Rate‐related left bundle branch block and cardiac memory in a patient with bradycardia: Case report and literature review

Abstract

Rate‐related left bundle branch block (LBBB) is a well‐studied phenomenon. Cardiac memory is another physiologic phenomenon in which T‐wave abnormalities occur in the absence of ischemia. The association between these 2 phenomena has been described in several case reports. A literature review was performed through Ovid and PubMed, where at total of 93 cases of rate‐related LBBB were identified. Cases were reviewed, and data were collected on rates of appearance and disappearance as well as the presence or absence of cardiac memory. There is some overlap in the rate at which LBBB appears. Cardiac memory is associated with rate‐related LBBB in several cases, but its true prevalence is unknown. Cardiac memory is a phenomenon that is well described in the literature but is often underrecognized in clinical practice. As a consequence of overlooking this phenomenon and not including cardiac memory in the differential when T‐wave abnormalities are observed, patients may be subjected to unnecessary invasive diagnostic testing.

1. INTRODUCTION

The presence of a left bundle branch block (LBBB) on electrocardiogram (ECG) convolutes the diagnosis of ischemia, as repolarization abnormalities associated with a LBBB can conceal or mimic an acute myocardial infarction. Occasionally, bundle branch blocks can arise intermittently, also making the diagnosis of ischemia challenging because the normally conducted beats now may show repolarization abnormalities, which is known as cardiac memory. There are several proposed mechanisms of transient bundle branch blocks: (1) a physiologic block during phase 3 and 4 of the action potential; (2) acceleration‐dependent block or phase 4 bradycardia‐dependent block due to disease of the His‐Purkinje system; and (3) retrograde concealment, in which retrograde penetration of the bundle branch renders it refractory to subsequent beats requiring a premature ventricular beat for termination.1

Cardiac memory is an ECG phenomenon often observed when abnormal left ventricular depolarization resolves. The T waves of normally conducted beats are abnormal, pointing in approximately the same direction as the QRS complexes with abnormal depolarization. This phenomenon has been observed after restoration of different forms of abnormal depolarization: intermittent or cured preexcitation in the Wolff‐Parkinson‐White syndrome, intermittent LBBB and intermittent ventricular pacing, and after, termination of ventricular tachycardia.2

In this article we present a case of a healthy middle‐aged female with postoperative nausea and bradycardia revealing a rate‐dependent LBBB with deeply inverted T waves associated with normally conducted complex on her ECG consistent with cardiac memory. We will address common clinical questions that occur in such scenarios.

2. CASE REPORT

A 50‐year‐old Caucasian female without significant medical history presented for an elective left knee arthroscopy for meniscectomy. A preoperative ECG showed a normal sinus rhythm, heart rate (HR) of 60 beats per minutes (bpm), and a LBBB consistent with her prior ECGs (Figure 1). The procedure was done under general anesthesia. She received propofol 150 mg, cisatracurium 8 mg, midazolam 2 mg, fentanyl 100 μg, cefazolin 2 g, dexamethasone 8 mg, ondansetron 4 mg, all intravenously, and lactated ringers solution (74.5 mL/5 min). Approximately 20 minutes into the procedure her HR decreased to 51 bpm and she was intravenously given glycopyrrolate (total of 0.8 mg), neostigmine 4 mg, and 2 doses of fentanyl 50 μg. The surgery was completed without any complications.

Figure 1.

Electrocardiogram from admission showing normal sinus rhythm and heart rate of 60 beats per minute with a left bundle branch block

During the immediate postoperative recovery she became bradycardic (HR 49 bpm) and she reported having nausea and paresthesias in her fingers. Atropine 1 mg was given intravenously, and a follow‐up ECG showed normal sinus rhythm, HR 64 bpm, with a LBBB. Her HR remained in the 60s overnight, and serial troponins were all negative (<0.02 ng/mL). The following morning she had another episode of nausea, which triggered an episode of bradycardia; this time her HR was in the 40s. Interestingly, her telemetry strip showed that she was having intermittent periods of wide QRS complexes with a LBBB pattern intermixed with periods of normal conduction with narrow QRS complexes. Two repeat ECGs were obtained and revealed a rather intriguing finding; when her HR was greater than 40 bpm she had a LBBB, and when her HR was 40 bpm or less she had normal conduction with disappearance of the LBBB (Figures 2 and 3). Additionally, new T wave inversions (TWI) in the precordial leads were observed at the lower HR. Due to concern for ischemia, the patient was transferred to a higher level of care for urgent coronary catheterization. The coronary angiogram showed normal coronary arteries without any atherosclerotic disease. The etiology of her bradycardia and ECG findings were thus attributed to a vagal response from postoperative nausea. She continued to have an LBBB present at HR greater than 40.

Figure 2.

Electrocardiogram showing sinus bradycardia with a left bundle bunch block at a heart rate of >40 beats per minute followed by normal conduction when the heart rate drops below 40 beats per minute. Inverted T waves are now seen in leads V4 to V6

Figure 3.

Electrocardiogram showing sinus bradycardia with normal conduction at a rate of 44 beats per minute. Inverted T waves present in the precordial leads are consistent with cardiac memory

3. DISCUSSION

Rate‐related LBBB was first described in 1913 by Dr. Thomas Lewis as an “unstable” bundle branch block.3 In 1973, Rosenbaum and colleagues described the mechanism of rate‐related LBBB with a presentation of 14 cases.4 Rosenbaum et al's proposed mechanism for a rate‐related LBBB was that it is tachycardia dependent, occurring only during tachycardia secondary to sudden or gradual failing of the left bundle branch. They explained that a rate‐related LBBB occurs in the setting of heart disease and disease of the His‐Purkinje system, which leads to failure of left bundle branch conduction at higher heart rates. They also postulated that a permanent LBBB is the result of physical blockage or death of the fascicle as opposed to infinitely long recovery periods. From observations of the 14 patients presented in their article, and with references to a much larger but unpublished dataset, they concluded that a 1500‐ to 1800‐ms cycle length (40–33 bpm) is likely the upper possible limit for an action potential in the fascicle.4

Our meta‐analysis reveals that the critical rate at which the LBBB becomes inactive and at which it again becomes active overlap depending on whether the cycle length is increasing or decreasing.4 This causes the LBBB to first appear at a higher heart rate than is required for it to disappear. During a LBBB, a retrograde impulse is conducted via the right bundle branch through septal fibers prior to invading the left bundle branch. As a result, the left bundle branch may be refractory to all supraventricular impulses due to retrograde invasion from the preceding beat. As this cycle continues, persistent left aberrant conduction results. If a fortunately timed premature ventricular beat occurs, or if the sinus rate slows, then the left bundle branch will no longer be refractory when the next supraventricular impulse arrives. This results in nonaberrant conduction.5

We performed an extensive literature search through Ovid and PubMed, where we found 93 reported cases of rate‐related LBBB since 1948. We reviewed each of these case reports, assessed all of the ECGs and rhythm strips for the presence of coronary artery disease, and recorded the HR at which the LBBB appeared and disappeared (Table 1). The rate of appearance and disappearance of a rate‐related LBBB in our case is the lowest (44 bpm) of any previous cases described in larger studies except for 1 of the 14 patients that Rosenbaum et al described in 1973 whose rate was 42 bpm.4

Table 1.

Summary of findings from a meta‐analysis of rate‐related left bundle branch case reports including rate of appearance and disappearance of the LBBB as well as the presence or absence of cardiac memory

Study	Agea	Gender	Appearance of LBBB (Heart Rate [Beats/Minute])	Disappearance of LBBB (Hear Rate [Beats/Minute])	Coronary Artery Disease	Cardiac Memory?
Rosenbaum et al4	57	M	100	60	Yes	Unk
Biliack and Denes10	Unk	Unk	88	76	Unk	Unk
	Unk	Unk	100	94	Unk	Unk
	Unk	Unk	104	100	Unk	No
Manohar and Young19	14 days	F	136	115	No	No
Byme and Filippone20	57	M	60	54	No	Yes
Haridas et al21	50	F	90	90	No	Unk
Jourdain et al22	52	F	120	85	No	No
Perin et al23	46	M	152	103	Yes	No
Heinsimer et al24	47	F	133	110	No	No
Towne et al25	37	F	150	140	Unk	Unk
Waxman et al26	Unk	Unk	150	140	Unk	Unk
	Unk	Unk	210	193	Unk	Unk
	Unk	Unk	210	190	Unk	Unk
	Unk	Unk	176	160	Unk	Unk
Mautner and Phillips27	62	M	70	70	No	No
Gholamrezanezhad et al28	48	F	169	Unk	No	No
	63	F	117	Unk	No	No
Strohmer29	61	F	93	50	No	No
Tyagi et al30	62	F	90	80	No	No
Xiao and Gibson31	49	F	92	83	No	Unk
Kurata et al32	67	F	70	60	No	Unk
Domino et al33	55	M	90	71	No	No
	67	F	97	95	No	No
Suzuki34	70	F	100	60	Yes	No
Virtanen et al35	44	M	125	75	No	No
	35	F	160	125	No	No
	44	M	91	71	No	No
	46	F	114	76	No	No
	55	M	99	76	No	No
	46	F	77	72	No	No
	42	F	73	Unk	No	No
Patil et al36	58	F	67	54	Yes	Likely
Wong et al37	56	F	145	120	No	Unk
Sung et al38	37	M	66	46	No	No
Vieweg et al39	48	M	110	110	No	No
Chung40	77	F	100	85	No	No
Saritas et al41	70	M	105	65	No	No
Van de Heyning et al42	63	M	85	85	Yes	Yes
Acikel et al43	43	M	100	65	No	Unk
Maury et al44	44	F	83	74	No	No
Arias et al45	32	F	79	71	No	No
Yonekura et al46	75	M	120	120	Yes	No
Vereckei and Tenczer47	72	F	88	71	Yes	No
Michelotti and Swiryn48	54	M	73	63	No	Yes
Gould et al49	70	F	58	50	No	Yes
Atlas et al50	46	M	71	50	No	Yes
Chapman51	30	M	180	Unk	No	No
Cannom et al52	45	M	112	Unk	Yes	No
	62	M	91	Unk	Unk	No
Takeshita et al53	56	M	83	80	No	Unk
Krikler and Lefevre54	38	M	104	94	No	No
	41	F	116	116	No	No
Bourassa et al55	42	M	95	88	No	Unk

Abbreviations: F, female; LBBB, left bundle branch block; M male; Unk, unknown.

^aAges are given in years unless indicated otherwise.

The latest explanation for the physiology of a rate‐dependent LBBB is that there may be several mechanisms involved, including a physiologic block during phase 3 of the action potential, acceleration dependent block, phase 4 bradycardia‐dependent block due disease of the His‐Purkinje system, or retrograde concealment in which retrograde penetration of the bundle branch renders it refractory to subsequent beats requiring a premature ventricular beat for termination.1, 5

Cardiac memory is another important electrophysiological phenomenon. Cardiac memory was a term first used by Rosenbaum et al to describe non–ischemic related TWI in patients with Wolf‐Parkinson‐White syndrome in 1982.6 The phenomenon, however, was first described earlier in 1969 by Chatterjee et al in the setting of TWI after artificial pacing.7 The cellular mechanisms responsible for cardiac memory have been extensively studied. Short‐term and long‐term cardiac memory share a common activation of altered ventricular stretch leading to increased angiotensin II synthesis and release. Short‐term memory results from internalization of a macromolecular complex, resulting in transient decrease in outward potassium current. Long‐term memory results from a signaling pathway impacting gene transcription and protein degradation.8, 9

Electrotonic load is a potential mechanism of action, which is perhaps what is seen in our case.10 Jeyaraj et al11 noted bundle‐branch block as a contributor to altered activation. Bradycardia altered the activation pathway; the disappearance of the LBBB supports this. With a new pathway, the myocytes are in a state of full depolarization, leading to the expression of deep T waves consistent with cardiac memory.11 Although these changes are commonly benign, there are several reports considering the arrhythmogenic potential of cardiac memory.11, 12 Cardiac memory is a diagnosis of exclusion, which unfortunately includes a broad differential including acute coronary syndrome.

Cardiac memory can imitate cardiac ischemia as seen in the aforementioned case as well as other case reports.13 One study demonstrated that cardiac memory–related TWI can be differentiated from ischemia‐related TWI based upon several criteria as described by Shvilkin et al in postpacing cardiac memory. These criteria include the combination of a positive T wave in lead aVL, a positive or isoelectric T wave in lead I, and maximal precordial TWI greater than TWI in lead III. This was shown to be 92% sensitive and 100% specific for cardiac memory.14 This study has not yet been prospectively validated, but could be a potentially useful tool in differentiating ischemia from benign cardiac memory TWI on ECG. In general, the TWI of normally conducted beats depends on the polarity of the QRS during abnormal depolarization and will be different for different causes of cardiac memory, such as following pacing, LBBB, ventricular tachycardia, or preexcitation. Though the criteria were described for postpacing cardiac memory, in the case at hand it occurred following resolution of LBBB.

The concurrence of a rate‐related LBBB and T‐wave inversions has only been described in a few case reports, and therefore, its true prevalence is not known. The Rosenbaum et al case series in 1973 mentions that a few of the 14 cases of rate‐related LBBB had deep precordial TWI.4 In 1978, Engel et al reviewed 46 ECGs, of which 33 had TWI during normal conduction, and 17 of those were found to have no coronary artery disease.15 In 1979, Abben et al reviewed 151 ECGs with a rate‐related LBBB, and found that 79 (63%) had TWI in the precordial leads, 21 (27%) had TWI in the frontal leads, and 8 (10%) had TWI in both.16 In 1992, a case series published by La Canna et al reports that 18 of 33 patients with rate‐related LBBB had deep TWI in the precordial leads.17 The prevalence of cardiac memory in cases of rate‐related LBBB is therefore, likely, fairly high.

Rate‐related LBBB block has been studied most extensively during an exercise tolerance test ETT. Studies have found that exercise‐induced LBBB may predict a higher risk of major cardiac events (19%) and death (29%) vs matched control groups (10%) and (25%).18 Additional studies have shown that, in the presence of normal coronaries, exercise‐induced LBBB has a favorable prognosis. However, on follow‐up, permanent LBBB may develop in addition to dyssynchrony of left ventricular contraction, heart failure, and rarely, atrioventricular block requiring pacemaker implantation.10

Because the TWI caused by cardiac memory can mimic ischemia, it may result in unnecessary testing as seen in this case. In 1 case, a patient was denied insurance coverage based on an ECG with TWI, which were actually due to cardiac memory.11 Unfortunately, the consequences of misdiagnosing ischemia as cardiac memory would be catastrophic. We therefore believe that ischemic workup of patients with TWI likely due to cardiac memory is still indicated at this time.

4. CONCLUSION

Future investigation to prospectively validate the Shvilkin et al criteria14 would be beneficial. For the time being, clinicians should be aware of the association of cardiac memory with rate‐related LBBB and perhaps consider noninvasive testing when cardiac memory is suspected to be the etiology of TWI on ECG.

Conflicts of interest

The authors declare no potential conflicts of interest.

Seibolt L, Maestas C, Lazkani M, Fatima U, Loli A, Chesser M. Rate‐related left bundle branch block and cardiac memory in a patient with bradycardia: Case report and literature review. Clin Cardiol. 2018;41:1097–1102. 10.1002/clc.22997