Contemporary data on the prevalence and incidence of aortic stenosis (AS) and aortic valve replacement (AVR) in patients initiating dialysis are scarce. This observational cohort study aimed to estimate (1) the point prevalence of AS and AVR at dialysis start and (2) the AS incidence and associated factors prospective to dialysis initiation. The study included 14,550 patients initiating dialysis registered in the Swedish Renal Registry between 2005 and 2018. AS was defined by the International Statistical Classification of Diseases, Tenth Revision codes, and AVR by the surgical procedure codes. Associations between covariates and outcomes were assessed in Cox regression models. The median age was 68 (57 to 77), 66% were males, and the point prevalence of AS and AVR was 3.4% and 1.1%, respectively. In those without known AS/AVR at dialysis initiation (n = 14,050), AS was diagnosed in 595 patients (incidence 16.3/1000 person-years, 95% confidence interval [CI] 15.1 to 17.7/1,000 person-years) during a median follow-up of 2.7 years (interquartile range 1.1 to 5.7). In adjusted Cox regression models, higher age (hazard ratio [HR] 1.03, 95% CI 1.02 to 1.04), male gender (HR 1.51, 95% CI 1.25 to 1.83), atrial fibrillation (HR 1.32, 95% CI 1.06 to 1.64), and hypertension (HR 1.65, 95% CI 1.03 to 2.65) were associated with incident AS. Peritoneal dialysis patients had a nonsignificant trend toward higher AS risk compared with hemodialysis (HR 1.18, 95% CI 0.99 to 1.40). About 20% of patients (n = 113) diagnosed with incident AS underwent AVR (incidence 3.1/1000 person-years, 95% CI 2.6 to 3.7/1,000). Only the male gender was associated with AVR (HR 2.07, 95% CI 1.30 to 3.30). In conclusion, the prevalence and incidence of AS and AVR in patients initiating dialysis are high. A fifth of newly diagnosed AS underwent AVR after dialysis onset.
Aortic stenosis (AS) is the most common valvular heart disease with a markedly higher prevalence with older age. After symptoms develop the prognosis is poor with a high mortality rate that exceeds 50% in untreated patients. Patients on dialysis have a particularly high risk of aortic valve calcification and AS. However, the reported incidence and prevalence of AS in the literature are based on smaller single-center studies with data collected several decades ago. In these studies, the prevalence in dialysis patients was 6% to 14.5% compared with 0.3% in the general population, and similarly, the incidence of AS was 3.3% per year compared with 0.44%. Fortunately, dialysis treatment has advanced over the decades, with a consequent improvement in survival, and therefore the risk for developing AS associated with dialysis may have changed. This study aimed to determine (1) the point prevalence of known AS and previous aortic valve replacement (AVR) at dialysis start, and (2) the AS incidence and associated factors for developing AS and AVR after dialysis initiation in a contemporary nationwide dialysis registry.
Methods
This study cohort consists of patients who initiated maintenance dialysis between January 1, 2005, and August 3, 2018, and was registered in the nationwide Swedish Renal Registry. The prospective analysis of the incidence of AS and AVR after dialysis initiation excluded patients with known AS diagnosis or AVR ( Figure 1 ).
Figure 1 includes 19 patients who were diagnosed with AS and underwent AVR during the same hospitalization.
The unique patient identification number given to all Swedish citizens was used for linkage with national registries. The Swedish Renal Registry was linked with the Swedish Drug Prescription registry to obtain redeemed medication at baseline (6-month before or 1 month after dialysis initiation). Furthermore, the Swedish Renal Registry was enriched with data on comorbidities through linkage with the National Patient Registry, and with the National Population Registry for data on vital status. The study protocol was approved by the regional ethics committee in Stockholm, Sweden.
The main outcome was AS (defined by International Statistical Classification of Diseases [ICD], Tenth Revision codes: I35.0 or I35.2). AS diagnosis based on ICD-codes has high accuracy (>90%) in Sweden, with the majority of patients having moderate or severe AS. In a majority of cases, this diagnosis is based on echocardiographic measurements. , The secondary outcome was AVR for AS identified by the procedure codes FMD00 (mechanical valve), FMD10 (biologic valve), FMD12 (transcatheter aortic valve implantation), or FMD13 (apical transcatheter aortic valve implantation). Patients were followed until the study outcome was reached, and were censored at the date of renal transplant, death, or at the end of data collection (August 3, 2018), whichever came first.
Comorbidities were identified from the patient registry (using ICD, Tenth Revision diagnostic codes), which has over 85% validity, and contains complete data since 1997 ( Supplementary Table 1 ). Hyperlipidemia was defined as the dispensation of lipid-lowering drugs 6 months before or 1 month after dialysis initiation. All patients at baseline were treated for secondary hyperparathyroidism with phosphate binders, calcium carbonate, and/or vitamin D.
Baseline characteristics are expressed for categorical data as number and percentage. Continuous variables are expressed as median and 25th to 75th percentile (interquartile range [IQR]). Unadjusted incidence rates (IRs) and cumulative IRs during the median follow-up period for AS and AVR were estimated and displayed as Kaplan-Meier graphs. Cox proportional hazard models were used to assess unadjusted and adjusted associations for incident AS and AVR. Adjustments were made for age (as a continuous variable), gender, ischemic heart disease, peripheral arterial disease, cerebrovascular diseases, heart failure, diabetes mellitus, hypertension, chronic obstructive pulmonary disease, hyperlipidemia, and dialysis modality. Because atrial fibrillation could have similar symptoms as AS and lead to an earlier investigation, a sensitivity analysis was performed to examine the incidence of AS after excluding all patients with atrial fibrillation before dialysis starts.
Biochemical data were obtained 6 months before and up to 1 month after dialysis initiation. Due to a high degree of missing values (83% to 89% for parathyroid hormone, phosphate, calcium, and albumin) multiple imputations were not performed. The association between biochemical data at dialysis start and risk of AS was only conducted in the subgroup of patients with complete data ( Supplementary Table 2 ).
Statistical analyses were performed using Stata version 15 (StataCorp. 2017, Stata Statistical Software: Release 15, StataCorp LLC, College Station, Texas).
Results
In total there were 14,550 patients included in the study ( Figure 1 ). The point prevalence of known AS and AVR at dialysis initiation was 3.4% (n = 500) and 1.1% (n = 155) respectively ( Supplementary Table 3 ). On multivariate logistic analysis age, hypertension, ischemic heart disease, atrial fibrillation, heart failure, and peripheral arterial disease remained independently associated with prevalent AS ( Supplementary Table 4 ). Of the 345 patients with known AS but without previous AVR at dialysis initiation, 25 underwent AVR during a median follow-up of 2.7 years. Patients with AS diagnosed before dialysis initiation had 20 times higher risk to undergo AVR after dialysis start (adjusted hazard ratio [HR] 20.35, 95% confidence interval [CI] 12.76 to 32.44, p <0.001), whereas patients with heart failure (HR 0.47, 95% CI 0.24 to 0.92, p = 0.029) or known peripheral arterial disease (HR 0.42, 95% CI 0.19 to 0.91, p = 0.029) were less likely to undergo AVR ( Supplementary Table 5 ).
Survival rates were significantly lower in patients with known AS and AVR before dialysis start. After adjustments, the presence of AS (adjusted HR 1.45, 95% CI 1.27 to 1.65, p <0.001) and previous AVR (adjusted HR 1.26; 95% CI 1.04 to 1.53, p = 0.038) were independently associated with a higher risk of death after the initiation of dialysis ( Supplementary Figure 1 ).
In the prospective analysis cohort (n = 14,050) the median age was 68 years (IQR 56 to 76) and 66.3% were males ( Table 1 ). In total, 595 patients (4.2%) developed AS during a median follow-up of 2.7 years (IQR 1.1 to 5.7) ( Table 1 ). The crude IR was 16.3 per 1,000 person-years (95% CI 15.1 to 17.7). The corresponding incidence was higher for men than for women (18.5 per 1,000 person-years, 95% CI 16.9 to 20.3 vs 11.8 per 1,000 person-years, 95% CI 10.1 to 13.9, p <0.001). Patients who developed AS during dialysis had significantly higher proportions of ischemic heart disease, previous coronary artery bypass graft surgery, and atrial fibrillation ( Table 1 ). In the sensitivity analysis, where patients with known atrial fibrillation diagnosis at dialysis initiation were excluded (n = 2,166), the incidence of new AS remained unchanged (15.0 per 1,000 person-years, 95% CI 13.75 to 16.44). Compared to hemodialysis, patients treated with peritoneal dialysis had a higher IR of AS (IR 17.90 per 1,000 person-year, 95% CI 15.62 to 20.54 vs IR 15.60 per 1,000 person-year, 95% CI 14.10 to 17.21, p = 0.039).
| Variable | Overall | Aortic stenosis | P-value | |
|---|---|---|---|---|
| No | Yes | |||
| (N = 14050) | (n = 13455) | (n = 595) | ||
| Men | 9313 (66.3%) | 8862 (65.9%) | 451 (75.8%) | <0.001 |
| Age, years (IQR) | 68.0 (56.0 – 76.0) | 68.0 (56.0 – 76.0) | 72.0 (64.0 – 78.0) | <0.001 |
| Diabetes mellitus | 5445 (38.8%) | 5212 (38.7%) | 233 (39.2%) | 0.84 |
| Hypertension | 13312 (94.7%) | 12736 (94.7%) | 576 (96.8%) | 0.021 |
| Hyperlipidemia | 634 (4.5%) | 602 (4.5%) | 32 (5.4%) | 0.30 |
| Coronary heart disease | 2692 (19.2%) | 2553 (19.0%) | 139 (23.4%) | 0.008 |
| Myocardial infarction | 1724 (12.3%) | 1636 (12.2%) | 88 (14.8%) | 0.056 |
| Prior PCI | 1094 (7.8%) | 1036 (7.7%) | 58 (9.7%) | 0.068 |
| Coronary bypass | 760 (5.4%) | 715 (5.3%) | 45 (7.6%) | 0.018 |
| Atrial fibrillation | 2166 (15.4%) | 2052 (15.2%) | 114 (19.2%) | 0.010 |
| Heart failure | 1904 (13.6%) | 1829 (13.6%) | 75 (12.6%) | 0.49 |
| COPD | 859 (6.1%) | 822 (6.1%) | 37 (6.2%) | 0.91 |
| Intracranial hemorrhage | 343 (2.4%) | 331 (2.5%) | 12 (2.0%) | 0.49 |
| Prior stroke | 1395 (9.9%) | 1328 (9.9%) | 67 (11.3%) | 0.27 |
| Peripheral arterial disease | 1545 (11.0%) | 1481 (11.0%) | 64 (10.8%) | 0.85 |
| Pulmonary embolism | 252 (1.8%) | 245 (1.8%) | 7 (1.2%) | 0.39 |
| Prior DVT | 209 (1.5%) | 198 (1.5%) | 11 (1.8%) | |
| Hemodialysis | 9459 (67.3%) | 9069 (67.4%) | 390 (65.5%) | 0.34 |
| Peritoneal dialysis | 4591 (32.7%) | 4386 (32.6%) | 205 (34.5%) | |
| Adult polycystic kidney disease | 949 (6.8%) | 919 (6.8%) | 30 (5.0%) | 0.055 |
| Diabetes Nephropathy | 3650 (26.0%) | 3496 (26.0%) | 154 (25.9%) | |
| Glomerulonephritis | 1801 (12.8%) | 1721 (12.8%) | 80 (13.4%) | |
| Hypertension | 2166 (15.4%) | 2050 (15.2%) | 116 (19.5%) | |
| Pyelonephritis | 380 (2.7%) | 368 (2.7%) | 12 (2.0%) | |
| Renovascular disease | 45 (0.3%) | 43 (0.3%) | 2 (0.3%) | |
| Unknown | 1879 (13.4%) | 1793 (13.3%) | 85 (14.3%) | |
| Aortic valve replacement after dialysis start | ||||
| Mechanical Aortic valve prosthesis | 71 (62.8%) | |||
| Biologic aortic valve prosthesis | 16 (14.2%) | |||
| Transfemoral TAVI | 22 (19.5%) | |||
| Transapical TAVI | 4 (3.5%) | |||
| Medication at dialysis start | ||||
| Lipid-lowering medication | 634 (4.5%) | 602 (4.5%) | 32 (5.4%) | 0.30 |
| ACEi/ARB | 818 (5.8%) | 782 (5.8%) | 36 (6.1%) | 0.81 |
| Beta-blockers | 1608 (11.4%) | 1534 (11.4%) | 74 (12.4%) | 0.44 |
| Diuretics | 1691 (12.0%) | 1632 (12.1%) | 59 (9.9%) | 0.10 |
| Calcium channel blockers | 1645 (11.7%) | 1584 (11.8%) | 61 (10.3%) | 0.26 |
| Antiplatelet treatment | 765 (5.45%) | 731 (5.43%) | 34 (5.87%) | 0.77 |
| Warfarin | 331 (2.4%) | 314 (2.3%) | 17 (2.9%) | 0.82 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
