Underutilization of left heart catheterization in kidney transplant patients presenting with non-ST segment elevation myocardial infarction

Background Cardiovascular disease (CVD) is the leading cause of mortality in kidney transplant (KT) patients. The perceived risk of contrast-induced nephropathy (CIN) may create a reluctance to perform coronary angiography in patients presenting with non-ST segment elevation myocardial infarction (NSTEMI). Methods National Inpatient Sample (NIS) Database was used to sample individuals presenting with NSTEMI. Patients were stratified into KT and Non-KT cohorts. Outcomes included left heart catheterization rates, mortality, arrhythmias, acute kidney injury/acute renal failure (AKI/ARF), and extended length of hospital stay (ELOS) (>72 h). Propensity matching (1:1 ratio) and regression analyses were performed. Results Out of 336,354 patients with NSTEMI, 742 patients were in the KT group. KT patients were less likely to have LHC relative to non-KT patients (22.0 % vs 18.3 %); a difference that persisted on post-match analysis (27.1 % vs 19.4 %). On pre-match analysis, KT transplant patients that underwent LHC had lower mortality (10.3 % vs 0.7 %), AKI/ARF (44.6 % vs 27.9 %), arrhythmias (30.4 % vs 20.6 %) and lower ELOS (58.6 % vs 41.9 %). Post-match, KT cohort patient that underwent LHC had lower arrhythmias (OR:0.60[0.38–0.96]), AKI/ARF (OR = 0.51[0.34–0.77]), ELOS (OR:0.49[0.34–0.73]). Conclusion KT patients underwent LHC much less frequently than their non-KT counterparts for NSTEMI. Coronary angiography and subsequent revascularization were associated with a significant decrease in morbidity and mortality. This theorized risk of CIN should not outweigh the benefit of LHC in KT patients.


Introduction
Kidney transplantation (KT) is the recommended treatment for endstage renal disease (ESRD) patients and is associated with improved outcomes [1].The survival benefit of kidney transplantation is mostly secondary to an overall decrease in cardiovascular disease (CVD) burden [2].The incidence of acute myocardial infarction (AMI) in the ESRD population in the USA has been declining and is likely secondary to improvements in primary prevention efforts [3,4].
Despite this risk reduction, KT patients remain at high risk for CVDrelated morbidity and mortality, compared to general population [5,6].In fact, CVD is the leading cause of mortality in KT patients with a functioning allograft [7].Specifically, AMI and its related complications remain the leading cause of death in this patient population [8,9].AMI encompasses ST-elevation myocardial infarction (STEMI) and non-STEMI (NSTEMI).The incidence of STEMI in KT patients is 1.3 times higher than that of the general population [3,10].At one year post-KT, the incidence of STEMI was 2.3 % and increased to 5.8 % in a 5year follow-up period [10][11][12].However, a history of KT is more likely to be associated with NSTEMI compared to STEMI upon presentation to the hospital [13].Recent reports showed that in-hospital mortality among KT patients with STEMI is declining, while that of NSTEMI patients remains unchanged [13,14].KT patients are more prone to develop acute kidney injury (AKI) after receiving iodinated contrast based on recent emerging data [15,16].Despite the elevated risk of CVD in KT patients, this may create a reluctance to perform coronary angiography due to the perceived risk of contrast-induced nephropathy.Thus, this can create a dilemma of optimally treating coronary artery disease (CAD) while maximizing kidney graft integrity.
KT patients represent a unique and vulnerable patient population.Nationwide data evaluating the outcomes and complications of NSTEMI in KT patients are limited.A better understanding of these outcomes can provide valuable prognostication information that could enhance cardiovascular care for these patients while maximizing graft integrity.Hence, we sought to study the rates of left heart catheterizations (LHC) in KT vs non-KT transplant patients and its effect on outcomes and mortality.

Methods
Individuals presenting with NSTEMI were sampled from National Inpatient Sample (NIS) database (2016-2018).The NIS database is the largest inpatient database in the United States and is funded through Healthcare Cost and Utilization Project (HCUP) [17].IRB is exempt as per the HCUP data use agreement since patient information is deidentified.
Comorbidities and baseline demographics were collected using International Classification of Disease 10 codes (Supplementary Table 1).Patients were stratified into KT and non-KT groups.Patients with an age <18, those with missing data, or those with chronic kidney disease 2 or higher, including ESRD, were excluded from the study (Fig. 1).
The primary outcome was LHC rates between the KT and non-KT transplant cohorts.Secondary outcomes were analyzed in the KT group and compared patients who underwent LHC to those who did not.Secondary outcomes included in-hospital mortality, arrhythmias (a composite of atrial fibrillation, atrial flutter, supraventricular tachycardia, atrioventricular blocks, bundle branch blocks, ventricular tachycardia), AKI (including post-LHC AKI/acute renal failure), acute heart failure and extended length of hospital stay (ELOS; defined as >72 h).
KT and non-KT groups were then matched 1:1 using Greedy propensity with R version 4.1.2(R Foundation for Statistical Computing, Vienna, Austria).The caliper for propensity score difference was 0.0000001.The two cohorts were matched on age, gender, race, CAD, carotid artery disease, chronic obstructive pulmonary disease, diabetes mellitus, dyslipidemia, hypertension, obesity, chronic heart failure, smoking, and peripheral vascular disease.After matching, the mean difference in propensity score was <0.1 which signified successful matching.A Chi-Square analysis was performed to confirm that the two groups were similar (p > 0.99).
Pre-match and post-match univariate analyses (chi-square and student's t-tests) were conducted for the primary outcome between the two cohorts.Continuous variables were analyzed using student t-tests and ANOVAs.Categorical variables were analyzed using chi-square analyses, and Fisher's exact tests.Only KT patients were included in the analysis for the secondary outcomes.Binary logistic regression analyses on pre-matched and post-matched KT cohorts were conducted.For the pre-matched cohorts, all the variables that were utilized for propensity score matching, and kidney transplant status were entered in the binary logistic regression model.For the post-match binary logistic regression model, only kidney transplant status was entered.A p-value <0.05 was considered statistically significant.All statistical analyses were done using IBM SPSS Statistics for Windows, version 28 (IBM Corp., Armonk, N.Y., USA) and R version 4.1.2(R Foundation for Statistical Computing, Vienna, Austria).

Results
Out of a total of 456,724 patients, 336,354 patients were included in the study.There were 335,612 non-KT patients and 742 KT patients.KT patients were younger, Caucasians, and had a higher prevalence of CAD, diabetes mellitus, and peripheral vascular disease.The non-KT group had more females and more prevalence of chronic obstructive pulmonary disease, obesity, and smoking (Table 1; Fig. 1).
KT patients who underwent LHC and those who did not were compared for secondary outcomes.Table 4 shows the differences in baseline characteristics.Patients who underwent LHC had a higher prevalence of CAD and dyslipidemia, and lower rates of chronic heart failure.Binary logistic regression analysis for secondary outcomes  revealed similar results pre-match and post-match (Table 5).Patients that underwent LHC had lower rates of arrhythmia [OR: 0.60 (0.38-0.96); p = 0.03] and extended LOS [OR: 0.49(0.33-0.73);p < 0.001].
One non-KT patient had a post-LHC AKI/ARF, whereas no KT patients had post-LHC acute kidney injury/acute renal failure.

Discussion
In this study, KT patients with NSTEMI were less likely to undergo LHC than their non-KT counterparts.LHC in KT patients presenting with NSTEMI was associated with less AKI, arrhythmias, less ELOS, and lower mortality.
CVD remains the leading cause of morbidity and mortality in KT patients [18].Over 10 % of KT patients experience an AMI within three years of transplantation [19].Traditional CAD risk factors such as diabetes mellitus and hypertension, are known to be more prevalent in KT patients.These factors are often the underlying etiologies of ESRD and further lead to adverse cardiovascular events [20].Previous studies also suggested that cardioprotective medications are either less effective or underutilized in KT individuals [20,21].
Previous cohort studies have evaluated the cardiovascular outcomes following KT.A 5-year follow-up study of KT recipients reported that 5.8 % of the KT group and 2.8 % of the non-KT group developed MI.Similar to our results, patients in the KT group had higher baseline comorbidities.It was concluded that KT was independently associated   with a 45 % higher risk of MI when compared to non-KT, with a predominance of NSTEMI.They showed that MI in KT patients was independently associated with a 15 % increased risk of mortality for patients with MI and non-KT [12].This study provided valuable trends but did not explore whether patients received an intervention or medical management.
Recent literature has focused on examining disparities in procedures based on age, sex, and race.Mcneil et al. coined the term "renalisim" to reflect the prejudice of coronary angiography underutilization after AMI in elderly patients with CKD compared to elderly patients without CKD [22].Studies have also examined the variation in the rates of cardiac procedures due to the presence of certain coexisting medical conditions [23,24].Certain coexisting conditions frequently resulted in conservative medical management, perhaps secondary to physicians' aversion to risk [22].Being a kidney transplant recipient is one of those conditions.In the study by Fox et al., KT patients were less likely to undergo coronary angiography or revascularization after AMI [21].This is likely due to the fear of inducing nephrotoxicity by radiocontrast media.Management of KT population requires a non-biased multidisciplinary approach with careful balancing of the potential benefit of coronary angiography with the risks of radiocontrast-associated nephropathy vs a more conservative medical approach.
Transplanted kidneys can be more susceptible to hemodynamic insults that lead to AKI.The absence of sympathetic innervation predisposes it to hyperfiltration injury [25].Another predisposing factor is the concomitant use of calcineurin inhibitors, which have nephrotoxic effects.These medications may cause afferent arteriolar vasoconstriction and tubular insults [26].Our study showed that KT patients who underwent LHC had paradoxically lower rates of AKI/acute renal failure.The likely explanation for this finding may be multifactorial.Management of hospitalized KT patients entails early consultation with transplant nephrologists that contributes to improved outcomes [27,28].This early collegial and collaborative management between nephrologists and cardiologists may lower the risk of cardio-renal syndromes and leads to optimization of fluid-dynamic management, hence reducing the risk of AKI in this cohort [29].
Our study has certain limitations.It is a retrospective design that may have led to certain selection biases.Thrombolysis in Myocardial Infarction (TIMI) and Global Registry of Acute Coronary Events (GRACE) score, commonly used to guide management in NSTEMI patients, could not be calculated.However, these limitations were largely eliminated by using propensity-score matching.The data's accuracy depends on coding practices that may vary between hospitals.However, the large size of the database reduces this potential limitation.NIS is an administrative database and lacks information on medication utilization and outpatient follow-up data.Data on functional status of KT patients was not available which might have influenced the treatment strategy.Propensity scores can only adjust for associations among observed covariates and the chosen treatment strategy but cannot discern what specific medications were utilized.Thus, there can be residual unmeasured confounders.Outcomes in the NIS are limited to index hospitalizations and hence only short-term outcomes could be analyzed.

Conclusion
The results of our study confirmed that KT patients were less likely to receive coronary angiography after NSTEMI compared to those without KT.This significant difference persisted even after 1:1 matching for baseline demographics and comorbidities.Coronary angiography and subsequent revascularization were associated with a significant decrease in morbidity and mortality in this population.Conventionally, the theoretical risk of nephrotoxicity secondary to cardiac angiography is one of the reasons this patient population received fewer coronary interventions.Strategies designed to decrease KT bias and allow KT patients to undergo coronary angiography for NSTEMI should be further explored as the benefit could outweigh the potential risks.

Fig. 2 .
Fig. 2. Post-match analysis outcomes of kidney transplant patients who underwent left heart catheterization (LHC) versus those who did not.

Table 1
Baseline characteristics and comorbidities.Values are N(%) or mean ± SD.

Table 2
Primary Outcome-Rates of left heart catheterization between KT and non-KT cohorts before and after matching.Values are N(%).KT = kidney transplant.

Table 3
Univariate analysis for secondary outcomes within the kidney transplant group.LHC = Left heart catheterization.

Table 4
Baseline Characteristics and comorbidities of Kidney Transplant cohort.Values are N(%) or mean ± SD.LHC = Left heart catheterization.