Abstract
Introduction:
Heart rate slowing agents are frequently prescribed to manage heart transplant (HTx) patients with the assumption that higher heart rate is a risk factor in cardiovascular disease.
Patients and Methods:
This prospective two-centre study investigated early progression of cardiac allograft vasculopathy (CAV) in 116 HTx patients. Examinations by coronary optical coherence tomography and 24-hour ambulatory ECG monitoring were performed both at baseline (1 month after HTx) and during follow up (12 months after HTx).
Results:
During the first post-HTx year, we observed a significant reduction in the mean coronary luminal area from 9.0±2.5 to 8.0±2.4 mm2 (p<0.001), and progression in mean intimal thickness (IT) from 106.5±40.4 to 130.1±53.0 μm (p<0.001). No significant relationship was observed between baseline and follow-up mean heart rates and IT progression (R=0.02, p=0.83; R=−0.13, p=0.18). We found a mild inverse association between beta blocker dosage at 12 months and IT progression (R=−0.20, p=0.035).
Conclusion:
Our study did not confirm a direct association between mean heart rate and progression of CAV. The role of beta blockers warrants further investigation, with our results indicating that they may play a protective role in early CAV development.
Keywords: cardiac allograft vasculopathy, heart rate, beta blocker, optical coherence tomography
INTRODUCTION
Heart transplantation (HTx) is accompanied by sympathetic and parasympathetic denervation, which typically results in increased heart rate (HR) due to negation of the vagus nerve effect1. Several epidemiological studies have demonstrated that faster HR is a risk factor for cardiovascular diseases, including coronary artery disease2–4. Faster HR results in decreased shear stress, leading to stimulation of mechanoreceptors located on the surface of the endothelium and upregulation of proatherogenic genes5, effects that can lead to endothelial dysfunction and affect graft outcomes5,6. However, reports on the relationship between elevated HR and cardiac allograft vasculopathy (CAV) development have reported contradictory and inconsistent results7–10. All the studies published on the subject suffer from significant limitations, mainly due to retrospective analysis, the use of coronary angiography as a CAV imaging modality, and the use of single ECG recording as a HR rate analysis tool7–10. Our aim was to perform a large prospective study using highly automated optical coherence tomography (OCT) image analysis and 24-hour ECG monitoring as a means of elucidating the relationship between HR and early progression of CAV.
METHODS
Patients
Between October 2014 and March 2017, 116 subjects, who survived first 12 months after HTx, from the Heart Centre at IKEM, Prague, Czech Republic and the Centre of Cardiovascular and Transplantation Surgery, Brno, Czech Republic were enrolled. The study (clinical trial NCT02503566) complies with the Declaration of Helsinki and was approved by the respective ethics committees. All HTx recipients ≥18 years of age were deemed eligible for inclusion in the study provided they were able and willing to give their informed consent. Exclusion criteria included chronic kidney disease stage ≥IV (glomerular filtration <30 mL/min), unfavourable post-HTx clinical conditions, such as episodes of severe rejection or nosocomial sepsis with prolonged antibiotic treatment during the first month, ongoing need for circulatory support using a ventricular assist device, and acute allograft failure.
Coronary angiography in donors was indicated according to our institutional protocol in men >45 years, and in women >50 years. Coronary angiography was also considered in case of abnormal ECG or echocardiogram, and in case of significant cumulation of risk factors for ischemic heart disease.
ECG monitoring
All patients were continuously monitored during the first month after HTx using ECG telemetry system, which allowed precise 24/7 assessment of HR and/or arrhythmias. 24-hour ECG Holter monitoring was performed twice under standardized conditions (during a hospital stay) at 1 month (M1) after HTx, shortly before the first post-surgical discharge, and at 12 months after HTx (M12).
24-hour ECG Holter recordings were performed using the GE MARS Ambulatory ECG Analysis System (Prague centre) and the Philips Zymed Holter Recorder (Brno centre). The following variables were assessed: mean, maximal, minimal HR, and the total number of atrial and ventricular premature complexes.
Heart rate slowing medication
Patients in our cohort were started on bradycardic medication within the first month after HTx solely at the discretion of the attending physician based on HR, and/or presence of any forms of tachyarrhythmias. The choice and dose of drugs was not prespecified. Patients were given either beta blockers (metoprolol or bisoprolol) and/or ivabradine. HR-slowing (non-dihydropyridine) calcium channel blockers were not administered. All patients were continuously monitored during the first month after HTx using ECG telemetry system, which allowed precise 24/7 assessment of heart rate, and/or arrhythmias. Likewise, during all follow-up visits within the first post-HTx year, bradycardic medication was titrated depending on resting ECG findings.
If bradycardic medication was administered, patients were given either beta blockers (metoprolol or bisoprolol) or ivabradine. For the purpose of analysis, the daily dose of bisoprolol was multiplied by a factor of 20 to obtain the metoprolol-equivalent dosage11.
Blood pressure monitoring and treatment
Everyday measurement of blood pressure under standardized conditions was performed during first month after HTx. Subsequently, blood pressure was routinely assessed during mandatory visits within the first post-HTx year. To tackle arterial hypertension, patients were treated with three main antihypertensive medications – ACE-inhibitors (ramipril, perindopril), dihydropyridine calcium channel blockers (amlodipine, lercanidipine) and beta blockers.
OCT method
Coronary OCT imaging was performed at M1 and M12 after HTx as part of routine surveillance cardiac catheterisation using a commercial intracoronary OCT system (ILUMIEN/DRAGONFLY OPTIS, St. Jude Medical, St. Paul, MN). A 54 mm-long segment of each HTx patient’s left anterior descending (LAD) artery, located within a proximal 100 mm segment, was imaged using automated pullback at 18 mm/sec and 10 frames/mm. Where the LAD artery exhibited unfavourable anatomical characteristics (small calibre, extreme tortuosity, muscle bridge), a 100 mm-long proximal segment of the left circumflex (LCx) or the right coronary artery (RCA) was imaged. Baseline and follow-up OCT-imaged segments mutually overlapped.
Image interpretation
For each frame of all OCT pullbacks, luminal, intimal, and medial surfaces were automatically segmented using a fully three-dimensional LOGISMOS graph-based approach developed at the University of Iowa,12,13 as previously documented14,15. Boundaries were identified as OCT brightness changes showing tissue interfaces between adjacent wall layers. Automatically identified borders were efficiently edited using our Just-Enough-Interaction method adapted for the OCT segmentation environment16,17. This technique allows segmentation errors to be corrected in a 3D fashion on a regional basis, an alternative to contour-by-contour/frame-by-frame manual retracing. This highly accurate, multi-layer model ensures quantitative CAV analysis of every frame of the imaged vessel for both baseline and follow-up image pullbacks. After identifying corresponding vascular landmarks, baseline and follow-up pullback pairs were co-registered, enabling location-specific and fully three-dimensional comparisons of layer-based changes using quantitative indices. Our 3D OCT layer analysis was presented in Figure 1, showing example cross-sectional OCT frames from baseline and follow-up image pullbacks together with ΔIT map for a typical non-progression and progression pullback pairs.
Figure 1: OCT layer analysis in 3D.
OCT layer analysis in 3D, showing example cross-sectional OCT frames from baseline and follow-up image pullbacks together with IT thickness maps and a map showing IT changes (ΔIT) for a typical non-progression (a, b, c, top) and a typical progression (d, e, f, bottom) image pullback pairs. (a,d) Cross-sectional OCT frames from a baseline (M1) pullback sequence – original and intimal & medial layer segmentation. (b,e) Corresponding OCT frame from the follow-up (M12) pullback, axially and rotationally registered. In (a,d) and (b,e) panels, white wedges depict automatically-identified exclusion regions in which layer thickness measurements are not possible; e.g., due to guidewire shadows (6 o’clock in (a,b); 11 o’clock in (d,e)), vessel branches (6–8 o’clock in (d,e)), imaging artifacts, or limited penetration depth of the OCT imaging modality. (c,f) Feature maps of baseline IT, follow-up IT, and intimal thickening (ΔIT) shown for the entire OCT pullback length as a color-coded vessel wall layer thickness, cylindrical vessel wall is shown in an unwrapped fashion. Baseline and follow-up maps are mutually registered and the shadowed areas represent non-measurable exclusion regions shown as wedges in the (a,b) panels.
Quantitative descriptors
Branches and ambiguous areas of the wall were excluded from analysis based on the consensus of two expert cardiologists. Full frames were excluded when appearing in only one of the two registered pullbacks.
The luminal, intimal, and medial layers for each frame were segmented and analysed to obtain average thickness and area measurements. The intima-to-media ratio was calculated for each frame of the analysed pullbacks by dividing frame-specific average intimal thickness (IT) by frame-specific average medial thickness. Mean intimal thickening (ΔIT) was determined as the average difference between IT at M1 and M12 for all co-registered vessel wall locations. While ΔIT can also be assessed locally, the reported ΔIT has always been averaged over the entire registered segment in this work. Maximal intimal thickening was determined as the maximal value of segmental intimal thickening, comparing all 3mm-long segments of the vessel wall. Our 3D OCT layer analysis is presented in Figure 1, showing example of cross-sectional OCT frames from baseline and follow-up image pullbacks together with ΔIT map for a typical non-progression and progression pullback pairs.
The cohort was divided into two groups based on ΔIT observed during the first post-HTx year: a non-progression group (70% of patients, ΔIT ≤0.028 mm) and a progression group (30% of patients, ΔIT >0.028 mm). The threshold of ΔIT=0.028mm was selected arbitrarily with respect natural clustering of the patient cohort.
Statistical analysis
Numerical variables were described as mean values ± standard deviations or median and interquartile ranges (IQR), where appropriate. The coronary morphological changes or patient characteristics changes from M1 to M12 were evaluated using Paired Student’s t-test or Wilcoxon signed-rank test, where appropriate. Categorical variables, presented as counts and percentages, were compared using Fisher’s exact test. Spearman’s rank correlation coefficient was used to evaluate the relationship between 12-month IT progression and HR-related variables. The generalised linear model was used to examine the association between donor age and HR in the presence of HR-slowing medication. Bonferroni correction was used to avoid false positives where several tests were performed simultaneously. R environment was employed for statistical computing, with a P-value of 0.05 used to denote statistical significance.
RESULTS
In total, 116 out of 121 originally recruited patients, were enrolled into the study (5 patients died prior to M12 follow-up visit). Patient characteristics are summarised in Table 1. In this population, 97 left anterior descending (LAD), 7 left circumflex (LCx), and 12 right coronary (RCA) arteries were analysed.
Table 1.
Patient characteristics
| Baseline (n=116) | Follow up (n=116) | p-value | |
|---|---|---|---|
| Age (years) | 50.3±12.5 | - | |
| Men | 90 (78%) | - | |
| Ischaemic cardiomyopathy | 23(20%) | - | |
| Cold ischaemia time (min) | 125 (±50.2) | - | |
| Donor age (years) | 41.2 (±11.7) | - | |
| Donor gender (men) | 79 (68%) | - | |
| Explosive brain death | 86 (74%) | - | |
| BMI (kg/m2) | 26.2 (±4.2) | 28.7 (±4.6) | <0.001 |
| CMV infection within 12 months | 13 (11%) | ||
| Mild rejection within 12 months | 87 (75%) | ||
| Severe rejection* within 12 months | 23 (20%) | ||
| Humoral rejection within 12 months | 3 (3%) | ||
| Systolic blood pressure (mmHg) | 125.9 (±15.9) | 127.2 (±13.7) | 0.41 |
| Diastolic blood pressure (mmHg) | 78.5 (±8.8) | 80.7 (±9.0) | 0.025 |
| Hemoglobin level (g/L) | 111.0 (±12.8) | 131.8 (±15.7) | <0.001 |
| LV ejection fraction (%) | 61.1 (±4.1) | 61.0 (±4.8) | 0.85 |
| Total cholesterol | 4.90 (±1.10) | 4.27 (±1.08) | <0.001 |
| HDL (mmol/l) | 1.51 (±0.42) | 1.23 (±0.44) | <0.001 |
| LDL (mmol/l) | 2.68 (±1.08) | 2.29 (±0.84) | <0.001 |
| TAG (mmol/l) | 1.55 (±0.67) | 1.75 (±1.09) | 0.07 |
| HbA1C (mmol/mol) | 40 (±7.8) | 46.4 (±16.9) | <0.001 |
| Aspirin | 68 (59%) | 80 (69%) | 0.13 |
| Statin | 92 (79%) | 99 (85%) | 0.42 |
| Steroids | 116 (100%) | 97 (84%) | <0.001 |
| Tacrolimus | 112 (97%) | 112 (97%) | 1 |
| Cyclosporin A | 4 (3.4%) | 4 (3.4%) | 1 |
| mTOR inhibitor (everolimus) | 0 (0%) | 11 (10%) | <0.001 |
| Mycophenolate mofetil | 113 (97.4%) | 81 (69.8%) | <0.001 |
| Bisoprolol | 10 (8.6%) | 16 (13.8%) | |
| Ivabradine | 2 (1.7%) | 16 (13.8%) | 0.001 |
| ACE-inhibitors | 22 (19%) | 30 (26%) | 0.20 |
| Dihydropyridine calcium channel blockers | 28 (24%) | 39 (34%) | 0.08 |
| ACE-inhibitors + Dihydropyridine calcium channel blockers | 7 (6%) | 16 (13.8%) | 0.047 |
BMI – body mass index; CMV – cytomegalovirus; HTx – heart transplant; LV – left ventricle; M – month;
min – minutes; mTOR – mammalian target of rapamycin; TAG - triacylglycerols
according to ISHLT classification (≥3A)
Cohort analysis
Based on the 116, M1–M12 pairs of registered OCT pullbacks, the median analyzable angular range was 264° (IQR 234–286°) per frame. No full frames were excluded after image registration; the median overlapping pullback length was 38.7 mm. A total of 43,012 registered frame pairs were analysed, resulting in an average co-registered intimal surface area of 239.7 mm2 per vessel pullback; local descriptors were available for an average of 78,617 co-registered wall locations per pullback.
During the first post-HTx year, we observed highly significant reduction in the mean coronary luminal area and increase in all indices related to IT (Table 2).
Table 2.
Optical coherence tomography findings
| M1 after HTx | M12 after HTx | Change (M12 - M1) | P-value (after Bonferroni correction) | |
|---|---|---|---|---|
| Mean luminal area (mm2) | 9.0±2.5 | 8.0±2.4 | −1.0±1.5 | <0.001 |
| Mean intimal thickness (μm) | 106.5±40.4 | 130.1±53.0 | 23.6± 30.3 | <0001 |
| Maximal intimal thickness (μm) | 174.5±92.3 | 223.3±106.9 | 79.1±71.3 | <0.001 |
| Mean medial thickness (μm) | 82.2±22.0 | 83.7±21.2 | 1.5±11.7 | 1 |
| Mean intimal+medial thickness (μm) | 188.7±58.6 | 213.8±70.2 | 25.1±37.7 | <0.001 |
| Max intimal+medial thickness (μm) | 274.3±107.7 | 324.5±122.0 | 87.0±79.3 | <0.001 |
| Mean intimal/medial ratio | 1.4±0.3 | 1.6±0.4 | 0.3±0.3 | <0.001 |
HTx – heart transplant; M – month; mm – millimetres; μm – micrometres
By separation the cohort into the non-progression and progression subgroups (with ΔIT dichotomy of 0.028 mm), we did not identify any significant CAV predictor among all patient-based, donor-related characteristics and non-immunological risk factors (Table 3).
Table 3.
Predictive factors of CAV progression, with the patient cohort divided into quartiles based on IT progression 12 mo after HTx
| Non-progression (70 %) n=81 | Progression (30 %) n=35 | P-value (after Bonferroni correction) | |
|---|---|---|---|
| Patient age (years) | 50.9±12.7 | 49.3±12.2 | 1 |
| Men (%) | 78 | 77 | 1 |
| History of ICMP (%) | 21 | 17 | 1 |
| Donor age (years) | 39.7±12.0 | 44.4±10.5 | 1 |
| Donor gender (men) (%) | 68 | 63 | 1 |
| Cold ischaemia time (min) | 137±55 | 129±52 | 1 |
| Explosive brain death (%) | 74 | 68 | 1 |
| Mild rejection (%) | 72 | 83 | 1 |
| Severe rejection*(%) | 16 | 29 | 1 |
| Humoral rejection (%) | 4 | 3 | 1 |
| CMV infection (%) | 10 | 14 | 1 |
| Systolic blood pressure (mmHg) M1/M12 | 126.7±16.6 | 124.1±14.3 | 1/1 |
| 126.8±14.2 | 127.9±12.8 | ||
| Diastolic blood pressure (mmHg) M1/M12 | 78.5±8.5 | 78.6±9.4 | 1/1 |
| 80.4±9.6 | 81.3±7.7 | ||
| BMI (kg/m2) M1/M12 | 26.4±3.9 | 25.9±5.0 | 1/1 |
| 29.1±4.4 | 27.4±4.9 | ||
| Total cholesterol (mmol/l) M1/M12 | 4.79±1.06 | 5.16±1.18 | 1/1 |
| 4.19±1.09 | 4.44±1.05 | ||
| HDL (mmol/l) M1/M12 | 1.50±0.43 | 1.55±0.41 | 1/0.83 |
| 1.16±0.37 | 1.38±0.55 | ||
| LDL (mmol/l) M1/M12 | 2.58±0.87 | 2.92±0.92 | 1/1 |
| 2.25±0.90 | 2.39±0.68 | ||
| TAG (mmol/l) M1/M12 | 1.57±0.73 | 1.52±0.51 | 1/1 |
| 1.76±0.93 | 1.73±1.40 | ||
| HbA1C (mmol/mol) M1/M12 | 39.7±7.8 | 40.9±7.8 | 1/1 |
| 46.5±17.4 | 46.2±15.9 | ||
| ACE-inhibitors (%) M1/M12 | 25.9 | 2.9 | 0.60/1 |
| 29.6 | 17.1 | ||
| Dihydropyridine calcium channel blockers (%) M1/M12 | 23.5 | 25.7 | 1/1 |
| 38.3 | 22.9 | ||
| ACE-inhibitors + Dihydropyridine calcium channel blockers (%) M1/M12 | 8.6 | 0 | 1/1 |
| 16.1 | 8.6 | ||
| Heart rate mean M1/M12 | 79 ± 9 | 81 ± 7 | 1/1 |
| 84 ± 9 | 82 ± 8 |
BMI – body mass index; CMV – cytomegalovirus; HbA1C – glycated haemoglobin; HTx – heart transplant;
ICMP – ischaemic cardiomyopathy; M – month; TAG – triacyglycerols; VAD – ventricular assist device.
according to ISHLT classification (≥3A)
Table 4 summarises 24-hour ECG Holter findings at M1 and M12 after HTx. No significant relationships were found between ΔIT and either baseline or follow-up Holter-based variables (Table 5).
Table 4.
Predictive factors of CAV progression, with the patient cohort divided into quartiles based on IT progression 12 mo after HTx
| M1 after HTx | M12 after HTx | Change (M12-M1) | P-value (after Bonferroni correction) | |
|---|---|---|---|---|
| HR min | 68±9 | 68±8 | −1±9 | 1 |
| HR max | 103±16 | 113±16 | 9±21 | <0.001 |
| HR mean | 80±8 | 83±8 | 2.5±9.9 | 0.06 |
| APC/hour | 1.9 (0.7~3.9) | 0.3 (0.1~0.9) | −1.1 (−2.5~−0.2) | <0.001 |
| VPC/hour | 0.5 (0.2~1.3) | 0.1 (0.0~0.5) | −0.2 (−0.8~0.08) | 0.181 |
APC – atrial premature complex; VPC – ventricular premature complex; HTx – heart transplant; M – month
Table 5.
Correlation between intimal thickness progression and individual 24- hour ECG Holter variables at M1 and M12
| M1 after HTx | M12 after HTx | |
|---|---|---|
| HR min | R=−0.00, p=0.99 | R=−0.15, p=0.13 |
| HR max | R=0.16, p=0.10 | R=−0.02, p=0.84 |
| HR mean | R=0.02, p=0.83 | R=−0.13, p=0.18 |
| APC/hour | R=0.18, p=0.053 | R=−0.03, p=0.79 |
| VPC/hour | R=0.06, p=0.54 | R=−0.12, p=0.23 |
APC – atrial premature complex; ECG – electrocardiogram; HR – heart rate; HTx – heart transplant; IT – intimal thickness; M – month; VPC – ventricular premature complex
Associations between HR-slowing medication and ΔIT at M12 after HTx are given in Table 6. There were no significant differences in ΔIT between treated and non-treated patients. However, we observed a mild inverse association between beta blocker dosage at 12 months and ΔIT (R=−0.20, p=0.035). It is well known that mean heart rate inversely correlates with donor age early after HTx (R=−0.26, p<0.01 at M1 in our cohort). This probably resulted in escalation of beta blocker therapy in patients after HTx from younger donors. Consequently, beta blocker dosage correlated with donor age even more tightly (R=−0.41, p<0.0001) at M12. However, donor age itself was not related to ΔIT, i.e. development of CAV. Also, recipient age was unrelated to both beta blocker dosage and ΔIT.
Table 6.
Associations between intimal thickness progression (ΔIT) and HR-slowing medication at M1 and M12
| M1 after HTx | M12 after HTx | |||
|---|---|---|---|---|
| Treatment=YES | Treatment=NO | Treatment=YES | Treatment=NO | |
| Beta blockers | 21.8±25.1(n=60) | 25.6±35.2 (n=56) | 21.3±28.3 (n=77) | 27.9±34.1 (n=39) |
| P | 0.50 | 0.32 | ||
| Dose of beta blockers (mg) vs. ΔIT | R=−0.04, p=0.687 | R=−0.20, p=0.035 | ||
| Ivabradine | 24.3±27.3 (n=2) | 23.6±30.5 (n=114) | 34.5±35.7 (n=16) | 21.9±29.2 (n=100) |
| P | 0.98 | 0.19 | ||
| Dose of ivabradine (mg) vs. ΔIT | R=0.02, p=0.826 | R=0.17, p=0.069 | ||
| Bradycardic medication (beta blocker, ivabradine, beta blocker + ivabradine) | 22.1±25.1 (n=61) | 25.3±35.4 (n=55) | 24.3±30.8(n=86) | 21.8±29.4(n=30) |
| p | 0.58 | 0.70 | ||
Numbers are ΔIT in micrometres
HTx- heart transplant; M- month; mg- milligram
DISCUSSION
The principal findings of our study can be summarised as follows: (1) there was significant progression of IT as early as within 12 months after HTx; (2) increased HR was not associated with early CAV development; (3) higher doses of beta blockers at 12 months after HTx are possibly a protective factor in IT progression.
CAV-related graft failure accounts for the majority of patient mortality within 5 to 10 years, surpassing the contributions of malignancies and infections18,19. Rapid development of CAV is diagnosed in a high proportion of HTx patients, occurring as early as within one year after the index procedure15,18. Early identification of such patients is of extreme importance, since it can favourably affect CAV-related allograft failure rates and the subsequent need for re-transplantations19. Furthermore, elevated resting HR after HTx might be an independent, and relatively easily modifiable, risk factor for the development of CAV or other systemic cardiovascular diseases20.
Rapid development is a hallmark of CAV, with major changes in IT typically observed within year 1, less so in subsequent years21. After registering overlapping baseline and follow-up images of OCT pullbacks, we were able to axially and rotationally match 387 frames (IQR: 321–438) per pullback on average. We detected significant intimal thickening as early as within one year after HTx, a finding in agreement with previous studies18,21. We want to stress out that our analysis is not using maximal intimal thickening – rather we employ intimal thickening averaged over the entire pullback length. Most of all previous studies illustrated CAV in a very simple manner, as they chose the thickest single site intimal thickness from the whole pullback as a CAV surrogate. Compared to those studies, our highly accurate and sophisticated 3D OCT analysis, enabled us to evaluate progression of IT frame by frame. Hence, we report detailed and accurate information about diffuse CAV progression of the whole vessel, as this approach better characterize pattern of this disease.
Despite the significant early IT progression observed in our relatively large cohort of 116 patients, no significant risk factors of CAV development were identified in respect of selected donor-related characteristics or non-immunological risk factors. This confirms the indisputable role of immune mechanisms in CAV development, such as donor-specific anti-HLA antibodies10,22, which are particularly prevalent in early stages of CAV disease and characterised by diffuse concentric fibrous intimal hyperplasia23. Traditional cardiovascular risk factors contribute to the CAV progression rather in the later stages of CAV development (beyond 1 year after HTx), which are characterized by the diffuse incorporation of lipids into the vessel wall and resemble classical processes in ischaemic heart disease24.
Autonomic denervation during the HTx procedure results in altered electrophysiological properties of the transplanted heart. During parasympathetic denervation, suppression of sinoatrial node automaticity is eliminated, and tachycardia becomes a typical post-HTx clinical feature1. Several mechanisms have recently been proposed as potentially responsible for various tachycardia-related detrimental effects. An elevated HR enhances the magnitude and frequency of the tensile stress imposed on the arterial wall and prolongs the exposure of coronary endothelium to the systolic low and oscillatory shear stress6. Consequently, causing arterial wall injury and endothelial cell dysfunction. In atherosclerotic phenotypes of coronary vessels, a faster HR increases the risk of plaque rupture and contributes to coronary thrombosis1,4.
Current data on the effects of increased resting HR on all-cause mortality in HTx patients and CAV development is contradictory. While there is relatively uniform evidence of tachycardia-related effects on mortality5,25–27, the impact on CAV development is less clear7–10. Due to small sample sizes, retrospective and observational approaches, and overall design, previous studies7–9have failed to produce convincing data on the role of tachycardia in CAV development. Such efforts are particularly hampered by the use of low-resolution coronary angiography as an imaging modality for CAV assessment and a reliance on single ambulatory ECG recordings for HR analysis. In contradiction to previous studies on this topic, recent data published by Liebo et al.10 reveal that a HR >95 beats/min 1 year after HTx is associated with increased incidence of CAV. In this context, the results of our prospective study – utilising 24-hour ECG Holter monitoring for HR assessment and OCT imaging for detailed and precise analysis of early ΔIT – are of major importance.
Our 24-hour ECG Holter monitoring results showed no significant changes in paired examinations. The one exception was maximal HR, which increased over time despite increased doses of bradycardic medication, perhaps indicating more vigorous activity of patients, or partial re-innervation as early as within one year after HTx. According to evidence from Awad et al.28, sympathetic re-innervation occurs within 5 to 6 months after HTx, mostly appearing in the left ventricular and sinoatrial node regions of the allograft.
None of our 24-hour ECG Holter variables, including mean HR, were associated with ΔIT 12 months after HTx. This finding thus refutes the hypothesis that high HR is involved in the development of CAV.
The association we observed between younger donor age and higher HR at M1 after HTx is in agreement with previous studies9,10. A likely explanation for such a finding is sinus node dysfunction, which increases in prevalence during ageing and commonly occurs in older adults, as demonstrated by the decline in intrinsic HR29.
Despite its detrimental effect on the exercise capacity of HTx patients30, beta-blocker treatment might be of benefit. Compared to other studies5,26, a relatively high proportion of patients in our cohort were prescribed beta blockers (51.7 % at M1, 66.4 % at M12). There are three possible explanations for this: (1) Assumption that increased HR predicts worse outcomes in cardiovascular disease5,25–27 could be associated with over-prescription of beta blockers to achieve resting HR <95–100/min; (2) Favoured beta blockers for the treatment of symptomatic atrial ectopy which can be indirectly suggested by higher frequency of atrial premature complexes on 24-hour ECG Holter observed at 1 month in patients treated with beta blockers; (3) Preference of beta blockers in combination treatment of arterial hypertension.
Interestingly, higher doses of beta blockers at 12 months after HTx were weakly associated with lower IT progression. We did not identify any factor that could indirectly be responsible for this relationship. Padilla et al. demonstrated that activation of the sympathetic nervous system leads to an immediate alteration in retrograde shear stress, which is associated with decreased endothelial function31. Their findings lend further weight to our speculation that the direct sympatholytic effect of beta blockers may play a role in preventing CAV development. In light of the negative association between higher HR and CAV progression, the role of beta blockers certainly warrants further investigation.
Study limitations
Our HTx patients were treated with beta blockers at the discretion of the attending physician based on HR, presence of any forms of arrhythmias, and hypertension. A randomised study would have been better in assessing HR effects on CAV progression.
Relatively high number of patients were treated with bradycardic medication. Our reliance on this treatment may have resulted in the narrow range of HR, possibly reducing the power of the study to detect the relationship between HR and CAV development.
First OCT imaging results were only available at M1 after HTx. Thus, we were unable to differentiate between donor-transmitted coronary artery disease and de novo progression of CAV during the first post-HTx month. Due to the additional need to administer angiographic contrast and the associated risk of renal function deterioration, only one vessel per patient was examined by OCT. Due to the limited depth of OCT tissue penetration (1.5 to 3 mm), layered coronary structures, especially the external elastic lamina, were not always visible in some patients with extensive donor-transmitted coronary disease, thus impacting on our ability to precisely detect changes in the coronary vasculature.
The patients with CAV progression were defined by arbitrary increase in mean intima thickness > 0.028 mm. At this stage of research, we can only rely on information about very early IT changes. The decision to use the 70%−30% split of the groups was motivated by the well-known percentages of long-term CAV effects without claiming that the same patients who demonstrate early IT thickening will ultimately be the same as those with advanced CAV later.
CONCLUSIONS
Our prospective 2-centre trial using highly automated 3D graph-based OCT analysis did not confirm a direct association between mean HR and 12-month progression of CAV. Nevertheless, the role of HR-slowing medication such as beta blockers warrants further investigation, with our results indicating that they may play a protective role in early CAV development.
Acknowledgements
This project was supported by research grants from the Czech Ministry of Health (16-27465A), MH-CZ-DRO (IKEM-IN 00023001), and NIH (R01-EB004640).
Footnotes
Disclosure statement
The authors have no conflicts of interest to disclose.
Data Availability Statement
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.
References
- 1.Sekar B, Critchley WR, Williams SG, et al. Should we consider heart rate reduction in cardiac transplant recipients? Clin Cardiol. 2013. February;36(2):68–73. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Palatini P, Thijs L, Staessen JA, et al. Predictive value of clinic and ambulatory heart rate for mortality in elderly subjects with systolic hypertension. Arch Intern Med. 2002;162(20):2313–2321. [DOI] [PubMed] [Google Scholar]
- 3.Cooney MT, Vartiainen E, Laatikainen T, et al. Elevated resting heart rate is an independent risk factor for cardiovascular disease in healthy men and women. Am Heart J. 2010;159(4):612–619.e3. [DOI] [PubMed] [Google Scholar]
- 4.Fox K, Borer JS, Camm AJ, et al. Resting heart rate in cardiovascular disease. J Am Coll Cardiol. 2007;50(9):823–830. [DOI] [PubMed] [Google Scholar]
- 5.Ciarka A, Lund LH, Van Cleemput J, et al. Effect of heart rate and use of beta blockers on mortality after heart transplantation. Am J Cardiol. 2016;118(12):1916–1921. [DOI] [PubMed] [Google Scholar]
- 6.Giannoglou GD, Chatzizisis YS, Zamboulis C, et al. Elevated heart rate and atherosclerosis: an overview of the pathogenetic mechanisms. Int J Cardiol. 2008;126(3):302–312. [DOI] [PubMed] [Google Scholar]
- 7.Ambrosi P, Kreitmann B, Habib G. Does heart rate predict allograft vasculopathy in heart transplant recipients? Int J Cardiol. 2010;145(2):256–257. [DOI] [PubMed] [Google Scholar]
- 8.Gullestad L, Ross H, Myers J, et al. Importance of decreased heart rate in predicting transplant coronary artery disease. Clin Transplant. 1997;11(6):628–632. [PubMed] [Google Scholar]
- 9.Olmetti F, Pinna GD, Maestri R, et al. Heart rate and cardiac allograft vasculopathy in heart transplant recipients. J Heart Lung Transplant. 2011;30(12):1368–1373. [DOI] [PubMed] [Google Scholar]
- 10.Liebo M, Newman J, Joshi A, et al. Elevated heart rate following heart transplantation is associated with increased graft vasculopathy and mortality. J Card Fail. 2019. April;25(4):249–256. [DOI] [PubMed] [Google Scholar]
- 11.Abraham WT. Switching to evidence-based once-daily beta-blockers for improved adherence to medication across the continuum of post-myocardial infarction left ventricular dysfunction and heart failure. Congest Heart Fail. 2008;14(5):272–280. [DOI] [PubMed] [Google Scholar]
- 12.Lee K, Zhang L, Abramoff MD, et al. Fast and memory efficient LOGISMOS graph search for intraretinal layer segmentation of 3D macular OCT scans. Proc SPIE. 2015;9413:94133X/1–6. [Google Scholar]
- 13.Zhang H, Abiose AK, Gupta D, et al. Novel indices for left-ventricular dyssynchrony characterization based on highly automated segmentation from real-time 3-D echocardiography. Ultrasound Med Biol. 2013;39(1):72–88. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Chen Z, Pazdernik M, Zhang H, et al. Quantitative 3D analysis of coronary wall morphology in heart transplant patients: OCT-assessed cardiac allograft vasculopathy progression. Med Image Anal. 2018;50:95–105. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Pazdernik M, Chen Z, Bedanova H, et al. Early detection of cardiac allograft vasculopathy using highly automated 3-dimensional optical coherence tomography analysis. J Heart Lung Transplant. 2018;37(8):992–1000. [DOI] [PubMed] [Google Scholar]
- 16.Sun S, Sonka M, Beichel RR. Graph-based IVUS segmentation with efficient computer-aided refinement. IEEE Trans Med Imaging. 2013;32(8):1536–1549. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Chen Z Novel quantitative description approaches assessing coronary morphology and development (PhD thesis). Iowa City, IA: University of Iowa; 2016. [Google Scholar]
- 18.Clemmensen TS, Holm NR, Eiskjær H, et al. Detection of early changes in the coronary artery microstructure after heart transplantation: A prospective optical coherence tomography study. J Heart Lung Transplant. 2017;37(4):486–495. [DOI] [PubMed] [Google Scholar]
- 19.Matsuo Y, Cassar A, Yoshino S, et al. Attenuation of cardiac allograft vasculopathy by sirolimus: Relationship to time interval after heart transplantation. J Heart Lung Transplant. 2013;32(8):784–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Ghafourian K, Kobashigawa JA. Slowing sinus tachycardia in heart transplant recipients: Is it time? Rev Esp Cardiol (Engl Ed). 2015;68(11):921–923. [DOI] [PubMed] [Google Scholar]
- 21.Tsutsui H, Ziada KM, Schoenhagen P, et al. Lumen loss in transplant coronary artery disease is a biphasic process involving early intimal thickening and late constrictive remodeling: Results from a 5-year serial intravascular ultrasound study. Circulation. 2001;104(6):653–657. [DOI] [PubMed] [Google Scholar]
- 22.Chih S, Chong AY, Mielniczuk LM, et al. Allograft vasculopathy: The Achilles’ heel of heart transplantation. J Am Coll Cardiol. 2016;68(1):80–91. [DOI] [PubMed] [Google Scholar]
- 23.Khandhar SJ, Yamamoto H, Teuteberg JJ, et al. Optical coherence tomography for characterization of cardiac allograft vasculopathy after heart transplantation (OCTCAV Study). J Heart Lung Transplant. 2013;32(6):596–602. [DOI] [PubMed] [Google Scholar]
- 24.Cassar A, Matsuo Y, Herrmann J, et al. Coronary atherosclerosis with vulnerable plaque and complicated lesions in transplant recipients: New insight into cardiac allograft vasculopathy by optical coherence tomography. Eur Heart J. 2013;34(33):2610–2617. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Castel MA, Roig E, Rios J, et al. Long-term prognostic value of elevated heart rate one year after heart transplantation. Int J Cardiol. 2013;168(3):2003–2007. [DOI] [PubMed] [Google Scholar]
- 26.Wachter SB, McCandless SP, Gilbert EM, et al. Elevated resting heart rate in heart transplant recipients: innocent bystander or adverse prognostic indicator? Clin Transplant. 2015;29(9):829–834. [DOI] [PubMed] [Google Scholar]
- 27.Bedanova H, Fila P, Horvath V, et al. Association between the heart rate early after heart transplantation and the long-term clinical outcomes. Bratisl Lek Listy. 2017;118(6):366–369. [DOI] [PubMed] [Google Scholar]
- 28.Awad M, Czer LS, Hou M, et al. Early denervation and later reinnervation of the heart following cardiac transplantation: A review. J Am Heart Assoc. 2016;5(11):e004070. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Moghtadaei M, Jansen HJ, Mackasey MJ, et al. The impacts of age and frailty on heart rate and sinoatrial node function. J Physiol. 2016;594(23):7105–7126. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Verani MS, Nishimura S, Mahmarian JJ, et al. Cardiac function after orthotopic heart transplantation: response to postural changes, exercise, and beta-adrenergic blockade. J Heart Lung Transplant. 1994;13(2):181–193. [PubMed] [Google Scholar]
- 31.Padilla J, Young CN, Simmons GH, et al. Increased muscle sympathetic nerve activity acutely alters conduit artery shear rate patterns. Am J Physiol Heart Circ Physiol. 2010;298(4):H1128–H1135. [DOI] [PMC free article] [PubMed] [Google Scholar]