYMORPHIC VENTRICULAR TACHYCARDIA
-t ...... !...... !.......
.
.
.
Figure 1. E l e c t r o c a r d i o g r a m s h o w s t y p i c a l split T w a v e s ( a r r o w h e a d ) o c c u r ring a f t e r i n j e c t i o n of t h e class III a n t i a r r h y t h m i c d r u g clofilium in ra b bit s (V3 l e a d of t h e s u r f a c e e l e c t r o c a r d i o g r a m ) .
!
I I I
Vl
.!~~'"iiill..!ii~i
.
...............
I l
. . . . . . . ; ....................i...... t
,
~
[ ~.u. [mm,/n~ ~J~n! "ti
i
1
-
diluted in 5% glucose. The class III antiarrhythmic agent clofilium (Research Biochemicals International, Natick, Mass), which is used as a reference compound, was injected (10 mg/kg, intravenous route, bolus injection), either 15 minutes after the start of methoxamine infusion or, in rabbits that did not receive methoxamine, at random. Electrocardiographic tracings were recorded for 45 minutes after clofilium injection. Interaction of contrast medium with methoxamine.-After 15 minutes of methoxamine infusion, the test solution was selectively injected into the right coronary artery (1.5 mL over 30 seconds [14]). ECG tracings were monitored for 30 minutes after contrast medium injection, and episodes of PVT were recorded. Each group consisted of 10 (in the ioxaglate group) or 11 (in the iohexol group) rabbits. Effects of sodium addition to isotonic iohexol.--Rabbits were randomly assigned to receive intracoronary injection of either isotonic iohexol alone or in combination with 150 mmol/L sodium according to the above-mentioned technique. Statistical Analysis Measured values are expressed as the mean + standard error of the mean. For each study, the homogeneity of the groups in terms of preinjection values (blood pressure, heart rate, QT interval) was checked by means of analysis of variance. With respect to blood pressure, heart rate, and QTc interval, the values at various specified times following the injection of test solutions were compared with preinjection values by using the paired Student t test adjusted by the number of measurements (Bonferroni adjustment). Frequencies of PVT occurrence were compared by using the Fisher exact test (15). Statistical analysis of blood pressure and QT interval was conducted by using the Number Cruncher Statistical Systems program version 5.0 (NCSS, Kaysville, Utah), and analysis of PVT occurrence was performed by using StatXact program version
3.0 (Cytel Software, Cambridge, Mass). A P value of .05 or less was considered to indicate a statistically significant difference. RESULTS Interaction of Clofilium with M e t h o x a m i n e Before clofilium injection, methoxamine infusion significantly increased blood pressure (102 m m Hg _+ 2 vs 84 mm Hg + 3 at baseline, P < .001) and decreased heart rate (249 beats per minute _+ 10 vs 261 beats per minute + 10 at baseline, P < .01). Separation of the T wave of the surface electrocardiogram into two separate peaks ("TU complex") was frequently observed after injection of clofilium alone or with methoxamine infusion and, in some cases, before any injection (Fig 1). Clofilium alone induced no PVT in any of the eight treated rabbits, whereas five of the eight methoxamine-infused rabbits showed PVT after clofilium injection (P = .03). One methoxamine-infused rabbit had lethal ventricular fibrillation after clofilium injection. The QTc interval was lengthened in the clofilium alone and clofilium- and methoxamine-infused groups (Table 1). In some cases, the QT interval was not measured because of repolarization-dependent blocks that occurred at measurement time points (Fig 2).
Interaction of Contrast M e d i u m with Methoxamine PVT was not induced by the positioning of the catheter in any of the coronary angiography studies. Hemodynamic effects of contrast medium alone.-Ioxaglate alone transiently decreased blood pressure (baseline, 85 m m Hg + 2; end of injection, 68 mm Hg _+ 7; P = .02), as did iohexol alone (baseline, 79 mm Hg + 3; end of injection [n = 4], 49 mm Hg + 15 ; P = .16). Baseline values did not significantly differ between the two groups (P = .10).
437
Table I Clofilium-induced Changes in QTc Intervals with and without Methoxamine QTc Interval Duration (msec) Treatment Clofilium Clofilium a n d methoxamine
Baseline
During Injection
326 + 10 (n = 8) 346+8 (n = 8)
311 + 19 (n = 8) 411 + 20 (n = 8)
5 minutes after Injection 365 + 26 (n = 7) 415+10 (n = 6 )*
15 minutes after Injection 373 _+ 29 (n = 6) 4 3 2 + 27 (n : 3)
30 minutes after Injection 413 + 46 (n = 4) 4 2 0 + 14 (n = 3)
40 minutes after Injection 370 + 34 (n = 6) 414 (n = 2)
N o t e . - - D a t a are g i v e n as means + standard error of the m e a n . *P< .05.
Table 2 Methoxamine-induced Changes in QT, QTc, and RR Intervals before loxaglate and lohexol Injections QT Interval (msec)
Treatment Methoxamine a n d clofilium
Baseline
15 minutes after Injection
QTc Interval (msec)
PValue
Baseline
15 minutes after Injection
RR Interval (msec)
PValue
Baseline
15 minutes after Injection
PValue
155_+4
170+6
<.01
324+ 9
346+8
<.05
230+ 8
241 + 9
<.05
147+6
164+7
<.01
299+8
318+8
<.01
243+ 12
266+ 12
<.05
160+5
172+4
<.01
320+8
333+6
<.01
250+_4
267+4
<.01
(n = 8) Methoxamine and ioxaglate (n : 10) Methoxamine a n d iohexol (n = 10)
N o t e . - - D a t a are given as means + standard error of t h e m e a n .
Figure 2. E l e c t r o c a r d i o g r a m shows typical r e p o l a r i z a t i o n - d e p e n d e n t b l o c k o c curring after injection of t h e class III antiarrhythmic drug clofilium in rabbits.
I
r:tft Intracoronary injection of ioxaglate was associated with a decrease in heart rate (baseline, 237 beats per minute + 7; end of injection, 205 beats per minute + 14; P -- .04), as was iohexol alone (baseline, 231 beats per minute + 10; 15 seconds after injection [n = 6], 172 beats per minute + 23, P -- .02; end of injection [n = 4], 191 beats per minute + 22, P = .06). Baseline values did not differ significantly between the two groups (P = .63). Effects of contrast medium on QTc intervaL--The difference in baseline QTc interval values for the groups receiving ioxaglate and iohexol was not statistically signifi-
438
F1 I - I - ' V F i I V 1 q ilqg"~:
cant (P = .48). In the methoxamine-infused rabbits, both the QT and QTc intervals were significantly lengthened before contrast medium injection compared with that at baseline. The RR interval was also significantly lengthened by methoxamine for both groups (Table 2). Ioxaglate lengthened the duration of the QTc interval in animals, regardless of whether they also received methoxamine (although the duration was not significantly lengthened in the latter group). Such an effect was still noticeable 3 minutes after the start of the contrast medium injection (Table 3). Because of the high frequency
Table 3 Effects of loxaglate and lohexol Injection with and without Methoxamine on QTc Interval QTc Interval Duration (msec) Treatment
Baseline
Ioxaglate
314 + 13 (n = 10) 298 _+10 (n = 11) 331 + 9 (n-- 10) 342 + 9 (n = 10)
Iohexol Ioxaglate and methoxamine Iohexol a n d methoxamine
End of Injection
1 minute after Injection
3 minutes after Injection
351 + 20 (n = 9)* 323 (n = 2) 361 + 20 (n = 9) 340 (n = 1)
340 + 7 (n = 9) 318+ 11 (n = 6) 334 + 9 (n = 10) 348 + 12 (n = 6)
334 + 8 (n = 9) 318+ 12 (n = 6) 336 + 8 (n = 10) 361 + 7 (n = 7)
N o t e . - - D a t a are g i v e n as m e a n s + s t a n d a r d error of t h e m e a n . *P<.05.
......
I
t t-'
' ..... i --
F 'L
' .... "
...... ii-]l
"
of PVT episodes during iohexol injection, only a small number of tracings allowed measurement of the QTc interval duration in this group, and statistical analysis was not possible. Effects of methoxamine infusion.--Methoxamine increased blood pressure before injection of ioxaglate (104 mm Hg + 4 vs 77 mm Hg + 3, P < .001) and iohexol (103 riam Hg + 3 vs 83 mm Hg _+ 3, P < .01) and concomitantly reduced the heart rate (ioxaglate: 230 beats per minute _+ 12 vs 247 beats per minute _+ 13, P < .001; iohexol: 224 beats per minute _+ 3 vs 240 beats per minute _+4, P < .001). ECG effects of contrast medium.--Iohexol alone induced a significantly greater frequency of PVT than did ioxaglate alone (P = .0006). Repolarization-dependent blocks were frequently observed. A typical ischemic tracing, with Pardee waves, was associated with selective contrast medium injection. During the injection period, PVT occurred in all iohexol cases and three of the 10 ioxaglate cases (Fig 3), often in association with bradycardia. The combination of methoxamine and ioxaglate did not increase the frequency of PVT when compared with that with ioxaglate alone (Table 4). The frequency of PVT was higher in the group that received methoxamine and iohexol than in the group that received methoxamine
t
Figure 3.
E l e c t r o c a r d i o g r a m shows PVT e p i s o d e after injection of iohexol into t h e right c o r o n a r y artery of a rabbit, The staff of this arrhythmic e p i s o d e is p r e c e d e d b y t h e o c c u r r e n c e of P a r d e e w a v e s with ST s e g m e n t including t h e T w a v e , w h i c h is i n d i c a t i v e of ischemia,
and ioxaglate (P = .003). In five of the rabbits that received iohexol alone, PVT turned into lethal ventricular fibrillation. This was the case in three methoxamine-infused rabbits that received iohexol and in one rabbit that received ioxaglate.
Effects of sodium addition to isotonic iohexol.--The addition of 150 mmol/L sodium to isotonic iohexol (145 mg of iodine per milliliter) significantly reduced the frequency of PVT (P = .0006) (Table 5). The high frequency of PVT in the group receiving iohexol alone (145 mg of iodine per milliliter) did not allow us to evaluate changes in the QTc interval at the end of injection (Table 6). Aortic blood pressure was not measurable during the arrhythmic episodes (ie, in animals receiving isotonic iohexol alone).
Two hypotheses were tested in this study: (a) Contrast medium-induced prolongation of ventricular repolarization may favor the occurrence of ventricular arrhythmia during coronary angiography procedures, and (b) such an effect may be related to either the absence or a high (150 mmol/L) concentration of sodium in the contrast medium solution.
439
Table 4 ECG Effects of loxaglate and lohexol Injection with and without Methoxamine No. of Rabbits with PVT
Treatment I o x a g l a t e (n = 10) Methoxamine and ioxaglate (n = 10) Iohexol (n = 11) M e t h o x a m i n e a n d iohexol (n = 10)
No. of Rabbits with Lethal FV
Time of PVT O c c u r r e n c e (sec)*
Duration of PVT (sec)*
3
1
14 + 9
19 (n = 2)
3 11 10
0 5 3
26 + 4 19 + 4 15 + 4
25 _+ 16 (n = 3) 32 + 8 (n = 6) 29 + 8 (n = 7)
* D a t a are g i v e n as m e a n s _+ s t a n d a r d error of t h e m e a n ,
Table 5 ECG Effects of Isotonic Iohexol with and without the Addition of 150 mmol/L Sodium No. of Rabbits with PVT
Treatment Iohexol (n = 10) Iohexol a n d 150 m m o l / L sodium (n = 10)
No. of Rabbits with Lethal FV
9 0
Time of PVT O c c u r r e n c e (sec)*
1 0
8+2 NA
Duration of PVT (sec)* 37 + 5 NA
N o t e . - - I s o t o n i c iohexol c o n t a i n e d 145 m g of iodine p e r milINiter, * D a t a are g i v e n as m e a n s + s t a n d a r d error of t h e m e a n . NA : not a p p l i c a b l e .
Table 6 Effects of Isotonic Iohexol with and without the Addition of 150 mmol/L Sodium on QTc Interval i
QTc Interval Duration (msec) Treatment Iohexol Iohexol a n d 150 m m o l / L s o d i u m
Baseline 299 ± 13 ( n = 10) 328 ± 11 ( n = 10)
End of Injection
1 minute after Injection
327 (n = 1) 366 + 20 (n=9)
323 + 6 (n = 9) 332 _+7 ( n = 10)
3 minutes after Injection 329 + 8 (n : 9) 325 + 8 ( n = 10)
N o t e , - - D a t a are g i v e n as m e a n s + s t a n d a r d error of t h e m e a n . Isotonic iohexol c o n t a i n e d 145 m g of iodine p e r milliliter.
Experimental Model The rabbit has been widely used for both in vitro (16) and in vivo (8,9,14,17) electrophysiologic studies. This species has few coronary collateral vessels, unlike dogs (13,18). Furthermore, the cardiac electrophysiology of rabbits is reported to have depolarizing and repolarizing currents that are similar to those of humans (19). The selective injection into the right coronary artery was chosen because in rabbits (14), as in humans (20), ventricular arrhythmias seem to be more frequent after injection into the right coronary artery than after injection into the left coronary artery. In our study, a transient reduction in heart rate was observed after the injection of both low-osmolar contrast
440
media. This result is consistent with data reported from other rabbit studies (14) and clinical trials (4) and is probably related to depression of the proximal conduction system or to a direct effect on the sinoatrial node. In this study, measurement of the QT interval was required. This parameter represents a useful approximation of action potential in the ventricle and is, therefore, worth measuring (21). The QT interval varies with heart rate in rabbits just as it does in many other species. Under our conditions, contrast medium reduced the heart rate. We used Bazett's correction since it has been shown that this procedure is beneficial in rabbits and that the corrected QT distribution is much better than the uncorrected data (21).
In our study, as in the original study in which the model was described (8), rabbits were sensitized with methoxamine. This agent is a specific %-adrenergic agonist that leads to the development of after-depolarizations (8). As shown in Table 2, comparison of baseline values and preinjection values (ie, just before injection of contrast medium or clofilium) shows that methoxamine alone significantly increased the QTc interval in all test groups. Furthermore, the heart rate was also significantly lowered in methoxamine-infused groups before the injection of the test substances. This phenomenon is probably a reflex adjustment to the methoxamine-related increase in blood pressure. It is well known that bradycardia very often precedes the clinical occurrence of PVT (8,22). This is a further element suggesting the clinical relevance of our model. Clofilium is a pure class III antiarrhythmic agent that prolongs monophasic action potential duration by means of the blockade of potassium channels (9,23). Compounds from this class have been associated with severe arrhythmias, including PVT (24). In our study, there was a significantly higher frequency of PVT in rabbits infused with methoxamine and injected with clofilium than in those receiving clofilium only. Our data are consistent with those of Carlsson et al (8) and Buchanan et al (9). A statistically significant lengthening of the QTc interval was observed in the clofilium and clofilium and methoxamine-infused animals, as described elsewhere (8,9). Thus, under our conditions, clofilium was clearly shown to be proarrhythmic, a phenomenon associated with pro' tracted prolongation of the QTc interval. The occurrence of split T waves (or U waves) on the electrocardiographic tracings of clofilium-treated rabbits is consistent with findings reported in the literature (8). U waves are considered to be associated with early after-depolarizations (8). The appearance of the PVT tracings observed in our study matches that described elsewhere (8,22). In the case of contrast medium, an electrocardiographic tracing highly suggestive of ischemia occurred immediately after the catheter was positioned, with the ST segment including the T wave (Pardee waves) (25). It thus appears that, in this case, ischemia is a superimposed factor when compared with the tracings obtained after methoxamine and clofilium administration. It has been suggested that ischemia may facilitate the occurrence of PVT (26). It can thus be suggested that, in studies involving selective intracoronary injection of contrast medium,
ischemia may be an additional sensitization factor for the occurrence of PVT.
Interaction of Contrast Medium with Methoxamine The frequency of PVT was significantly lower with ioxaglate alone than with iohexol alone. A lower frequency of arrhythmia (ventricular fibrillation) for ioxaglate versus iohexol has already been shown in the same species (14). In our study, five of 11 cases of PVT in the iohexol group and one of three cases of PVT in the ioxaglate group turned into lethal ventricular fibrillation. In humans, PVT often ends spontaneously but sometimes degenerates into ventricular fibrillation (22). It has been claimed that ioxaglate (152 mmol of sodium per liter), which in some dog studies markedly lengthened monophasic action potential duration in contrast medium-perfused myocardium, may be proarrhythmic, similar to class III antiarrhythmic drugs, which also lengthen ventricular repolarization (6,7). We therefore tested ioxaglate in experimental conditions that allow detection of the arrhythmogenic potential of class III agents (ie, by injecting this contrast medium in methoxamine-infused rabbits). It is worth noting that, under these severe conditions, ioxaglate did not cause a higher frequency of major arrhythmias in animals whose QTc and RR intervals were markedly lengthened (Table 2), although it lengthened the repolarization period (Table 4). Therefore, our data do not support the hypothesis that an arrhythmogenic effect of this contrast medium may occur as a consequence of lengthening of ventricular repolarization. There is a strong body of experimental evidence from studies on dogs (27-29), pigs (30), isolated rat hearts (31), and isolated sheep myocardial fibers and Purkinje fibers (32) that shows ioxaglate is associated with a lower proarrhythmic risk than are nonionic low-osmolality contrast media. In a study in dogs, it was observed that iohexol, which caused the fewest regional electrophysiologic effects, paradoxically caused ventricular fibrillation in 100% of infusions (33). The clinical relevance of monophasic action potential duration increase in the particular case of contrast medium thus appears to be debatable. Injection of iohexol in methoxamineqnfused rabbits was associated with a similar frequency of PVT as that in the group receiving iohexol alone (ie, all animals showed PVT episodes in both groups). Given the maximum frequency of PVT induced with iohexol alone, a putative potentiation by methoxamine obviously could not be evidenced.
441
Effects of Sodium Addition to Isotonic Iohexol On the basis of their severe (34) dog model, Pedersen et al (6) also suggested that the addition of 150 mmol/L sodium to contrast medium solutions may be deleterious and lead to arrhythmias. In their response to a letter to the editor written by Id6e et al (34), Pedersen et al (35) indicated that either the absence of sodium or the presence of 150 mmol/L sodium in a contrast medium solution is proarrhythmogenic. To investigate this hypothesis, we compared the proarrhythmic potential of isotonic iohexol (without sodium) and iohexol and 150 mmol/L sodium (thus made hypertonic with an osmolality of 635 mOsm/kg). This concentration of sodium clearly protected the rabbits from the onset of PVT (Table 5). Our results, thus, are not fully consistent with the data of Pedersen et al (6). Results of several dog studies have confirmed the value of adding a physiologic concentration of sodium to nonionic contrast medium (27,36,37). Results of in vitro studies also confirmed the value of adding a physiologic concentration of sodium to contrast medium solutions. This was shown by using both isolated rabbit (16) and rat (31) heart models. The addition of 145-150 mmol/L sodium to a nonionic contrast medium solution, however, would lead to a substantial increase in the osmolality of the solution (ie, of 300 mOsm/kg) that could, in turn, have detrimental hemodynamic consequences. Sodium being a counterion in the case of the ionic low-osmolar contrast medium ioxaglate, the osmolality of the solution is kept to acceptable values. It is interesting that isotonicity per se did not abolish the l~ropensity of iohexol to cause arrhythmias. The relative role of this parameter thus appears to be less important than that of sodium. The occurrence of ventricular fibrillation and other serious arrhythmias during coronary angiography or interventional procedures is uncommon. Numerous parameters play an obvious role in their occurrence, including site of injection, wedging of the catheter and subsequent occlusion of a vessel, myocardial ischemia, coronary anatomy, and type and duration of the procedure. In clinical practice, the role of contrast medium in the development of arrhythmia is, at most, limited (5). In conclusion, our results have shown that ioxaglate, an ionic low-osmolar contrast medium, led to a lower frequency of PVT than did iohexol, a nonionic contrast medium, when selectively injected into the right coronary artery of rabbits. Although ioxaglate injection resulted in prolongation of the repolarization period, this agent did
442
not cause a higher frequency of arrhythmia when injected during infusion of methoxamine, a compound known to reveal the proarrhythmic potential of certain agents such as clofilium, as confirmed in our study. The addition of 150 mmol/L sodium to isotonic iohexol completely abolished the occurrence of arrhythmia observed with the contrast agent alone. This confirms the favorable role of this cation in the prevention of arrhythmias during coronary procedures. ~,CKNOWLEDGMENT
The authors thank Fiona Feleton, MD, and Dominique Debize-Henderson, MSc, for reviewing the English version of this manuscript, and Huguette Lamoulie for her help in the preparation of the manuscript. EFERENCE 1. Morris TW. A review of coronary arteriography-and contrast media-induced ventricular fibrillation. Acta Radio11995; 36(suppl 399): 100-104. 2. Ajluni SC, Rizik DG, Grines CL. Medical and peripheral vascular complications following coronary interventional procedures. In: Freed M, Grines C, eds. Manual of interventional cardiology. Birmingham, Mich: Physician's Press, 1992; 247-266, 3. Surawicz B. Ventricular fibrillation. Am J Cardio11971; 28:268-287. 4. Hirshfeld JW. Radiographic contrast agents. In: Marcus ML, Schelbert HR, Skorton DJ, Wolf GL, Braunwald E, eds. Cardiac imaging. Philadelphia, Pa: Saunders, 1991; 162-181. 5. Armstrong SJ, Murphy KP, Wilde P, Hartnell GG. Ventricular fibrillation in coronary angiography: what is the role of contrast medium? Eur Heart J 1989; 10:892-895. 6. Pedersen HK, Jacobsen EA, Refsum H. Contrast media-induced ventricular fibrillation: an experimental study of the effects of dimeric contrast media during w e d g e d catheter injection in dogs. A c a d Radie11994; 1:136-144. 7. Pedersen HK, Jacobsen EA, Mortensen E, Refsum H. Contrast medium-induced ventricular fibrillation: arrhythmogenic mechanisms and the role of antiarrhythmic drugs in dogs, A c a d Radiol 1995; 2:1082-1088. 8. Carlsson L, Almgren O, Duker G. QTU-prolongation and torsades de pointes induced by putative class III antiarrhythmic agents in the rabbit: etiology and interventions, J Cardiovasc Pharmacol 1990; 16:276-285. 9. Buchanan LV, Kabell G, Brunden MN, Gibson JK. Comparative assessment of ibutilide, D-sotalol, clofilium, E-4031, and UK-68,798 in a rabbit model of proarrhythmia. J Cardiovasc Pharmacol 1993; 220:540-549. 10. Carlsson L, Amos J, Andersson B, Drews L, Duker G, Wadstedt G. Electrophysiological characterization of the prokinetic agents cisapride and mosapride in vivo and in vitro: implications for proarrhythmic potential? J Pharmacol Exp Ther 1997; 282:220227. 11. Bazett HC. An analysis of the time relations of electrocardiograms. Heart 1920; 7:353-370. 12. Sandoe E, Sigurd B. Arrhythmia: a guide to clinical electrocardielegy. Bingen, Germany: Publishing Partners Verlag, 1990. 13. Walker MJA, Curtis M J, Hearse DJ, et al. The Lambeth conventions: guidelines for the study of arrhythmias in ischemia, infarction, and reperfusion. Cardiovasc Res 1988; 22:447-455. 14. Donadieu AM, Hartl C, Cardinal A, Bonnemain B. Incidence of ventricular fibrillation during coronary arteriography in the rabbit: a comparative study of isotonic ioxaglate and iohexol. Acta Radio11987; 22:106-110.
15'. Weber E. Nonparametric methods. In: Delaunots AL, ed. Biostatistics in pharmacology, Oxford, England: Pergamon, 1973; 889969, 16. Baath L, AIm6n T, Reducing the risk of ventricular fibrillation by adding sodium to ionic and non-ionic contrast media with low iodine concentration: coronary perfusion of the isolated rabbit heart with meglumine diatrizoate or iopentol at 140 mgl/mL and 0-154 mmol Na+/L. Acta Radio11989; 30:207-212, 17, Manley BS, Chong EM, Walton C, Economides AP, C o b b e SM. An animal model for the chronic study of ventricular repolarisation and refractory period, Cardiovasc Res 1989; 23:16-20, 18, FIores NA, Davies RL, Penny W J, Sheridan DJ. Coronary microangiography in the guinea pig, rabbit and ferret. Int J Cardio11984; 6:459-471, 19. Barrett TD, MacLeod BA, Walker MJA. A model of myocardial ischemia for the simultaneous assessment of electrophysiological changes and arrhythmias in intact rabbits. J Pharmacol Toxicol Methods 1997; 37:27-36. 20. Lehmann KG, Chen YCJ. Reduction of ventricular arrhythmias by atropine during coronary arteriography. Am J Cardio11989; 63:447-451. 21, Hayes E, Pugsley MK, Penz WP, Adaikan G, Walker MJA, Relationship between QT and RR intervals in rats, guinea pigs, rabbits and primates. J Pharmacol Toxicol Methods 1994; 32:201-207. 22, Weissenburger J, Davy JM, Ch6zalviel F. Experimental models of torsades de pointes, Fundam Clin Pharmaco11993; 7:29-38, 23. Kopia GA, Eller BT, Patterson E, Shea M J, Lucchesi BR, Antiarrhythmic and electrophysiologic actions of clofilium in experimental canine models, Eur J Pharmaco11985; 116:49-61, 24. Stratmann HG, Kennedy HL. Torsades de pointes associated with drugs and toxins: recognition and management. Am Heart J 1987; 113:1470-1482. 25. Pardee HEB, An electrocardiographic sign of coronary artery obstruction. Ann Intern Med 1920; 26:244-257. 26. Gough WB, El-Sherif N. The differential response of normal and ischaemic Purkinje fibres to clofilium, D-sotalol and bretylium, Cardiovasc Res 1989; 23:554-559.
27, Morris TW, Hayakawa K, Sahler LG, Ekholm S. Incidence of fibrillation with isotonic contrast media for intra-arterial coronary digital subtraction angiography. Diagn Imaging Clin Med 1986; 55:109-113, 28. Morris TW. Ventricular fibrillation during right coronary arteriogra~ phy with ioxaglate, iohexol, and iopamidol in dogs. Invest Radiol 1988; 23:205-208. 29, Doucet D, Id6e JM, Donadieu AM, Bonnemain B. Hemodynamio and electrophysiologic effects of low osmolar contrast agents during coronary angiography in the dog and the rabbit. Ann Radio11989; 32:43-48. 30. Besjakov J, Chai CM, Baath L, AIm6n T, Comparison between oxygenated iohexol solution enriched with electrolytes and other low-osmotic contrast media: electrophysiologic and hemodynamic effects during perfusion of the left coronary artery in the pig, Invest Radio11994; 29(suppl 2):$241-248. 31, Paulin S, Gladstone S, Fossel ET. Cardiac side effects of different contrast agents: a comparative study in isolated perfused rat hearts, In: Felix R, Fischer HW, Kormano M, et al, eds. Contrast media from the past to the future. Stuttgart, Germany: Thieme, 1987; 85-96. 32, Fillette F, Nassif G. Electrophysiological effects of contrast media, Can J Cardio11987; 3(suppl A):I 1A-15A. 33. Jacobsen EA, KIow NE, Refsum H. Role of sodium addition to nonionic contrast medium in preventing ventricular fibrillation during coronary arteriography in dogs. A c a d Radio11994; 1:261269. 34, Id6e JM, Matalon C, Koeltz B, Bonnemain B, Lefevre T. Contrast medium-induced ventricular fibrillation: arrhythmogenic mechanisms and the role of antiarrhythmic drugs in dogs (letter). A c a d Radio11996; 3:781-783. 35, Pedersen HK, Jacobsen EA, Mortensen E, Refsum H. Authors" reply to letter to the editor. A c a d Radio11996; 3:783-785, 36. Ralston WH, Robbins MS, Mosier LD, Barco SJ, Adams MD, Hopkins RM The effect of sodium on the fibrillatory potential of ioversol, Invest Radio11988; 23(suppl 1):$140-$143. 37. Hayakawa K, Yamashita K. Low-osmolality contrast media-induced ventricular fibrillation, Invest Radio11989; 24:298-301.
443
Rationale and Objectives. The authors: (a) Compared the proarrhythmic effects of i0xaglate (!52 mmol/L sodium) and iohexol (no sodium) in a rabbit model and (b) as;sessed the effect of adding 150 mmot/L:sodium to iso:tonic iohexo!. : : M a t e r i a l s and Metlaods; Either ioxagtate:(320 mg of iodine:per milliliter) or iohexot (350 m g o f iodine per mitli:liter) was selectively injected into the right Coronary aritery (t.5 mL over 30 seconds) of 10 rabbits, some of which also received the %,adrenergic receptor agonist :::methoxamine.To validate the model;: the class IlI antiar:::rhythmic agent clofilium Was injected intravenously during methoxamine infusion. Frontal electrocardiography was performed continuously todetect polymorphic ventricular tachycardia (PVT): In a second study; the authors '::i assessed the frequency of arrhyttmaiaS:after hnjection of :::isotonic iohexot Solution (145 mg of iodine per miIlili: ter), either atone or with i50 mxr~olFLsodium. : : :Results. Methoxamine :significantly iengthened the QT, Q T c , and RR:intervalS (P < .05). The use of clofilium :: alone induced no PVT, whereas five of:eight methoxa: ~ne-infused rabbits developed PVT: after: Ciofilium injection (P = .03). Both contrast media prolonged the repolarization period. IohexoI alone induced a higher fiequency of PVT than did ioxaglate a!one (P = .0006). Methoxamine infusion did not Potentiate the frequency of p\,r-f in the ioxagiate-injected rabbits. The addition of Sodium to isotonic iohexol prevented: the occurrence of :PVT (P = :.0006). :: :.... :: :: :::Conclusion, Although ioxagiate prolonged :the repolari ization period, itdid not Causea higher frequency of ar::: :rhythmia When injected in association with: methoxa: mine. Iohexol, which contains no Sodium, induced a t'Agh :frequencyof arrhythrnia. The addition of a physiologic Concentration of sodium to isotonic iohexoI can prevent :::i:ventricuiax arrhythmias: :i ::: : :Key Words: Angiographyi e;ntrast :media; contrast media, comparative studies; heart, an'hythrrfia; iohexoI; ? ;loxagtate. i::i:: : :::: :::::i: : :::: :i
Iodinated contrast medium-induced arrhythmia may occur during coronary angiography or percutaneous transluminal coronary angioplasty (1,2). The most serious arrhythmia that may occur during such procedures is ventricular fibrillation, that is, chaotic asynchronous fractionated activity of the heart (3). Ventricular fibrillation generally occurs as a consequence of the inherent properties of the contrast medium, the patient's clinical status, and operator technique (4,5). The contrast medium-induced component of such arrhythmia is still unclear. Sodium concentration is generally considered to be an important contrast medium-related parameter (4). Pedersen et al (6) found a marked lengthening in monophasic action potential duration when ioxaglate, an ionic low-osmolar contrast medium containing 152 mmol/L sodium, was injected into the anterior descending branch of the left coronary artery during a simulated wedged-catheter situation in dogs. They found that ioxaglate caused a high frequency of ventricular fibrillation in their particular model. They suggested that excessive lengthening of ventricular repolarization induced by this contrast medium may be arrhythmogenic, as seen in patients with long QT syndrome and after using some class III antiarrhythmic drugs (6,7). They hypothesized that such excessive lengthening of ventricular repolarization (and the consecutive proarrhythmic risk) may be due to the high sodium concentration of the ioxaglate solution. To investigate this hypothesis, we compared the ef-
Acad Radio11998; 5:435-443 1From the Departments of Pharmacology (J.MI., C.B,B., V,G., C.M,), Biostatistics (P.Z.), and Clinical Research (S.G.), Laboratoire Guerbet, BP 50400, F-95943 Roissy-Charles-de-Gaulle Cedex, Aulnay-sous-Bois, France; and the Department of Interventional Cardiology, Institut Cardiovasculaire Paris-Sud, Massy, France (T.L.). Received June 2, 1997; revision requested August 1; revision received August 18; acc e p t e d September 20. Address reprint requests to J.M.I. © AUR, 1998
435
fects of ioxaglate with those of a nonionic contrast medium, iohexol (which contains no sodium), on a recent and validated rabbit model of arrhythmia. This model, which consists of animal sensitization and QT interval prolongation with methoxamine, revealed that some drugs, such as class III antiarrhythmic drugs or cisapride, are associated with polymorphic ventricular tachycardia (PVT), which results from prolonged cardiac repolarization (8-10). If the hypothesis of Pedersen et al (6) is correct, ioxaglate should induce a higher frequency of arrhythmia when associated with methoxamine than when injected alone. The same authors also found that adding 150 mmol/L sodium to a nonionic contrast medium led to ventricular fibrillation (6). To investigate this eventuality, we evaluated the electrocardiographic consequences of the addition of 150 mmol/L sodium to iohexol. To avoid any possible interference with osmolality, we used an isotonic iohexol solution.
Animal Preparation Seventy-seven male New Zealand white rabbits (Elevage Cunicole de Lantages, Sainte Savine, France; mean weight + standard error of the mean, 3.0 kg + 0.1) were anesthetized by means of intramuscular injection of 5.0 mg/kg midazolam (Hypnovel; Roche, Neuilly-surSeine, France) and 50 mg/kg ketamine (Imalgene; RhtneMtrieux, Lyon, France), and a tracheostomy was performed. From the femoral artery, a catheter (16 gauge; Vygon, Ecouen, France) was advanced into the aorta for blood pressure recordings. The marginal vein of each ear was catheterized to allow infusion of methoxamine and the validation agent when necessary. The left common carotid artery was isolated surgically and catheterized for right coronary injection via a preshaped 3-F polyethylene catheter (Guerbet Biomtdical, Louvres, France). A few seconds before the injection, the catheter was advanced selectively into the right coronary artery. The catheter was positioned and the injection performed under fluoroscopic control (Stenoscop; GE Medical Systems, Issy-les-Moulineaux, France). Immediately after the injection, the catheter was withdrawn to the aortic arch.
Electrophysiologic Measurements Frontal I and II and precordial V 3 leads were recorded at electrocardiography with an AT-3 recorder (Schiller,
436
Baar, Switzerland). Monitoring was continuous. The QT interval was corrected for heart rate changes by using the following equation (11): QTc = QT x ~](1,000/RR interval), where QTc, QT, and RR intervals are measured in milliseconds. PVT was defined as a series of rapid repetitive polymorphic QRS complexes with characteristic continuous undulation around the isoelectric line usually associated with a pause or slowing of heart rate and was considered to have occurred if four or more closely coupled repetitive extrasystoles with this particular pattern were observed (12,13).
Hemodynamic Measurements Aortic blood pressure was measured with a Spectramed Statham P 23 XL transducer (Gould Electronique, Ballainvilliers, France) and was continuously recorded and displayed on a Gould ES 1000 polygraph and a Gould V 1000 monitor, respectively. Continuous data acquisition and treatment were carried out by means of HEM software (Notocord, Croissy-sur-Seine, France). All experiments were carried out in compliance with the European Economic Community directive (86/609/ EEC) on animal welfare.
Test Solutions Sodium meglumine ioxaglate (Hexabrix; Laboratoire Guerbet, Aulnay-sous-Bois, France) and iohexol (Omnipaque; Nycomed, Oslo, Norway) were tested in their commercially available forms (ioxaglate: 320 mg of iodine per milliliter, osmolality at 37°C = 600 mOsm/kg; iohexol: 350 mg of iodine per milliliter, osmolality at 37°C = 880 mOsm/kg). In a second series of experiments, iohexol was diluted in distilled water to reach isotonicity (osmolality at 37°C = 308 mOsm/kg; iodine concentration, 145 mg of iodine per milliliter) and, in another test, 150 mmol/L sodium was added to this solution (osmolality at 37°C = 635 mOsm/kg).
Experimental Protocol After surgery, animals were allowed to equilibrate for 10 minutes before baseline measurements were obtained and the study was started. Time points of PVT occurrence and the duration of the arrhythmic episode were measured for all experiments. Validation of the experimental modeL--Rabbits (eight per group) were sensitized according to the technique used by Carlsson et al (8), which involves continuous intravenous infusion (10 gg/kg/min over 30 minutes) of methoxamine (Sigma, Saint Quentin-Fallavier, France)
vol 5, No 6, June 1998
I!! ii,¸.If¸ I
CONTRAST MEDIUM-INDUCED POLYMORPHIC VENTRICULAR TACHYCARDIA
-t ...... !...... !.......
.
.
.
Figure 1. E l e c t r o c a r d i o g r a m s h o w s t y p i c a l split T w a v e s ( a r r o w h e a d ) o c c u r ring a f t e r i n j e c t i o n of t h e class III a n t i a r r h y t h m i c d r u g clofilium in ra b bit s (V3 l e a d of t h e s u r f a c e e l e c t r o c a r d i o g r a m ) .
!
I I I
Vl
.!~~'"iiill..!ii~i
.
...............
I l
. . . . . . . ; ....................i...... t
,
~
[ ~.u. [mm,/n~ ~J~n! "ti
i
1
-
diluted in 5% glucose. The class III antiarrhythmic agent clofilium (Research Biochemicals International, Natick, Mass), which is used as a reference compound, was injected (10 mg/kg, intravenous route, bolus injection), either 15 minutes after the start of methoxamine infusion or, in rabbits that did not receive methoxamine, at random. Electrocardiographic tracings were recorded for 45 minutes after clofilium injection. Interaction of contrast medium with methoxamine.-After 15 minutes of methoxamine infusion, the test solution was selectively injected into the right coronary artery (1.5 mL over 30 seconds [14]). ECG tracings were monitored for 30 minutes after contrast medium injection, and episodes of PVT were recorded. Each group consisted of 10 (in the ioxaglate group) or 11 (in the iohexol group) rabbits. Effects of sodium addition to isotonic iohexol.--Rabbits were randomly assigned to receive intracoronary injection of either isotonic iohexol alone or in combination with 150 mmol/L sodium according to the above-mentioned technique. Statistical Analysis Measured values are expressed as the mean + standard error of the mean. For each study, the homogeneity of the groups in terms of preinjection values (blood pressure, heart rate, QT interval) was checked by means of analysis of variance. With respect to blood pressure, heart rate, and QTc interval, the values at various specified times following the injection of test solutions were compared with preinjection values by using the paired Student t test adjusted by the number of measurements (Bonferroni adjustment). Frequencies of PVT occurrence were compared by using the Fisher exact test (15). Statistical analysis of blood pressure and QT interval was conducted by using the Number Cruncher Statistical Systems program version 5.0 (NCSS, Kaysville, Utah), and analysis of PVT occurrence was performed by using StatXact program version
3.0 (Cytel Software, Cambridge, Mass). A P value of .05 or less was considered to indicate a statistically significant difference. RESULTS Interaction of Clofilium with M e t h o x a m i n e Before clofilium injection, methoxamine infusion significantly increased blood pressure (102 m m Hg _+ 2 vs 84 mm Hg + 3 at baseline, P < .001) and decreased heart rate (249 beats per minute _+ 10 vs 261 beats per minute + 10 at baseline, P < .01). Separation of the T wave of the surface electrocardiogram into two separate peaks ("TU complex") was frequently observed after injection of clofilium alone or with methoxamine infusion and, in some cases, before any injection (Fig 1). Clofilium alone induced no PVT in any of the eight treated rabbits, whereas five of the eight methoxamine-infused rabbits showed PVT after clofilium injection (P = .03). One methoxamine-infused rabbit had lethal ventricular fibrillation after clofilium injection. The QTc interval was lengthened in the clofilium alone and clofilium- and methoxamine-infused groups (Table 1). In some cases, the QT interval was not measured because of repolarization-dependent blocks that occurred at measurement time points (Fig 2).
Interaction of Contrast M e d i u m with Methoxamine PVT was not induced by the positioning of the catheter in any of the coronary angiography studies. Hemodynamic effects of contrast medium alone.-Ioxaglate alone transiently decreased blood pressure (baseline, 85 m m Hg + 2; end of injection, 68 mm Hg _+ 7; P = .02), as did iohexol alone (baseline, 79 mm Hg + 3; end of injection [n = 4], 49 mm Hg + 15 ; P = .16). Baseline values did not significantly differ between the two groups (P = .10).
437
Table I Clofilium-induced Changes in QTc Intervals with and without Methoxamine QTc Interval Duration (msec) Treatment Clofilium Clofilium a n d methoxamine
Baseline
During Injection
326 + 10 (n = 8) 346+8 (n = 8)
311 + 19 (n = 8) 411 + 20 (n = 8)
5 minutes after Injection 365 + 26 (n = 7) 415+10 (n = 6 )*
15 minutes after Injection 373 _+ 29 (n = 6) 4 3 2 + 27 (n : 3)
30 minutes after Injection 413 + 46 (n = 4) 4 2 0 + 14 (n = 3)
40 minutes after Injection 370 + 34 (n = 6) 414 (n = 2)
N o t e . - - D a t a are g i v e n as means + standard error of the m e a n . *P< .05.
Table 2 Methoxamine-induced Changes in QT, QTc, and RR Intervals before loxaglate and lohexol Injections QT Interval (msec)
Treatment Methoxamine a n d clofilium
Baseline
15 minutes after Injection
QTc Interval (msec)
PValue
Baseline
15 minutes after Injection
RR Interval (msec)
PValue
Baseline
15 minutes after Injection
PValue
155_+4
170+6
<.01
324+ 9
346+8
<.05
230+ 8
241 + 9
<.05
147+6
164+7
<.01
299+8
318+8
<.01
243+ 12
266+ 12
<.05
160+5
172+4
<.01
320+8
333+6
<.01
250+_4
267+4
<.01
(n = 8) Methoxamine and ioxaglate (n : 10) Methoxamine a n d iohexol (n = 10)
N o t e . - - D a t a are given as means + standard error of t h e m e a n .
Figure 2. E l e c t r o c a r d i o g r a m shows typical r e p o l a r i z a t i o n - d e p e n d e n t b l o c k o c curring after injection of t h e class III antiarrhythmic drug clofilium in rabbits.
I
r:tft Intracoronary injection of ioxaglate was associated with a decrease in heart rate (baseline, 237 beats per minute + 7; end of injection, 205 beats per minute + 14; P -- .04), as was iohexol alone (baseline, 231 beats per minute + 10; 15 seconds after injection [n = 6], 172 beats per minute + 23, P -- .02; end of injection [n = 4], 191 beats per minute + 22, P = .06). Baseline values did not differ significantly between the two groups (P = .63). Effects of contrast medium on QTc intervaL--The difference in baseline QTc interval values for the groups receiving ioxaglate and iohexol was not statistically signifi-
438
F1 I - I - ' V F i I V 1 q ilqg"~:
cant (P = .48). In the methoxamine-infused rabbits, both the QT and QTc intervals were significantly lengthened before contrast medium injection compared with that at baseline. The RR interval was also significantly lengthened by methoxamine for both groups (Table 2). Ioxaglate lengthened the duration of the QTc interval in animals, regardless of whether they also received methoxamine (although the duration was not significantly lengthened in the latter group). Such an effect was still noticeable 3 minutes after the start of the contrast medium injection (Table 3). Because of the high frequency
Table 3 Effects of loxaglate and lohexol Injection with and without Methoxamine on QTc Interval QTc Interval Duration (msec) Treatment
Baseline
Ioxaglate
314 + 13 (n = 10) 298 _+10 (n = 11) 331 + 9 (n-- 10) 342 + 9 (n = 10)
Iohexol Ioxaglate and methoxamine Iohexol a n d methoxamine
End of Injection
1 minute after Injection
3 minutes after Injection
351 + 20 (n = 9)* 323 (n = 2) 361 + 20 (n = 9) 340 (n = 1)
340 + 7 (n = 9) 318+ 11 (n = 6) 334 + 9 (n = 10) 348 + 12 (n = 6)
334 + 8 (n = 9) 318+ 12 (n = 6) 336 + 8 (n = 10) 361 + 7 (n = 7)
N o t e . - - D a t a are g i v e n as m e a n s + s t a n d a r d error of t h e m e a n . *P<.05.
......
I
t t-'
' ..... i --
F 'L
' .... "
...... ii-]l
"
of PVT episodes during iohexol injection, only a small number of tracings allowed measurement of the QTc interval duration in this group, and statistical analysis was not possible. Effects of methoxamine infusion.--Methoxamine increased blood pressure before injection of ioxaglate (104 mm Hg + 4 vs 77 mm Hg + 3, P < .001) and iohexol (103 riam Hg + 3 vs 83 mm Hg _+ 3, P < .01) and concomitantly reduced the heart rate (ioxaglate: 230 beats per minute _+ 12 vs 247 beats per minute _+ 13, P < .001; iohexol: 224 beats per minute _+ 3 vs 240 beats per minute _+4, P < .001). ECG effects of contrast medium.--Iohexol alone induced a significantly greater frequency of PVT than did ioxaglate alone (P = .0006). Repolarization-dependent blocks were frequently observed. A typical ischemic tracing, with Pardee waves, was associated with selective contrast medium injection. During the injection period, PVT occurred in all iohexol cases and three of the 10 ioxaglate cases (Fig 3), often in association with bradycardia. The combination of methoxamine and ioxaglate did not increase the frequency of PVT when compared with that with ioxaglate alone (Table 4). The frequency of PVT was higher in the group that received methoxamine and iohexol than in the group that received methoxamine
t
Figure 3.
E l e c t r o c a r d i o g r a m shows PVT e p i s o d e after injection of iohexol into t h e right c o r o n a r y artery of a rabbit, The staff of this arrhythmic e p i s o d e is p r e c e d e d b y t h e o c c u r r e n c e of P a r d e e w a v e s with ST s e g m e n t including t h e T w a v e , w h i c h is i n d i c a t i v e of ischemia,
and ioxaglate (P = .003). In five of the rabbits that received iohexol alone, PVT turned into lethal ventricular fibrillation. This was the case in three methoxamine-infused rabbits that received iohexol and in one rabbit that received ioxaglate.
Effects of sodium addition to isotonic iohexol.--The addition of 150 mmol/L sodium to isotonic iohexol (145 mg of iodine per milliliter) significantly reduced the frequency of PVT (P = .0006) (Table 5). The high frequency of PVT in the group receiving iohexol alone (145 mg of iodine per milliliter) did not allow us to evaluate changes in the QTc interval at the end of injection (Table 6). Aortic blood pressure was not measurable during the arrhythmic episodes (ie, in animals receiving isotonic iohexol alone).
Two hypotheses were tested in this study: (a) Contrast medium-induced prolongation of ventricular repolarization may favor the occurrence of ventricular arrhythmia during coronary angiography procedures, and (b) such an effect may be related to either the absence or a high (150 mmol/L) concentration of sodium in the contrast medium solution.
439
Table 4 ECG Effects of loxaglate and lohexol Injection with and without Methoxamine No. of Rabbits with PVT
Treatment I o x a g l a t e (n = 10) Methoxamine and ioxaglate (n = 10) Iohexol (n = 11) M e t h o x a m i n e a n d iohexol (n = 10)
No. of Rabbits with Lethal FV
Time of PVT O c c u r r e n c e (sec)*
Duration of PVT (sec)*
3
1
14 + 9
19 (n = 2)
3 11 10
0 5 3
26 + 4 19 + 4 15 + 4
25 _+ 16 (n = 3) 32 + 8 (n = 6) 29 + 8 (n = 7)
* D a t a are g i v e n as m e a n s _+ s t a n d a r d error of t h e m e a n ,
Table 5 ECG Effects of Isotonic Iohexol with and without the Addition of 150 mmol/L Sodium No. of Rabbits with PVT
Treatment Iohexol (n = 10) Iohexol a n d 150 m m o l / L sodium (n = 10)
No. of Rabbits with Lethal FV
9 0
Time of PVT O c c u r r e n c e (sec)*
1 0
8+2 NA
Duration of PVT (sec)* 37 + 5 NA
N o t e . - - I s o t o n i c iohexol c o n t a i n e d 145 m g of iodine p e r milINiter, * D a t a are g i v e n as m e a n s + s t a n d a r d error of t h e m e a n . NA : not a p p l i c a b l e .
Table 6 Effects of Isotonic Iohexol with and without the Addition of 150 mmol/L Sodium on QTc Interval i
QTc Interval Duration (msec) Treatment Iohexol Iohexol a n d 150 m m o l / L s o d i u m
Baseline 299 ± 13 ( n = 10) 328 ± 11 ( n = 10)
End of Injection
1 minute after Injection
327 (n = 1) 366 + 20 (n=9)
323 + 6 (n = 9) 332 _+7 ( n = 10)
3 minutes after Injection 329 + 8 (n : 9) 325 + 8 ( n = 10)
N o t e , - - D a t a are g i v e n as m e a n s + s t a n d a r d error of t h e m e a n . Isotonic iohexol c o n t a i n e d 145 m g of iodine p e r milliliter.
Experimental Model The rabbit has been widely used for both in vitro (16) and in vivo (8,9,14,17) electrophysiologic studies. This species has few coronary collateral vessels, unlike dogs (13,18). Furthermore, the cardiac electrophysiology of rabbits is reported to have depolarizing and repolarizing currents that are similar to those of humans (19). The selective injection into the right coronary artery was chosen because in rabbits (14), as in humans (20), ventricular arrhythmias seem to be more frequent after injection into the right coronary artery than after injection into the left coronary artery. In our study, a transient reduction in heart rate was observed after the injection of both low-osmolar contrast
440
media. This result is consistent with data reported from other rabbit studies (14) and clinical trials (4) and is probably related to depression of the proximal conduction system or to a direct effect on the sinoatrial node. In this study, measurement of the QT interval was required. This parameter represents a useful approximation of action potential in the ventricle and is, therefore, worth measuring (21). The QT interval varies with heart rate in rabbits just as it does in many other species. Under our conditions, contrast medium reduced the heart rate. We used Bazett's correction since it has been shown that this procedure is beneficial in rabbits and that the corrected QT distribution is much better than the uncorrected data (21).
In our study, as in the original study in which the model was described (8), rabbits were sensitized with methoxamine. This agent is a specific %-adrenergic agonist that leads to the development of after-depolarizations (8). As shown in Table 2, comparison of baseline values and preinjection values (ie, just before injection of contrast medium or clofilium) shows that methoxamine alone significantly increased the QTc interval in all test groups. Furthermore, the heart rate was also significantly lowered in methoxamine-infused groups before the injection of the test substances. This phenomenon is probably a reflex adjustment to the methoxamine-related increase in blood pressure. It is well known that bradycardia very often precedes the clinical occurrence of PVT (8,22). This is a further element suggesting the clinical relevance of our model. Clofilium is a pure class III antiarrhythmic agent that prolongs monophasic action potential duration by means of the blockade of potassium channels (9,23). Compounds from this class have been associated with severe arrhythmias, including PVT (24). In our study, there was a significantly higher frequency of PVT in rabbits infused with methoxamine and injected with clofilium than in those receiving clofilium only. Our data are consistent with those of Carlsson et al (8) and Buchanan et al (9). A statistically significant lengthening of the QTc interval was observed in the clofilium and clofilium and methoxamine-infused animals, as described elsewhere (8,9). Thus, under our conditions, clofilium was clearly shown to be proarrhythmic, a phenomenon associated with pro' tracted prolongation of the QTc interval. The occurrence of split T waves (or U waves) on the electrocardiographic tracings of clofilium-treated rabbits is consistent with findings reported in the literature (8). U waves are considered to be associated with early after-depolarizations (8). The appearance of the PVT tracings observed in our study matches that described elsewhere (8,22). In the case of contrast medium, an electrocardiographic tracing highly suggestive of ischemia occurred immediately after the catheter was positioned, with the ST segment including the T wave (Pardee waves) (25). It thus appears that, in this case, ischemia is a superimposed factor when compared with the tracings obtained after methoxamine and clofilium administration. It has been suggested that ischemia may facilitate the occurrence of PVT (26). It can thus be suggested that, in studies involving selective intracoronary injection of contrast medium,
ischemia may be an additional sensitization factor for the occurrence of PVT.
Interaction of Contrast Medium with Methoxamine The frequency of PVT was significantly lower with ioxaglate alone than with iohexol alone. A lower frequency of arrhythmia (ventricular fibrillation) for ioxaglate versus iohexol has already been shown in the same species (14). In our study, five of 11 cases of PVT in the iohexol group and one of three cases of PVT in the ioxaglate group turned into lethal ventricular fibrillation. In humans, PVT often ends spontaneously but sometimes degenerates into ventricular fibrillation (22). It has been claimed that ioxaglate (152 mmol of sodium per liter), which in some dog studies markedly lengthened monophasic action potential duration in contrast medium-perfused myocardium, may be proarrhythmic, similar to class III antiarrhythmic drugs, which also lengthen ventricular repolarization (6,7). We therefore tested ioxaglate in experimental conditions that allow detection of the arrhythmogenic potential of class III agents (ie, by injecting this contrast medium in methoxamine-infused rabbits). It is worth noting that, under these severe conditions, ioxaglate did not cause a higher frequency of major arrhythmias in animals whose QTc and RR intervals were markedly lengthened (Table 2), although it lengthened the repolarization period (Table 4). Therefore, our data do not support the hypothesis that an arrhythmogenic effect of this contrast medium may occur as a consequence of lengthening of ventricular repolarization. There is a strong body of experimental evidence from studies on dogs (27-29), pigs (30), isolated rat hearts (31), and isolated sheep myocardial fibers and Purkinje fibers (32) that shows ioxaglate is associated with a lower proarrhythmic risk than are nonionic low-osmolality contrast media. In a study in dogs, it was observed that iohexol, which caused the fewest regional electrophysiologic effects, paradoxically caused ventricular fibrillation in 100% of infusions (33). The clinical relevance of monophasic action potential duration increase in the particular case of contrast medium thus appears to be debatable. Injection of iohexol in methoxamineqnfused rabbits was associated with a similar frequency of PVT as that in the group receiving iohexol alone (ie, all animals showed PVT episodes in both groups). Given the maximum frequency of PVT induced with iohexol alone, a putative potentiation by methoxamine obviously could not be evidenced.
441
Effects of Sodium Addition to Isotonic Iohexol On the basis of their severe (34) dog model, Pedersen et al (6) also suggested that the addition of 150 mmol/L sodium to contrast medium solutions may be deleterious and lead to arrhythmias. In their response to a letter to the editor written by Id6e et al (34), Pedersen et al (35) indicated that either the absence of sodium or the presence of 150 mmol/L sodium in a contrast medium solution is proarrhythmogenic. To investigate this hypothesis, we compared the proarrhythmic potential of isotonic iohexol (without sodium) and iohexol and 150 mmol/L sodium (thus made hypertonic with an osmolality of 635 mOsm/kg). This concentration of sodium clearly protected the rabbits from the onset of PVT (Table 5). Our results, thus, are not fully consistent with the data of Pedersen et al (6). Results of several dog studies have confirmed the value of adding a physiologic concentration of sodium to nonionic contrast medium (27,36,37). Results of in vitro studies also confirmed the value of adding a physiologic concentration of sodium to contrast medium solutions. This was shown by using both isolated rabbit (16) and rat (31) heart models. The addition of 145-150 mmol/L sodium to a nonionic contrast medium solution, however, would lead to a substantial increase in the osmolality of the solution (ie, of 300 mOsm/kg) that could, in turn, have detrimental hemodynamic consequences. Sodium being a counterion in the case of the ionic low-osmolar contrast medium ioxaglate, the osmolality of the solution is kept to acceptable values. It is interesting that isotonicity per se did not abolish the l~ropensity of iohexol to cause arrhythmias. The relative role of this parameter thus appears to be less important than that of sodium. The occurrence of ventricular fibrillation and other serious arrhythmias during coronary angiography or interventional procedures is uncommon. Numerous parameters play an obvious role in their occurrence, including site of injection, wedging of the catheter and subsequent occlusion of a vessel, myocardial ischemia, coronary anatomy, and type and duration of the procedure. In clinical practice, the role of contrast medium in the development of arrhythmia is, at most, limited (5). In conclusion, our results have shown that ioxaglate, an ionic low-osmolar contrast medium, led to a lower frequency of PVT than did iohexol, a nonionic contrast medium, when selectively injected into the right coronary artery of rabbits. Although ioxaglate injection resulted in prolongation of the repolarization period, this agent did
442
not cause a higher frequency of arrhythmia when injected during infusion of methoxamine, a compound known to reveal the proarrhythmic potential of certain agents such as clofilium, as confirmed in our study. The addition of 150 mmol/L sodium to isotonic iohexol completely abolished the occurrence of arrhythmia observed with the contrast agent alone. This confirms the favorable role of this cation in the prevention of arrhythmias during coronary procedures. ~,CKNOWLEDGMENT
The authors thank Fiona Feleton, MD, and Dominique Debize-Henderson, MSc, for reviewing the English version of this manuscript, and Huguette Lamoulie for her help in the preparation of the manuscript. EFERENCE 1. Morris TW. A review of coronary arteriography-and contrast media-induced ventricular fibrillation. Acta Radio11995; 36(suppl 399): 100-104. 2. Ajluni SC, Rizik DG, Grines CL. Medical and peripheral vascular complications following coronary interventional procedures. In: Freed M, Grines C, eds. Manual of interventional cardiology. Birmingham, Mich: Physician's Press, 1992; 247-266, 3. Surawicz B. Ventricular fibrillation. Am J Cardio11971; 28:268-287. 4. Hirshfeld JW. Radiographic contrast agents. In: Marcus ML, Schelbert HR, Skorton DJ, Wolf GL, Braunwald E, eds. Cardiac imaging. Philadelphia, Pa: Saunders, 1991; 162-181. 5. Armstrong SJ, Murphy KP, Wilde P, Hartnell GG. Ventricular fibrillation in coronary angiography: what is the role of contrast medium? Eur Heart J 1989; 10:892-895. 6. Pedersen HK, Jacobsen EA, Refsum H. Contrast media-induced ventricular fibrillation: an experimental study of the effects of dimeric contrast media during w e d g e d catheter injection in dogs. A c a d Radie11994; 1:136-144. 7. Pedersen HK, Jacobsen EA, Mortensen E, Refsum H. Contrast medium-induced ventricular fibrillation: arrhythmogenic mechanisms and the role of antiarrhythmic drugs in dogs, A c a d Radiol 1995; 2:1082-1088. 8. Carlsson L, Almgren O, Duker G. QTU-prolongation and torsades de pointes induced by putative class III antiarrhythmic agents in the rabbit: etiology and interventions, J Cardiovasc Pharmacol 1990; 16:276-285. 9. Buchanan LV, Kabell G, Brunden MN, Gibson JK. Comparative assessment of ibutilide, D-sotalol, clofilium, E-4031, and UK-68,798 in a rabbit model of proarrhythmia. J Cardiovasc Pharmacol 1993; 220:540-549. 10. Carlsson L, Amos J, Andersson B, Drews L, Duker G, Wadstedt G. Electrophysiological characterization of the prokinetic agents cisapride and mosapride in vivo and in vitro: implications for proarrhythmic potential? J Pharmacol Exp Ther 1997; 282:220227. 11. Bazett HC. An analysis of the time relations of electrocardiograms. Heart 1920; 7:353-370. 12. Sandoe E, Sigurd B. Arrhythmia: a guide to clinical electrocardielegy. Bingen, Germany: Publishing Partners Verlag, 1990. 13. Walker MJA, Curtis M J, Hearse DJ, et al. The Lambeth conventions: guidelines for the study of arrhythmias in ischemia, infarction, and reperfusion. Cardiovasc Res 1988; 22:447-455. 14. Donadieu AM, Hartl C, Cardinal A, Bonnemain B. Incidence of ventricular fibrillation during coronary arteriography in the rabbit: a comparative study of isotonic ioxaglate and iohexol. Acta Radio11987; 22:106-110.
15'. Weber E. Nonparametric methods. In: Delaunots AL, ed. Biostatistics in pharmacology, Oxford, England: Pergamon, 1973; 889969, 16. Baath L, AIm6n T, Reducing the risk of ventricular fibrillation by adding sodium to ionic and non-ionic contrast media with low iodine concentration: coronary perfusion of the isolated rabbit heart with meglumine diatrizoate or iopentol at 140 mgl/mL and 0-154 mmol Na+/L. Acta Radio11989; 30:207-212, 17, Manley BS, Chong EM, Walton C, Economides AP, C o b b e SM. An animal model for the chronic study of ventricular repolarisation and refractory period, Cardiovasc Res 1989; 23:16-20, 18, FIores NA, Davies RL, Penny W J, Sheridan DJ. Coronary microangiography in the guinea pig, rabbit and ferret. Int J Cardio11984; 6:459-471, 19. Barrett TD, MacLeod BA, Walker MJA. A model of myocardial ischemia for the simultaneous assessment of electrophysiological changes and arrhythmias in intact rabbits. J Pharmacol Toxicol Methods 1997; 37:27-36. 20. Lehmann KG, Chen YCJ. Reduction of ventricular arrhythmias by atropine during coronary arteriography. Am J Cardio11989; 63:447-451. 21, Hayes E, Pugsley MK, Penz WP, Adaikan G, Walker MJA, Relationship between QT and RR intervals in rats, guinea pigs, rabbits and primates. J Pharmacol Toxicol Methods 1994; 32:201-207. 22, Weissenburger J, Davy JM, Ch6zalviel F. Experimental models of torsades de pointes, Fundam Clin Pharmaco11993; 7:29-38, 23. Kopia GA, Eller BT, Patterson E, Shea M J, Lucchesi BR, Antiarrhythmic and electrophysiologic actions of clofilium in experimental canine models, Eur J Pharmaco11985; 116:49-61, 24. Stratmann HG, Kennedy HL. Torsades de pointes associated with drugs and toxins: recognition and management. Am Heart J 1987; 113:1470-1482. 25. Pardee HEB, An electrocardiographic sign of coronary artery obstruction. Ann Intern Med 1920; 26:244-257. 26. Gough WB, El-Sherif N. The differential response of normal and ischaemic Purkinje fibres to clofilium, D-sotalol and bretylium, Cardiovasc Res 1989; 23:554-559.
27, Morris TW, Hayakawa K, Sahler LG, Ekholm S. Incidence of fibrillation with isotonic contrast media for intra-arterial coronary digital subtraction angiography. Diagn Imaging Clin Med 1986; 55:109-113, 28. Morris TW. Ventricular fibrillation during right coronary arteriogra~ phy with ioxaglate, iohexol, and iopamidol in dogs. Invest Radiol 1988; 23:205-208. 29, Doucet D, Id6e JM, Donadieu AM, Bonnemain B. Hemodynamio and electrophysiologic effects of low osmolar contrast agents during coronary angiography in the dog and the rabbit. Ann Radio11989; 32:43-48. 30. Besjakov J, Chai CM, Baath L, AIm6n T, Comparison between oxygenated iohexol solution enriched with electrolytes and other low-osmotic contrast media: electrophysiologic and hemodynamic effects during perfusion of the left coronary artery in the pig, Invest Radio11994; 29(suppl 2):$241-248. 31, Paulin S, Gladstone S, Fossel ET. Cardiac side effects of different contrast agents: a comparative study in isolated perfused rat hearts, In: Felix R, Fischer HW, Kormano M, et al, eds. Contrast media from the past to the future. Stuttgart, Germany: Thieme, 1987; 85-96. 32, Fillette F, Nassif G. Electrophysiological effects of contrast media, Can J Cardio11987; 3(suppl A):I 1A-15A. 33. Jacobsen EA, KIow NE, Refsum H. Role of sodium addition to nonionic contrast medium in preventing ventricular fibrillation during coronary arteriography in dogs. A c a d Radio11994; 1:261269. 34, Id6e JM, Matalon C, Koeltz B, Bonnemain B, Lefevre T. Contrast medium-induced ventricular fibrillation: arrhythmogenic mechanisms and the role of antiarrhythmic drugs in dogs (letter). A c a d Radio11996; 3:781-783. 35, Pedersen HK, Jacobsen EA, Mortensen E, Refsum H. Authors" reply to letter to the editor. A c a d Radio11996; 3:783-785, 36. Ralston WH, Robbins MS, Mosier LD, Barco SJ, Adams MD, Hopkins RM The effect of sodium on the fibrillatory potential of ioversol, Invest Radio11988; 23(suppl 1):$140-$143. 37. Hayakawa K, Yamashita K. Low-osmolality contrast media-induced ventricular fibrillation, Invest Radio11989; 24:298-301.
443