- Email: [email protected]
Relation of Vasoplegia in the Absence of Primary Graft Dysfunction to Mortality Following Cardiac Transplantation
Accepted Manuscript
Relation of Vasoplegia in the Absence of Primary Graft Dysfunction to Mortality Following Cardiac Transplantation Fayez S. Raza , Andy Y. Lee , Aayla K. Jamil , Huanying Qin , Joost Felius , Aldo E. Rafael , Gonzalo V. Gonzalez-Stawinski , Shelley A. Hall , Susan M. Joseph , Brian Lima , Amarinder S. Bindra PII: DOI: Reference:
S0002-9149(18)31734-X https://doi.org/10.1016/j.amjcard.2018.08.035 AJC 23496
To appear in:
The American Journal of Cardiology
Received date: Revised date: Accepted date:
14 May 2018 8 August 2018 14 August 2018
Please cite this article as: Fayez S. Raza , Andy Y. Lee , Aayla K. Jamil , Huanying Qin , Joost Felius , Aldo E. Rafael , Gonzalo V. Gonzalez-Stawinski , Shelley A. Hall , Susan M. Joseph , Brian Lima , Amarinder S. Bindra , Relation of Vasoplegia in the Absence of Primary Graft Dysfunction to Mortality Following Cardiac Transplantation, The American Journal of Cardiology (2018), doi: https://doi.org/10.1016/j.amjcard.2018.08.035
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT 1 Relation of Vasoplegia in the Absence of Primary Graft Dysfunction to Mortality Following Cardiac Transplantation
Fayez S. Raza MD,a Andy Y. Lee MD,a Aayla K. Jamil MPH,b Huanying Qin MS,b Joost Felius PhD,b
CR IP T
Aldo E. Rafael MD,a,b,c Gonzalo V. Gonzalez-Stawinski MD,a,b,c Shelley A. Hall MD,a,b Susan M. Joseph MD,a,b Brian Lima MD,a,b,c and Amarinder S. Bindra MDa,b a
Center for Advanced Heart and Lung Disease, Baylor University Medical Center, Dallas, TX
b
Annette C. and Harold C. Simmons Transplant Institute, Baylor Scott & White Research Institute,
c
AN US
Dallas, TX
Department of Cardiac and Thoracic Surgery, Baylor University Medical Center, Dallas, TX
Running Head: Vasoplegia Following Heart Transplantation Word count: (Abstract + main text) 2254
M
Disclosures: None
Funding: This work was funded in part by the Baylor Health Care System Foundation (Dallas, TX)
ED
Corresponding Author: Amarinder Bindra, MD
PT
Center for Advanced Heart and Lung Disease Baylor University Medical Center
CE
3410 Worth Street, Suite 250 Dallas, TX 75246
AC
Phone (214) 820 6856; FAX (214) 820 1474; Email [email protected]
This work was presented as a poster at the 2017 Annual Meeting of the International Society for Heart and Lung Transplantation (San Diego, CA, April 2017).
ACCEPTED MANUSCRIPT 2 ABSTRACT Vasoplegia following cardiac transplantation is associated with increased morbidity and mortality. Previous studies have not accounted for primary graft dysfunction (PGD). The definition of vasoplegia is based on pressor requirement at 48 hours, many PGD parameters may have normalized after the initial 24
CR IP T
hours on inotropes. We surmised that the purported negative effects of vasoplegia following transplantation may in part be driven by PGD. We reviewed 240 consecutive adult cardiac transplants at our center between 2012- 2016. The severity of vasoplegia was evaluated as a risk factor for 1-year survival, and the analysis was repeated for the subgroup of 177 patients who did not develop PGD.
AN US
Overall, 63 (26%) of patients developed mild, moderate, or severe PGD. Among those without PGD, vasoplegia was associated with length of stay but not with short- or long-term mortality. Moderate/severe vasoplegia occurred in 35 (15%) patients and was associated with higher short-term mortality, length of stay, and PGD. Multivariate logistic regression identified body mass index ≥35 kg/m2, LVAD prior to
M
transplantation, and use of ECMO as joint risk factors for vasoplegia. In patients without PGD, only LVAD prior to transplantation was associated with vasoplegia. In conclusion, our results show that, in the
ED
sizeable subgroup of patients with no signs of PGD, vasoplegia had a much more modest impact on posttransplant morbidity and no significant effect on 1- and 3-year survival. This suggests that PGD may be a
PT
confounder when assessing vasoplegia as a risk factor for adverse outcomes.
AC
CE
Key words: Cardiac transplantation; Vasoplegia; Allograft dysfunction; Outcomes
ACCEPTED MANUSCRIPT 3 INTRODUCTION Vasoplegia is a known complication of surgery, especially cardiac surgery. Loss of systemic vascular tone and resistance lead to a vasodilatory state and profound hypotension. The prevalence of vasoplegia is estimated at up to 25% in patients who are placed on cardiopulmonary bypass.1,2 Vasoplegia following
CR IP T
orthotopic heart transplantation is associated with increased morbidity3,4 and mortality.4,5 However, prior studies on postoperative vasoplegia in this population have not specifically accounted for the role of primary graft dysfunction (PGD). PGD presents a challenge postoperatively and likely shares common physiological pathways with vasoplegia. However, the nature of the possible interplay between PGD and
AN US
vasoplegia is currently unknown. With the recently introduced consensus definition of PGD, which
requires PGD be diagnosed within 24 hours of the transplant operation, it may now be better feasible to distinguish between vasoplegia and PGD. Thus, we sought to examine vasoplegia after cardiac transplantation and assess the role that PGD may play in its effects on clinical outcomes.
M
METHODS
ED
This was a retrospective study at a high-volume institution of adult patients receiving isolated orthotopic heart transplantation between November 2012 and March 2016. Data collection was approved
PT
by the Institutional Review Board of Baylor University Medical Center Dallas, and informed consent was waived. Repeat transplantations and multi-organ transplantations were excluded from analysis.
CE
Examined recipient characteristics were demographics, etiology of heart failure, comorbidities, UNOS listing status, mechanical circulatory support prior to transplantation, hospitalization, ICU stay
AC
prior to transplantation, cytomegalovirus (CMV) status, prior sternotomies, subjective global assessment, and time on the wait list. Donor characteristics included demographics, cause of death, hemodynamics, CMV status, and prior use of cardiopulmonary resuscitation. In addition, procedural characteristics included donor-recipient size match (by body weight ratio, body weight percent difference, and predicted heart mass percent difference6-8), organ travel distance, and cold ischemic time.
ACCEPTED MANUSCRIPT 4 Vasoplegia was defined as the occurrence of the need for intravenous vasopressors within 48 hours of surgery for more than 24 hours to maintain a mean arterial pressure >70 mmHg, following Esmailian et al.3,9 Intravenous pressors i.e. vasopressin, norepinephrine, epinephrine were included in the definition of vasoplegia. Inotropes were not included. The severity categories were none, mild, and
CR IP T
moderate-to-severe, with mild vasoplegia requiring one vasopressor (vasopressin, norepinephrine or high dose [>5µg/min] epinephrine), and moderate-to-severe vasoplegia requiring 2 or more vasopressors. PGD was defined according to the 2014 International Society for Heart and Lung Transplantation (ISHLT) consensus definition,10 which requires that PGD be diagnosed within 24 hours after the completion of the
AN US
transplantation procedure. The ISHLT criteria distinguish between left (or bi-) ventricular graft
dysfunction (PGD-LV) and right ventricular graft dysfunction (PGD-RV), and offer a grading scale for the severity of PGD-LV (mild, moderate, or severe). It was not feasible to incorporate the distinction between PGD-LV and PGD-RV in our study, as our clinical approach to patients with evidence of PGD
M
precluded such analysis.11 Additionally, no right ventricular assist devices (RVADs) were placed in our cohort because extracorporeal membrane oxygenation (ECMO) therapy has become our preferred
ED
therapeutic modality for all patients who require support beyond maximum inotropic therapy. The primary outcome was 1-year mortality. Secondary outcomes included duration of stay in
PT
hospital and ICU, severe rejection, postoperative complications, infections,12 blood product utilization, and 3-year survival. All outcomes were compared by vasoplegia severity (none vs. mild vs.
CE
moderate/severe). In order to elucidate the possible confounding of the results by PGD, we repeated the
AC
data analysis for the subgroup of patients without PGD (mild, moderate or severe10). Continuous variables were compared across groups by the Wilcoxon rank-sum test. Categorical
variables were compared across groups by Chi-squared or Fisher’s exact test. Survival was calculated by the Kaplan-Meier method and survival comparisons were performed using the log-rank test. To identify risk factors for vasoplegia, an initial set of candidate risk factors was first analyzed with univariate logistic regression. Subsequently, a stepwise multivariate model for moderate/severe PGD was built using
ACCEPTED MANUSCRIPT 5 the factors from the univariate analysis that yielded a P-value <0.2. All factors retained in the multivariate model were statistically significant (P<0.05). Multivariate models based on different variable sets were compared using the C-statistic. All statistical analyses were performed using SAS software (version 9.4; SAS Institute, Cary, NC, USA).
CR IP T
RESULTS
A total of 240 patients met the eligibility criteria and were included in the analysis. The majority of recipients were male (181; 75%) and White (162; 68%), and half of the patients (115; 48%) had
ischemic cardiomyopathy. Postoperative vasoplegia was diagnosed in 58 patients (24%) and categorized
AN US
as mild (n=23; 9.6%) or moderate/severe (n=35; 15%). Recipient and donor characteristics grouped by vasoplegia severity category are shown in Table 1. Recipient variables showing a significant association with the severity of vasoplegia were creatinine level, presence of a left ventricular assist device (LVAD) prior to transplantation, use of extracorporeal membrane oxygenation (ECMO) prior to transplantation,
M
and prior median sternotomy procedures. Additionally, vasoplegia was associated with the
ED
donor/recipient body weight ratio, body weight difference, and predicted heart mass difference (Table 1). Other postoperative outcomes are presented in Table 2. The overall incidence of moderate/severe
PT
PGD in the study cohort was 26% (63 of 240 patients), with 30 patients (13%) developing mild, 15 patients (6.3%) moderate, and 18 patients (7.5%) severe PGD according to ISHLT criteria. As expected,
CE
the presence of PGD was positively correlated with the severity of vasoplegia (P<0.0001). Other outcomes associated with vasoplegia severity were 30-day/in-hospital mortality, length of stay in ICU,
AC
length of stay in the hospital, inotrope score, blood product utilization, reoperation for bleeding, dialysis, and incidence of stroke, pneumonia, and infections (Table 2). Separate analysis of the findings in the subgroup of 177 patients who did not develop PGD revealed that vasoplegia in the absence of PGD was associated with length of stay in ICU, length of stay in the hospital, inotrope score, incidence of pneumonia, and the use of packed red blood cells and cryoprecipitate (Table 3). Note that 30-day/inhospital mortality was zero in this subgroup, and that there were no cases requiring dialysis.
ACCEPTED MANUSCRIPT 6 One-year survival was significantly poorer for increasing vasoplegia severity level (Figure 1A), and this effect persisted at 3-year follow-up (Figure 1B) as determined by Kaplan-Meier analysis. However, when limiting this analysis to the subgroup with no PGD, the association between 1- and 3-year survival and the severity level of vasoplegia vanished (Figure 2.)
CR IP T
Table 4 shows the results of the logistic regression that was performed to identify predictors of moderate/severe vasoplegia. For the entire cohort, the candidate factors yielding P<0.2 on univariate logistic regression were: recipient BMI ≥35 kg/m2, UNOS status 1A, use of LVAD prior to
transplantation, prior sternotomy, pre-operative creatinine level, and an undersized donor heart (by
AN US
predicted heart mass; either as continuous or dichotomous variable). The multivariate logistic regression model identified BMI ≥35 kg/m2 and the presence of LVAD prior to transplantation as joint risk factors for vasoplegia. Recipients with BMI ≥35 kg/m2 were at 2.7 times the risk for vasoplegia than those with lower BMI. Similarly, patients with an LVAD were at 2.5 times the risk for vasoplegia. When repeated in
M
patients without PGD, the multivariate logistic regression model identified only LVAD prior to
ED
transplantation (odds ratio 4.2) as a factor associated with vasoplegia (Table 4). DISCUSSION
PT
In this cohort, the overall incidence of vasoplegia post-transplant was 24%, with moderate to severe vasoplegia diagnosed in 15% of patients and mild vasoplegia seen in 9.5%. These numbers are in
CE
line with prior data.1 We found that recipients who had mechanical support (LVAD or ECMO) had a higher chance of having postoperative vasoplegia consistent with prior studies.3,4 Possible mechanisms
AC
that may explain these findings have been postulated before. These include that the removal of assist devices increases the technical complexity and procedural lengths of cases. In addition, temporary assist devices may alter vascular regulatory mechanisms that may favor a vasodilatory state.13 Variables not associated with postoperative vasoplegia included race, gender, diabetes mellitus, subjective global assessment, or CMV status. Not surprisingly, donor characteristics were not associated with recipient vasoplegia suggesting that the transplanted organ does not “bring” it.
ACCEPTED MANUSCRIPT 7 The presence of vasoplegia was meaningful in predicting outcomes. Our results showed that in patients with moderate/severe vasoplegia there was a significant association with ICU time, total length of hospital stay, inotrope score, need for dialysis, reoperation for bleeding, rate of PGD, and higher 30-day mortality. These variables may have ramifications for planning the hospital course of these patients. Our
CR IP T
findings are consistent with prior data, which have shown worse outcomes for patients with vasoplegia following cardiac surgery.2 However, a more recent paper found no significant difference in mortality at the 1 year mark.13
The etiology of vasoplegia after cardiac surgery is not fully understood but likely multifactorial.
AN US
Generally, vasoplegia is a result of activation of vasodilatory mechanisms and a resistance to
vasopressors. Patients who undergo cardiac transplantation may have additional factors that may predispose them further to the development of vasoplegia. Advanced heart failure is associated with a pro-inflammatory state with an increased number of circulating inflammatory cytokines and cells. In
M
addition, there is a more pronounced release of pro-inflammatory cytokines after cardiac transplant surgery as compared to conventional cardiac surgery.13,14 These physiological states may therefore lead to
ED
a more pronounced decrease in vascular tone and result in vasoplegia. Blood pressure is determined by the cardiac output and the systemic vascular resistance. A drop in either of these parameters without a
PT
compensatory increase in the other will lead to a decrease pressure. A decrease in blood pressure below a certain threshold will ultimately lead to end organ ischemia which if not corrected can lead to
CE
dysfunction. Thus, vasopressors are an attractive drug of choice in vasoplegic patients who have an inappropriately low systemic vascular resistance (SVR).
AC
The body’s regulation of SVR is complex with local and systemic neuro-hormonal systems.
Cardiac surgery in general places a large burden of “stress” on the body. It is not known to what degree removal of the recipient heart, implantation of a donor heart, as well as being placed on cardiopulmonary bypass adds to this stress state. It is also not known how PGD affects SVR. PGD may potentially lead to an inflammatory response, which can affect SVR and, consequently, vasoplegia. Vasoplegia and the need for vasopressors can also lead to PGD.10 Some risk factors for PGD are also risk factors for vasoplegia.
ACCEPTED MANUSCRIPT 8 The relationship between vasoplegia and PGD has not been well studied. This may be due in part to the lack of a consensus definition of vasoplegia as some the circulating definitions include adequate myocardial pump function and arbitrary 24 and 48 hour time cutoffs. In our group moderate to severe PGD, as per the ISHLT definition, occurred in 26% of the patients. The presence of PGD correlated with
CR IP T
the presence and severity of vasoplegia. Interestingly, when we excluded patients with PGD patients from our analysis, there was no association between vasoplegia and 30-day and 1-year survival. These findings suggest that the consequences of vasoplegia alone may be far less deleterious in the absence of PGD. Both PGD and vasoplegia may actually be manifestations of several common physiological pathways,
AN US
intertwined with each other. However, this is a poorly understood area that needs a significant amount of further study at both the basic science and clinical levels.
This study has the limitations of a single-center, retrospective analysis. Although we receive patient referrals from across the country, our population may not apply to other centers. In addition, our
M
study does not account for possible confounders during the actual transplant operation. Lastly, in reviewing the literature there is a lack of uniformity for the accepted definitions of vasoplegia.
ED
In conclusion, the recent consensus definition of PGD allowed us to better separate vasoplegia from PGD. We confirmed that vasoplegia and PGD are common after orthotopic heart transplantation.
PT
Vasoplegia appeared to be an independent risk factor for morbidity and mortality at 1 year in this population. However, in patients without PGD, vasoplegia did not have nearly as much impact on
CE
outcomes. Our results suggest therefore, that PGD may be a possible confounder when assessing vasoplegia as a risk factor for transplant outcomes.
AC
ACKNOWLEDGEMENTS Funding: Funded in part by the Baylor Health Care System Foundation Disclosures: None
ACCEPTED MANUSCRIPT 9 1. Fischer GW, Levin MA. Vasoplegia during cardiac surgery: current concepts and management. Semin Thorac Cardiovasc Surg 2010;22:140-144. 2. Omar S, Zedan A, Nugent K. Cardiac vasoplegia syndrome: pathophysiology, risk factors and treatment. Am J Med Sci 2015;349:80-88.
CR IP T
3. Chan JL, Kobashigawa JA, Aintablian TL, Li Y, Perry PA, Patel JK, Kittleson MM, Czer LS, Zarrini P, Velleca A, Rush J, Arabia FA, Trento A, Esmailian F. Vasoplegia after heart transplantation: outcomes at 1 year. Interactive CardioVascular and Thoracic Surgery 2017;25:212-217.
4. Patarroyo M, Simbaqueba C, Shrestha K, Starling RC, Smedira N, Tang WH, Taylor DO. Pre-
AN US
operative risk factors and clinical outcomes associated with vasoplegia in recipients of orthotopic heart transplantation in the contemporary era. J Heart Lung Transplant 2012;31:282-287. 5. Byrne JG, Leacche M, Paul S, Mihaljevic T, Rawn JD, Shernan SK, Mudge GH, Stevenson LW. Risk factors and outcomes for 'vasoplegia syndrome' following cardiac transplantation. Eur J Cardiothorac
M
Surg 2004;25:327-332.
6. Bluemke DA, Kronmal RA, Lima JA, Liu K, Olson J, Burke GL, Folsom AR. The relationship of left
ED
ventricular mass and geometry to incident cardiovascular events: the MESA (Multi-Ethnic Study of Atherosclerosis) study. J Am Coll Cardiol 2008;52:2148-2155.
PT
7. Kawut SM, Lima JA, Barr RG, Chahal H, Jain A, Tandri H, Praestgaard A, Bagiella E, Kizer JR, Johnson WC, Kronmal RA, Bluemke DA. Sex and race differences in right ventricular structure and
CE
function: the multi-ethnic study of atherosclerosis-right ventricle study. Circulation 2011;123:25422551.
AC
8. Reed RM, Netzer G, Hunsicker L, Mitchell BD, Rajagopal K, Scharf S, Eberlein M. Cardiac size and sex-matching in heart transplantation : size matters in matters of sex and the heart. JACC Heart Fail 2014;2:73-83.
9. Esmailian F, Perry P, Luu M, Patel J, Kittleson M, Czer L, Aintablian T, Zarrini P, Velleca A, Rush J, Arabia F, Kobashigawa JA. Vasoplegia After Heart Transplantation: Unraveling the Enigma. The Journal of Heart and Lung Transplantation;35:S166-S167.
ACCEPTED MANUSCRIPT 10 10. Kobashigawa J, Zuckermann A, Macdonald P, Leprince P, Esmailian F, Luu M, Mancini D, Patel J, Razi R, Reichenspurner H, Russell S, Segovia J, Smedira N, Stehlik J, Wagner F, Consensus Conference Participants. Report from a consensus conference on primary graft dysfunction after cardiac transplantation. J Heart Lung Transplant 2014;33:327-340.
CR IP T
11. Squiers JJ, Saracino G, Chamogeorgakis T, MacHannaford JC, Rafael AE, Gonzalez-Stawinski GV, Hall SA, DiMaio JM, Lima B. Application of the International Society for Heart and Lung
Transplantation (ISHLT) criteria for primary graft dysfunction after cardiac transplantation: outcomes from a high-volume centre. Eur J Cardiothorac Surg 2017;51:263-270.
AN US
12. Rajagopal K, Lima B, Petersen RP, Mesis RG, Daneshmand MA, Lemaire A, Felker GM, Hernandez AF, Rogers JG, Lodge AJ, Milano CA. Infectious complications in extended criteria heart transplantation. J Heart Lung Transplant 2008;27:1217-1221.
13. Chemmalakuzhy J, Costanzo MR, Meyer P, Piccione W, Kao W, Winkel E, Saltzberg M, Heroux A,
M
Parrillo J. Hypotension, acidosis, and vasodilatation syndrome post-heart transplant: prognostic variables and outcomes. J Heart Lung Transplant 2001;20:1075-1083.
ED
14. Wan S, Marchant A, DeSmet JM, Antoine M, Zhang H, Vachiery JL, Goldman M, Vincent JL, LeClerc JL. Human cytokine responses to cardiac transplantation and coronary artery bypass grafting.
CE
PT
J Thorac Cardiovasc Surg 1996;111:469-477.
AC
FIGURE LEGENDS
ACCEPTED MANUSCRIPT
AN US
CR IP T
11
Figure 1: Kaplan-Meier analysis of 1-year (A) and 3-year (B) survival by severity of vasoplegia for the
AC
CE
PT
ED
M
entire cohort. The effect of vasoplegia severity was significant both at 1 year and at 3 years (P<0.001).
ACCEPTED MANUSCRIPT 12 Figure 2: Kaplan-Meier analysis of 1-year (A) and 3-year (B) survival by severity of vasoplegia for the subgroup of patients with no PGD.
Table 1. Preoperative Patient Characteristics by Severity of Vasoplegia (N=240) Vasoplegia Severity Variable
P-value
Mild (n=23)
Moderate/Severe (n=35)
60 [54, 66]
60 [51, 65]
0.88
19 (83%)
27 (77%)
0.64
4 (17%)
5 (14%)
0.44
121 (67%)
18 (78%)
23 (66%)
26 (14%)
1 (4%)
7 (20%)
85 (47%)
13 (57%)
17 (49%)
0.67
32 (18%)
2 (9%)
4 (11%)
0.36
1.4 [1.1, 1.6]
1.5 [1.3, 1.9]
1.5 [1.3, 1.8]
0.012
28 [26, 31]
29 [24, 36]
29 [25, 34]
0.47
2.07 [1.92, 2.24]
2.13 [2.01. 2.30]
2.16 [1.94, 2.30]
0.19
36 (20%)
5 (22%)
11 (31%)
On intensive care prior to transplant
87 (48%)
9 (39%)
10 (29%)
0.078
LVAD pre-transplant
44 (26%)
10 (48%)
15 (56%)
0.003
ECMO pre-transplant
4 (2%)
4 (19%)
5 (19%)
0.0008
56 (31%)
11 (48%)
17 (49%)
0.064
1.97 [1.14, 3.13]
2.12 [1.51, 3.04]
1.97 [1.41, 3.02]
0.75
Previous sternotomy
77 (45%)
12 (57%)
20 (74%)
0.015
Wait list time (days)
22.5 [8, 104]
38 [7, 141]
23 [9, 58]
0.79
Recipient Variables 59 [52, 65]
Men
135 (74%)
Black
35 (19%)
AN US
Age (years)
White Other Ischemic cardiomyopathy Diabetes mellitus
M
Creatinine (mg/dL)
CE
PT
Hospitalized
ED
Body mass index (kg/m2) Body surface area (m2)
UNOS status 1A
Pulmonary vascular resistance (Wood units)
AC
CR IP T
None (n=182)
Subjective global assessment:
0.15
Well nourished
112 (69%)
17 (85%)
18 (72%)
Moderately malnourished
48 (29%)
2 (10%)
5 (20%)
3 (2%)
1 (5%)
2 (8%)
Severely malnourished
ACCEPTED MANUSCRIPT 13
Donor Variables 34 [23, 44]
34 [24, 42]
31 [26, 44]
0.98
Men
81 (45%)
8 (35%)
19 (54%)
0.33
Black
29 (16%)
2 (9%)
7 (20%)
0.061
White
145 (80%)
18 (78%)
22 (63%)
Other
8 (4%)
3 (13%)
Cause of death: 75 (41%)
Head trauma
55 (30%)
Other
52 (29%)
Previous cardiopulmonary resuscitation Body mass index (kg/m2) Donor/recipient weight ratio
Ejection fraction (%) Cytomegalovirus mismatch
PT
Distance from center (mi)
ED
Left ventricular wall thickness (cm)
Total ischemic time (min)
0.19 16 (46%)
10 (44%)
10 (29%)
9 (39%)
9 (26%)
47 (26%)
3 (13%)
7 (20%)
0.31
29 [25, 34]
28.5 [23, 32]
27 [24, 34]
0.81
0.96 [0.82, 1.13]
0.80 [0.72, 0.94]
0.86 [0.77, 1.15]
0.020
3 [-13, 18]
20 [6, 28]
14 [-15, 23]
0.020
7 [-6, 19]
17.5 [10, 33]
11 [-7, 26]
0.016
1.00 [0.90, 1.20]
1.10 [0.89, 1.20]
1.06 [0.90, 1.20]
0.99
M
Total body weight (% difference)
6 (17%)
4 (17%)
AN US
Anoxia
Predicted heart mass (% difference)
CR IP T
Age (years)
62.5 [55, 67.5]
60 [55, 65]
60 [54, 69]
0.49
72 (40%)
5 (22%)
12 (35%)
0.22
287 [38, 747]
362 [192, 904]
241 [38, 431]
0.17
221 [171, 268]
234 [202, 274]
209 [182, 273]
0.35
CE
Abbreviations: ECMO, extracorporeal membrane oxygenation; UNOS, United Network for Organ Sharing; LVAD, left ventricular assist device
AC
Table 2. Postoperative Outcomes by Severity of Vasoplegia (N=240)
Variable
Vasoplegia Severity None (n=182)
Mild (n=23)
Moderate/Severe (n=35)
Intensive care length of stay (days)
3 [2, 4]
5 [3, 8]
10.5 [5, 30]
Hospital length of stay (days)
8 [6, 12]
13 [9, 17]
18 [12, 45]
ACCEPTED MANUSCRIPT 14
Inotrope score
15.5 [11.5, 21.5]
19.5 [16, 30]
31 [26.5, 40]
None
150 (82%)
10 (44%)
17 (49%)
Mild
19 (10%)
7 (30%)
4 (11%)
Moderate
7 (3.8%)
1 (4.3%)
7 (20%)
Severe
6 (3.3%)
5 (22%)
7 (20%)
Dialysis
1 (0.5%)
1 (4.3%)
5 (14%)
Reoperate for bleeding
15 (8.2%)
4 (17%)
11 (31%)
Stroke
4 (2.2%)
1 (4.3%)
2 (5.7%)
Pneumonia
3 (1.6%)
2 (8.7%)
11 (31%)
Bacterial infection
14 (7.7%)
4 (17%)
13 (37%)
3 (1.6%)
2 (8.7%)
9 (26%)
4 (2.2%)
1 (4.3%)
4 (11%)
0
0
3 (8.6%)
5 (2.7%)
2 (8.7%)
2 (5.7%)
6 (3.3%)
1 (4.3%)
7 (20%)
3 (1.6%)
1 (4.3%)
1 (2.9%)
1 [0, 2]
1 [0, 2]
3 [1, 6]
0 [0, 1]
0 [0, 1]
0 [0, 2]
0 [0, 0]
0 [0, 0]
0 [0, 0]
0 [0, 0]
0 [0, 1]
0 [0, 2]
1 (0.5%)
2 (8.7%)
4 (11%)
AN US
CR IP T
Primary graft dysfunction:
Fungal infection Sepsis Mediastinitis
Bloodstream infection
Blood product utilization: Packed red blood cells
Cryoprecipitate
CE
Platelets
PT
Fresh frozen plasma
ED
Severe rejection (3R) within 3 months
M
Wound infection
30-day/in-hospital mortality
AC
Table 3. Postoperative Outcomes by Severity of Vasoplegia in the Absence of Primary Graft Dysfunction (N=177).
Variable
Intensive care length of stay (days) Hospital length of stay (days)
Vasoplegia Severity None (n=182)
Mild (n=23)
Moderate/Severe (n=35)
2 [2, 3]
5 [3, 6]
7 [4, 15]
7.5 [6, 10]
11 [9, 15]
18 [13, 24]
ACCEPTED MANUSCRIPT 15
Inotrope score
14.5 [10.5, 19.5]
19 [16, 20.5]
31 [29.5, 38.5]
0
0
0
Reoperate for bleeding
9 (6.0%)
1 (10%)
4 (24%)
Stroke
3 (2.0%)
1 (10%)
1 (5.9%)
Pneumonia
3 (2.0%)
0
3 (18%)
Bacterial infection
10 (6.7%)
1 (10%)
3 (18%)
Fungal infection
2 (1.3%)
1 (10%)
1 (5.9%)
Sepsis
3 (2.0%)
0
0
0
0
0
Wound infection
4 (2.7%)
1 (10%)
1 (5.9%)
Bloodstream infection
3 (2.0%)
0
0
Severe rejection (3R) within 3 months
3 (2.0%)
0
1 (5.9%)
Packed red blood cells
1 [0, 2]
0 [0, 1]
2 [1, 4]
0 [0, 0]
0 [0, 2]
0 [0, 3]
0 [0, 0]
0 [0, 0]
0 [0, 0]
0 [0, 0]
0 [0, 0]
0 [0, 0]
0
0
CR IP T
Dialysis
AN US
Mediastinitis
Fresh frozen plasma Cryoprecipitate Platelets
ED
M
30-day/in-hospital mortality 0 Table 4. Univariate and Multivariate Predictors of Moderate or Severe Vasoplegia.
PT
Variable
All Patients
Odds ratio
Patients with no p dysfunc P-value
[95% Confidence interval]
Odds ratio [95% Confidence interval]
CE
Univariate analyses Male gender
1.34 [0.65, 2.73]
0.43
0.94 [0.37, 2.39]
1.03 [0.53, 2.01]
0.92
0.90 [0.36, 2.29]
3.39 [1.57, 7.32]
0.002*
2.40 [0.78, 7.38]
UNOS 1A status
2.07 [1.13, 3.78]
0.019*
3.89 [1.65, 9.16]
LVAD prior to transplant
2.91 [1.58, 5.35]
0.0006*
3.68 [1.59, 8.57]
ECMO prior to transplant
2.13 [0.35, 13.1]
0.41
5.74 [0.35, 94.6]
Creatinine (continuous scale)
2.61 [1.21, 5.63]
0.015*
3.00 [0.93, 9.66]
Creatinine ≥2.0 mg/dL
1.65 [0.63, 4.32]
0.31
1.28 [0.26, 6.28]
Hospitalized prior to transplant
1.52 [0.77, 3.01]
0.23
1.92 [0.76, 4.84]
On intensive care prior to transplant
0.49 [0.14, 1.73]
0.27
0.70 [0.15, 3.23]
Recipient age ≥65 y
AC
Recipient body mass index ≥35 kg/m
2
ACCEPTED MANUSCRIPT 16 2.75 [1.46, 5.20]
0.0018*
2.55 [1.07, 6.03]
Donor age ≥50 y
0.82 [0.34, 2.01]
0.67
1.13 [0.35, 3.61]
Donor male gender
1.08 [0.59, 1.95]
0.81
1.22 [0.54, 2.76]
Donor ejection fraction <50%
0.69 [0.15, 3.25]
0.64
0.48 [0.06, 3.90]
Donor undersized ≥30% by body weight
1.40 [0.51, 3.84]
0.51
1.31 [0.35, 4.94]
Donor undersized by predicted heart mass (continuous scale)
1.02 [1.00, 1.03]
0.043*
1.00 [0.98, 1.02]
Donor undersized ≥30% by predicted heart mass
1.81 [0.79, 4.17]
0.16*
1.68 [0.51, 5.55]
Donor left ventricular wall thickness >1.3 cm
0.69 [0.27, 1.76]
0.43
1.18 [0.36, 3.82]
Pulmonary vascular resistance (Woods)
1.01 [0.84, 1.22]
0.90
1.05 [0.82, 1.34]
Donor sequence number >80
0.59 [0.28, 1.27]
0.18*
0.69 [0.25, 1.96]
2.65 [1.18, 5.98]
0.019
Multivariate analyses
AN US
Recipient body mass index ≥35 kg/m2
CR IP T
Prior sternotomy
LVAD prior to transplant
2.47 [1.30, 4.69]
UNOS 1A status
0.0056
*Variables with a P-value <0.2 in the univariate analysis were selected entered into the stepwise
M
multivariate regression model.
Abbreviations: ECMO, extracorporeal membrane oxygenation; UNOS, United Network for
AC
CE
PT
ED
Organ Sharing; LVAD, left ventricular assist device
4.24 [1.76, 10.2]
Relation of Vasoplegia in the Absence of Primary Graft Dysfunction to Mortality Following Cardiac Transplantation Fayez S. Raza , Andy Y. Lee , Aayla K. Jamil , Huanying Qin , Joost Felius , Aldo E. Rafael , Gonzalo V. Gonzalez-Stawinski , Shelley A. Hall , Susan M. Joseph , Brian Lima , Amarinder S. Bindra PII: DOI: Reference:
S0002-9149(18)31734-X https://doi.org/10.1016/j.amjcard.2018.08.035 AJC 23496
To appear in:
The American Journal of Cardiology
Received date: Revised date: Accepted date:
14 May 2018 8 August 2018 14 August 2018
Please cite this article as: Fayez S. Raza , Andy Y. Lee , Aayla K. Jamil , Huanying Qin , Joost Felius , Aldo E. Rafael , Gonzalo V. Gonzalez-Stawinski , Shelley A. Hall , Susan M. Joseph , Brian Lima , Amarinder S. Bindra , Relation of Vasoplegia in the Absence of Primary Graft Dysfunction to Mortality Following Cardiac Transplantation, The American Journal of Cardiology (2018), doi: https://doi.org/10.1016/j.amjcard.2018.08.035
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT 1 Relation of Vasoplegia in the Absence of Primary Graft Dysfunction to Mortality Following Cardiac Transplantation
Fayez S. Raza MD,a Andy Y. Lee MD,a Aayla K. Jamil MPH,b Huanying Qin MS,b Joost Felius PhD,b
CR IP T
Aldo E. Rafael MD,a,b,c Gonzalo V. Gonzalez-Stawinski MD,a,b,c Shelley A. Hall MD,a,b Susan M. Joseph MD,a,b Brian Lima MD,a,b,c and Amarinder S. Bindra MDa,b a
Center for Advanced Heart and Lung Disease, Baylor University Medical Center, Dallas, TX
b
Annette C. and Harold C. Simmons Transplant Institute, Baylor Scott & White Research Institute,
c
AN US
Dallas, TX
Department of Cardiac and Thoracic Surgery, Baylor University Medical Center, Dallas, TX
Running Head: Vasoplegia Following Heart Transplantation Word count: (Abstract + main text) 2254
M
Disclosures: None
Funding: This work was funded in part by the Baylor Health Care System Foundation (Dallas, TX)
ED
Corresponding Author: Amarinder Bindra, MD
PT
Center for Advanced Heart and Lung Disease Baylor University Medical Center
CE
3410 Worth Street, Suite 250 Dallas, TX 75246
AC
Phone (214) 820 6856; FAX (214) 820 1474; Email [email protected]
This work was presented as a poster at the 2017 Annual Meeting of the International Society for Heart and Lung Transplantation (San Diego, CA, April 2017).
ACCEPTED MANUSCRIPT 2 ABSTRACT Vasoplegia following cardiac transplantation is associated with increased morbidity and mortality. Previous studies have not accounted for primary graft dysfunction (PGD). The definition of vasoplegia is based on pressor requirement at 48 hours, many PGD parameters may have normalized after the initial 24
CR IP T
hours on inotropes. We surmised that the purported negative effects of vasoplegia following transplantation may in part be driven by PGD. We reviewed 240 consecutive adult cardiac transplants at our center between 2012- 2016. The severity of vasoplegia was evaluated as a risk factor for 1-year survival, and the analysis was repeated for the subgroup of 177 patients who did not develop PGD.
AN US
Overall, 63 (26%) of patients developed mild, moderate, or severe PGD. Among those without PGD, vasoplegia was associated with length of stay but not with short- or long-term mortality. Moderate/severe vasoplegia occurred in 35 (15%) patients and was associated with higher short-term mortality, length of stay, and PGD. Multivariate logistic regression identified body mass index ≥35 kg/m2, LVAD prior to
M
transplantation, and use of ECMO as joint risk factors for vasoplegia. In patients without PGD, only LVAD prior to transplantation was associated with vasoplegia. In conclusion, our results show that, in the
ED
sizeable subgroup of patients with no signs of PGD, vasoplegia had a much more modest impact on posttransplant morbidity and no significant effect on 1- and 3-year survival. This suggests that PGD may be a
PT
confounder when assessing vasoplegia as a risk factor for adverse outcomes.
AC
CE
Key words: Cardiac transplantation; Vasoplegia; Allograft dysfunction; Outcomes
ACCEPTED MANUSCRIPT 3 INTRODUCTION Vasoplegia is a known complication of surgery, especially cardiac surgery. Loss of systemic vascular tone and resistance lead to a vasodilatory state and profound hypotension. The prevalence of vasoplegia is estimated at up to 25% in patients who are placed on cardiopulmonary bypass.1,2 Vasoplegia following
CR IP T
orthotopic heart transplantation is associated with increased morbidity3,4 and mortality.4,5 However, prior studies on postoperative vasoplegia in this population have not specifically accounted for the role of primary graft dysfunction (PGD). PGD presents a challenge postoperatively and likely shares common physiological pathways with vasoplegia. However, the nature of the possible interplay between PGD and
AN US
vasoplegia is currently unknown. With the recently introduced consensus definition of PGD, which
requires PGD be diagnosed within 24 hours of the transplant operation, it may now be better feasible to distinguish between vasoplegia and PGD. Thus, we sought to examine vasoplegia after cardiac transplantation and assess the role that PGD may play in its effects on clinical outcomes.
M
METHODS
ED
This was a retrospective study at a high-volume institution of adult patients receiving isolated orthotopic heart transplantation between November 2012 and March 2016. Data collection was approved
PT
by the Institutional Review Board of Baylor University Medical Center Dallas, and informed consent was waived. Repeat transplantations and multi-organ transplantations were excluded from analysis.
CE
Examined recipient characteristics were demographics, etiology of heart failure, comorbidities, UNOS listing status, mechanical circulatory support prior to transplantation, hospitalization, ICU stay
AC
prior to transplantation, cytomegalovirus (CMV) status, prior sternotomies, subjective global assessment, and time on the wait list. Donor characteristics included demographics, cause of death, hemodynamics, CMV status, and prior use of cardiopulmonary resuscitation. In addition, procedural characteristics included donor-recipient size match (by body weight ratio, body weight percent difference, and predicted heart mass percent difference6-8), organ travel distance, and cold ischemic time.
ACCEPTED MANUSCRIPT 4 Vasoplegia was defined as the occurrence of the need for intravenous vasopressors within 48 hours of surgery for more than 24 hours to maintain a mean arterial pressure >70 mmHg, following Esmailian et al.3,9 Intravenous pressors i.e. vasopressin, norepinephrine, epinephrine were included in the definition of vasoplegia. Inotropes were not included. The severity categories were none, mild, and
CR IP T
moderate-to-severe, with mild vasoplegia requiring one vasopressor (vasopressin, norepinephrine or high dose [>5µg/min] epinephrine), and moderate-to-severe vasoplegia requiring 2 or more vasopressors. PGD was defined according to the 2014 International Society for Heart and Lung Transplantation (ISHLT) consensus definition,10 which requires that PGD be diagnosed within 24 hours after the completion of the
AN US
transplantation procedure. The ISHLT criteria distinguish between left (or bi-) ventricular graft
dysfunction (PGD-LV) and right ventricular graft dysfunction (PGD-RV), and offer a grading scale for the severity of PGD-LV (mild, moderate, or severe). It was not feasible to incorporate the distinction between PGD-LV and PGD-RV in our study, as our clinical approach to patients with evidence of PGD
M
precluded such analysis.11 Additionally, no right ventricular assist devices (RVADs) were placed in our cohort because extracorporeal membrane oxygenation (ECMO) therapy has become our preferred
ED
therapeutic modality for all patients who require support beyond maximum inotropic therapy. The primary outcome was 1-year mortality. Secondary outcomes included duration of stay in
PT
hospital and ICU, severe rejection, postoperative complications, infections,12 blood product utilization, and 3-year survival. All outcomes were compared by vasoplegia severity (none vs. mild vs.
CE
moderate/severe). In order to elucidate the possible confounding of the results by PGD, we repeated the
AC
data analysis for the subgroup of patients without PGD (mild, moderate or severe10). Continuous variables were compared across groups by the Wilcoxon rank-sum test. Categorical
variables were compared across groups by Chi-squared or Fisher’s exact test. Survival was calculated by the Kaplan-Meier method and survival comparisons were performed using the log-rank test. To identify risk factors for vasoplegia, an initial set of candidate risk factors was first analyzed with univariate logistic regression. Subsequently, a stepwise multivariate model for moderate/severe PGD was built using
ACCEPTED MANUSCRIPT 5 the factors from the univariate analysis that yielded a P-value <0.2. All factors retained in the multivariate model were statistically significant (P<0.05). Multivariate models based on different variable sets were compared using the C-statistic. All statistical analyses were performed using SAS software (version 9.4; SAS Institute, Cary, NC, USA).
CR IP T
RESULTS
A total of 240 patients met the eligibility criteria and were included in the analysis. The majority of recipients were male (181; 75%) and White (162; 68%), and half of the patients (115; 48%) had
ischemic cardiomyopathy. Postoperative vasoplegia was diagnosed in 58 patients (24%) and categorized
AN US
as mild (n=23; 9.6%) or moderate/severe (n=35; 15%). Recipient and donor characteristics grouped by vasoplegia severity category are shown in Table 1. Recipient variables showing a significant association with the severity of vasoplegia were creatinine level, presence of a left ventricular assist device (LVAD) prior to transplantation, use of extracorporeal membrane oxygenation (ECMO) prior to transplantation,
M
and prior median sternotomy procedures. Additionally, vasoplegia was associated with the
ED
donor/recipient body weight ratio, body weight difference, and predicted heart mass difference (Table 1). Other postoperative outcomes are presented in Table 2. The overall incidence of moderate/severe
PT
PGD in the study cohort was 26% (63 of 240 patients), with 30 patients (13%) developing mild, 15 patients (6.3%) moderate, and 18 patients (7.5%) severe PGD according to ISHLT criteria. As expected,
CE
the presence of PGD was positively correlated with the severity of vasoplegia (P<0.0001). Other outcomes associated with vasoplegia severity were 30-day/in-hospital mortality, length of stay in ICU,
AC
length of stay in the hospital, inotrope score, blood product utilization, reoperation for bleeding, dialysis, and incidence of stroke, pneumonia, and infections (Table 2). Separate analysis of the findings in the subgroup of 177 patients who did not develop PGD revealed that vasoplegia in the absence of PGD was associated with length of stay in ICU, length of stay in the hospital, inotrope score, incidence of pneumonia, and the use of packed red blood cells and cryoprecipitate (Table 3). Note that 30-day/inhospital mortality was zero in this subgroup, and that there were no cases requiring dialysis.
ACCEPTED MANUSCRIPT 6 One-year survival was significantly poorer for increasing vasoplegia severity level (Figure 1A), and this effect persisted at 3-year follow-up (Figure 1B) as determined by Kaplan-Meier analysis. However, when limiting this analysis to the subgroup with no PGD, the association between 1- and 3-year survival and the severity level of vasoplegia vanished (Figure 2.)
CR IP T
Table 4 shows the results of the logistic regression that was performed to identify predictors of moderate/severe vasoplegia. For the entire cohort, the candidate factors yielding P<0.2 on univariate logistic regression were: recipient BMI ≥35 kg/m2, UNOS status 1A, use of LVAD prior to
transplantation, prior sternotomy, pre-operative creatinine level, and an undersized donor heart (by
AN US
predicted heart mass; either as continuous or dichotomous variable). The multivariate logistic regression model identified BMI ≥35 kg/m2 and the presence of LVAD prior to transplantation as joint risk factors for vasoplegia. Recipients with BMI ≥35 kg/m2 were at 2.7 times the risk for vasoplegia than those with lower BMI. Similarly, patients with an LVAD were at 2.5 times the risk for vasoplegia. When repeated in
M
patients without PGD, the multivariate logistic regression model identified only LVAD prior to
ED
transplantation (odds ratio 4.2) as a factor associated with vasoplegia (Table 4). DISCUSSION
PT
In this cohort, the overall incidence of vasoplegia post-transplant was 24%, with moderate to severe vasoplegia diagnosed in 15% of patients and mild vasoplegia seen in 9.5%. These numbers are in
CE
line with prior data.1 We found that recipients who had mechanical support (LVAD or ECMO) had a higher chance of having postoperative vasoplegia consistent with prior studies.3,4 Possible mechanisms
AC
that may explain these findings have been postulated before. These include that the removal of assist devices increases the technical complexity and procedural lengths of cases. In addition, temporary assist devices may alter vascular regulatory mechanisms that may favor a vasodilatory state.13 Variables not associated with postoperative vasoplegia included race, gender, diabetes mellitus, subjective global assessment, or CMV status. Not surprisingly, donor characteristics were not associated with recipient vasoplegia suggesting that the transplanted organ does not “bring” it.
ACCEPTED MANUSCRIPT 7 The presence of vasoplegia was meaningful in predicting outcomes. Our results showed that in patients with moderate/severe vasoplegia there was a significant association with ICU time, total length of hospital stay, inotrope score, need for dialysis, reoperation for bleeding, rate of PGD, and higher 30-day mortality. These variables may have ramifications for planning the hospital course of these patients. Our
CR IP T
findings are consistent with prior data, which have shown worse outcomes for patients with vasoplegia following cardiac surgery.2 However, a more recent paper found no significant difference in mortality at the 1 year mark.13
The etiology of vasoplegia after cardiac surgery is not fully understood but likely multifactorial.
AN US
Generally, vasoplegia is a result of activation of vasodilatory mechanisms and a resistance to
vasopressors. Patients who undergo cardiac transplantation may have additional factors that may predispose them further to the development of vasoplegia. Advanced heart failure is associated with a pro-inflammatory state with an increased number of circulating inflammatory cytokines and cells. In
M
addition, there is a more pronounced release of pro-inflammatory cytokines after cardiac transplant surgery as compared to conventional cardiac surgery.13,14 These physiological states may therefore lead to
ED
a more pronounced decrease in vascular tone and result in vasoplegia. Blood pressure is determined by the cardiac output and the systemic vascular resistance. A drop in either of these parameters without a
PT
compensatory increase in the other will lead to a decrease pressure. A decrease in blood pressure below a certain threshold will ultimately lead to end organ ischemia which if not corrected can lead to
CE
dysfunction. Thus, vasopressors are an attractive drug of choice in vasoplegic patients who have an inappropriately low systemic vascular resistance (SVR).
AC
The body’s regulation of SVR is complex with local and systemic neuro-hormonal systems.
Cardiac surgery in general places a large burden of “stress” on the body. It is not known to what degree removal of the recipient heart, implantation of a donor heart, as well as being placed on cardiopulmonary bypass adds to this stress state. It is also not known how PGD affects SVR. PGD may potentially lead to an inflammatory response, which can affect SVR and, consequently, vasoplegia. Vasoplegia and the need for vasopressors can also lead to PGD.10 Some risk factors for PGD are also risk factors for vasoplegia.
ACCEPTED MANUSCRIPT 8 The relationship between vasoplegia and PGD has not been well studied. This may be due in part to the lack of a consensus definition of vasoplegia as some the circulating definitions include adequate myocardial pump function and arbitrary 24 and 48 hour time cutoffs. In our group moderate to severe PGD, as per the ISHLT definition, occurred in 26% of the patients. The presence of PGD correlated with
CR IP T
the presence and severity of vasoplegia. Interestingly, when we excluded patients with PGD patients from our analysis, there was no association between vasoplegia and 30-day and 1-year survival. These findings suggest that the consequences of vasoplegia alone may be far less deleterious in the absence of PGD. Both PGD and vasoplegia may actually be manifestations of several common physiological pathways,
AN US
intertwined with each other. However, this is a poorly understood area that needs a significant amount of further study at both the basic science and clinical levels.
This study has the limitations of a single-center, retrospective analysis. Although we receive patient referrals from across the country, our population may not apply to other centers. In addition, our
M
study does not account for possible confounders during the actual transplant operation. Lastly, in reviewing the literature there is a lack of uniformity for the accepted definitions of vasoplegia.
ED
In conclusion, the recent consensus definition of PGD allowed us to better separate vasoplegia from PGD. We confirmed that vasoplegia and PGD are common after orthotopic heart transplantation.
PT
Vasoplegia appeared to be an independent risk factor for morbidity and mortality at 1 year in this population. However, in patients without PGD, vasoplegia did not have nearly as much impact on
CE
outcomes. Our results suggest therefore, that PGD may be a possible confounder when assessing vasoplegia as a risk factor for transplant outcomes.
AC
ACKNOWLEDGEMENTS Funding: Funded in part by the Baylor Health Care System Foundation Disclosures: None
ACCEPTED MANUSCRIPT 9 1. Fischer GW, Levin MA. Vasoplegia during cardiac surgery: current concepts and management. Semin Thorac Cardiovasc Surg 2010;22:140-144. 2. Omar S, Zedan A, Nugent K. Cardiac vasoplegia syndrome: pathophysiology, risk factors and treatment. Am J Med Sci 2015;349:80-88.
CR IP T
3. Chan JL, Kobashigawa JA, Aintablian TL, Li Y, Perry PA, Patel JK, Kittleson MM, Czer LS, Zarrini P, Velleca A, Rush J, Arabia FA, Trento A, Esmailian F. Vasoplegia after heart transplantation: outcomes at 1 year. Interactive CardioVascular and Thoracic Surgery 2017;25:212-217.
4. Patarroyo M, Simbaqueba C, Shrestha K, Starling RC, Smedira N, Tang WH, Taylor DO. Pre-
AN US
operative risk factors and clinical outcomes associated with vasoplegia in recipients of orthotopic heart transplantation in the contemporary era. J Heart Lung Transplant 2012;31:282-287. 5. Byrne JG, Leacche M, Paul S, Mihaljevic T, Rawn JD, Shernan SK, Mudge GH, Stevenson LW. Risk factors and outcomes for 'vasoplegia syndrome' following cardiac transplantation. Eur J Cardiothorac
M
Surg 2004;25:327-332.
6. Bluemke DA, Kronmal RA, Lima JA, Liu K, Olson J, Burke GL, Folsom AR. The relationship of left
ED
ventricular mass and geometry to incident cardiovascular events: the MESA (Multi-Ethnic Study of Atherosclerosis) study. J Am Coll Cardiol 2008;52:2148-2155.
PT
7. Kawut SM, Lima JA, Barr RG, Chahal H, Jain A, Tandri H, Praestgaard A, Bagiella E, Kizer JR, Johnson WC, Kronmal RA, Bluemke DA. Sex and race differences in right ventricular structure and
CE
function: the multi-ethnic study of atherosclerosis-right ventricle study. Circulation 2011;123:25422551.
AC
8. Reed RM, Netzer G, Hunsicker L, Mitchell BD, Rajagopal K, Scharf S, Eberlein M. Cardiac size and sex-matching in heart transplantation : size matters in matters of sex and the heart. JACC Heart Fail 2014;2:73-83.
9. Esmailian F, Perry P, Luu M, Patel J, Kittleson M, Czer L, Aintablian T, Zarrini P, Velleca A, Rush J, Arabia F, Kobashigawa JA. Vasoplegia After Heart Transplantation: Unraveling the Enigma. The Journal of Heart and Lung Transplantation;35:S166-S167.
ACCEPTED MANUSCRIPT 10 10. Kobashigawa J, Zuckermann A, Macdonald P, Leprince P, Esmailian F, Luu M, Mancini D, Patel J, Razi R, Reichenspurner H, Russell S, Segovia J, Smedira N, Stehlik J, Wagner F, Consensus Conference Participants. Report from a consensus conference on primary graft dysfunction after cardiac transplantation. J Heart Lung Transplant 2014;33:327-340.
CR IP T
11. Squiers JJ, Saracino G, Chamogeorgakis T, MacHannaford JC, Rafael AE, Gonzalez-Stawinski GV, Hall SA, DiMaio JM, Lima B. Application of the International Society for Heart and Lung
Transplantation (ISHLT) criteria for primary graft dysfunction after cardiac transplantation: outcomes from a high-volume centre. Eur J Cardiothorac Surg 2017;51:263-270.
AN US
12. Rajagopal K, Lima B, Petersen RP, Mesis RG, Daneshmand MA, Lemaire A, Felker GM, Hernandez AF, Rogers JG, Lodge AJ, Milano CA. Infectious complications in extended criteria heart transplantation. J Heart Lung Transplant 2008;27:1217-1221.
13. Chemmalakuzhy J, Costanzo MR, Meyer P, Piccione W, Kao W, Winkel E, Saltzberg M, Heroux A,
M
Parrillo J. Hypotension, acidosis, and vasodilatation syndrome post-heart transplant: prognostic variables and outcomes. J Heart Lung Transplant 2001;20:1075-1083.
ED
14. Wan S, Marchant A, DeSmet JM, Antoine M, Zhang H, Vachiery JL, Goldman M, Vincent JL, LeClerc JL. Human cytokine responses to cardiac transplantation and coronary artery bypass grafting.
CE
PT
J Thorac Cardiovasc Surg 1996;111:469-477.
AC
FIGURE LEGENDS
ACCEPTED MANUSCRIPT
AN US
CR IP T
11
Figure 1: Kaplan-Meier analysis of 1-year (A) and 3-year (B) survival by severity of vasoplegia for the
AC
CE
PT
ED
M
entire cohort. The effect of vasoplegia severity was significant both at 1 year and at 3 years (P<0.001).
ACCEPTED MANUSCRIPT 12 Figure 2: Kaplan-Meier analysis of 1-year (A) and 3-year (B) survival by severity of vasoplegia for the subgroup of patients with no PGD.
Table 1. Preoperative Patient Characteristics by Severity of Vasoplegia (N=240) Vasoplegia Severity Variable
P-value
Mild (n=23)
Moderate/Severe (n=35)
60 [54, 66]
60 [51, 65]
0.88
19 (83%)
27 (77%)
0.64
4 (17%)
5 (14%)
0.44
121 (67%)
18 (78%)
23 (66%)
26 (14%)
1 (4%)
7 (20%)
85 (47%)
13 (57%)
17 (49%)
0.67
32 (18%)
2 (9%)
4 (11%)
0.36
1.4 [1.1, 1.6]
1.5 [1.3, 1.9]
1.5 [1.3, 1.8]
0.012
28 [26, 31]
29 [24, 36]
29 [25, 34]
0.47
2.07 [1.92, 2.24]
2.13 [2.01. 2.30]
2.16 [1.94, 2.30]
0.19
36 (20%)
5 (22%)
11 (31%)
On intensive care prior to transplant
87 (48%)
9 (39%)
10 (29%)
0.078
LVAD pre-transplant
44 (26%)
10 (48%)
15 (56%)
0.003
ECMO pre-transplant
4 (2%)
4 (19%)
5 (19%)
0.0008
56 (31%)
11 (48%)
17 (49%)
0.064
1.97 [1.14, 3.13]
2.12 [1.51, 3.04]
1.97 [1.41, 3.02]
0.75
Previous sternotomy
77 (45%)
12 (57%)
20 (74%)
0.015
Wait list time (days)
22.5 [8, 104]
38 [7, 141]
23 [9, 58]
0.79
Recipient Variables 59 [52, 65]
Men
135 (74%)
Black
35 (19%)
AN US
Age (years)
White Other Ischemic cardiomyopathy Diabetes mellitus
M
Creatinine (mg/dL)
CE
PT
Hospitalized
ED
Body mass index (kg/m2) Body surface area (m2)
UNOS status 1A
Pulmonary vascular resistance (Wood units)
AC
CR IP T
None (n=182)
Subjective global assessment:
0.15
Well nourished
112 (69%)
17 (85%)
18 (72%)
Moderately malnourished
48 (29%)
2 (10%)
5 (20%)
3 (2%)
1 (5%)
2 (8%)
Severely malnourished
ACCEPTED MANUSCRIPT 13
Donor Variables 34 [23, 44]
34 [24, 42]
31 [26, 44]
0.98
Men
81 (45%)
8 (35%)
19 (54%)
0.33
Black
29 (16%)
2 (9%)
7 (20%)
0.061
White
145 (80%)
18 (78%)
22 (63%)
Other
8 (4%)
3 (13%)
Cause of death: 75 (41%)
Head trauma
55 (30%)
Other
52 (29%)
Previous cardiopulmonary resuscitation Body mass index (kg/m2) Donor/recipient weight ratio
Ejection fraction (%) Cytomegalovirus mismatch
PT
Distance from center (mi)
ED
Left ventricular wall thickness (cm)
Total ischemic time (min)
0.19 16 (46%)
10 (44%)
10 (29%)
9 (39%)
9 (26%)
47 (26%)
3 (13%)
7 (20%)
0.31
29 [25, 34]
28.5 [23, 32]
27 [24, 34]
0.81
0.96 [0.82, 1.13]
0.80 [0.72, 0.94]
0.86 [0.77, 1.15]
0.020
3 [-13, 18]
20 [6, 28]
14 [-15, 23]
0.020
7 [-6, 19]
17.5 [10, 33]
11 [-7, 26]
0.016
1.00 [0.90, 1.20]
1.10 [0.89, 1.20]
1.06 [0.90, 1.20]
0.99
M
Total body weight (% difference)
6 (17%)
4 (17%)
AN US
Anoxia
Predicted heart mass (% difference)
CR IP T
Age (years)
62.5 [55, 67.5]
60 [55, 65]
60 [54, 69]
0.49
72 (40%)
5 (22%)
12 (35%)
0.22
287 [38, 747]
362 [192, 904]
241 [38, 431]
0.17
221 [171, 268]
234 [202, 274]
209 [182, 273]
0.35
CE
Abbreviations: ECMO, extracorporeal membrane oxygenation; UNOS, United Network for Organ Sharing; LVAD, left ventricular assist device
AC
Table 2. Postoperative Outcomes by Severity of Vasoplegia (N=240)
Variable
Vasoplegia Severity None (n=182)
Mild (n=23)
Moderate/Severe (n=35)
Intensive care length of stay (days)
3 [2, 4]
5 [3, 8]
10.5 [5, 30]
Hospital length of stay (days)
8 [6, 12]
13 [9, 17]
18 [12, 45]
ACCEPTED MANUSCRIPT 14
Inotrope score
15.5 [11.5, 21.5]
19.5 [16, 30]
31 [26.5, 40]
None
150 (82%)
10 (44%)
17 (49%)
Mild
19 (10%)
7 (30%)
4 (11%)
Moderate
7 (3.8%)
1 (4.3%)
7 (20%)
Severe
6 (3.3%)
5 (22%)
7 (20%)
Dialysis
1 (0.5%)
1 (4.3%)
5 (14%)
Reoperate for bleeding
15 (8.2%)
4 (17%)
11 (31%)
Stroke
4 (2.2%)
1 (4.3%)
2 (5.7%)
Pneumonia
3 (1.6%)
2 (8.7%)
11 (31%)
Bacterial infection
14 (7.7%)
4 (17%)
13 (37%)
3 (1.6%)
2 (8.7%)
9 (26%)
4 (2.2%)
1 (4.3%)
4 (11%)
0
0
3 (8.6%)
5 (2.7%)
2 (8.7%)
2 (5.7%)
6 (3.3%)
1 (4.3%)
7 (20%)
3 (1.6%)
1 (4.3%)
1 (2.9%)
1 [0, 2]
1 [0, 2]
3 [1, 6]
0 [0, 1]
0 [0, 1]
0 [0, 2]
0 [0, 0]
0 [0, 0]
0 [0, 0]
0 [0, 0]
0 [0, 1]
0 [0, 2]
1 (0.5%)
2 (8.7%)
4 (11%)
AN US
CR IP T
Primary graft dysfunction:
Fungal infection Sepsis Mediastinitis
Bloodstream infection
Blood product utilization: Packed red blood cells
Cryoprecipitate
CE
Platelets
PT
Fresh frozen plasma
ED
Severe rejection (3R) within 3 months
M
Wound infection
30-day/in-hospital mortality
AC
Table 3. Postoperative Outcomes by Severity of Vasoplegia in the Absence of Primary Graft Dysfunction (N=177).
Variable
Intensive care length of stay (days) Hospital length of stay (days)
Vasoplegia Severity None (n=182)
Mild (n=23)
Moderate/Severe (n=35)
2 [2, 3]
5 [3, 6]
7 [4, 15]
7.5 [6, 10]
11 [9, 15]
18 [13, 24]
ACCEPTED MANUSCRIPT 15
Inotrope score
14.5 [10.5, 19.5]
19 [16, 20.5]
31 [29.5, 38.5]
0
0
0
Reoperate for bleeding
9 (6.0%)
1 (10%)
4 (24%)
Stroke
3 (2.0%)
1 (10%)
1 (5.9%)
Pneumonia
3 (2.0%)
0
3 (18%)
Bacterial infection
10 (6.7%)
1 (10%)
3 (18%)
Fungal infection
2 (1.3%)
1 (10%)
1 (5.9%)
Sepsis
3 (2.0%)
0
0
0
0
0
Wound infection
4 (2.7%)
1 (10%)
1 (5.9%)
Bloodstream infection
3 (2.0%)
0
0
Severe rejection (3R) within 3 months
3 (2.0%)
0
1 (5.9%)
Packed red blood cells
1 [0, 2]
0 [0, 1]
2 [1, 4]
0 [0, 0]
0 [0, 2]
0 [0, 3]
0 [0, 0]
0 [0, 0]
0 [0, 0]
0 [0, 0]
0 [0, 0]
0 [0, 0]
0
0
CR IP T
Dialysis
AN US
Mediastinitis
Fresh frozen plasma Cryoprecipitate Platelets
ED
M
30-day/in-hospital mortality 0 Table 4. Univariate and Multivariate Predictors of Moderate or Severe Vasoplegia.
PT
Variable
All Patients
Odds ratio
Patients with no p dysfunc P-value
[95% Confidence interval]
Odds ratio [95% Confidence interval]
CE
Univariate analyses Male gender
1.34 [0.65, 2.73]
0.43
0.94 [0.37, 2.39]
1.03 [0.53, 2.01]
0.92
0.90 [0.36, 2.29]
3.39 [1.57, 7.32]
0.002*
2.40 [0.78, 7.38]
UNOS 1A status
2.07 [1.13, 3.78]
0.019*
3.89 [1.65, 9.16]
LVAD prior to transplant
2.91 [1.58, 5.35]
0.0006*
3.68 [1.59, 8.57]
ECMO prior to transplant
2.13 [0.35, 13.1]
0.41
5.74 [0.35, 94.6]
Creatinine (continuous scale)
2.61 [1.21, 5.63]
0.015*
3.00 [0.93, 9.66]
Creatinine ≥2.0 mg/dL
1.65 [0.63, 4.32]
0.31
1.28 [0.26, 6.28]
Hospitalized prior to transplant
1.52 [0.77, 3.01]
0.23
1.92 [0.76, 4.84]
On intensive care prior to transplant
0.49 [0.14, 1.73]
0.27
0.70 [0.15, 3.23]
Recipient age ≥65 y
AC
Recipient body mass index ≥35 kg/m
2
ACCEPTED MANUSCRIPT 16 2.75 [1.46, 5.20]
0.0018*
2.55 [1.07, 6.03]
Donor age ≥50 y
0.82 [0.34, 2.01]
0.67
1.13 [0.35, 3.61]
Donor male gender
1.08 [0.59, 1.95]
0.81
1.22 [0.54, 2.76]
Donor ejection fraction <50%
0.69 [0.15, 3.25]
0.64
0.48 [0.06, 3.90]
Donor undersized ≥30% by body weight
1.40 [0.51, 3.84]
0.51
1.31 [0.35, 4.94]
Donor undersized by predicted heart mass (continuous scale)
1.02 [1.00, 1.03]
0.043*
1.00 [0.98, 1.02]
Donor undersized ≥30% by predicted heart mass
1.81 [0.79, 4.17]
0.16*
1.68 [0.51, 5.55]
Donor left ventricular wall thickness >1.3 cm
0.69 [0.27, 1.76]
0.43
1.18 [0.36, 3.82]
Pulmonary vascular resistance (Woods)
1.01 [0.84, 1.22]
0.90
1.05 [0.82, 1.34]
Donor sequence number >80
0.59 [0.28, 1.27]
0.18*
0.69 [0.25, 1.96]
2.65 [1.18, 5.98]
0.019
Multivariate analyses
AN US
Recipient body mass index ≥35 kg/m2
CR IP T
Prior sternotomy
LVAD prior to transplant
2.47 [1.30, 4.69]
UNOS 1A status
0.0056
*Variables with a P-value <0.2 in the univariate analysis were selected entered into the stepwise
M
multivariate regression model.
Abbreviations: ECMO, extracorporeal membrane oxygenation; UNOS, United Network for
AC
CE
PT
ED
Organ Sharing; LVAD, left ventricular assist device
4.24 [1.76, 10.2]