Abstract
OBJECTIVES
The anti-fibrillatory effect of potassium is well recognized from experimental models. There have, however, been very few clinical reports on the use of potassium to convert ventricular fibrillation (VF) after cardioplegic arrest.
METHODS
In total, 8465 adult patients undergoing cardiac operations on cardiopulmonary bypass (CPB) and with cold antegrade crystalloid cardioplegic arrest were consecutively enrolled in a database. Patients with VF after removal of the aortic clamp were given 20 mmol potassium, and if needed an extra 10 mmol, in the perfusion line and the conversion rate was registered. Preoperative and intraoperative factors possibly related to the occurrence of post-ischaemic VF were assessed.
RESULTS
Of these, 1721 (20%) patients had VF and 1366 of these (79%) were successfully treated with potassium infusion. Only 355 (21%) patients (4% of all operations) had direct-current countershock. The need for pacing was lower in the treatment group compared with the non-treatment group (P <0.001). Multivariate analysis revealed as the main findings that age, gender, amount of cardioplegia related to body mass index (BMI), and blood transfusion during the time of CPB had a highly significant (P <0.001) impact on reducing the rate of post-arrest VF. Somewhat contrary to expectation, left ventricular hypertrophy (LVH) was not a significant factor (P = 0.32) for post-arrest VF. No conversion by potassium was significant for age (P <0.001), gender (P <0.001) and LVH (P <0.001), but not for blood transfusion during CPB (P = 0.38) and for the ratio of cardioplegia–BMI (P = 0.26).
CONCLUSIONS
The results from this register study demonstrate that potassium infusion is an effective and convenient first-hand measure to convert post declamping VF on CPB.
Keywords: Cardiopulmonary bypass, Defibrillation, Potassium, Ventricular fibrillation
INTRODUCTION
Application of paddles into the pericardial sac with direct-current (DC) countershock is the traditional way of treating post-ischaemic ventricular fibrillation (VF) during on-pump cardiac surgery. It has been brought to our knowledge that in many heart surgery centres, VF conversion with administration of potassium in the perfusion line has been the first-hand alternative for several years. We also favour this option because it implies no interruption of the operative procedure and avoids the possibility of inducing conceivable minor myocardial necrosis and damage to constructed anastomosis and grafts [1, 2]. In the literature, the use of potassium to convert VF during cardiopulmonary bypass (CPB) has been seldom addressed. In 1984, Robicsek [3] reported on 12 patients who had reclamping before recardioplegia; and in 1995, Øvrum et al. [2] described the results from 100 patients who had potassium treatment comparable with the present study. Recently, 3 more patients with post-arrest VF storm who had potassium injection into the aortic root were described [4]. The paper was commented on and it was maintained that secondary cross-clamping and blood cardioplegia could serve just as well in this situation [5], as also reported by others [6]. Our study is much larger concerning the patient population where potassium conversion of post declamping VF is the routine. The purpose of the study was to describe the efficacy of potassium in converting post-arrest VF and to find possible significant differences between patients with and without post-ischaemic VF during CPB. The study is observational and not a comparative investigation between conversion by potassium vs electrical conversion, as the latter represents patients with failed conversion after primary treatment with potassium infusion.
MATERIALS AND METHODS
The study was approved by the institutional review board.
On January 1, 1999 a patient database was established at the Feiring Heart Clinic. Preoperative, intraoperative and postoperative data were consecutively and prospectively registered. In the present study, all patients who had cardiac operations in our institution from January 1999 to December 2006 were included and their relevant characteristics are shown in Table 1. Postoperative data are not included in this report.
Table 1:
Patient characteristics (n = 8465)
| Male (%) | 6352 (75.0) |
| Female (%) | 2113 (25.0) |
| Age years mean ± SD (range) | 66.9 ± 10.1 (29–91) |
| Body mass index mean ± SD (range) | 26.6 ± 3.9 (15–49) |
| EuroSCORE mean ± SD (range) | 4.7 ± 3.2 (0–24) |
| Elective surgery (%) | 6128 (72.4) |
| Ejection fraction mean ± SD (range) | 64.7 ± 13.9 (12–96) |
| Number of stenosed vessels mean ± SD | 2.5 ± 0.9 |
| Left main stenosis (%) | 1961 (23.2) |
| Heart failure treatment (%) | 1189 (14.0) |
| Hypertension (%) | 4243 (50.1) |
| Lung disease (%) | 1186 (14.0) |
| Smoking no/yes/former (%) | 3556 (42.0) 1849 (21.8) 3260 (38.5) |
| Diabetes (%) | 1369 (16.2) |
| Number myocardial infarction mean ± SD | 0.62 ± 0.8 |
| Sinus rhythm (%) | 7958 (94.0) |
| Atrial fibrillation (%) | 410 (4.8) |
| Pacemaker rhythm (%) | 53 (0.6) |
| Other rhythms (%) | 44 (0.5) |
| Left ventricular hypertrophy (%) | 1704 (20.1) |
| Pulmonary hypertension (%) | 334 (3.9) |
| Haemoglobin g/dl mean ± SD (range) | 14.1 ± 1.4 (7.3–20.2) |
| Creatinine µmol/l mean ± SD (range) | 95.9 ± 22.5 (25–343) |
| Coronary bypass graft surgery (%) | 6943 (82.0) |
| Aortic valve surgery (%) | 1352 (16.0) |
| Ascending aortic surgery (%) | 120 (1.4) |
| Mitral valve surgery (%) | 50 (0.6) |
| Redo surgery (%) | 344 (4.1) |
| CPB minutes mean ± SD (range) | 52.7 ± 24.1 (9–322) |
| Arrest time minutes mean ± SD (range) | 33.3 ± 17.8 (5–186) |
| Cardioplegia ml mean ± SD (range) | 956.6 ± 335.5 (300–3800) |
| Cardioplegia/body mass index mean | 35.9 |
| Betablocker (%) | 6712 (79.3) |
| Calcium antagonist (%) | 1858 (21.9) |
| ACE inhibitor (%) | 3096 (36.6) |
| Statin (%) | 6432 (76.0) |
| Diuretics (%) | 1601 (18.9) |
| Digitalis (%) | 294 (3.5) |
| Steroid (%) | 271 (3.2) |
ACE: angiotensin-converting enzyme; CPB: cardiopulmonary bypass; SD: standard deviation.
Left ventricular hypertrophy was diagnosed by electrocardiography and/or echocardiography.
Operation
Anaesthesia and operative procedures were standardized throughout the period. Anaesthesia was inducted with isoflurane, thiopental sodium, diazepam, fentanyl and pancuronium and prolonged with propofol infusion. Adequate blood pressure regulation was obtained with glycerolnitrate and ephedrine. The operations consisted of coronary bypass graft surgery, valvular procedures (replacement or plasties when appropriate), aortic surgery or combinations of these (Table 1). All operations were performed on CPB with non-pulsatile flow at moderate hypothermia (except a few aortic replacement procedures in circulatory arrest) and with cold antegrade crystalloid cardiac arrest. Retrograde administration of cardioplegia and blood cardioplegia were not used. Venting was routinely performed via a combined cardioplegic/venting line in the ascending aorta and by way of the aortic ostium during aortic valve procedures and the left atrial incision when mitral surgery was undertaken. The CPB circuit had a hard shell reservoir. Vacuum-assisted venous drainage was done in 1480 (17.5%) patients, and only 261 (3.1%) had ultrafiltration. The ventilator was deactivated during CPB. When judged necessary, blood as erythrocyte concentrate (SAG) was transfused. During CPB, the normative haemoglobin level for transfusion was <7 g/dl. Anti-arrhythmic drugs were not given before release of the aortic clamp.
Cardioplegia
We used a modified St. Thomas crystalloid solution containing per millilitre 0.8 mmol magnesium chloride, 0.8 mmol potassium chloride, 0.5 mmol procain hydrochloride with sterile water and sodium acid to pH 3.5–4.0, of which 20 ml was mixed with 1000 ml Ringer's acetate. This final solution was administered cold and antegradely through the cardioplegic line in the proximal aorta or directly into the left and right coronary ostium whenever the aorta was opened, and repeated every 20 min during the time of cross-clamping.
Cardioplegia–BMI unit
In our analysis, the amount of cardioplegia was related to the body mass index (BMI) by using a cardioplegia–BMI ratio in the calculations (cardioplegia/BMI).
Administration of potassium after aortic clamp removal
If VF was sustained after declamping of the aorta, 20 mmol (20 ml) of potassium chloride was given through the sampling line into the venous oxygenator reservoir and infused into the proximal aorta via the arterial line. If VF conversion did not occur, 10 mmol more potassium chloride was added. This implies that the potassium infusion was performed without any prior sophisticated calculations or consideration of potassium values in all patients with post-arrest VF because of simplicity. If not successful, DC conversion was carried out as a second-hand procedure and always with success. The typical response to potassium infusion was a few seconds of asystole followed by resumed supraventricular rhythm—mostly sinus.
Subgroup study of serum potassium
In the database, systemic serum potassium values were not registered. To overcome this limitation, 150 patients, 50 without post-arrest VF, 50 with VF and successful potassium-induced conversion and 50 with unsuccessful potassium conversion (DC converted), were randomly selected using a random number generator. Preoperative serum potassium and the last on-pump potassium value before aortic declamping were retrospectively obtained from laboratory files and perfusion records.
Statistical analysis
Continuous variables are depicted as mean ± standard deviation (SD) and tested for normal distribution by the Shapiro–Wilk test. If no deviation from normal distribution was observed, differences in means were tested by unpaired t-tests. Otherwise, the Mann–Whitney test was used. Fisher’s exact test was used for analysing categorical data, with the χ2 test only used when the permutations in Fisher's test exceeded the limit of the statistical package. In no case were both tests applied. In the retrospective subgroup study of serum potassium, the difference in measurements was assessed by repeated-measure analysis of variance. The logistic regression model was used in the multivariate analysis. All analyses were conducted with the use of STATA software version 10 (College Station, TX, USA).
RESULTS
Of the 8465 patients 1721 (20.3%) had post-arrest VF, and 1366 (79.1%) were successfully converted by potassium. Only 355 (20.9%) patients (4.2% of all patients) were treated by electrical defibrillation. In the conversion group, 87.9% were treated with 20 mmol potassium and 12.1% with 30 mmol potassium. No side-effects of the potassium infusion were registered. In particular, there were no observations of severe blood pressure elevation. Temporary use of a pacemaker was needed in 178 (10.3%) patients in the VF group and in 960 (14.2%) patients in the non-VF group (P <0.001). Table 2 shows several other differences between the post-arrest VF group and the non-VF patients.
Table 2:
Comparison between patients with and without post-arrest ventricular fibrillation
| Variable | VF (n = 1721) | Non-VF (n = 6744) | P-value |
|---|---|---|---|
| Gender female/male (%) | 269/1452 (15.6/84.4) | 1844/4900 (27.3/72.7) | <0.001 |
| Age (years) | 65.1 ± 10.1 (33–90) | 67.4 ± 10.1 (29–91) | 0.0001 |
| Body mass index | 27.5 ± 4.1 (17–46) | 26.5 ± 3.9 (15–49) | <0.0001 |
| EuroSCORE | 4.1 ± 3.0 (0–16) | 4.9 ± 3.3 (0–24) | 0.0001 |
| Urgency (%) | 486 (28.2) | 1851 (27.4) | 0.526 |
| Ejection fraction | 64.6 ± 14.0 (15–96) | 64.7 ± 13.8 (12–96) | 0.059 |
| Stenosed vessels | 2.57 ± 0.8 | 2.52 ± 0.8 | 0.001 |
| Left main stenosis (%) | 399 (23.2) | 1562 (23.1) | 1.0 |
| Heart failure treatment (%) | 234 (13.6) | 955 (14.2) | 0.56 |
| Hypertension (%) | 894 (51.9) | 3349 (49.7) | 0.094 |
| Lung disease (%) | 251 (14.6) | 935 (13.9) | 0.437 |
| Smoking no/yes/former (%) | 627/406/688 (36.4/23.6/40.0) | 2729/1443/2572 (40.5/21.4/38.1) | 0.007 |
| Diabetes (%) | 254 (14.8) | 1115 (16.5) | 0.078 |
| Myocardial infarction (%) | 804 (46.7) | 3242 (48.1) | 0.317 |
| Sinus rhythm (%) | 1638 (95.2) | 6320 (93.7) | 0.103 |
| Left ventricular hypertrophy (%) | 278 (16.2) | 1426 (21.1) | <0.001 |
| Pulmonary hypertension (%) | 59 (3.4) | 275 (4.1) | 0.238 |
| Cardiopulmonary bypass (min) | 49.4 ± 22.9 (11–276) | 53.6 ± 24.3 (9–322) | 0.001 |
| Arrest (min) | 29.9 ± 16.7 (5–186) | 34.2 ± 17.9 (5–186) | 0.001 |
| AVR(P) (%) | 175 (10.2) | 1177 (17.5) | <0.001 |
| Redo (%) | 82 (4.8) | 262 (3.9) | 0.101 |
| Cardioplegia–BMI | 32.9 | 36.7 | 0.001 |
| CPB SAG (%) | 211 (12.3) | 1140 (16.9) | <0.001 |
| Haemoglobin (g/dl) | 14.3 ± 1.3 (8.9–18.1) | 14.0 ± 1.3 (7.3–20.2) | 0.001 |
| Creatinine (µmol/l) | 94.5 ± 33.6 (43–231) | 96.2 ± 23.3 (25–343) | 0.129 |
| Beta blocker (%) | 1394 (81.0) | 5318 (78.9) | 0.049 |
| Calcium blocker (%) | 433 (25.2) | 1425 (21.1) | <0.001 |
| ACE inhibitor (%) | 644 (37.4) | 2452 (36.4) | 0.416 |
| Digitalis (%) | 38 (2.2) | 256 (3.8) | 0.001 |
| Statin (%) | 1330 (77.3) | 5102 (75.7) | 0.164 |
| Diuretics (%) | 305 (17.7) | 1296 (19.2) | 0.168 |
| Steroid (%) | 55 (3.2) | 216 (3.2) | 0.444 |
| Nitroglycerin (%) | 1342 (78.0) | 5052 (74.9) | 0.008 |
| Aspirin (%) | 1466 (85.2) | 5691 (84.4) | 0.433 |
| Clopidogrel (%) | 274 (15.9) | 1239 (18.4) | 0.047 |
| Warfarin (%) | 126 (7.3) | 510 (7.6) | 0.759 |
| LMWH (%) | 333 (19.3) | 1264 (18.7) | 0.581 |
| Mannitol (%) | 143 (8.3) | 834 (12.4) | 0.254 |
| Ultrafiltration (%) | 42 (2.4) | 219 (3.2) | 0.64 |
| VAVD (%) | 191 (11.1) | 1289 (19.1) | <0.001 |
| Need pacemaker (%) | 178 (10.3) | 960 (14.2) | <0.001 |
ACE: angiotensin-converting enzyme; AVR(P): aortic valve replacement(plasty); BMI: body mass index; CPB SAG: transfusion of packed red cells during CPB; LMWH: low-molecular-weight heparin; SAG: packed red cells; VAVD: vacuum-assisted venous drainage; VF: ventricular fibrillation.
Multivariate analysis
After multivariate analysis, age, gender, amount of cardioplegia–BMI and blood transfusion given strictly during CPB emerged as important factors for the occurrence of post-arrest VF. Left ventricular hypertrophy (LVH) was not a significant factor, nor was aortic clamp time. No conversion by potassium was, however, significant for LVH, and also for age, gender, but not for blood transfusion during CPB and cardioplegia–BMI (Table 3).
Table 3:
Multivariate analysis
| Variable | Ventricular fibrillation |
No conversion by potassium infusion |
||
|---|---|---|---|---|
| Odds ratio (95% confidence interval, 95% CI) | P-value | Odds ratio (95% CI) | P-value | |
| Age/5 years | 0.94 (0.92–0.98) | <0.001 | 0.90 (0.85–0.95) | <0.001 |
| Male gender | 1.72 (1.48–2.00) | <0.001 | 2.57 (1.80–3.67) | <0.001 |
| Cardioplegia–BMI (ml/m2/kg) | 0.98 (0.974–0.985) | <0.001 | 0.99 (0.99–1.00) | 0.26 |
| Blood transfusion (yes/no) | 0.65 (0.53–0.81) | <0.001 | 0.83 (0.54–1.27) | 0.38 |
| Left ventricular hypertrophy | 1.08 (0.93–1.27) | 0.32 | 1.68 (1.28–2.21) | <0.001 |
Age
Figure 1 shows the age distribution related to the number of post-arrest VF and the potassium-induced conversion rate. The post-arrest VF was significantly lower in the older age group.
Figure 1:
Age distribution of patients and the rate of post-arrest ventricular fibrillation and potassium-induced conversion.
Gender
Regarding gender, it was found that the rate of post-arrest VF was 12.7% for females and 22.9% for males, and the respective potassium conversion rate was 85.6and 77.8%. In the female group, 90.5% converted with 20 mmol potassium and 9.5% after an extra dose of 10 mmol potassium. For males, the percentages were 85.3 and 14.7. For females, the VF rate was lower than for males all through the age span and the potassium conversion rate was higher except for the age group 46–55 years (Fig. 2 and Table 3). The differences between the genders remained when corrected for BMI in the logistic regression.
Figure 2:
Gender and the rate of ventricular fibrillation and potassium-induced conversion.
Cardioplegia
Figure 3 illustrates the relationship between the amount of cardioplegia–BMI and the occurrence of post-arrest VF and the rate of conversion by potassium. Lower cardioplegi–BMI gave a significantly higher frequency of post-arrest VF (Table 3). The only exception to this was male patients with BMI under 20 and over 40, but they represent only 1.5% of all males.
Figure 3:
Cardioplegia–BMI ratio and the rate of ventricular fibrillation and potassium-induced conversion.
Blood transfusion during cardiopulmonary bypass
Blood transfusion during CPB reduced VF considerably (Table 3). Without transfusion, 22.1% had VF, decreasing to 10.2% for those with transfusion. For females and males, the percentages were, respectively, 15.3 and 8.7 and 23.6 and 12.8. Overall, blood transfusion reduced the risk of post-arrest VF by 31.6%. The potassium-induced conversion rate was 79.1% for patients with no blood transfusions and 77.7% for those with transfusions. These respective percentages were 85.9 and 84.7 for females and 78.1 and 69.0 for males. In the female group, 38.5% received transfusion during CPB compared with 7.1% for males. There was a linear increase in blood transfusion with rising age.
Aortic cross-clamp time
Cross-clamp time was an apparent significant predictor for post-arrest VF, but not when confounders were included. The rate of VF was higher in patients with very short clamp times (<15 min).
Left ventricular hypertrophy
Figure 4 demonstrates that the VF rate and conversion rate in male and female patients with LVH were lower when compared with patients diagnosed with no LVH (Table 3).
Figure 4:
Left ventricular hypertrophy and ventricular fibrillation rate and potassium-induced conversion in male and female patients.
Serum potassium
The results from the retrospective subgroup study of serum potassium in blood tests taken preoperatively and during CPB before removal of the aortic cross-clamp are shown in Table 4. There were no significant intergroup differences in potassium values at the two time-points.
Table 4:
Potassium values randomly sampled in 50 patients without post-arrest ventricular fibrillation [VF] (Group A), 50 patients with post-arrest VF successfully converted by potassium (Group B) and 50 patients with post-arrest VF with no conversion after potassium infusion (Group C)
| Group A | Group B | Group C | P-value | |
|---|---|---|---|---|
| Preoperative potassium mean ± SD | 3.74 ± 0.32 | 3.72 ± 0.32 | 3.69 ± 0.33 | 0.133 |
| Pre declamping potassium mean ± SD | 4.38 ± 0.44 | 4.33 ± 0.44 | 4.38 ± 0.47 | 0.30 |
DISCUSSION
The ability of potassium to convert VF is not a new observation. Already in 1961, it was demonstrated that intracardiac injection of potassium might induce cardioversion in a clinical resuscitation setting [7]. In 1984, Robicsek [3] reported a series of 12 patients having persistent post-arrest VF treated with reclamping and reinstitution of cardioplegia with subsequent cardiac arrest and pacing in the reperfusion phase. Eleven years later, there was an article describing 100 patients who had potassium treatment nearly the same way as in our study [2]. One study on three patients with VF storm and potassium injection in the aortic root was published in 2011 [4]. In experimental models, the anti-fibrillatory effect of hyperkalaemia on sustained VF has been demonstrated years ago [8]. The exact mechanism of action has remained unclear, but a recent report gives new insight into the electrophysiological effects of hyperkalaemia on VF dynamics, including organization and termination beyond the universal principle of depolarization [9]. Recently, there was also a study on the use of systemic hyperkalaemia to convert intractable VF during CPB in a porcine model [10]. It was reported that potassium can offset VF unresponsive to electrical defibrillation due to prolonged (15 min) cardiac arrest. There have been few published observational studies on the topic. To our knowledge, the present study represents the largest so far.
In our study, 20.3% of the patients had post-ischaemic VF and 79.9% of these had successful VF conversion by infusion of potassium into the perfusion line. The conversion rate is similar to that of the previous report [2]. We also agree that the advantage of potassium infusion compared with the somewhat more cumbersome conventional defibrillation is that the operative procedure can proceed without interruption, and there are definitely no risks of injury to grafts by paddles and minor myocardial necrosis due to the electrical current [1]. No side-effects of the potassium treatment were observed. The need for transient pacing was not higher in the potassium-treated group compared with the rest of the patients.
By adding 10 mmol extra potassium, 12.1% more patients were converted. It might be that even more potassium could have implied a yet higher cardioversion rate, hypothetically 100%. We believe that the option should be tested. In a porcine model with intractable VF after simulating an un-witnessed cardiac arrest of 15 min, induction of hyperkalaemia converted all [10]. This was described as an exciting and ‘disruptive concept’ [11].
Potassium is essential in most cardioplegic solutions to keep the heart arrested [12]. Systemic hyperkalaemia has also been used innovatively to induce cardiac standstill during valve surgery in patients with the patent internal mammary graft [13]. In our register, serum potassium values were not included as a parameter. To overcome this limitation, a retrospective registration of potassium measurements performed preoperatively and on-pump before release of the aortic cross-clamp was done in a random sample of 150 patients, revealing no significant differences between patients with and without post-arrest VF and between the patients with successful and non-successful potassium-induced conversion (Table 4).
The exact mechanism of VF following removal of the aortic clamp is not completely understood. The aetiology is without doubt multi-factorial. Poor myocardial protection, transient electrolyte and acid disturbances and gradients, and conduction abnormalities among others, play a role.
We found that age, gender, amount of cardioplegia (also when related to BMI) and blood transfusion (packed red cells, SAG) given strictly during CPB were statistically related to the occurrence of reperfusion VF. Older patients had a lower post-arrest VF rate than younger ones. This is in some contrast to experimental animal studies showing ischaemic intolerance in the aged heart [14, 15]. However, in animal studies, improvement has also been reported in this respect with senescence [16, 17]. In a clinical setting, it can be speculated whether the phenomenon of preconditioning is operating [18, 19]. The older patients in the present study might have been exposed to more repeated episodes of myocardial ischaemia and thus have become more tolerant to cardioplegic ischaemia manifested by a lower rate of post-cardioplegic reperfusion VF. However, this remains as a pure speculation as there are no data substantiating the theory.
We found that females had a lower occurrence of post-arrest VF than males. Sex differences have also been demonstrated in experimental studies using Langendorff perfusion [14, 15]. These studies are in agreement with clinical reports showing that women have worse outcomes after cardiac surgery [20, 21]. The publications showing inferior outcomes for women have mostly been on patients undergoing cardiac surgery prior to year 2000. More recent reports challenge these previous results, demonstrating a clear trend to better prognosis for female patients [22, 23].
It was reassuring to document that more cardioplegia, also when related to BMI, gave a seemingly better cardiac protection by lowering the post-arrest VF rate. It raises the question whether some of our patients had inadequate cardioplegic protection by receiving less-than-required cardioplegia resulting in post declamping VF. If the issue is an adequate amount of cardioplegia, the problem can be solved. The message is, as learned from long experience, that good protection of the myocardium is of paramount importance.
A quite surprising and novel observation was that transfusion of erythrocyte concentrate (SAG) during the time of CBP reduced the VF rate considerably and independently. The exact time-point of transfusion during CPB was not registered in the database (solely that the transfusions were performed during the time of CPB). The notion is that most of the transfusions were performed prior to aortic clamp removal because of dilution with low haematocrit during extracorporeal circulation. SAG-blood has containments that might influence reperfusion arrhythmias. It is known that the potassium content increases with storage time. If such blood is given close to the declamping time, then it will be understandable that the potassium load might suppress the occurrence of malignant reperfusion arrhythmias. This possibility was, to some degree, eliminated by comparing potassium values from patients with and without VF and finding that there were no significant differences. The calcium homeostasis is also affected by blood transfusion and might have a role [24]. Other blood containments can also influence the reperfusion phase. A more detailed investigation of the observed blood transfusion effect is warranted.
LVH was not a predictor for post-arrest VF, but was for the potassium-induced conversion rate. This is not quite as expected [3–5]. LVH was diagnosed by electrocardiography and/or echocardiography. It cannot be totally ruled out that some patients were not correctly diagnosed.
Cross-clamping time should be as short as possible. Paradoxically, we found that the shortest ischaemic times had the highest VF rates. These patients mostly had one- and two-vessel disease with expected rapid constructions of the peripheral anastomoses and with a tendency to relatively less cardioplegia. Short arrest time also implies that the cardioplegic solution might not fully concur its effects.
We used cold crystalloid cardioplegia in every patient. It remains open whether the findings can be transferred to surgical units, which prefer to use blood cardioplegia. Yet, in a recent randomized study between cold blood and cold crystalloid cardioplegia in aortic valve patients, the occurrence of post-arrest VF was not significantly different [25].
It is concluded that infusion of potassium to a high extent converted post-declamping VF. We believe that the study establishes the practice and advice it as a routine measure instead of first-hand electroconversion by paddles.
Conflict of interest: none declared.
REFERENCES
- 1.Runsjö M, Källner A, Källner G, Rosenquist M, Bergfeldt L. Myocardial injury after electrical therapy for cardiac arrhythmias assessed by troponin-T release. Am J Cardiol. 1997;79:1241–5. doi: 10.1016/s0002-9149(97)00090-8. [DOI] [PubMed] [Google Scholar]
- 2.Øvrum E, Tangen G, Holen EÅ, Ringdal M-AL, Istad R. Conversion of postischemic ventricular fibrillation with intraaortic infusion of potassium chloride. Ann Thorac Surg. 1995;60:156–9. doi: 10.1016/s0003-4975(95)00327-4. [DOI] [PubMed] [Google Scholar]
- 3.Robicsek F. Biochemical termination of sustained fibrillation occurring after artificially induced ischemic arrest. J Thorac Cardiovasc Surg. 1984;87:143–5. [PubMed] [Google Scholar]
- 4.Watanabe G, Noriyoshi Y, Shigeyuki T, Yamaguchi S. Potassium-induced cardiac resetting technique for persistent ventricular tachycardia and fibrillation after aortic declamping. Ann Thorac Surg. 2011;91:619–20. doi: 10.1016/j.athoracsur.2010.07.075. [DOI] [PubMed] [Google Scholar]
- 5.Turkoz R, Özker E, Turkoz A. Secondary cross-clamping and blood cardioplegia for refractory ventricular fibrillation. Ann Thorac Surg. 2011;92:403. doi: 10.1016/j.athoracsur.2011.03.077. [DOI] [PubMed] [Google Scholar]
- 6.Gadhinglajkar SV, Sreedhar R, Varma PK. Controlled aortic root perfusion: a novel method to treat refractory arrhythmias after aortic valve replacement. J Cardiothorac Vasc Anesth. 2004;18:197–200. doi: 10.1053/j.jvca.2004.01.009. [DOI] [PubMed] [Google Scholar]
- 7.Weinstock L, Clark JH. Successful treatment of ventricular fibrillation with intracardiac potassium chloride. Am J Cardiol. 1961;7:742–5. doi: 10.1016/0002-9149(61)90462-3. [DOI] [PubMed] [Google Scholar]
- 8.Koller ML, Riccio ML, Gilmour RF., Jr Effects of (K) on electrical restitution and activation dynamics during ventricular fibrillation. Am J Physiol Heart Circ Physiol. 2000;279:H2665–72. doi: 10.1152/ajpheart.2000.279.6.H2665. [DOI] [PubMed] [Google Scholar]
- 9.Pandit SV, Warren M, Mironov S, Tolkacheva EG, Kalifa J, Berenfeld O, et al. Mechanisms underlying the antifibrillatory action of hyperkalemia in guinea pig hearts. Biophys J. 2010;98:2091–101. doi: 10.1016/j.bpj.2010.02.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Liakopoulos OJ, Allen BS, Buckberg GD, Hristov N, Tan Z, Villablanca JP, et al. Resuscitation after prolonged cardiac arrest: role of cardiopulmonary bypass and systemic hyperkalemia. Ann Thorac Surg. 2010;89:1972–80. doi: 10.1016/j.athoracsur.2010.02.052. [DOI] [PubMed] [Google Scholar]
- 11.Conte J. Invited commentary. Ann Thorac Surg. 2010;89:1979–80. doi: 10.1016/j.athoracsur.2010.04.005. [DOI] [PubMed] [Google Scholar]
- 12.Vinten-Johansen J, Thourani VH. Myocardial protection: an overview. J Extra Corpor Technol. 2000;32:38–48. [PubMed] [Google Scholar]
- 13.Fujita T, Kobayashi J, Nakajima H, Toda K. Systemic hyperkalemia and mild hypothermia for valve surgery in patients with patent internal mammary artery graft. Interact CardioVasc Thorac Surg. 2010;11:3–5. doi: 10.1510/icvts.2010.233262. [DOI] [PubMed] [Google Scholar]
- 14.McCully JD, Toyoda Y, Wakiyama H, Rousou AJ, Parker RA, Levitsky S. Age- and gender-related differences in ischemia/reperfusion injury and cardioprotection: effects of diazoxide. Ann Thorac Surg. 2006;82:117–23. doi: 10.1016/j.athoracsur.2006.03.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.McCully JD, Rousou AJ, Parker RA, Levitsky S. Age- and gender-related differences in mitochondrial oxygen consumption and calcium with cardioplegia and diazoxide. Ann Thorac Surg. 2007;83:1102–9. doi: 10.1016/j.athoracsur.2006.10.059. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Boucher F, Tanguy S, Toufektsian MC, Besse S, Tressallet N, Favier A. Age-dependent changes in myocardial susceptibility to zero flow ischemia and reperfusion in isolated perfused rat hearts:relation to antioxidant status. Mech Ageing Dev. 1998;103:301–16. doi: 10.1016/s0047-6374(98)00050-5. [DOI] [PubMed] [Google Scholar]
- 17.Willems L, Zatta A, Holmgren K, Ashton KJ, Headrick JP. Age-related changes in ischemic tolerance in male and female mouse hearts. J Moll Cell Cardiol. 2005;38:245–56. doi: 10.1016/j.yjmcc.2004.09.014. [DOI] [PubMed] [Google Scholar]
- 18.Murray CE, Jennings RB, Reimer KA. Preconditioning with ischemia: a delay of lethal cell injury in ischemic myocardium. Circulation. 1986;74:1124–36. doi: 10.1161/01.cir.74.5.1124. [DOI] [PubMed] [Google Scholar]
- 19.Yellon DM, Alkulaif AM, Pugsley WB. Preconditioning the human myocardium. Lancet. 1993;342:276–7. doi: 10.1016/0140-6736(93)91819-8. [DOI] [PubMed] [Google Scholar]
- 20.Edwards FH, Carey JS, Grover FL, Bero JW, Hartz RS. Impact of gender on coronary bypass operative mortality. Ann Thorac Surg. 1998;66:125–31. doi: 10.1016/s0003-4975(98)00358-0. [DOI] [PubMed] [Google Scholar]
- 21.Abramov D, Tamariz MG, Sever JY, Christakis GT, Bhatnager G, Heenan AL, et al. The influence of gender on the outcome of coronary artery bypass surgery. Ann Thorac Surg. 2000;70:800–5. doi: 10.1016/s0003-4975(00)01563-0. [DOI] [PubMed] [Google Scholar]
- 22.Guru V, Fremes SE, Austin PC, Blackstone EH, Tu JV. Gender difference in outcome after hospital discharge from coronary artery bypass grafting. Circulation. 2006;113:507–16. doi: 10.1161/CIRCULATIONAHA.105.576652. [DOI] [PubMed] [Google Scholar]
- 23.Mølstad P. Coronary heart disease in women: less extensive disease and improved long-term survival compared to men. Scand Cardiovasc J. 2009;43:10–6. doi: 10.1080/14017430802455833. [DOI] [PubMed] [Google Scholar]
- 24.Aguilera IM, Vaughan RS. Calcium and the anaesthetist. Anaesthesia. 2000;55:779–90. doi: 10.1046/j.1365-2044.2000.01540.x. [DOI] [PubMed] [Google Scholar]
- 25.Øvrum E, Tangen G, Tølløfsrud S, Øystese R, Ringdal MA, Istad R. Cold blood versus cold crystalloid cardioplegia: a prospective randomised study of 345 aortic valve patients. Eur J Cardiothorac Surg. 2010;38:745–9. doi: 10.1016/j.ejcts.2010.03.052. [DOI] [PubMed] [Google Scholar]