- Email: [email protected]
Permanent Pacemaker Implantation After Isolated Aortic Valve Replacement: Incidence, Indications, and Predictors
CARDIOVASCULAR
Permanent Pacemaker Implantation After Isolated Aortic Valve Replacement: Incidence, Indications, and Predictors Sam Dawkins, MB BS, Alex R. Hobson, MB BS, Paul R. Kalra, MD, Augustine T. M. Tang, DM, FRCS Ed (C-Th), James L. Monro, FRCS, and Keith D. Dawkins, MD, FRCP Wessex Cardiac Unit, Southampton University Hospital, Southampton, United Kingdom
Background. Conducting system defects are common in patients with aortic valve disease. Aortic valve replacement may result in further conduction abnormalities and necessitate permanent pacemaker implantation (PPM). We sought to identify the contemporary incidence and predictors for early postoperative PPM in patients undergoing isolated aortic valve replacement. Methods. Data were analyzed from 354 consecutive patients undergoing isolated aortic valve replacement at a referral cardiac unit during a 30-month period; data were unavailable on 4 patients and a further 8 had undergone preoperative PPM. Results for the remaining 342 patients (97%; mean age, 67 ⴞ 14 years), of whom 212 were males, are presented. The major indications for aortic valve replacement were valvular stenosis (n ⴝ 224), regurgitation (n ⴝ 70), or infective endocarditis (n ⴝ 25). Preoperative conducting system disease was present in 26% of patients.
Results. In-hospital mortality was 1.8% (6 of 342 patients). Postoperatively 29 patients (8.5%) required early PPM, of which 26 were during the index admission. Patients with preoperative conducting system disease (16% versus 6%; p ⴝ 0.004) and valvular regurgitation (16% versus 7%; p ⴝ 0.01) were more likely to require PPM as opposed to those without. Preoperative conducting system disease was the only independent predictor of PPM (p < 0.01); the relative risk of PPM requirement in this group was 2.88 (95% confidence interval, 1.31 to 6.33). Conclusions. Permanent pacemaker implantation requirement after aortic valve replacement is a common occurrence, and should be discussed as part of the preoperative consent process. Preexisting conducting disease and preoperative aortic regurgitation were predictors of PPM requirement. (Ann Thorac Surg 2008;85:108 –12) © 2008 by The Society of Thoracic Surgeons
A
The most recent series (reported in 2003) is a prospective cohort study performed in 276 patients (mean age, 57.5 ⫾ 14 years) undergoing AVR in a single center [8]. Patients were included irrespective of whether the primary indication for surgery was severe aortic valve disease or the need for revascularization with concomitant aortic valve disease. Permanent pacemaker implantation was indicated in 3.2% of patients. Identifying patients at increased risk of developing conduction system abnormalities that are likely to require postoperative PPM would be of substantial clinical benefit, facilitating the planning of postoperative care. It also has important implications in the context of informed preoperative consent. The aim of the current study was to establish the contemporary incidence of early postoperative PPM requirement in patients undergoing isolated AVR. In addition, we sought to identify clinical criteria that might predict the need for early postoperative PPM.
ortic valvular stenosis and regurgitation are associated with abnormalities of conduction, including higher degrees of atrioventricular (AV) block [1, 2]. Aortic valve replacement (AVR) can result in the development of further conduction abnormalities [3], which may be associated with an increased risk of sudden death [4]. Higher degrees of AV block, although often reversible, may also necessitate permanent pacemaker implantation (PPM) [5]. Studies conducted in the late 1970s suggested that early PPM was required in approximately 6% of patients undergoing isolated AVR [6]. Advances in cardiac surgical techniques and improved availability have resulted in surgery being offered to a wider range of patients. Elderly patients, an increasing proportion of the adult population, gain substantial benefits from cardiac surgery [7], and this has resulted in a significant change in the demographics of patients undergoing AVR. Caution is therefore required when extrapolating data from older case series.
Accepted for publication Aug 13, 2007. Address correspondence to Dr Dawkins, Wessex Cardiac Unit, Southampton University Hospital, Southampton SO16 6YD, United Kingdom; e-mail: [email protected].
© 2008 by The Society of Thoracic Surgeons Published by Elsevier Inc
Patients and Methods Patients Data were analyzed from 354 consecutive patients undergoing isolated AVR at the Wessex Cardiac Unit (a referral 0003-4975/08/$34.00 doi:10.1016/j.athoracsur.2007.08.024
DAWKINS ET AL PERMANENT PACEMAKER IMPLANTATION AFTER AVR
109
Table 1. Characteristics of 342 Patients Undergoing Isolated Aortic Valve Replacement All Patients (n ⫽ 342) PPM Subset (n ⫽ 29) No PPM (n ⫽ 313) p Value (PPM vs. No PPM)
Characteristic Age (y) Sex (male) Conduction abnormalities Indications for AVR AS AR Mixed Endocarditis Others Elective/emergency Bioprosthetic/mechanical Prosthetic AV diameter (mm) Cross-clamp time (min) Bypass time (min) Lowest operative temperature (°C) AR ⫽ aortic regurgitation; permanent pacemaker.
67 ⫾ 14 212 (62%) 89 (20%)
66 ⫾ 13 19 (65%) 14 (48%)
193 (62%) 75 (22%)
NS NS 0.004
224 (66%) 70 (20%) 20 (6%) 25 (7%) 3 (1%) 283/80 195/147
16 (55%) 11 (38%) 0 (0%) 2 (7%) 0 (0%) 25/4 16/13 24 ⫾ 4 54 ⫾ 23 72 ⫾ 34 30.6 ⫾ 1.9
208 (61%) 59 (17%) 20 (6%) 23 (7%) 3 (1%) 258/76 179/134 23 ⫾ 3 51 ⫾ 20 66 ⫾ 25 31.2 ⫾ 2.5
NS 0.01 NS NS NS NS NS 0.03 NS NS NS
AS ⫽ aortic stenosis;
AV ⫽ aortic valve;
cardiac center) during a 30-month period from January 2000 to July 2002. Patients were identified from a customdesigned cardiac surgical database (based on FoxPro), which is maintained by a dedicated data clerk. Eight patients had undergone preoperative PPM, and data were not available in a further 4 patients. The remaining 342 patients (mean age, 67 ⫾ 14 years [⫾ standard deviation]) formed the study population; 212 (62%) were males. Baseline demographic details are shown in Table 1. The major indications for AVR were predominant valvular stenosis (n ⫽ 224, 66%), predominant regurgitation (n ⫽ 70, 20%), or infective endocarditis (n ⫽ 25, 7%). The majority of surgical procedures were performed on an elective basis (90%), and only one was a redo. Comprehensive perioperative data were collected from the database and from surgical or anesthetic notes. One hundred ninety-five patients (57%) received bioprosthetic valves, of which four were stentless (three homograft, one Toronto stentless porcine valve). The standard approach for myocardial protection within the unit includes intermittent, cold, anterograde blood cardioplegia supplemented with topical cooling. The mean length of follow-up was 114 ⫾ 192 days (range, 2 to 938 days). All patients underwent standard 12-lead resting electrocardiography before surgery. Analyses of the recordings were performed by one of two experienced cardiologists. Preoperative conducting system disease was defined by the presence of at least one of the following abnormalities: first-degree AV block (PR interval ⬎200 milliseconds), left bundle-branch block, right bundlebranch block, or left anterior hemiblock (mean QRS axis, ⫺30 degrees, or greater). Postoperatively, resting electrocardiographs were repeated and interpreted using the same criteria. The first electrocardiograph in the notes after day 1 was used. The requirement for PPM was determined by the attending cardiologist and was based on the general
AVR ⫽ aortic valve replacement;
NS ⫽ not significant;
PPM ⫽
policy of waiting until at least the fifth postoperative day. Uniformly accepted indications within the department are the continued presence of complete heart block, symptomatic bradycardia, or the need to prevent undue bradycardia while controlling tachyarrhythmias at day 7. The actual decision and timing of PPM was, therefore, determined on an individual patient needs basis. All systems were implanted using a transvenous approach. Preoperative clinical characteristics thought likely to influence the conducting system and several operative variables were analyzed to determine whether any were predictive of postoperative PPM requirement. This audit was approved by the Southampton & South West Hampshire Research Ethics Committee.
Statistical Analysis All results are presented as mean ⫾ standard deviation. Data were analyzed using Stat View 4.5 (Abacus Concepts Inc, Berkeley, CA). Analysis of variance and 2 were used as appropriate. The receiver-operating characteristic curve for prosthetic aortic valve diameter was drawn (MedCalc 5.0, MedCalc Inc, Mariakerke, Belgium). The cutoff level that resulted in the highest product of sensitivity and specificity was considered the optimum for predicting PPM requirement. Logistic regression analysis was performed to determine independent predictors of PPM requirement. A probability value of less than 0.05 was considered significant.
Results Preoperative Electrocardiographic Findings A preoperative resting electrocardiograph demonstrated that the majority of patients were in sinus rhythm (n ⫽ 300, 88%); atrial fibrillation was present in 40 patients (12%) and complete AV block in a further 2. Other
CARDIOVASCULAR
Ann Thorac Surg 2008;85:108 –12
CARDIOVASCULAR
110
DAWKINS ET AL PERMANENT PACEMAKER IMPLANTATION AFTER AVR
Ann Thorac Surg 2008;85:108 –12
Table 2. Preoperative Electrocardiographic Findings in 342 Patients Undergoing Isolated Aortic Valve Replacementa
(mechanical or bioprosthetic), lowest operative temperature, and total cross-clamp or bypass time. Logistic regression analysis demonstrated that the presence of preoperative conducting system disease was the only independent predictor of requirement for postoperative PPM (p ⫽ 0.0087). In the presence of preoperative conducting system disease, the relative risk of PPM requirement was 2.88 (95% confidence interval, 1.31 to 6.33).
Rhythm Sinus rhythm Atrial fibrillation CHB Conducting abnormalities LAHB 1st-degree AV block LBBB RBBB a
Number of Patients (Total n ⫽ 342) 300 (88%) 40 (12%) 2 (1%) 34 (10%) 23 (7%) 23 (7%) 17 (5%)
Five patients had more than one conducting system abnormality.
AV ⫽ atrioventricular; CHB ⫽ complete atrioventricular block; LAHB ⫽ left anterior hemiblock (mean QRS axis ⱖ30 degree); LBBB ⫽ left bundle-branch block; RBBB ⫽ right bundle-branch block.
conducting system abnormalities were present in 89 patients (26%; Table 2). Postoperatively, 29 patients (8.5%) required a permanent pacemaker (PPM). Of these, 23 patients received DDD and 6 patients received VVI pacemakers at the discretion of the attending cardiologist. The indications for PPM were persistent complete AV block (n ⫽ 25), sinus bradycardia (n ⫽ 3), and atrial fibrillation (required to prevent bradycardia while controlling faster rates, n ⫽ 1). After surgery, the mean time to PPM implantation was 13 days (range, 6 to 57 days). Twenty-six patients (90%) received PPM during the index admission, a mean duration of 11 days after aortic valve surgery (range, 6 to 25 days).
Predictors of Permanent Pacemaker Implantation Requirement Patients with aortic regurgitation as an indication for surgery were more likely to require PPM than those with other indications (16% versus 7%; p ⫽ 0.01; Table 1). The size of the aortic valve prosthesis was associated with PPM requirement; receiver-operating characteristic analysis demonstrated that the best cutoff prosthetic aortic valve diameter for separating patients according to PPM requirement was 24 mm (area under curve ⫽ 0.60; 95% confidence interval, 0.55 to 0.65). At this point sensitivity was 53% and specificity 68%. A cutoff value of 23 mm gave a sensitivity of 80% with a low specificity of 32%. Patients with aortic regurgitation received larger prosthetic valves (24.9 ⫾ 2.8 mm) as compared with the rest of the study group (22.9 ⫾ 2.4 mm; p ⬍ 0.0001). More patients with electrocardiographic evidence of preoperative conducting system disease needed PPM as compared with those with normal conduction (16% versus 6%; p ⫽ 0.004). No difference was seen when each abnormality of conduction was considered alone, although individual numbers were small. The following variables were not associated with an increased requirement for permanent pacemaker: age, sex, aortic stenosis or endocarditis as indication for surgery, elective or emergency procedure, valve type
Comment We have shown that PPM is a common requirement after isolated AVR in a contemporary population (50% older than 70 years); approximately 1 in every 12 patients undergoing this procedure will require early PPM. The vast majority (90%) of these occurred during the index admission with potential adverse impact on hospitalization times. Histologic abnormalities of the conducting system are common in patients with aortic valve disease, and suggestions regarding their cause have varied from mechanical (elevated left ventricular pressures) and ischemic factors, to exaggerated age-related change and primary degenerative disease of the conducting system [9, 10]. Aortic valve disease, and aortic regurgitation in particular, results in fibrous thickening of the endocardium of the ventricular septum [9]. Subsequent impingement on underlying conducting tissue is plausible and may explain our finding that aortic regurgitation was associated with PPM. Follath and Ginks [3] demonstrated that intraventricular conduction defects were common after aortic valve surgery (26%). The intimate relationship between the aortic valve and conducting tissue has led to the suggestion that direct trauma at the time of surgery might be involved, occurring as a consequence of injury from sutures, pressure from residual calcific material, or impingement of the prosthetic valvular seat on conducting tissue [9]. Indeed continuous sutures rather than interrupted sutures have been shown to increase the incidence of postoperative AV conduction abnormalities and PPM [11]. The development of conduction defects after cardiac surgery may also be associated with total bypass and cross-clamp time, degree of myocardial cooling [12, 13], and method of cardioplegia [14]. These presumably reflect severity of ischemic damage to the conducting system. In our study these factors did not impact on postoperative PPM requirement, which may be related, at least in part, to advances in myocardial preservation during cardiopulmonary bypass. Nair and colleagues [15] showed that in the presence of calcific aortic stenosis, mitral annular calcification was associated with a greater need for PPM after valve replacement. Although in the current study we did not calculate an index of valvular calcification, aortic regurgitation and not stenosis was associated with PPM. Patients with aortic regurgitation received larger valve prostheses than patients with other defects. Although it is possible that larger prostheses were associated with greater operative
DAWKINS ET AL PERMANENT PACEMAKER IMPLANTATION AFTER AVR
111
Table 3. Summary of Previous Studies Investigating the Incidence of Permanent Pacemaker After Valvular Surgery Author Keefe et al [6] Jaeger et al [18] Del Rizzo et al [16] Morell et al [20] Lewis et al [17] Kolh et al [19] Limongelli et al [8]
Date Reported
Number Studied
Isolated AVR
Comments
Incidence of PPM
1985 1994 1996 1996 1998 1999 2003
102 256 3,448 46 558 83 276
Yes No No No No No No
Isolated AVR Repeat valvular surgery Open heart surgery AVR ⫾ CABG in the elderly Reoperative cardiac surgery AVR ⫾ CABG in the elderly AVR ⫾ CABG
6% 7.9% 1.3% 6.5% 9.7% 6% 3.2%
AVR ⫽ aortic valve replacement;
CABG ⫽ coronary artery bypass grafting;
trauma, a larger aortic root may also be indicative of more severe valvular disease and hence preoperative exposure to more unfavorable mechanical effects on the conducting system. Whether this renders the conducting system susceptible to ischemic damage during cardiac surgery is uncertain. In this study a conduction abnormality identified on a preoperative electrocardiograph was the only independent predictor of PPM; it is likely that such patients are more vulnerable to the adverse effects of surgery. From the current sample size it was not possible to identify a specific conduction abnormality that presented the greatest risk. Few previous studies have evaluated the incidence of PPM after valvular surgery. A summary is shown in Table 3. Keefe and coworkers [6] evaluated 102 consecutive patients undergoing isolated AVR at a single institution during 1977 to 1978. Postoperative complete heart block was common, occurring in 18 patients (18%). In the majority of cases this was a transient phenomenon. Six patients (6%) required PPM before discharge, and a further 3 patients (3%) required late PPM (median follow-up, 4.2 years; time of late PPM not specified). The investigators were unable to demonstrate any predictive clinical or operative variables for the early development of postoperative complete AV block. Del Rizzo and associates [16] evaluated all patients undergoing open heart surgery at their center during a 5.5-year period. Of the 3,448 patients included, 1.3% required PPM. Multivariate logistic regression analysis showed that aortic valve surgery (odds ratio, 8.23; p ⫽ 0.001), the absence of preoperative sinus rhythm (odds ratio, 5.60; p ⫽ 0.001); postoperative myocardial infarction (odds ratio, 3.46; p ⫽ 0.024), and female sex (odds ratio, 2.52; p ⫽ 0.003) were independent predictors of PPM. Aortic valve surgery in this study included combined procedures; data pertaining to the incidence of PPM in isolated AVR were not reported. Intuitively more complex cardiac surgery will be more likely to adversely affect the conducting system; indeed, the need for PPM is increased after reoperative cardiac surgery (9.7%) [17]. In this clinical setting several factors were found to strongly relate to increased need: valve replacement (including AVR), preoperative endocarditis, number of reoperations, age, and degree of hypothermia during bypass. Increased total aortic cross-clamp and bypass times were also associated with PPM. A further
PPM ⫽ permanent pacemaker.
study showed that in patients surviving repeat valvular surgery, 7.9% required PPM [18]. Some studies have looked specifically at AVR in the elderly. Kolh and associates [19] studied 83 octogenarians undergoing AVR between 1992 and 1997 (with coexistent coronary artery bypass graft surgery in 25%) and showed that PPM was required in 5 patients (6%). Morell and colleagues [20] studied a group of 46 elderly patients (range, 75 to 88 years) undergoing AVR (and coronary artery bypass graft surgery in 57%). In this group, 6.5% required PPM insertion. In a more recent cohort study of 276 patients undergoing AVR [8], 3.2% required PPM. Multivariate logistic regression analysis identified preoperative aortic regurgitation, myocardial infarction, pulmonary hypertension, and postoperative electrolyte imbalance as predictors of postoperative PPM requirement. This group was on average 10 years younger than in our study and also included patients undergoing additional interventions such as coronary artery bypass grafting. In patients in whom the primary indication was the requirement for revascularization, it is likely that the associated aortic valve disease was less severe. In addition, patients with preoperative atrial fibrillation were excluded in this study. These important differences are likely to explain the comparatively low requirement for PPM in this study. In common with the present study, aortic regurgitation was found to predict the requirement for PPM. In this study the vast majority (90%) of implants occurred during the index surgical admission, and it is likely that this contributes to protracted hospitalization (mean time to implant after surgery, 11 days). Habicht and coworkers [21] looked at the incidence of late conduction system abnormalities after AVR in 100 consecutive cases. They reported that 13.7% of operative survivors with normal perioperative electrocardiographs exhibited late conduction disorders. These occurred from 3 to 102 months after surgery. Only 1 patient in this study required PPM 82 months after AVR. Whether this can be directly attributed to AVR is debatable. In our study 3 patients received PPM during our follow-up period (114 ⫾ 192 days) after initial hospital discharge (the latest, 57 days after AVR). Our study was primarily designed to investigate early PPM requirement (especially on the index admission). It is a limitation of this study that we do not have more comprehensive follow-up data, which
CARDIOVASCULAR
Ann Thorac Surg 2008;85:108 –12
CARDIOVASCULAR
112
DAWKINS ET AL PERMANENT PACEMAKER IMPLANTATION AFTER AVR
may have identified additional patients who required late PPM implantation. Previous studies have shown that in patients receiving PPM after AVR, the requirement for PPM persists. Sniezek and associates [22] studied 22 patients receiving PPM after AVR. In 1 patient the indication for PPM (complete heart block) disappeared 24 months after AVR. In all other patients the requirement for PPM persisted for the duration of follow-up (range, 9 to 196 months). It therefore appears that the vast majority of PPM are implanted during the index admission and that the indication for PPM requirement again persists in the majority. Contemporary surgical techniques have resulted in a gratifyingly low in-hospital mortality of 1.8% for this large consecutive series of patients undergoing isolated AVR. Nevertheless, the requirement for early PPM implantation after isolated AVR is common. Although the majority of patients are likely to accept pacing as an additional intervention, this should be fully discussed preoperatively as part of the informed consent process. No single variable appears to reliably identify all patients who will require a postoperative PPM. Although careful monitoring should be performed on all patients, particular attention should be paid to patients with preexisting conducting system disease and those undergoing surgery for aortic regurgitation, particularly after the insertion of large valve prosthesis, because they appear to be at greater risk.
References 1. Alblaza SG, Blanco G, Maranhao V, et al. Calcific aortic valvular disease associated with complete heart block. Case reports of successful correction. Dis Chest 1968;54:457– 60. 2. Jensen B, Sigurd B. Atrioventricular block in aortic insufficiency. Mechanism, ECG features and clinical consequences. Acta Med Scand 1972;192:391– 4. 3. Follath F, Ginks WR. Changes in the QRS complex after aortic valve replacement. B Heart J 1972;34:553– 60. 4. Thomas JL, Dickstein RA, Parker FB Jr, et al. Prognostic significance of the development of left bundle conduction defects following aortic valve replacement. J Thorac Cardiovasc Surg 1982;84:382– 6. 5. Gannon PG, Sellers RD, Kanjuhm VI, Edwards JE, Lillehei CW. Complete heart block following replacement of the aortic valve. Circulation 1966;33(Suppl 1):I-152– 61.
Ann Thorac Surg 2008;85:108 –12
6. Keefe DL, Griffin JC, Harrison DC, Stinson EB. Atrioventricular conduction abnormalities in patients undergoing isolated aortic or mitral valve replacement. Pacing Clin Electrophysiol 1985;8:393– 8. 7. Dalrymple-Hay MJ, Alzetani A, Aboel-Nazar S, Haw M, Livesey S, Monro J. Cardiac surgery in the elderly. Eur J Cardiothorac Surg 1999;15:61– 6. 8. Limongelli G, Ducceschi V, D’Andrea A, et al. Risk factors for pacemaker implantation following aortic valve replacement: a single center experience. Heart 2003;89:901– 4. 9. Fukuda T, Hawley RL, Edwards JE. Lesions of conduction tissue complicating aortic valvular replacement. Chest 1976; 69:605–14. 10. Davies M, Harris A. Pathological basis of primary heart block. Br Heart J 1969;31:219 –26. 11. Totaro P, Calamai G, Montesi G, Barzaghi C, Vaccari M. Continuous suture technique and impairment of the atrioventricular conduction after aortic valve replacement. J Card Surg 2000;15:418 –22. 12. Baerman JM, Kirsh MM, de Buitleir M, et al. Natural history and determinants of conduction defects following coronary artery bypass surgery. Ann Thorac Surg 1987;44:150 –3. 13. Caretta Q, Mercanti CA, De Nardo D, et al. Ventricular conduction defects and atrial fibrillation after coronary artery bypass grafting. Multivariate analysis of preoperative, intraoperative and postoperative variables. Eur Heart J 1991; 12:1107–11. 14. Flack JE 3rd, Hafer J, Engelman RM, Rousou JA, Deaton DW, Pekow P. Effect of normothermic blood cardioplegia on postoperative conduction abnormalities and supraventricular arrhythmias. Circulation 1992;86(5 Suppl):II-385–92. 15. Nair CK, Sketch MH, Ahmed I, et al. Calcific valvular aortic stenosis with and without mitral annular calcium. Am J Cardiol 1987;60:865–70. 16. Del Rizzo DF, Nishimura S, Lau C, Sever J, Goldman BS. Cardiac pacing following surgery for acquired heart disease. J Card Surg 1996;11:332– 40. 17. Lewis JW Jr, Webb CR, Pickard SD, Lehman J, Jacobsen G. The increased need for a permanent pacemaker after reoperative cardiac surgery. J Thorac Cardiovasc Surg 1998;116: 74 – 81. 18. Jaeger FJ, Trohman RG, Brener S, Loop F. Permanent pacing following repeat cardiac valve surgery. Am J Cardiol 1994; 74:505–7. 19. Kolh P, Lahaye L, Gerard P, Limet R. Aortic valve replacement in the octogenarians: perioperative outcome and clinical follow-up. Eur J Cardiothorac Surg 1999;16:68 –73. 20. Morell V, Daggett W, Pezzella A, Moran J, Bitran D. Aortic stenosis in the elderly: result of aortic valve replacement. J Cardiovasc Surg (Torino) 1996;37:33–5. 21. Habicht J, Scherr P, Zerkowski H, Hoffmann A. Late conduction defects following aortic valve replacement. J Heart Valve Dis 2000;9:629 –32. 22. Sniezek M, Malecka B, Bednarek J, et al. Patients history following artificial aortic valve and pacemaker implantation. Przegl Lek 2004;61:718 –21.
Permanent Pacemaker Implantation After Isolated Aortic Valve Replacement: Incidence, Indications, and Predictors Sam Dawkins, MB BS, Alex R. Hobson, MB BS, Paul R. Kalra, MD, Augustine T. M. Tang, DM, FRCS Ed (C-Th), James L. Monro, FRCS, and Keith D. Dawkins, MD, FRCP Wessex Cardiac Unit, Southampton University Hospital, Southampton, United Kingdom
Background. Conducting system defects are common in patients with aortic valve disease. Aortic valve replacement may result in further conduction abnormalities and necessitate permanent pacemaker implantation (PPM). We sought to identify the contemporary incidence and predictors for early postoperative PPM in patients undergoing isolated aortic valve replacement. Methods. Data were analyzed from 354 consecutive patients undergoing isolated aortic valve replacement at a referral cardiac unit during a 30-month period; data were unavailable on 4 patients and a further 8 had undergone preoperative PPM. Results for the remaining 342 patients (97%; mean age, 67 ⴞ 14 years), of whom 212 were males, are presented. The major indications for aortic valve replacement were valvular stenosis (n ⴝ 224), regurgitation (n ⴝ 70), or infective endocarditis (n ⴝ 25). Preoperative conducting system disease was present in 26% of patients.
Results. In-hospital mortality was 1.8% (6 of 342 patients). Postoperatively 29 patients (8.5%) required early PPM, of which 26 were during the index admission. Patients with preoperative conducting system disease (16% versus 6%; p ⴝ 0.004) and valvular regurgitation (16% versus 7%; p ⴝ 0.01) were more likely to require PPM as opposed to those without. Preoperative conducting system disease was the only independent predictor of PPM (p < 0.01); the relative risk of PPM requirement in this group was 2.88 (95% confidence interval, 1.31 to 6.33). Conclusions. Permanent pacemaker implantation requirement after aortic valve replacement is a common occurrence, and should be discussed as part of the preoperative consent process. Preexisting conducting disease and preoperative aortic regurgitation were predictors of PPM requirement. (Ann Thorac Surg 2008;85:108 –12) © 2008 by The Society of Thoracic Surgeons
A
The most recent series (reported in 2003) is a prospective cohort study performed in 276 patients (mean age, 57.5 ⫾ 14 years) undergoing AVR in a single center [8]. Patients were included irrespective of whether the primary indication for surgery was severe aortic valve disease or the need for revascularization with concomitant aortic valve disease. Permanent pacemaker implantation was indicated in 3.2% of patients. Identifying patients at increased risk of developing conduction system abnormalities that are likely to require postoperative PPM would be of substantial clinical benefit, facilitating the planning of postoperative care. It also has important implications in the context of informed preoperative consent. The aim of the current study was to establish the contemporary incidence of early postoperative PPM requirement in patients undergoing isolated AVR. In addition, we sought to identify clinical criteria that might predict the need for early postoperative PPM.
ortic valvular stenosis and regurgitation are associated with abnormalities of conduction, including higher degrees of atrioventricular (AV) block [1, 2]. Aortic valve replacement (AVR) can result in the development of further conduction abnormalities [3], which may be associated with an increased risk of sudden death [4]. Higher degrees of AV block, although often reversible, may also necessitate permanent pacemaker implantation (PPM) [5]. Studies conducted in the late 1970s suggested that early PPM was required in approximately 6% of patients undergoing isolated AVR [6]. Advances in cardiac surgical techniques and improved availability have resulted in surgery being offered to a wider range of patients. Elderly patients, an increasing proportion of the adult population, gain substantial benefits from cardiac surgery [7], and this has resulted in a significant change in the demographics of patients undergoing AVR. Caution is therefore required when extrapolating data from older case series.
Accepted for publication Aug 13, 2007. Address correspondence to Dr Dawkins, Wessex Cardiac Unit, Southampton University Hospital, Southampton SO16 6YD, United Kingdom; e-mail: [email protected].
© 2008 by The Society of Thoracic Surgeons Published by Elsevier Inc
Patients and Methods Patients Data were analyzed from 354 consecutive patients undergoing isolated AVR at the Wessex Cardiac Unit (a referral 0003-4975/08/$34.00 doi:10.1016/j.athoracsur.2007.08.024
DAWKINS ET AL PERMANENT PACEMAKER IMPLANTATION AFTER AVR
109
Table 1. Characteristics of 342 Patients Undergoing Isolated Aortic Valve Replacement All Patients (n ⫽ 342) PPM Subset (n ⫽ 29) No PPM (n ⫽ 313) p Value (PPM vs. No PPM)
Characteristic Age (y) Sex (male) Conduction abnormalities Indications for AVR AS AR Mixed Endocarditis Others Elective/emergency Bioprosthetic/mechanical Prosthetic AV diameter (mm) Cross-clamp time (min) Bypass time (min) Lowest operative temperature (°C) AR ⫽ aortic regurgitation; permanent pacemaker.
67 ⫾ 14 212 (62%) 89 (20%)
66 ⫾ 13 19 (65%) 14 (48%)
193 (62%) 75 (22%)
NS NS 0.004
224 (66%) 70 (20%) 20 (6%) 25 (7%) 3 (1%) 283/80 195/147
16 (55%) 11 (38%) 0 (0%) 2 (7%) 0 (0%) 25/4 16/13 24 ⫾ 4 54 ⫾ 23 72 ⫾ 34 30.6 ⫾ 1.9
208 (61%) 59 (17%) 20 (6%) 23 (7%) 3 (1%) 258/76 179/134 23 ⫾ 3 51 ⫾ 20 66 ⫾ 25 31.2 ⫾ 2.5
NS 0.01 NS NS NS NS NS 0.03 NS NS NS
AS ⫽ aortic stenosis;
AV ⫽ aortic valve;
cardiac center) during a 30-month period from January 2000 to July 2002. Patients were identified from a customdesigned cardiac surgical database (based on FoxPro), which is maintained by a dedicated data clerk. Eight patients had undergone preoperative PPM, and data were not available in a further 4 patients. The remaining 342 patients (mean age, 67 ⫾ 14 years [⫾ standard deviation]) formed the study population; 212 (62%) were males. Baseline demographic details are shown in Table 1. The major indications for AVR were predominant valvular stenosis (n ⫽ 224, 66%), predominant regurgitation (n ⫽ 70, 20%), or infective endocarditis (n ⫽ 25, 7%). The majority of surgical procedures were performed on an elective basis (90%), and only one was a redo. Comprehensive perioperative data were collected from the database and from surgical or anesthetic notes. One hundred ninety-five patients (57%) received bioprosthetic valves, of which four were stentless (three homograft, one Toronto stentless porcine valve). The standard approach for myocardial protection within the unit includes intermittent, cold, anterograde blood cardioplegia supplemented with topical cooling. The mean length of follow-up was 114 ⫾ 192 days (range, 2 to 938 days). All patients underwent standard 12-lead resting electrocardiography before surgery. Analyses of the recordings were performed by one of two experienced cardiologists. Preoperative conducting system disease was defined by the presence of at least one of the following abnormalities: first-degree AV block (PR interval ⬎200 milliseconds), left bundle-branch block, right bundlebranch block, or left anterior hemiblock (mean QRS axis, ⫺30 degrees, or greater). Postoperatively, resting electrocardiographs were repeated and interpreted using the same criteria. The first electrocardiograph in the notes after day 1 was used. The requirement for PPM was determined by the attending cardiologist and was based on the general
AVR ⫽ aortic valve replacement;
NS ⫽ not significant;
PPM ⫽
policy of waiting until at least the fifth postoperative day. Uniformly accepted indications within the department are the continued presence of complete heart block, symptomatic bradycardia, or the need to prevent undue bradycardia while controlling tachyarrhythmias at day 7. The actual decision and timing of PPM was, therefore, determined on an individual patient needs basis. All systems were implanted using a transvenous approach. Preoperative clinical characteristics thought likely to influence the conducting system and several operative variables were analyzed to determine whether any were predictive of postoperative PPM requirement. This audit was approved by the Southampton & South West Hampshire Research Ethics Committee.
Statistical Analysis All results are presented as mean ⫾ standard deviation. Data were analyzed using Stat View 4.5 (Abacus Concepts Inc, Berkeley, CA). Analysis of variance and 2 were used as appropriate. The receiver-operating characteristic curve for prosthetic aortic valve diameter was drawn (MedCalc 5.0, MedCalc Inc, Mariakerke, Belgium). The cutoff level that resulted in the highest product of sensitivity and specificity was considered the optimum for predicting PPM requirement. Logistic regression analysis was performed to determine independent predictors of PPM requirement. A probability value of less than 0.05 was considered significant.
Results Preoperative Electrocardiographic Findings A preoperative resting electrocardiograph demonstrated that the majority of patients were in sinus rhythm (n ⫽ 300, 88%); atrial fibrillation was present in 40 patients (12%) and complete AV block in a further 2. Other
CARDIOVASCULAR
Ann Thorac Surg 2008;85:108 –12
CARDIOVASCULAR
110
DAWKINS ET AL PERMANENT PACEMAKER IMPLANTATION AFTER AVR
Ann Thorac Surg 2008;85:108 –12
Table 2. Preoperative Electrocardiographic Findings in 342 Patients Undergoing Isolated Aortic Valve Replacementa
(mechanical or bioprosthetic), lowest operative temperature, and total cross-clamp or bypass time. Logistic regression analysis demonstrated that the presence of preoperative conducting system disease was the only independent predictor of requirement for postoperative PPM (p ⫽ 0.0087). In the presence of preoperative conducting system disease, the relative risk of PPM requirement was 2.88 (95% confidence interval, 1.31 to 6.33).
Rhythm Sinus rhythm Atrial fibrillation CHB Conducting abnormalities LAHB 1st-degree AV block LBBB RBBB a
Number of Patients (Total n ⫽ 342) 300 (88%) 40 (12%) 2 (1%) 34 (10%) 23 (7%) 23 (7%) 17 (5%)
Five patients had more than one conducting system abnormality.
AV ⫽ atrioventricular; CHB ⫽ complete atrioventricular block; LAHB ⫽ left anterior hemiblock (mean QRS axis ⱖ30 degree); LBBB ⫽ left bundle-branch block; RBBB ⫽ right bundle-branch block.
conducting system abnormalities were present in 89 patients (26%; Table 2). Postoperatively, 29 patients (8.5%) required a permanent pacemaker (PPM). Of these, 23 patients received DDD and 6 patients received VVI pacemakers at the discretion of the attending cardiologist. The indications for PPM were persistent complete AV block (n ⫽ 25), sinus bradycardia (n ⫽ 3), and atrial fibrillation (required to prevent bradycardia while controlling faster rates, n ⫽ 1). After surgery, the mean time to PPM implantation was 13 days (range, 6 to 57 days). Twenty-six patients (90%) received PPM during the index admission, a mean duration of 11 days after aortic valve surgery (range, 6 to 25 days).
Predictors of Permanent Pacemaker Implantation Requirement Patients with aortic regurgitation as an indication for surgery were more likely to require PPM than those with other indications (16% versus 7%; p ⫽ 0.01; Table 1). The size of the aortic valve prosthesis was associated with PPM requirement; receiver-operating characteristic analysis demonstrated that the best cutoff prosthetic aortic valve diameter for separating patients according to PPM requirement was 24 mm (area under curve ⫽ 0.60; 95% confidence interval, 0.55 to 0.65). At this point sensitivity was 53% and specificity 68%. A cutoff value of 23 mm gave a sensitivity of 80% with a low specificity of 32%. Patients with aortic regurgitation received larger prosthetic valves (24.9 ⫾ 2.8 mm) as compared with the rest of the study group (22.9 ⫾ 2.4 mm; p ⬍ 0.0001). More patients with electrocardiographic evidence of preoperative conducting system disease needed PPM as compared with those with normal conduction (16% versus 6%; p ⫽ 0.004). No difference was seen when each abnormality of conduction was considered alone, although individual numbers were small. The following variables were not associated with an increased requirement for permanent pacemaker: age, sex, aortic stenosis or endocarditis as indication for surgery, elective or emergency procedure, valve type
Comment We have shown that PPM is a common requirement after isolated AVR in a contemporary population (50% older than 70 years); approximately 1 in every 12 patients undergoing this procedure will require early PPM. The vast majority (90%) of these occurred during the index admission with potential adverse impact on hospitalization times. Histologic abnormalities of the conducting system are common in patients with aortic valve disease, and suggestions regarding their cause have varied from mechanical (elevated left ventricular pressures) and ischemic factors, to exaggerated age-related change and primary degenerative disease of the conducting system [9, 10]. Aortic valve disease, and aortic regurgitation in particular, results in fibrous thickening of the endocardium of the ventricular septum [9]. Subsequent impingement on underlying conducting tissue is plausible and may explain our finding that aortic regurgitation was associated with PPM. Follath and Ginks [3] demonstrated that intraventricular conduction defects were common after aortic valve surgery (26%). The intimate relationship between the aortic valve and conducting tissue has led to the suggestion that direct trauma at the time of surgery might be involved, occurring as a consequence of injury from sutures, pressure from residual calcific material, or impingement of the prosthetic valvular seat on conducting tissue [9]. Indeed continuous sutures rather than interrupted sutures have been shown to increase the incidence of postoperative AV conduction abnormalities and PPM [11]. The development of conduction defects after cardiac surgery may also be associated with total bypass and cross-clamp time, degree of myocardial cooling [12, 13], and method of cardioplegia [14]. These presumably reflect severity of ischemic damage to the conducting system. In our study these factors did not impact on postoperative PPM requirement, which may be related, at least in part, to advances in myocardial preservation during cardiopulmonary bypass. Nair and colleagues [15] showed that in the presence of calcific aortic stenosis, mitral annular calcification was associated with a greater need for PPM after valve replacement. Although in the current study we did not calculate an index of valvular calcification, aortic regurgitation and not stenosis was associated with PPM. Patients with aortic regurgitation received larger valve prostheses than patients with other defects. Although it is possible that larger prostheses were associated with greater operative
DAWKINS ET AL PERMANENT PACEMAKER IMPLANTATION AFTER AVR
111
Table 3. Summary of Previous Studies Investigating the Incidence of Permanent Pacemaker After Valvular Surgery Author Keefe et al [6] Jaeger et al [18] Del Rizzo et al [16] Morell et al [20] Lewis et al [17] Kolh et al [19] Limongelli et al [8]
Date Reported
Number Studied
Isolated AVR
Comments
Incidence of PPM
1985 1994 1996 1996 1998 1999 2003
102 256 3,448 46 558 83 276
Yes No No No No No No
Isolated AVR Repeat valvular surgery Open heart surgery AVR ⫾ CABG in the elderly Reoperative cardiac surgery AVR ⫾ CABG in the elderly AVR ⫾ CABG
6% 7.9% 1.3% 6.5% 9.7% 6% 3.2%
AVR ⫽ aortic valve replacement;
CABG ⫽ coronary artery bypass grafting;
trauma, a larger aortic root may also be indicative of more severe valvular disease and hence preoperative exposure to more unfavorable mechanical effects on the conducting system. Whether this renders the conducting system susceptible to ischemic damage during cardiac surgery is uncertain. In this study a conduction abnormality identified on a preoperative electrocardiograph was the only independent predictor of PPM; it is likely that such patients are more vulnerable to the adverse effects of surgery. From the current sample size it was not possible to identify a specific conduction abnormality that presented the greatest risk. Few previous studies have evaluated the incidence of PPM after valvular surgery. A summary is shown in Table 3. Keefe and coworkers [6] evaluated 102 consecutive patients undergoing isolated AVR at a single institution during 1977 to 1978. Postoperative complete heart block was common, occurring in 18 patients (18%). In the majority of cases this was a transient phenomenon. Six patients (6%) required PPM before discharge, and a further 3 patients (3%) required late PPM (median follow-up, 4.2 years; time of late PPM not specified). The investigators were unable to demonstrate any predictive clinical or operative variables for the early development of postoperative complete AV block. Del Rizzo and associates [16] evaluated all patients undergoing open heart surgery at their center during a 5.5-year period. Of the 3,448 patients included, 1.3% required PPM. Multivariate logistic regression analysis showed that aortic valve surgery (odds ratio, 8.23; p ⫽ 0.001), the absence of preoperative sinus rhythm (odds ratio, 5.60; p ⫽ 0.001); postoperative myocardial infarction (odds ratio, 3.46; p ⫽ 0.024), and female sex (odds ratio, 2.52; p ⫽ 0.003) were independent predictors of PPM. Aortic valve surgery in this study included combined procedures; data pertaining to the incidence of PPM in isolated AVR were not reported. Intuitively more complex cardiac surgery will be more likely to adversely affect the conducting system; indeed, the need for PPM is increased after reoperative cardiac surgery (9.7%) [17]. In this clinical setting several factors were found to strongly relate to increased need: valve replacement (including AVR), preoperative endocarditis, number of reoperations, age, and degree of hypothermia during bypass. Increased total aortic cross-clamp and bypass times were also associated with PPM. A further
PPM ⫽ permanent pacemaker.
study showed that in patients surviving repeat valvular surgery, 7.9% required PPM [18]. Some studies have looked specifically at AVR in the elderly. Kolh and associates [19] studied 83 octogenarians undergoing AVR between 1992 and 1997 (with coexistent coronary artery bypass graft surgery in 25%) and showed that PPM was required in 5 patients (6%). Morell and colleagues [20] studied a group of 46 elderly patients (range, 75 to 88 years) undergoing AVR (and coronary artery bypass graft surgery in 57%). In this group, 6.5% required PPM insertion. In a more recent cohort study of 276 patients undergoing AVR [8], 3.2% required PPM. Multivariate logistic regression analysis identified preoperative aortic regurgitation, myocardial infarction, pulmonary hypertension, and postoperative electrolyte imbalance as predictors of postoperative PPM requirement. This group was on average 10 years younger than in our study and also included patients undergoing additional interventions such as coronary artery bypass grafting. In patients in whom the primary indication was the requirement for revascularization, it is likely that the associated aortic valve disease was less severe. In addition, patients with preoperative atrial fibrillation were excluded in this study. These important differences are likely to explain the comparatively low requirement for PPM in this study. In common with the present study, aortic regurgitation was found to predict the requirement for PPM. In this study the vast majority (90%) of implants occurred during the index surgical admission, and it is likely that this contributes to protracted hospitalization (mean time to implant after surgery, 11 days). Habicht and coworkers [21] looked at the incidence of late conduction system abnormalities after AVR in 100 consecutive cases. They reported that 13.7% of operative survivors with normal perioperative electrocardiographs exhibited late conduction disorders. These occurred from 3 to 102 months after surgery. Only 1 patient in this study required PPM 82 months after AVR. Whether this can be directly attributed to AVR is debatable. In our study 3 patients received PPM during our follow-up period (114 ⫾ 192 days) after initial hospital discharge (the latest, 57 days after AVR). Our study was primarily designed to investigate early PPM requirement (especially on the index admission). It is a limitation of this study that we do not have more comprehensive follow-up data, which
CARDIOVASCULAR
Ann Thorac Surg 2008;85:108 –12
CARDIOVASCULAR
112
DAWKINS ET AL PERMANENT PACEMAKER IMPLANTATION AFTER AVR
may have identified additional patients who required late PPM implantation. Previous studies have shown that in patients receiving PPM after AVR, the requirement for PPM persists. Sniezek and associates [22] studied 22 patients receiving PPM after AVR. In 1 patient the indication for PPM (complete heart block) disappeared 24 months after AVR. In all other patients the requirement for PPM persisted for the duration of follow-up (range, 9 to 196 months). It therefore appears that the vast majority of PPM are implanted during the index admission and that the indication for PPM requirement again persists in the majority. Contemporary surgical techniques have resulted in a gratifyingly low in-hospital mortality of 1.8% for this large consecutive series of patients undergoing isolated AVR. Nevertheless, the requirement for early PPM implantation after isolated AVR is common. Although the majority of patients are likely to accept pacing as an additional intervention, this should be fully discussed preoperatively as part of the informed consent process. No single variable appears to reliably identify all patients who will require a postoperative PPM. Although careful monitoring should be performed on all patients, particular attention should be paid to patients with preexisting conducting system disease and those undergoing surgery for aortic regurgitation, particularly after the insertion of large valve prosthesis, because they appear to be at greater risk.
References 1. Alblaza SG, Blanco G, Maranhao V, et al. Calcific aortic valvular disease associated with complete heart block. Case reports of successful correction. Dis Chest 1968;54:457– 60. 2. Jensen B, Sigurd B. Atrioventricular block in aortic insufficiency. Mechanism, ECG features and clinical consequences. Acta Med Scand 1972;192:391– 4. 3. Follath F, Ginks WR. Changes in the QRS complex after aortic valve replacement. B Heart J 1972;34:553– 60. 4. Thomas JL, Dickstein RA, Parker FB Jr, et al. Prognostic significance of the development of left bundle conduction defects following aortic valve replacement. J Thorac Cardiovasc Surg 1982;84:382– 6. 5. Gannon PG, Sellers RD, Kanjuhm VI, Edwards JE, Lillehei CW. Complete heart block following replacement of the aortic valve. Circulation 1966;33(Suppl 1):I-152– 61.
Ann Thorac Surg 2008;85:108 –12
6. Keefe DL, Griffin JC, Harrison DC, Stinson EB. Atrioventricular conduction abnormalities in patients undergoing isolated aortic or mitral valve replacement. Pacing Clin Electrophysiol 1985;8:393– 8. 7. Dalrymple-Hay MJ, Alzetani A, Aboel-Nazar S, Haw M, Livesey S, Monro J. Cardiac surgery in the elderly. Eur J Cardiothorac Surg 1999;15:61– 6. 8. Limongelli G, Ducceschi V, D’Andrea A, et al. Risk factors for pacemaker implantation following aortic valve replacement: a single center experience. Heart 2003;89:901– 4. 9. Fukuda T, Hawley RL, Edwards JE. Lesions of conduction tissue complicating aortic valvular replacement. Chest 1976; 69:605–14. 10. Davies M, Harris A. Pathological basis of primary heart block. Br Heart J 1969;31:219 –26. 11. Totaro P, Calamai G, Montesi G, Barzaghi C, Vaccari M. Continuous suture technique and impairment of the atrioventricular conduction after aortic valve replacement. J Card Surg 2000;15:418 –22. 12. Baerman JM, Kirsh MM, de Buitleir M, et al. Natural history and determinants of conduction defects following coronary artery bypass surgery. Ann Thorac Surg 1987;44:150 –3. 13. Caretta Q, Mercanti CA, De Nardo D, et al. Ventricular conduction defects and atrial fibrillation after coronary artery bypass grafting. Multivariate analysis of preoperative, intraoperative and postoperative variables. Eur Heart J 1991; 12:1107–11. 14. Flack JE 3rd, Hafer J, Engelman RM, Rousou JA, Deaton DW, Pekow P. Effect of normothermic blood cardioplegia on postoperative conduction abnormalities and supraventricular arrhythmias. Circulation 1992;86(5 Suppl):II-385–92. 15. Nair CK, Sketch MH, Ahmed I, et al. Calcific valvular aortic stenosis with and without mitral annular calcium. Am J Cardiol 1987;60:865–70. 16. Del Rizzo DF, Nishimura S, Lau C, Sever J, Goldman BS. Cardiac pacing following surgery for acquired heart disease. J Card Surg 1996;11:332– 40. 17. Lewis JW Jr, Webb CR, Pickard SD, Lehman J, Jacobsen G. The increased need for a permanent pacemaker after reoperative cardiac surgery. J Thorac Cardiovasc Surg 1998;116: 74 – 81. 18. Jaeger FJ, Trohman RG, Brener S, Loop F. Permanent pacing following repeat cardiac valve surgery. Am J Cardiol 1994; 74:505–7. 19. Kolh P, Lahaye L, Gerard P, Limet R. Aortic valve replacement in the octogenarians: perioperative outcome and clinical follow-up. Eur J Cardiothorac Surg 1999;16:68 –73. 20. Morell V, Daggett W, Pezzella A, Moran J, Bitran D. Aortic stenosis in the elderly: result of aortic valve replacement. J Cardiovasc Surg (Torino) 1996;37:33–5. 21. Habicht J, Scherr P, Zerkowski H, Hoffmann A. Late conduction defects following aortic valve replacement. J Heart Valve Dis 2000;9:629 –32. 22. Sniezek M, Malecka B, Bednarek J, et al. Patients history following artificial aortic valve and pacemaker implantation. Przegl Lek 2004;61:718 –21.