Abstract
Background:
Patients with chronic renal failure (CRF) on hemodialysis are at increased risk for developing infective endocarditis (IE). However, outcomes of surgical treatment for IE in these patients have not been well studied.
Methods:
Between 1997 to 2017, 539 patients underwent surgery for IE. Of these patients, 125 were on hemodialysis for end-stage renal disease (ESRD) and 414 had no history of CRF. Primary endpoints compared in this study were short- and long-term survival.
Results:
Preoperatively, dialysis patients had higher incidences of diabetes (43% vs. 18%), hypertension (79% vs. 49%), congestive heart failure (63% vs. 48%), cardiogenic shock (13% vs. 5.3), and sepsis (29% vs. 18%), all p<0.05. Postoperatively, they experienced higher rates of prolonged mechanical ventilation (54% vs. 22%), pneumonia (17% vs. 5.6%), sepsis (6.4% vs. 1.0%), cardiac arrest (7.2% vs. 1.7%), gastrointestinal events (14% vs. 5.1%), and operative mortality (14% vs. 5.8%), all p<0.05. Five- and ten-year survival were significantly worse in the dialysis group at 29% and 16%, respectively, compared to 72% and 53% in non-CRF patients (p<0.0001). ESRD was a risk factor for both short- (OR=2.0) and long-term mortality (HR=2.7). Rates of reoperation in dialysis patients were very low: five- and ten-year incidences were 0% and 2.0%, respectively.
Conclusions:
In ESRD patients with IE, poor postoperative outcomes emphasized the importance of prevention and raised the question of whether indications for surgery in the general population are appropriate for patients who are dialysis-dependent. Additionally, low rates of reoperation supported the use of bioprosthetic valves in these patients.
Despite the near eradication of rheumatic heart disease in the United States, the prevalence of infective endocarditis (IE) has continued to increase1. With a growing number of chronically ill patients receiving intensive and invasive therapies, healthcare-associated procedures have been recognized as an important etiology for IE. Within this category, hemodialysis has been identified as a major risk factor. Patients on chronic dialysis experience an incidence of IE at rates 18 times higher than the general population2. This is predominantly thought to be a result of transient bacteremia occurring from repetitive vascular access, though accelerated valvular disease and uremia-related immune system deficiencies are also considered to be contributing factors2–4. Meanwhile, the overall incidence of chronic renal failure (CRF) and end-stage renal disease (ESRD) has been on the rise. With the expanding prevalence of obesity and diabetes, as well as a decline in ESRD mortality, the projected number of patients with dialysis-dependent renal failure has been estimated to increase between 29%−68% by 20305.
Not only are patients with ESRD at higher risk for developing IE, they also experience higher mortality compared to the general population once infected. Among dialysis patients, the in-hospital mortality has been reported at 52%, with an even greater mortality among patients requiring valve surgery6. In general, it has been well-established that the outcomes associated with cardiac surgery are inferior in the setting of ESRD7,8. Likewise, studies specifically examining the surgical treatment of IE have shown increased perioperative morbidity and mortality in this population9–11. However, current literature on this topic has been limited, and further investigation is needed to ascertain the impact of renal failure on operative outcomes in patients with IE and its role in the risk stratisfaction of surgical candidates.
We therefore examined the short- and long-term survival of patients with and without ESRD following surgical intervention for IE at a high-volume cardiovascular center. We hypothesized that patients on chronic hemodialysis would have inferior outcomes and lower survival compared to those who were not.
Patients and Methods
This study was approved by the Institutional Review Board at Michigan Medicine on March 31st, 2018 (HUM00142927), and a waiver of informed consent was obtained.
Patient Selection
Between the years 1997 to 2017, 539 patients underwent surgery for IE at the Michigan Medicine Cardiovascular Center. Valvular involvement in these patients included the aortic, mitral, tricuspid, or multiple valves. Of these patients, 125 (23%) were on hemodialysis for ESRD (dialysis group) and 414 (77%) had normal kidney function with no history of chronic renal failure (non-CRF group).
Data Collection
Pre-, intra-, and post-operative data elements were obtained from the Society of Thoracic Surgery from Michigan Medicine’s Cardiac Surgery Data Warehouse. Investigators supplemented data collection through review of medical records and operative reports. All living patients with unknown medical status through record review at the time of data collection (n=344) were mailed Institutional Review Board-approved surveys examining current health status and subsequent cardiac reoperation. Survival data were obtained through the National Death Index database through December 31st, 201812 and supplemented with medical record review and questionnaire response via mailed survey or phone call. Out of 539 patients, there was follow-up regarding reoperation and long-term events for 459 (85%), of which 262 (49%) had died. 80 (15%) patients did not respond to the survey and were lost to follow-up. These patients were censored to the date of last known status of reoperation.
Definitions
Primary endpoints in this study were operative mortality and long-term survival at five and ten years. Operative mortality included all deaths occurring during the hospitalization in which the operation was performed and after discharge from the hospital before the thirtieth postoperative day. Elective operations were defined as surgeries scheduled days to weeks in advance for patients with stable cardiac function. Urgent operations were performed during the same admission in which the patients were diagnosed with IE after allowing for medical stabilization and preoperative workup. Emergency operations were performed without delay. The definitions of variables including cardiogenic shock, chronic lung disease, and prolonged mechanical ventilation were based on those provided by the Society of Thoracic Surgeons13. Postoperative gastrointestinal events included bowel ischemia, perforation, pancreatitis, colitis, gastrointestinal bleeding, ulceration, ileus, and cholecystitis.
Statistical Analysis
Data are presented as median (interquartile range, 25th percentile, 75th percentile) for continuous data and n (%) for categorical data. Univariate comparisons between dialysis and non-CRF patients were performed using chi-square tests for categorical data and Wilcoxon rank-sum tests for continuous data. Kolmogorov-Smirnov D and Cramer-von Mises tests revealed a non-normal distribution of the data. Therefore, conservative non-parametric tests were performed to analyze continuous variables. Multivariable logistic regression was used to calculate odds ratios for risk factors for operative mortality, adjusting for ESRD, age, gender, prosthetic valve endocarditis, liver disease, congestive heart failure, cardiogenic shock, postoperative sepsis, and decade of operation. Cox regression was used to calculate hazard ratios for risk factors for late mortality, adjusting for ESRD, age, gender, intravenous drug use, coronary artery disease, chronic obstructive pulmonary disease, congestive heart failure, cardiogenic shock, liver disease, and postoperative sepsis. The Kaplan-Meier method was used for crude survival curves estimating time to death since operation. Survival comparisons between the dialysis and non-CRF groups were made using a log-rank test. Gray’s test was performed to determine the incidence of reoperation, adjusting for death as a competing factor. P-values of less than 0.05 (2-tailed) were considered statistically significant.
Results
Demographics and Preoperative Data
The median age (54 vs. 52 years old) and percentage of intravenous drug users were similar between the dialysis and non-CRF groups. ESRD patients had significantly higher rates of diabetes, hypertension, and congestive heart failure. They also presented with higher incidences of sepsis and cardiogenic shock. However, there was no significant difference between the two groups regarding valve involvement or history of prior cardiovascular surgery (Table 1). Dialysis patients were most commonly infected by staphylococcal species, primarily Staphylococcus aureus (39%). In contrast, the most common causative organism in non-dialysis patients was Streptococcus (29%) (Table 2).
Table 1.
Preoperative and Demographic Data
| Variable | Hemodialysis (n=125) | Non-Renal Failure (n=414) | p-value |
|---|---|---|---|
| Age (median) | 54 (45, 64) | 52 (41, 64) | 0.12 |
| Female Sex | 43 (34) | 118 (29) | 0.22 |
| IV Drug Use | 21 (17) | 61 (15) | 0.57 |
| Coronary Artery Disease | 32 (26) | 79 (19) | 0.13 |
| Diabetes | 54 (43) | 75 (18) | <0.0001 |
| Dyslipidemia | 59 (47) | 123 (30) | 0.0004 |
| Hypertension | 99 (79) | 202 (49) | <0.0001 |
| History of Endocarditis | 11 (8.8) | 69 (17) | 0.03 |
| Current Smoker | 50 (40) | 102 (25) | 0.001 |
| Lung Disease | 0.18 | ||
| Moderate | 6 (4.8) | 13 (3.1) | 0.41 |
| Severe | 9 (7.2) | 12 (2.9) | 0.04 |
| Pneumonia | 18 (14) | 34 (8.2) | 0.06 |
| Liver Disease | 12 (9.6) | 31 (7.5) | 0.45 |
| Stroke | 26 (21) | 78 (19) | 0.70 |
| Congestive Heart Failure | 79 (63) | 199(48) | 0.003 |
| NYHA Class | 0.002 | ||
| Class I | 2(1.7) | 38(10) | 0.0008 |
| Class II | 10 (8.7) | 45 (12) | 0.24 |
| Class III | 33 (29) | 88 (23) | 0.51 |
| Class IV | 41 (36) | 88(23) | 0.03 |
| Unknown | 39(31) | 155 (37) | 0.21 |
| Sepsis | 36 (29) | 76(18) | 0.02 |
| Cardiogenic Shock | 16 (13) | 22 (5.3) | 0.008 |
| Arrhythmia | 27 (22) | 48 (12) | 0.008 |
| Aortic Insufficiency | 0.29 | ||
| Moderate | 19 (15) | 36 (8.7) | 0.04 |
| Severe | 39 (31) | 129 (31) | 1.0 |
| Aortic Stenosis | 32 (26) | 145 (35) | 0.05 |
| Aortic Aneurysms | 5 (4.0) | 32 (7.7) | 0.15 |
| Previous Cardiovascular Surgery | |||
| Previous CABG | 16 (13) | 36 (8.7) | 0.17 |
| Previous Aortic Valve Surgery, | 21 (21) | 88 (21) | 0.90 |
| including TAVR | |||
| revious Mitral Valve Surgery | 13 (10.4) | 38 (9.2) | 0.68 |
| Previous Aortic Surgery | 4 (3.2) | 25 (6.0) | 0.22 |
| Incidence | 0.62 | ||
| First cardiovascular surgery | 77 (62) | 273 (66) | 0.39 |
| First cardiovascular reoperation | 39 (31) | 105 (25) | 0.21 |
| Second or more cardiovascular | 9 (7.2) | 36 (8.7) | 0.60 |
| reoperation | |||
| STS Risk Score* | |||
| Morbidity | 53% (35%, 75%) | 24% (15%, 38%) | <0.0001 |
| Mortality | 11% (5%, 38%) | 2% (1%, 5%) | <0.0001 |
Data presented as median (25th, 75th percentile) for continuous data and n (%) for categorical data. Abbreviations: CABG, coronary artery bypass graft; NYHA, New York Heart Association; STS, Society of Thoracic Surgeons; TAVR, transcatheter aortic valve replacement.
STS risk scores were available for only 21 (17%) dialysis patients and 70 (17%) non-CRF patients.
Table 2.
Characteristics of Endocarditis
| Variable | Hemodialysis (n=125) | Non-Renal Failure (n=414) | p-value |
|---|---|---|---|
| Endocarditis Type | |||
| Aortic Valve Only | 58 (46) | 187(45) | 0.81 |
| Mitral Valve Only | 35 (28) | 126 (30) | 0.61 |
| Tricuspid Valve Only | 5 (4.0) | 33 (8.0) | 0.13 |
| Pulmonic Valve Only | 1 (0.8) | 2(0.5) | 0.67 |
| Multiple Valve | 26(21) | 66(16) | 0.20 |
| Root Abscess | 37 (30) | 118(29) | 0.82 |
| Pseudoaneurysms | 1 (0.8) | 6(1.5) | 1.0 |
| Microbiology | |||
| Staphylococcus aureus | 49 (39) | 84 (20) | <0.0001 |
| Coagulase-Negative | 29 (23) | 39 (9.4) | <0.0001 |
| Staphylococcus | |||
| Enterococcus | 19(15) | 59(14) | 0.79 |
| Streptococcus | 11(8.8) | 119(29) | <0.0001 |
| Gram-Negative Rods | 2(1.6) | 19 (4.6) | 0.13 |
| Fungus | 4 (3.2) | 15 (3.6) | 0.82 |
| Other | 1 (0.8) | 11(2.7) | 0.22 |
| Polymicrobial | 1 (0.8) | 5(1.2) | 0.70 |
| Not Identified | 9 (7.2) | 63(15) | 0.02 |
Data presented as n (%).
Intraoperative Data
The proportion of elective, urgent, and emergent operations was similar between dialysis and non-CRF patients. Although dialysis patients underwent more coronary artery bypass grafting, the two groups had similar rates of all valve procedures and other concomitant surgeries. There was no significant difference in the proportion of dialysis and non-CRF patients receiving bioprosthetic, mechanical, or homograft valve replacements. The majority of patients in both groups had bioprosthetic valves placed (81% and 86% of all valves replaced in dialysis and non-CRF patients, respectively). Rates of intraoperative mortality were similar in the two groups (1.6% vs 1.9% in dialysis and non-CRF patients, respectively, p=1.0). (Table 3).
Table 3.
Operative Data
| Variable | Hemodialysis (n=125) | Non-Renal Failure (n=414) | p-value |
|---|---|---|---|
| Status | 0.27 | ||
| Elective | 5 (4.0) | 31 (7.5) | 0.22 |
| Urgent | 108 (86) | 333 (80) | 0.15 |
| Emergent | 12 (9.6) | 50 (12) | 0.52 |
| Circulatory Arrest | 4 (3.2) | 26 (6.3) | 0.27 |
| Cardiopulmonary Bypass Time (min) | 183 (126, 269) | 165 (103, 245) | 0.06 |
| Cross-Clamp Time (min) | 153 (106, 225) | 136 (85, 199) | 0.01 |
| Aortic Valve Procedure | |||
| Repair | 1 (0.8) | 6 (1.5) | 1.0 |
| Replacement | 78 (60) | 248 (62) | 0.68 |
| Freestyle Valves | 39 (31) | 134 (32) | 0.81 |
| Total Root Replacement | 4 (3.0) | 23 (5.2) | 0.45 |
| Modified Inclusion | 32 (24) | 107 (24) | 1.0 |
| Subcoronary | 3 (2.2) | 4 (0.9) | 0.08 |
| Mitral Valve Procedure | |||
| Repair | 32 (26) | 126 (30) | 0.31 |
| Replacement | 34 (27) | 91 (22) | 0.23 |
| Tricuspid Valve Procedure | |||
| Repair | 27 (22) | 62 (15) | 0.10 |
| Replacement | 4 (3.2) | 23 (5.6) | 0.36 |
| Pulmonic Valve Procedure | |||
| Repair | 1 (0.8) | 1 (0.2) | 0.40 |
| Replacement | 1 (0.8) | 1 (0.2) | 0.40 |
| Total Number of Valves Implanted | 117 | 363 | |
| Type of Valve Replacement* | |||
| Bioprosthetic | 101 (81) | 312 (86) | 0.17 |
| Mechanical | 7 (5.6) | 22 (6.1) | 0.86 |
| Homograft | 9 (7.2) | 29 (8.0) | 0.78 |
| Concomitant Procedures | |||
| CABG | 19 (15) | 33 (8.0) | 0.02 |
| Aortic Root Procedure | 47 (38) | 166 (40) | 0.68 |
| Ascending Aorta Procedure | 10 (8.0) | 50 (12) | 0.26 |
| Aortic Hemiarch Procedure | 1 (0.8) | 5 (1.2) | 1.0 |
| Total Aortic Arch Procedure | 0 (0) | 2 (0.5) | 1.0 |
| Intraoperative Mortality | 2 (1.6) | 8 (1.9) | 1.0 |
Data presented as median (25th, 75th percentile) for continuous data and n (%) for categorical data.
Data presented as number of valves (% of total number of valves replaced).
Perioperative Outcomes
Postoperatively, dialysis patients required significantly more red blood cell units and ventilation hours. They experienced higher rates of prolonged mechanical ventilation, pneumonia, sepsis, cardiac arrest, and gastrointestinal events. In-hospital mortality was significantly higher in the dialysis group at 12%, compared to 5.3% in non-CRF group (p=0.01) (Table 4). ESRD was found to be an significant risk factor for operative mortality after multivariable analysis, with an odds ratio (OR) of 2.03 (95% CI: 1.03–4.01; p=0.04) Operation in the first decade (1997–2007) was also a significant risk factor. (Table 5). Overall, the operative mortality in patients who underwent surgery for IE decreased from 9.5% in the first decade to 6.7% in the second decade (2008–2017).
Table 4.
Postoperative Outcomes
| Variable | Hemodialysis (n=125) | Non-Renal Failure (n=414) | p-value |
|---|---|---|---|
| Red Blood Cell Units | 3 (0.5, 6) | 1 (0, 3) | <0.0001 |
| Ventilation Hours | 24 (6.7, 94.7) | 5.3 (0, 18.3) | <0.0001 |
| Reoperation for Bleeding | 1 (0.8) | 13 (3.1) | 0.21 |
| Planned Delayed Sternal Closure | 4 (3.2) | 3 (0.7) | 0.05 |
| Sternal Dehiscence | 2 (1.6) | 0 (0) | 0.05 |
| Sepsis | 8 (6.4) | 4 (1.0) | 0.002 |
| Positive Blood Cultures | 9 (7.2) | 4 (1.0) | 0.0005 |
| Stroke | 4 (3.2) | 4 (1.0) | 0.09 |
| Prolonged Ventilation | 67 (54) | 91 (22) | <0.0001 |
| Pneumonia | 21 (17) | 23 (5.6) | 0.0002 |
| Device | |||
| Pacemaker | 7 (5.6) | 30 (7.2) | 0.69 |
| ICD | 1 (0.8) | 4 (1.0) | 1.0 |
| Cardiac Arrest | 9 (7.2) | 7 (1.7) | 0.004 |
| Multi-System Organ Failure | 4 (3.2) | 4 (1.0) | 0.09 |
| Gastrointestinal Event | 18 (14) | 21 (5.1) | 0.001 |
| Atrial Fibrillation | 38 (30) | 101 (24) | 0.20 |
| In-Hospital Mortality | 15 (12) | 22 (5.3) | 0.01 |
| Causes of In-Hospital Mortality | |||
| Heart Failure | 5 (4.0) | 11 (2.7) | 0.44 |
| Septic Shock | 9 (7.2) | 6 (1.4) | 0.0006 |
| Respiratory Failure | 0 (0) | 2 (0.5) | 0.44 |
| Other | 1 (0.8) | 3 (0.7) | 0.34 |
| Operative Mortality | 18 (14) | 24 (5.8) | 0.004 |
| Reasons for Reoperation | |||
| Structural Valve Degeneration | 1 (0.8) | 15 (3.6) | 0.10 |
| Prosthetic Valve Endocarditis | 1 (0.8) | 7 (1.7) | 0.47 |
| Other | 0 (0) | 6 (1.4) | 0.18 |
Data presented as median (25th, 75th percentile) for continuous data and n (%) for categorical data. Abbreviations: ICD, implanted cardioverter defibrillator.
Table 5.
Risk Factors for Operative Mortality
| Variable | Odds Ratio | p-value |
|---|---|---|
| Renal Failure on Dialysis | 2.31 (1.15–4.62) | 0.02 |
| Age | 1.02 (0.99–1.04) | 0.19 |
| Gender (Female) | 1.24 (0.62–2.51) | 0.54 |
| Prosthetic Valve Endocarditis | 1.86 (0.89–3.90) | 0.10 |
| Liver Disease | 3.07 (1.11–8.53) | 0.03 |
| Congestive Heart Failure | 1.28 (0.63–2.58) | 0.50 |
| Cardiogenic Shock | 1.37 (0.46–4.02) | 0.57 |
| Postoperative Sepsis | 8.21 (2.25–29.9) | 0.001 |
| First Decade (1997–2007) | 2.19 (1.06–4.50) | 0.03 |
Long-Term Outcomes
The median follow-up time of survival was 4.2 years for the whole cohort (95% CI: 1.6–8.5 years). Kaplan-Meier analysis showed significantly worse long-term survival in dialysis patients: five-and ten-year survival were 29% and 16%, respectively, compared to 72% and 53% in non-CRF patients (p<0.0001) (Figure 1). ESRD was a significant risk factor for long-term mortality with a hazard ratio (HR) of 2.67 (95% CI: 2.02–3.54; p<0.0001) (Table 6). The rates of reoperation among dialysis patients was very low: five- and ten-year incidences, after adjusting for death, were 0% (95% CI: 0.08–4.6%) and 2.0% (95% CI: 1.5–22%), respectively. In non-CRF patients, the incidence of reoperation after five and ten years were 2.8% (95% CI: 1.5–4.8%) and 3.5% (95% CI: 1.5–5.9%) (Figure 2). Causes for reoperation, namely structural valve degeneration and prostetic valve endocarditis, were not significantly different between the dialysis and non-CRF groups (Table 4).
Figure 1.
Kaplan-Meier long-term survival following surgical management of IE in dialysis (red) and non-CRF (blue) patients. 5- and 10-year survival were 28.6% (95% CI: 20.2%, 37.6%) and 15.9% (95% CI: 7.5%, 27.2%) in the dialysis group, respectively, compared to 72.1% (95% CI: 67.3%, 76.3%) and 52.8% (95% CI: 46.7%, 58.5%) in the non-CRF group.
Table 6.
Risk Factors for Long-Term Mortality
| Variable | Odds Ratio | p-value |
|---|---|---|
| Renal Failure on Dialysis | 2.67 (2.02–3.54) | <0.0001 |
| Age | 1.02 (1.01–1.03) | <0.0001 |
| Gender (Female) | 1.16 (0.89–1.51) | 0.29 |
| IV Drug Use | 1.54 (1.05–2.25) | 0.03 |
| Coronary Artery Disease | 1.05 (0.78–1.40) | 0.76 |
| Chronic Obstructive Pulmonary Disease | 0.97 (0.65–1.44) | 0.86 |
| Congestive Heart Failure | 1.59 (1.23–2.06) | 0.0005 |
| Cardiogenic Shock | 0.71 (0.43–1.15) | 0.16 |
| Liver Disease | 2.19 (1.34–3.57) | 0.002 |
| Postoperative Sepsis | 2.41 (1.24–4.68) | 0.009 |
Figure 2.
Cumulative incidence function for reoperation after adjusting for death as a competing factor. 5- and 10-year incidence of reoperation were 0% (95% CI: 0–0%) and 2.0% (95% CI: 0.1–9.7%) in the dialysis group, respectively, and 2.8% (95% CI: 1.5–4.8%) and 3.5% (95% CI: 1.5–5.9%) in the non-CRF group.
Comment
This series represents a 20-year surgical experience in ESRD patients on dialysis treated for IE. We found that postoperatively, patients on chronic hemodialysis experienced higher rates of prolonged mechanical ventilation, pneumonia, sepsis, cardiac arrest, and gastrointestinal events. ESRD was associated with higher mortality at 30 days, as well as five and ten years after surgery.
According to the available literature, 25–30% of all patients with acute endocarditis require surgical intervention with an overall operative mortality of 8–16%14–16. However, there is a dearth of information on the outcomes of dialysis-dependent patients treated surgically for IE. In a previous study, Dohmen et al reported that the 30-day mortality was 42% among a sample of 45 dialysis patients following aortic valve replacement for endocarditis9. More recently, Raza et al reported an in-hospital mortality of 13% among 144 dialysis patients, with five- and ten-year survival rates of 25% and 8.4%, respectively10. Our results were consistent with these later findings. We found that the operative mortality in dialysis patients was 14%, compared to 5.8% in patients without CRF (p=0.004). Long-term survival was significantly worse in this patient population with five- and ten-year survival rates of 29% and 16%, compared to 72% and 53% in non-CRF patients. Notably, these figures were also substantially lower than the five and ten-year survival rates of the age-matched general dialysis patient population in the United States, which are 57% and 31%, respectively17. However, operative outcomes have improved over time with implementation of a multidisciplinary team approach, as well as earlier and more aggressive surgical treatment. This was demonstrated by a decrease in the operative mortality between the first and second decades from 9.5% to 6.7%, with increased risk associated with operation in the first decade (OR=2.19).
The inferior outcomes seen in dialysis patients were most likely multifactorial. In agreement with previous reports, we found that the most common causative organism in dialysis patients was Staphylococcus aureus3,4. The higher incidence of staphylococcal infection in these patients indicates that the primary source of bacteremia is the intravascular access site. In contrast, non-CRF patients were most commonly infected by less aggressive Streptococcus species. Moreover, ESRD patients presented with more severe disease and harbored more co-morbidities. Given that CRF is highly associated with cardiovascular disease18, it may be expected that patients in the dialysis group had higher rates of diabetes, dyslipidemia, hypertension, congestive heart failure, NYHA IV symptoms, cardiogenic shock, and arrhythmia preoperatively. CRF is also a known risk factor for infection19, and dialysis patients experienced higher rates of sepsis both before and after surgery. Nonetheless, the proportion of elective, urgent, and emergent operations was similar between the two groups, suggesting that a history of renal failure did not affect the intent to treat or time until surgery. Furthermore, multivariable analysis revealed that ESRD was a significant risk factor for both early and late mortality.
Given the rising prevalence of both ESRD and endocarditis, the identification of methods to address the high morbidity and mortality experienced by this patient population is becoming crucial, particularly in those who require surgical intervention. It has been shown that surgery improves survival in patients with IE, and this finding has been consistent among those who are dialysis-dependent9–11. The sicker presentation and poorer operative outcomes seen in ESRD patients raises the question of whether accepted indications for valve surgery in the general population are applicable to patients on chronic hemodialysis. We found that the STS score for predicted morbidity and mortality were very consistent with our results. It may be a useful tool for stratifying the operative risk in this sick patient population.
Other prospective investigations should endeavor to characterize the long-term outcomes of placing biologic versus mechanical valve prostheses in dialysis patients. Guidelines in the past have recommended against using bioprosthetic valves in these patients due to dialysis-related accelerated valve calcification and degeneration. However, further studies on this topic have demonstrated comparable five-year survival rates among ESRD patients who received either valve type20. Meanwhile, patients with mechanical valves exhibited higher rates of thromboembolism and hemorrhage21,22. In our study, the majority of patients received bioprosthetic valves. 81% of the valve replacements performed in dialysis patients were done using bioprosthetic valves. These bioprostheses included stentless valves (45%) at aortic valve positions and stented valves (55%) at all valve positions. Multiple reports have noted the poor long-term survival of ESRD patients, many of whom are unlikely to outlive their prosthesis lifespan or require reoperation for valve deterioration23,24. Our findings were consistent with this observation. The rate of reoperation was very low in our dialysis group, with five- and ten-year incidences of 0% and 2.0%, respectively. Therefore, the use of bioprostheses may be a reasonable choice for valve replacement in this patient population.
This study is limited by its single center experience and retrospective nature. It has all the limitations associated with a retrospective study. Due to the association of renal failure with cardiovascular disease and an immunocompromised state, it was not possible to completely eliminate confounding variables from the baseline patient characteristics. Additionally, we did not have 100% follow-up on reoperation, therefore the rate of reoperation could be underestimated. The causes of long-term mortality were also largely unavailable.
Conclusion
Patients on hemodialysis for ESRD had significantly worse short- and long-term outcomes following surgery for IE. The prevention of IE in this patient population is of utmost importance. In dialysis-dependent patients who underwent surgical intervention, bioprothesis was a reasonable choice for valve replacement.
Acknowledgments
This project was supported by the Michigan Medicine Cardiac Surgery Department. Special thanks to Francis D. Pagani, MD, Jonathan Haft, MD, Richard L. Prager, MD, Steven F. Bolling, MD, Himanshu J. Patel, MD, G. Michael Deeb, MD, and members of the Yang Lab. Dr. Yang is supported by the NHLBI of NIH K08HL130614 and R01HL141891, and the Phil Jenkins and Darlene & Stephen J. Szatmari Funds.
Footnotes
Publisher's Disclaimer: 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.
Meeting Presentation: The Society of Thoracic Surgeons, 2020
Classifications: Adult; Cardiac; Co-morbidity; Endocarditis; Heart Valves, Multiple; Kidney, Renal Function Failure, dialysis.
References
- 1.Pant S, Patel NJ, Deshmukh A, et al. Trends in infective endocarditis incidence, microbiology, and valve replacement in the United States from 2000 to 2011. J Am Coll Cardiol. 2015;65:2070–2076. [DOI] [PubMed] [Google Scholar]
- 2.Abbott KC, Agodoa LY. Hospitalizations for bacterial endocarditis after initiation of chronic dialysis in the United States. Nephron. 2002;91:203–209. [DOI] [PubMed] [Google Scholar]
- 3.Chaudry MS, Carlson N, Gislason GH, et al. Risk of infective endocarditis in patients with end stage renal disease. Clin J Am Soc Nephrol. 2017;12:1814–1822. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Bhatia N, Agrawal S, Garg A, et al. Trends and outcomes of infective endocarditis in patients on dialysis. Clinical cardiology. 2017;40:423–429. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.McCullough KP, Morgenstern H, Saran R, et al. Projecting ESRD incidence and prevalence in the United States through 2030. J Am Soc Nephrol. 2019;30:127–135. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Spies C, Madison JR, Schatz IJ. Infective endocarditis in patients with end-stage renal disease: clinical presentation and outcome. Arch Intern Med. 2004;164:71–75. [DOI] [PubMed] [Google Scholar]
- 7.Horst M, Mehlhorn U, Hoerstrup SP, et al. Cardiac surgery in patients with end-stage renal disease: 10-year experience. Ann Thorac Surg. 2000;69:96–101. [DOI] [PubMed] [Google Scholar]
- 8.Rahmanian PB, Adams DH, Castillo JG, et al. Early and late outcome of cardiac surgery in dialysis-dependent patients: single-center experience with 245 consecutive patients. J Thorac Cardiovasc Surg. 2008;135:915–922. [DOI] [PubMed] [Google Scholar]
- 9.Dohmen PM, Binner C, Mende M, et al. Outcome of aortic valve replacement for active infective endocarditis in patients on chronic hemodialysis. Ann Thorac Surg. 2015;99:532–538. [DOI] [PubMed] [Google Scholar]
- 10.Raza S, Hussain ST, Rajeswaran J, et al. Value of surgery for infective endocarditis in dialysis patients. J Thorac Cardiovasc Surg. 2017;154:61–70. [DOI] [PubMed] [Google Scholar]
- 11.Chaudry MS, Gislason GH, Kamper AL, et al. The impact of hemodialysis on mortality risk and cause of death in Staphylococcus aureus endocarditis. BMC Nephrology. 2018;19:216. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Centers for Disease Control and Prevention; National Center for Health Statistics. National death index. Available at http://www.cdc.gov/nchs/ndi/index.htm. December 31, 2018.
- 13.The Society of Thoracic Surgeons. Adult Cardiac Surgery Database Data Collection. Available at: https://www.sts.org/registries-research-center/sts-national-database/adult-cardiac-surgery-database/data-collection. Accessed March 22, 2020.
- 14.Alexiou C, Langley SM, Stafford H, Lowes JA, Livesey SA, Monro JL. Surgery for active culture-positive endocarditis: Determinants of early and late outcome. Ann Thorac Surg. 2000;69:1448–1454. [DOI] [PubMed] [Google Scholar]
- 15.Prendergast BD. The changing face of infective endocarditis. Heart. 2006;92:879–885. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Said SM, Abdelsattar ZM, Schaff HV, et al. Outcomes of surgery for infective endocarditis: a single-centre experience of 801 patients. Eur J Cardiothorac Surg. 2018;53:435–439. [DOI] [PubMed] [Google Scholar]
- 17.United States Renal Data System. 2019. ADR reference tables. Available at: https://www.usrds.org/reference.aspx. Accessed March 15, 2020.
- 18.Sarnak MJ, Levey AS, Schoolwerth AC, et al. Kidney disease as a risk factor for development of cardiovascular disease: a statement from the American Heart Association Councils on Kidney in Cardiovascular Disease, High Blood Pressure Research, Clinical Cardiology, and Epidemiology and Prevention. Circulation. 2003;108:2154–2169. [DOI] [PubMed] [Google Scholar]
- 19.Dalrymple LS, Katz R, Kestenbaum B, et al. The risk of infection-related hospitalization with decreased kidney function. Am J Kidney Dis. 2012;59:356–363. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Filsoufi F, Chikwe J, Castillo JG, et al. Prosthesis type has minimal impact on survival after valve surgery in patients with moderate to endstage renal failure. Nephrol Dial Transplant. 2008;23:3613–21. [DOI] [PubMed] [Google Scholar]
- 21.Brinkman WT, Williams WH, Guyton RA, Jones EL, Craver JM. Valve replacement in patients on chronic renal dialysis: implications for valve prosthesis selection. Ann Thorac Surg. 2002;74:37–42. [DOI] [PubMed] [Google Scholar]
- 22.Toole JM, Stroud MR, Kratz JM, et al. Valve surgery in renal dialysis patients. J Heart Valve Dis. 2006;15:453–8. [PubMed] [Google Scholar]
- 23.Bianchi G, Solinas M, Bevilacqua S, et al. Are bioprostheses associated with better outcome than mechanical valves in patients with chronic kidney disease requiring dialysis who undergo valve surgery? Interact Cardiovasc Thorac Surg. 2012;15:473–483. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Manghelli JL, Carter DI, Khiabani AJ, et al. A 20-year multicenter analysis of dialysis-dependent patients who had aortic or mitral valve replacement: implications for valve selection. J Thorac Cardiovasc Surg. 2019;158:805–13. [DOI] [PMC free article] [PubMed] [Google Scholar]