Abnormal invasive hemodynamics in heart transplant recipients: A single-center, retrospective study

Abstract

BACKGROUND

Few studies have quantified invasive hemodynamic parameters in post heart transplant recipients.

AIM

To report the incidence of abnormal hemodynamics in heart transplant recipients at 1-year and 3-year post-transplant and determine if there was any correlation with recipient and donor characteristics.

METHODS

Data from 279 consecutive heart transplant recipients from 2007 through 2020 were analyzed. Clinical variables regarding both recipients and donors as well as hemodynamic variables obtained via right heart catheterization during 1-year and 3-year annual testing were recorded. Simple and multiple linear regression tests were used to determine how recipient and donor variables influenced hemodynamic parameters at 1-year and 3-year.

RESULTS

Data were available for 260 patients and 224 patients at 1-year and 3-year post-transplant respectively. At 1-year, abnormal hemodynamic parameters were common with 24% patients having right atrial pressure (RAP) > 10 mmHg, 52% with mean pulmonary artery pressure > 20 mmHg, and 12% with pulmonary capillary wedge pressure (PCWP) > 18 mmHg. Similar abnormalities were noted at 3-year post-transplant. Recipient body mass index (BMI) demonstrated the strongest correlation with all 3 variables at both 1-year and 3-year by multivariate linear regression analysis (P < 0.001 for both). Both donor age and predicted heart mass difference between recipient and donor were significantly linked to RAP and PCWP at 1-year but did not predict any variables at 3-year post-transplant.

CONCLUSION

Abnormal hemodynamics are common at 1-year and 3-year post-transplant and are associated with recipients with high BMI.

Key Words: Heart transplantation; Right heart catheterization; Obesity; Heart failure; Hemodynamics

Core Tip: Our research demonstrates that abnormal invasive hemodynamics are commonly observed in heart transplant recipients at 1-year and 3-year post-transplantation. Obesity as reflected by body mass index was strongly associated with elevated right atrial, mean pulmonary artery, and pulmonary capillary wedge pressure in transplant recipients. Further research is needed to determine if the development of abnormal hemodynamics predicts transplant mortality and if behavioral modifications and pharmacologic therapy can reverse observed hemodynamic abnormalities.

INTRODUCTION

As an option for patients with end-stage heart failure, heart transplantation is associated with excellent survival[1]. Transplant recipients however often experience limitations in their functional capacity as reflected by reduction in peak oxygen consumption with exercise[2,3]. This limitation exists even with allograft systolic function that appears normal on echocardiography, leading to speculation that chronotropic incompetence, diastolic dysfunction, and right ventricular dysfunction may all have a role[4-6]. Previous studies have examined abnormalities of diastolic function in heart transplant patients, primarily as assessed by echocardiography, with few examining invasive hemodynamics post-transplant[7,8]. Tallaj et al[9] reported an incidence of diastolic dysfunction of 12% at one-year post-transplant defined by elevated right atrial or pulmonary capillary wedge pressure (PCWP). However, there has been limited analysis of invasive hemodynamics beyond the first year post-transplant[10]. The purpose of this study was to report the incidence of abnormal hemodynamics in heart transplant recipients at 1-year and 3-year post-transplant. Furthermore, we sought to determine if there were donor or recipient clinical factors that correlated with hemodynamic variables.

MATERIALS AND METHODS

We retrospectively reviewed the medical records of consecutive patients undergoing orthotopic heart transplantation at our institution from January 2009 through December 2020. Combined heart/lung transplants were excluded from analysis but patients undergoing heart/kidney transplantation were included (5% of heart transplant recipients). One patient underwent re-transplantation during the study period and the second transplant was excluded from analysis. Re-transplantation occurred 5 years following the original transplant. Baseline recipient characteristics were recorded and clinical parameters from the donor were obtained from DonorNet, an electronic database maintained by the United Network for Organ Sharing (UNOS). Given the time period of the study, all donors were procured as donation after brain death and traditional cold storage technique was utilized. Predicted heart mass (PHM) was calculated for both recipients and donors according to accepted methods[11]. PHM difference was calculated as follows: Donor PHM–recipient PHM)/recipient PHM × 100.

Our institution’s post-transplant protocol has patients undergo combined left and right heart catheterization on a yearly basis for 5 years starting at 1-year post-transplantation. The following hemodynamics were recorded during right heart catheterization: (1) Right atrial pressure (RAP); (2) Pulmonary artery pressure (PAP) including systolic, diastolic and mean; and (3) PCWP. Cardiac output and index were included if performed via the thermodilution method; indirect Fick method was not included in our analysis. Hemodynamic data were included in our analysis if they were obtained within 30 days of the patient’s 1-year transplant anniversary, and within 60 days of the 3-year anniversary. The severity of coronary artery vasculopathy (CAV) was graded according to the 2010 International Society for Heart and Lung Transplantation guidelines[12]. The presence of donor specific antibodies (DSA) was noted. Body weight and body mass index (BMI) of the transplant recipient was noted at the time of transplant, and subsequently at 1-year and 3-year post-transplant. We initially sought to examine the incidence of multiple episodes of treated acute cellular rejection and any episode of antibody mediated rejection in our study population. However, the number of survivors who experienced these episodes was too small for meaningful analysis. The research was approved by the University of Kentucky Institutional Review Board.

Statistical analysis

Continuous variables were reported as mean ± SD. Comparisons between variables measured at different time intervals were made using parametric t tests. Simple and multiple linear regression tests were used to determine how recipient and donor age, donor and recipient BMI, ischemic time, recipient CAV, DSA, and PHM difference between donor and recipient influenced hemodynamic variables at years 1 and 3. Data were analyzed in MATLAB (Ver2023B, The MathWorks, Natick, MA, United States). A P value < 0.05 was considered statistically significant.

RESULTS

A total of 279 heart transplant recipients were identified during the study period (Table 1). Clinical follow-up at 1-year post-transplant was available for 261 patients (18 deaths within 1-year; 93% survival). Hemodynamic data was available for 260 of these 261 patients. At 3-year post-transplant follow-up, clinical data was available for 230 patients (41 deaths; 5 transferred to another institution; and 3 lost to follow-up). Hemodynamic data was available for 224 of these patients at 3-year follow-up.

Table 1 Clinical characteristics of recipients and cardiac donors at the time of transplant, n (%).

Recipients	n = 279
Age (years)	52 ± 13
Male sex	221 (79)
White	232 (83)
Black	41 (15)
Hispanic	4 (2)
Asian	2 (1)
Height (cm)	175 ± 10
Weight (kg)	86 ± 18
BMI (kg/m2)	28 ± 5
Ischemic cardiomyopathy	107 (38)
Presence of left ventricular assist device/total artificial heart	108 (39)
Donors
Age (years)	34 ± 11
Age ≥ 40 years	86 (31)
Male sex	168 (60)
Height (cm)	174 ± 10
Weight (kg)	90 ± 22
BMI (kg/m2)	30 ± 8
Ischemic time (minute)	219 ± 64
Male recipient/male donor	154 (55)
Male recipient/female donor	67 (24)
Female recipient/female donor	44 (16)
Female recipient/male donor	14 (5)
Donor-recipient PHM difference	1 ± 16
PHM difference < -15%	31 (11)

Numerical variables reported as mean and SD. BMI: Body mass index; PHM: Predicted heart mass.

Our patient population was predominantly male and given our regional demographics was primarily White with elevated BMI. Based upon BMI > 30 kg/m², 99 patients (35% study population) were classified as obese at the time of transplant, with another 99 patients with BMI 25 to 30 kg/m² (overweight). Five patients were under-weight (BMI < 18.5 kg/m²) at the time of transplant. Durable mechanical circulatory support was present in 39% of patients, primarily left ventricular assist devices.

Donor age was > 40 years in slightly more than 30% of our patient population with 10% having donor age > 50 years. Based upon BMI, obesity was prevalent in our donor population. Donors with BMI > 35 kg/m² was used in 22% of cases with 11% having BMI > 40 kg/m². Female donor to male recipient occurred in 24% of our transplant cohort. Under-sizing the donor heart by greater than 15% based upon PHM difference has been found to be an adverse prognostic variable in heart transplantation and was found in 11% of our patients. One and 3-year survival in our cohort for these patients (n = 31) was 94% and 87%, respectively.

Weight and BMI at 1-year did not demonstrate a significant increase from the time of transplant (Table 2). Three-year body weight and BMI were both significantly increased compared to both time of transplant and 1-year. At 3 years, 107 patients (47% of patients with available 3-year data) experienced weight gain of 10 kg with 76 patients (33%) having over 20 kg weight gain from transplant. Depressed left ventricular ejection fraction (< 50%) was observed in very few patients: 7 patients (2%) at 1-year and 5 patients (2%) at 3-year. Significant CAV (grade 2 or 3) was also quite uncommon.

Table 2 Clinical outcomes of heart transplant recipients at 1-year and 3-year post-transplant.

	At transplant	1-year post transplant (n = 260)	3-year post-transplant (n = 230)
Survival	N/A	93%	85%
Weight (kg)	86 ± 18	88 ± 19	91 ± 20a
Body mass index	28 ± 5	28 ± 5	29 ± 6a
> 30 kg/m2	35%	36%	40%
Left ventricular ejection fraction (%)	N/A	59 ± 5	60 ± 6
< 50%	N/A	2%	2%
Presence of donor specific antibodies	N/A	23%	26%
Coronary artery vasculopathy
Grade 0	N/A	80%	75%
Grade 1	N/A	16%	18%
Grade 2-3	N/A	0%	4%

^aP < 0.05.

N/A: Not applicable.

Invasive hemodynamics

Right heart catheterization results are shown in Table 3. Elevated PAPs were common with slightly more than 50% having a mean PAP > 20 mmHg at 1-year and 3-year post-transplant. Elevation of PCWP was more common than elevation of RAP. PCWP was the only hemodynamic variable that was significantly increased at 3-year post-transplant compared to 1-year (P = 0.02). PCWP ≥ 18 mmHg was observed in 12% of patients at 1-year and 15% at 3-year. There were no differences in hemodynamic variables between male and female recipients, at both 1-year and 3-year. There was no significant difference in any hemodynamic variable in male recipients whether they received a male vs female donor.

Table 3 Hemodynamics by right heart catheterization at 1-year and 3-year post-transplant.

	1-year post transplant (n = 260)	3-year post-transplant (n = 224)
Right atrial pressure (mm Hg)	7.2 ± 4.1	7.2 ± 4.2
< 10 mmHg	76%	72%
10-14 mmHg	19%	23%
≥ 15 mmHg	5%	5%
Systolic PAP (mm Hg)	31.6 ± 8.5	30.7 ± 7.6
Diastolic PAP (mm Hg)	14.0 ± 4.7	14.5 ± 5.1
Mean PAP (mm Hg)	21.4 ± 5.8	21.3 ± 5.7
> 20 mmHg	52%	55%
Pulmonary capillary wedge pressure (mm Hg)	11.8 ± 4.7	12.3 ± 4.7a
≤ 12 mmHg	62%	55%
13-17 mmHg	26%	30%
≥ 18 mmHg	12%	15%
Cardiac output (L/minute)	5.9 ± 1.4	5.6 ± 1.2
Cardiac index (L/minute/m2)	2.9 ± 0.6	2.6 ± 0.5
Pulmonary vascular resistance (Wood units)	1.7 ± 0.7	1.7 ± 0.7
> 2 Wood units	27%	29%

^aP < 0.05 3-year vs 1-year post-transplant.

PAP: Pulmonary artery pressure.

Univariate and multivariate analysis

We investigated correlation between clinical variables to hemodynamic variables at 1-year and 3-year post-transplant. The strongest correlation was observed with recipient BMI which had significant correlation with RAP, mean PAP, and PCWP at both 1-year and 3-year respectively (P < 0.001 for all) (Figure 1). Weaker but still significant correlations with mean PAP and PCWP were observed with both donor age and PHM difference at both 1-year and 3-year. Donor age also demonstrated significant correction with RAP at both time points. Donor BMI had a significant correlation with mean PAP at 3-year, but with no other variables.

Figure 1 Correlations between invasive hemodynamics and body mass index. A-F: Mean pulmonary artery pressure, pulmonary capillary wedge pressure and mean right atrial pressure plotted against body mass index at year 1 (top row) and year 3 (bottom row). Panels show correlation coefficients and P values. BMI: Body mass index; PAP: Pulmonary artery pressure; PCWP: Pulmonary capillary wedge pressure; RAP: Right atrial pressure.

With multivariate analysis, BMI remained a significant predictor of RAP, mean PAP, and PCWP at both 1-years and 3-year post-transplant (P < 0.001 for all) (Table 4). Both donor age and PHM difference were significantly linked to RAP and PCWP at 1-year but were not associated with any hemodynamic variable at 3-year. At 3-year post-transplant, the presence of CAV was associated with mean PAP and the presence of DSA were associated with PCWP. Recipient age, donor BMI, and ischemic time were not associated with any hemodynamic variables at either 1-year or 3-year post-transplant.

Table 4 P values for multivariate analysis for correlation between clinical variables and right atrial pressure, mean pulmonary artery pressure, and pulmonary capillary wedge pressure at 1-year and 3-year post transplant.

	1-year			3-year
	RAP	Mean PAP	PCWP	RAP	Mean PAP	PCWP
Recipient age	0.524	0.318	0.485	0.927	0.054	0.720
Recipient BMI	< 0.001	< 0.001	< 0.001	< 0.001	< 0.001	< 0.001
Donor age	0.022	0.057	0.017	0.288	0.591	0.168
Donor BMI	0.385	0.342	0.354	0.839	0.896	0.852
Ischemic time	0.300	0.972	0.496	0.961	0.403	0.847
PHM difference	0.036	0.120	0.008	0.455	0.664	0.377
Coronary artery vasculopathy	0.312	0.076	0.037	0.443	0.023	0.237
Donor specific antibodies	0.337	0.803	0.519	0.086	0.114	0.041

BMI: Body mass index; PAP: Pulmonary artery pressure; PCWP: Pulmonary capillary wedge pressure; RAP: Right atrial pressure; PHM: Predicted heart mass.

DISCUSSION

In this paper we demonstrate that abnormal hemodynamics including elevation of RAP, mean PAP, and PCWP are common in heart transplant recipients at both 1-year and 3-year post-transplant. On multivariate analysis, recipient BMI was significantly correlated with RAP, mean PAP, and PCWP at both 1-year and 3-year post-transplant. While donor age was significantly linked to RAP and PCWP, and almost (P = 0.057) with mean PAP at 1-year, it was not significant with any variable at 3-year post-transplant. Our results indicate that obesity may have a prominent role in the development of abnormal hemodynamics in heart transplant recipients.

Heart transplantation has demonstrated excellent survival with median survival close to 12 years[13]. In the United States, UNOS maintains a database that allows for clinical donor and recipient characteristics to be analyzed to predict recipient survival, episodes of rejection, and opportunistic infections. Measures of diastolic dysfunction, cardiopulmonary performance, and invasive hemodynamics are not recorded and may not be routinely assessed by all transplant programs. Heart transplant recipients may have sub-optimal exercise tolerance after transplant despite normal left ventricular function and absence of CAV[14]. Research that has analyzed invasive hemodynamics in post-transplant patients are few in number. Tallaj et al[9] used hemodynamics to define diastolic dysfunction in heart transplant recipients and found an incidence of 12% at 1-year. Another analysis utilized hemodynamic measurements during exercise to observe that diastolic dysfunction was common in patients 5 years post-transplant but was limited to 25 patients[8]. Previous research has not correlated abnormal hemodynamics with clinical recipient and donor variables. Our paper is the largest series to report on invasive hemodynamic variables in heart transplant recipients beyond the first post-transplant year.

There is a known association between obesity and the incidence of heart failure with preserved ejection fraction (HFpEF)[15]. The diagnostic criteria of HFpEF has evolved over the years, but the contemporary definition requires the presence of normal left ventricular systolic function with elevated natriuretic peptides and evidence of congestion. Congestion can be demonstrated via elevation in left-sided filling pressures at rest (≥ 15 mmHg) or with exercise (≥ 20 mmHg)[16]. Although we do not have data regarding symptom burden, based upon hemodynamics at rest, approximately 20% of our patient have potential HFpEF at 3-year post-transplant. It is therefore not surprising that obesity as characterized by BMI was strongly predictive of elevated filling pressures in our study population. For those patients who have normal pressures at rest, provocative maneuvers such as saline loading or exercise can lead to an exaggerated increase in PCWP, indicative of HFpEF[16]. Although we did not perform these maneuvers in our study, other researchers have demonstrated abnormal hemodynamic response to exercise and saline loading in heart transplant recipients[17,18]. More widespread use of exercise hemodynamics may lead to future investigative efforts in heart transplant recipients.

Previous studies have suggested increased mortality in male heart transplant recipients with a female donor vs male, and this increased risk is apparent in the early post-transplant period[19,20]. However, more recent analysis has suggested that the issue is one of proper-size matching as opposed to sex-mismatch. Significant under-sizing as calculated by PHM has been shown to be significantly associated with post-transplant mortality[21,22]. When accounting for differences in PHM, female donor to male recipient was not associated with increased mortality[23]. Although our single-center analysis did not focus on mortality, we found an association between PHM difference and abnormal hemodynamics at 1-year. No difference was observed in hemodynamics between male and female donors. Further research is needed to determine the impact of under-sized and sex mismatch donors on functional status and hemodynamics in post-transplant recipients.

Donor age was significantly associated with hemodynamics at 1-year but was no longer significant at 3 years post-transplant. Older heart donors have been associated with lower 1-year and 3-year survival, but to our knowledge there is no data showing relationship with invasive hemodynamics[24]. Dolan et al[25] demonstrated that higher donor age was associated with increased edema and fibrosis as seen with cardiac magnetic resonance imaging in heart transplant recipients. It is possible that this contributes to abnormal hemodynamics during the first year post-transplant, but this association is no longer apparent over time. Further research is necessary to determine if using older donors in a transplant population with increasing incidence of obesity may increase the risk of developing myocardial structural changes consistent with HFpEF.

Obesity is becoming more common in patients awaiting heart transplantation, reflecting trends in society as a whole. An analysis of the UNOS registry demonstrates that BMI of both transplanted patients and cardiac donors have been increasing over time with the average BMI of heart transplant recipients in the United States at 28 kg/m²[26]. Higher BMI is associated with increased risk of post-transplant mortality leading many transplant programs to have BMI criteria for transplant candidacy[27]. Conversely, donor BMI, even if > 40 kg/m², has not been associated with increased post-transplant mortality[28]. Our analysis demonstrates that recipient BMI is more predictive than donor BMI with abnormal hemodynamics.

Weight gain is common following heart transplantation. A meta-analysis of 10 studies showed an average weight gain of 7.1 kg in the first year after transplant[29]. Few studies have examined weight gain over a longer time period. Prior to transplant, many patients suffer from cardiac cachexia due to end-stage heart failure and in those, weight gain is a positive development following transplantation. BMI is an imperfect measure of nutritional status as it does not measure the degree of muscle mass nor the presence (or absence) of central obesity. However, given that a significant percentage (30%) of our patients had a BMI > 30 kg/m² at the time of transplant, it is evident that obesity was a major comorbidity in our study population. Very few patients in our cohort were underweight at the time of transplant by BMI criteria. Although weight gain in the first-year post-transplant may be caused in part by corticosteroids, the majority of patients at our transplant program are weaned off these medications by 1-year post-transplant. Our patients experienced more weight gain after the first year as opposed to during the first post-transplant year. Considerable interest has been generated recently with the introduction of weight loss pharmacologic therapy and their success in improving cardiovascular outcomes, especially in HFpEF[30]. Given the prevalence of obesity and cardiovascular risk factors in post-transplant patients, these medications may have an important role in the management of transplant recipients.

Limitations

This is a retrospective study from a single institution. Our data is focused on hemodynamic variables and does not incorporate patient outcomes such as survival or quality of life indicators. With continued follow-up, we plan to investigate if year 3 hemodynamics are predictive of subsequent clinical outcomes. Due to the time period of the analysis, the newer methods of organ preservation as well as donation from cardiac death (DCD) were not utilized. Although survival by DCD thus far appears comparable to traditional brain death donors, prospective research will be needed to determine if there are any differences in allograft quality. We did not examine the effect of exercise on hemodynamics. It is possible that some of our patients with normal filling pressures at rest would have abnormal elevation with exercise suggestive of HFpEF.

CONCLUSION

Despite overall excellent survival, a significant proportion of our transplant patients demonstrated elevated PAP and right and left-sided filling pressures at rest. The strong association with recipient BMI highlights the importance that obesity may be playing in abnormal intracardiac hemodynamics of transplant recipients. The role of multidisciplinary care teams to successfully manage the obesity epidemic and educate transplant recipients on proper diet and exercise training, as well as consideration of potential pharmacologic therapy may be explored in future studies.