c
  c 



   

 

     

  

Ofer Meri, M.D.; Michael Ilan, M.D.; Avraham Oren, M.D.; Daniel Fink, M.D.; Maher Deeb, M.D.;
Dani Bitran, M.D.; Shuli Silberman, M.D.

Posted: 03/16/2009; Pacing Clin Electrophysiol. 2009;32(1):7-12. © 2009 Blackwell Publishing


 
 

°


ð
 Conduction disturbances requiring permanent pacemaker implantation after heart
surgery occur in about 1.5% of patients. Early pacemaker implantation may reduce morbidity
and postoperative hospital stay. We reviewed our experience with patients undergoing surgery
to try and identify predictors for pacemaker requirements and patients who will remain
pacemaker dependent.
˜   We performed a retrospective review of 4,999 patients undergoing surgery between
the years 1993 and 2005. Patient age was 64 ± 12 years, and 71% were males. Coronary bypass
was performed in 4,071 (81%), aortic valve replacement in 675 (14%), and mitral valve
replacement in 968 (18%) patients.
  Seventy-two patients (1.4%) required implantation of a permanent pacemaker after
surgery. Indications for pacemaker implantation included complete atrioventricular block in 59,
symptomatic bradycardia/slow atrial fibrillation in nine, second-degree atrioventricular block in
two, and other conduction disturbances in two patients. Predictors for pacemaker requirement
by multivariate analysis were left bundle branch block and aortic valve replacement (P < 0.001).
Late follow-up was available in 58 patients, at 72 ± 32 months. Thirty-seven (63%) were
pacemaker dependent. Predictors for late pacemaker dependency were third-degree
atrioventricular block after surgery and preoperative left bundle branch block (P < 0.001).
   Patients at high risk for pacemaker implantation after heart surgery include those
with preexisting conduction disturbances, and those undergoing aortic valve replacement. Of
those receiving a pacemaker, about one-third will recover at late follow-up. For patients in the
high-risk group who are pacemaker dependent after surgery, we recommend implanting a
permanent pacemaker at 5 days after surgery, thus enabling early mobilization and early
discharge.

m  
 

The incidence of conduction disturbance requiring pacing after open heart operations is
between 10% and 15%.[1-4] Most of these will recover, however, 1-3% of patients will require
permanent pacemaker implantation.[5-7] We reviewed our experience with permanent
pacemaker (PPM) implantation following cardiac surgery. Our aim was to identify predictors for
PPM requirement and identify patients who will remain pacemaker dependent over long term.

˜  

c
 


Between the years 1993 and 2005, 4,999 patients underwent heart surgery in our department.
Patient age was 64 ± 12 years and 3,539 (71%) were males. Isolated coronary bypass grafting
(CABG) was performed in 3,448 (69%), valve surgery in 831 (17%), and 623 (12%) patients
underwent combined CABG and valve surgery. All patients in this series had their pacemaker
implanted within the same hospitalization as their surgery.

°  




Patient data were collected on the standard computerized Society of Thoracic Surgeons (STS)
database (Summit Medical Systems, Minneapolis, MN, USA) and reviewed according to a
predetermined protocol. Collected data included demographic, preoperative, operative, and
early postoperative data (Appendix I). This information was collected during hospitalization.
Since detailed electrocardiogram (ECG) data were not routinely collected on the database, we
reviewed the preoperative ECG of 920 consecutive patients who did not require a permanent
pacemaker who served as a control group, as well as the preoperative ECG of all patients
undergoing aortic valve replacement (AVR). Electrocardiogram data of patients receiving a
permanent pacemaker were compared to that of the control group. Late follow-up was
obtained from outpatient records, medical summaries, or contact with the family physician.
Stepwise logistic regression analysis was used to identify predictors for permanent pacemaker
requirement. Data from the pacemaker clinic were used to determine pacemaker dependency.
The underlying rhythm was obtained by programming the pacemaker to VVI rate of 40 beats
per minute during the patient's visit to the pacemaker outpatient clinic.







Patients were examined in the outpatient clinic for dependency by lowering pacemaker rate to
40 bpm for a short duration of up to 10 seconds. Patients were considered pacemaker
dependent if they continued to be paced at a pacemaker ventricular rate of 40 bpm. Patients
were considered nonpacemaker dependent if they had sinus rhythm or atrial fibrillation with an
adequate ventricular response at pacemaker rate of 40 bpm.

 

Seventy-two patients (1.4%) required permanent pacemaker implantation. Indications for PPM
implantation included: (i) complete atrioventricular (AV) block in 59; of these, 35 were
immediate in onset and persistent and 24 were either delayed onset or intermittent; (ii)
second-degree AV block in two; (iii) severe bradycardia (sinus rhythm or atrial fibrillation with a
ventricular response <50 bpm) in nine; (iv) first-degree AV block with left bundle branch block
(LBBB) and bradycardia in one; and (v) alternating LBBB and right bundle branch block (RBBB) in
one patient. The choice of pacemaker implanted was based upon the underlying conduction
disturbance and the patient's needs. There were no pacemaker-related complications such as
implantation site infection in this group of patients.

Table 1 shows the clinical profile of patients requiring a permanent pacemaker compared with
those not requiring a pacemaker. Patients requiring a pacemaker were older and in higher New
York Heart Association (NYHA) functional class. Preoperatively, a larger number of these
patients were receiving antiarrhythmic medication, although by multivariate analysis this did
not emerge as a predictor for permanent pacemaker implantation. All antiarrhythmic
medication was discontinued after surgery, and resumed only if conduction was normal, so that
none of the patients in this series were receiving AV nodal blocking agents. Table 2shows the
preoperative ECG characteristics of patients requiring a permanent pacemaker compared to the
control group. Patients requiring a pacemaker had a higher incidence of conduction
disturbances preoperatively, especially LBBB and AV block. The need for pacemaker
implantation was significantly higher in patients undergoing AVR in comparison with patients
undergoing CABG or mitral valve replacement (5.7%, 1%, and 1.8%, respectively; P = 0.0001).
Patients in the PPM group had longer cardiopulmonary bypass times as well as longer ischemic
times. Postoperatively, these patients had a longer intensive care stay, and there was a
tendency for higher creatinine kinase (CK-MB) values, reflecting some degree of myocardial
injury. Left main stem coronary stenosis or proximal left anterior descending stenosis were not
associated with an increased need for PPM, nor was surgery of the tricuspid valve.

Timing of implantation ranged from 2 to 75 days (mean 13). Reasons for delayed implantation
included a fluctuating conduction disorder, questionable indication, or active infection. Two
patients received a pacemaker on postoperative days 2 and 3 due to lack of underlying rhythm
and unreliable function of the temporary pacing wires. Of note, both patients were not
pacemaker dependent at time of follow-up. One patient received a pacemaker on
postoperative day 75. This patient had a prolonged stay in the intensive care unit, and
implantation was delayed due to active infection. Postoperative day of implantation was similar
for patients who were pacemaker dependent and nondependent at late follow-up. Univariate
analysis showed no correlation between early (ч 5 days) or late implantation to pacemaker
dependency at late follow-up. Predictors for PPM requirement by multivariate analysis were
preoperative LBBB, AVR, and elevated CK-MB ( Table 3 ).

Rc  

Of the 72 patients requiring PPM, follow-up was available in 58 (81%), at an interval of 72 ± 32

months. Forty-nine were alive at the time of follow-up, and follow-up was available in all of
them. Nine had died but data relating to their pacemaker before death were available. In 14
patients there was no-follow up (operative mortality in four, late mortality with no available
data in 10). Of the 58 patients, 37 (63%) were pacemaker dependent, and 21 (37%) were
nonpacemaker dependent. By multivariate analysis, preoperative LBBB and persistent
postoperative third-degree AV block were predictors for late pacemaker dependency ( Table 4).
Of 26 patients with third-degree AV block, 22 remained pacemaker dependent. Four were
nondependent, in two of whom the PPM was inserted very earlyͶdays 2 and 3. Only two
patients who had a third-degree atrioventricular block (AVB) after day 5 were nonpacemaker
dependent on late follow-up.

°
   

Within the group of patients undergoing aortic valve replacement, the incidence of permanent
pacemaker implantation was 5.7%. By univariate analysis, lower minimum temperature (P =
0.02), longer bypass time (P = 0.02), and lower preoperative aortic valve gradient (P = 0.004)
were found to predict the need for pacemaker implantation in this subgroup. Overall, patients
undergoing AVR had a postoperative rate of AVB of 4% (26/672) compared with 0.6%
(28/4,327) in patients undergoing non-AVR surgery (P < 0.0001).

 

 
cc cc c! 

N 72 4,927

c





Age 67 ± 10 64 ± 11 0.0106

Age >70 years 32 (44%) 1,521 (31%) 0.01

Male 44 (61%) 3,495 (71%) 0.07

Hypertension 39 (54%) 2,717 (55%) ns

Diabetes 21 (29%) 1,662 (34%) ns

COPD 32 (44%) 2,627 (53%) ns

Renal failure (creatinine >1.5) 9 (13%) 461 (9%) ns

Old stroke 5 (7%) 469 (10%) ns

Peripheral vascular disease 10 (14%) 671 (14%) ns

Pulmonary hypertension 25 (35%) 960 (19%) 0.0013

NYHA class IIIʹIV 39 (57%) 1,573 (35%) <0.0001

LV function

Normal 55 (76%) 3,809 (77%) ns

Moderately impaired 12 (17%) 839 (17%) ns

Severely impaired 5 (7%) 279 (6%) ns

Left main stenosis (>50%) 4 (9%) 797 (20%) ns

Proximal LAD stenosis (>90%) 17 (40%) 2,009(50%) ns

ɴ-blocking agents 37 (51%) 2,781 (56%) ns

Digoxin 4 (5%) 308 (6%) ns

Antiarrhythmics 15 (21%) 328 (7%) <0.0001






Elective 64 (89%) 4,609 (94%) ns

Reoperation 8 (11%) 318 (6%) ns

CABG 42 (58%) 4,029 (82%) <0.0001

AVR 38 (53%) 634 (13%) <0.0001

MVR 18 (25%) 942 (19%) ns

TV repair 3 (4%) 133 (3%) ns

BPT (minutes) 141 ± 57 102 ± 50 <0.0001

lCT (minutes) 92 ± 41 64 ± 36 <0.0001

lCT >120 minutes 16 (23%) 307 (6%) 0.0001

Minimum temp (degree centigrade) 28 ± 4.7 30 ± 3 <0.0001

Cold cardioplegia 69 (95%) 4,413 (90%) 0.05

c 




Perioperative MI 4 (6%) 120 (2%) 0.09

Elevated CK-MB (>100) 8 (19%) 287 (7%) 0.009

ICU stay (days) 7 ± 12 3.1 ± 7 <0.0001

Hospital stay (days) 24 ± 23 12 ± 11 <0.0001

Mortality 4 (6%) 263 (5%) ns

Abbreviations: PP = permanent pacemaker; non-PP = nonpermanent pacemaker; COPD =

chronic obstructive lung disease; NYHA = New York Heart Association; LV = left ventricle; LAD =
left anterior descending coronary artery; CABG = coronary artery bypass grafting; AVR = aortic
valve replacement; MVR = mitral valve replacement/repair; TV = tricuspid valve; BPT =
cardiopulmonary bypass time; lCT = aortic cross-clamp time; MI = myocardial infarction; ICU =
intensive care unit.

 c
 

cc cc c! 

N 72 920

Any BBB 29 (40%) 91 (10%) <0.0001

LBBB 18 (25%) 24 (2.6%) <0.0001

RBBB 11 (15%) 67 (7%) 0.015

Bifascicular block 3 (4%) 9 (1%) 0.017

AVB 11 (15%) 54 (6%) 0.01

1-degree AVB 10 (14%) 52 (5.5%) 0.005

2-degree AVB 1 (1%) 2 (0.2%) 0.08

Abbreviations: PP = permanent pacemaker; non-PP = nonpermanent pacemaker; BBB = bundle

branch block; LBBB = left bundle branch block; RBBB = right bundle branch block; AVB =
atrioventricular block.

 c

c
  

°  



" #$%  c! 

LBBB 13 7ʹ25 <0.0001

AVR 8 5ʹ13 <0.0001

Elevated CK-MB 2 1ʹ4 0.07

Abbreviations: OR = odds ratio; CI = confidence interval; LBBB = left bundle branch block; AVR = aortic
valve replacement.

&   






 

The need for permanent pacemaker after cardiac surgery can be attributed to various factors.
These include ischemia resulting in injury to conduction tissue. This may be more common in
patients with proximal left anterior descending artery and septal artery disease,[8,9] although
this was not the case in our patients. Another cause is direct injury to the conduction system.
Surgery of the aortic valve in particular is in proximity to the conduction tissue. Local edema
may disrupt conduction, usually temporarily, but removal of penetrating calcium or insertion of
deep stitches may cause permanent damage.[10] Requirement of PPM after AVR may be as high
as 8.5%, especially in patients with preoperative conduction disturbance.[11]

We found an overall need for PPM implantation of 1.4% in patients undergoing coronary,
valvular, or combined surgery. This is in accordance with 0.8-6% cited by others.[1-4] Of those
patients receiving a pacemaker, about one-third recovered and were not pacemaker dependent
at the time of examination. Glikson et al.[12] determined pacemaker dependency by repeated
transtelephonic transmission to the pacemaker clinic, programming down the pacemaker in the
pacemaker clinic, and by other means of prolonged monitoring. In addition, they studied the
reliability of the dependency status by 24-hour continuous monitoring in a subgroup of 20
patients and found good concordance between the clinical evaluation of dependency and the
ambulatory monitoring results. In our patients, those who did not exhibit a spontaneous
rhythm with the pacemaker programmed to VVI at a rate of 40 bpm were considered
pacemaker dependent. We did not wait for patients to become symptomatic, and for clinical
purposes this seemed sufficient. We do not have complete data as to the type of escape, but
since the cutoff point was 40 bpm, this supports narrow QRS escape.

From a practical viewpoint, the question remains which patients should have a permanent
pacemaker implanted, and at what time interval after surgery? Patients at high risk are the
elderly, patients undergoing aortic valve replacement, and those with preexisting conduction
disturbances.

Patients with complete AV block are often totally pacemaker dependent, and therefore are
kept in the intensive care unit for close monitoring. Often they receive a temporary
transvenous pacemaker to serve as a backup, especially if the temporary epicardial pacemaker
threshold begins to increase. Under such circumstances, patient mobilization, comfort, and
safety are compromised. Insertion of a permanent pacemaker enables early mobilization, thus
reducing morbidity associated with prolonged bed stay. Kim
 ., in a series of patients
undergoing valve surgery, found that all those requiring a pacemaker developed AVB within the
first postoperative day.[13] At late follow-up, 56% remained pacemaker dependent. If AVB did
not resolve by 48 hours, they recommend implantation of a pacemaker by day 7. Berdajs

.[14] found a 4% incidence of third-degree AV block in patients undergoing mitral valve surgery
in whom a pacemaker was implanted at a mean interval of 4 days after surgery. Their policy is
to implant a pacemaker in any patient with block who is otherwise ready for discharge from the
intensive care unit, and not beyond 1 week. According to the American College of
Cardiology/American Heart Association guidelines, permanent pacemaker implantation is
indicated for third-degree and advanced second-degree AV block associated with postoperative
AV block that is not expected to resolve (class I). The decision to implant, as well as timing, is
left to the physician's discretion.[15] Our policy of early implantation is in accordance with these
data.

Another issue to consider is that of resource utilization. The risks and cost of pacemaker
implantation should be weighed against those of prolonged occupation of intensive care beds,
as well as prolonged hospital stay.

Most patients with conduction disturbance requiring pacing after surgery will have recovery of
their cardiac rhythm. Our experience shows that of patients receiving a PPM, about one-third
will recover at a later stage. However, one cannot guarantee such recovery, and we cannot
accurately determine which patients will definitely recover. Despite a more liberal approach to
early implantation in our patients, we did not see a correlation between timing of pacemaker
implantation to late dependency. Others have found a 65%[16] and 41%[17] rate of late
dependency after CABG, and 49% after AVR. In our series, late dependency was 67% and 61%
after non-AVR and AVR, respectively. Reported dependency in patients in whom AVB was the
indication for a pacemaker was 41% for CABG and 49% for AVR[17] compared with 74% and 73%,
respectively, in our patients. Moreover, Onalan
  [17] found a 23% rate of new pacemaker
dependency at late follow-up compared to baseline. We did not check dependency at baseline,
and our late results may include those who may have not been pacemaker dependent at
discharge, but eventually did become completely dependent. One must keep in mind, however,
the pacemaker clinic examination is only a "spot check," and we do not know if these patients
are not intermittently dependent. We do not know how many cases of sudden death after AVR
can be attributed to late onset of fatal conduction disturbances. While preoperative LBBB was
not a predictor for pacemaker dependency in other studies,[12,16,17] we did find it to be a
predictor in our patients.

We therefore recommend pacemaker implantation after cardiac surgery for patients who are
pacemaker dependent after surgery and whose spontaneous rhythm does not recover by
postoperative day 5. This is especially true for patients at high risk for conduction disturbances,
namely (i) patients with preexisting conduction disturbance, especially LBBB, and (ii) type of
surgery, especially on the aortic valve. This will enable early mobilization, early discharge from
hospital, and will offer the patient a "safety net" in case the potentially vulnerable conduction
tissue will fail at a later time.

Limitations of this study include its retrospective nature. While all clinical data were collected
prospectively, detailed ECG data were not complete. We therefore chose a consecutive group
of 920 patients to serve as control, and collected their ECG data retrospectively. Although we
expect these patients to be representative of the whole group, we cannot exclude some degree
of bias. As for the definition of pacemaker dependency, this was determined by assessment in
the outpatient clinic over a brief period of interrogation. Longer monitoring may reveal more
patients with escape rhythm that may be defined as nondependent, although for practical
purposes, they still remain pacemaker dependent.

 




Patients at high risk for conduction disturbances requiring permanent pacemaker implantation
after cardiac surgery include those with preexisting LBBB, and those undergoing aortic valve
replacement. We believe that early implantation enables faster patient recovery and early
discharge. Overall, about two-thirds will remain pacemaker dependent.