Factors associated with subendocardial ischemia risk in patients on hemodialysis

Introdução: Distúrbios do metabolismo ósseo (DMO) e alterações da função vascular contribuem para a elevada mortalidade de pacientes em hemodiálise. A disfunção vascular, um novo marcador de aterosclerose, pode contribuir para este risco. Apesar de associada a aumento de mortalidade na população geral, a avaliação de tal disfunção ainda não foi realizada de modo amplo em pacientes em hemodiálise. Métodos: Neste estudo transversal, pacientes em hemodiálise foram submetidos à avaliação da vasodilatação mediada por fluxo, razão de viabilidade subendocárdica (RVSE) e índice de duração de ejeção, como estimativas de avaliação dos marcadores de DMO sobre disfunção vascular. Resultados: Uma coorte pareada com (n = 16) e sem (n = 11) hiperparatireoidismo secundário (HPTS) grave foi estudada. Adicionalmente, o tempo transcorrido do diagnóstico de HPTS grave também foi avaliado. Pacientes com HPTS grave apresentaram menores valores de RVSE e maiores valores de índice de duração de ejeção, apontando maior risco cardiovascular. Baixa RVSE também foi associada à pressão arterial diastólica (r = 0,435, p = 0,049), níveis séricos de 25-Vitamina D (r = 0,479, p = 0,028) e maior tempo transcorrido desde diagnóstico de HPTS grave, definido como tempo em que o paciente permaneceu com valores de paratormônio superiores a 500 pg/ml até realização de cirurgia de paratireoidectomia ou término do estudo (r = -0,642, p = 0,027). Em regressão logística stepwise entre RVSE e variáveis independentes, menor RVSE foi independentemente associado a menores valores de 25-Vitamina D (p = 0,005), sexo feminino (p = 0,012) e maior tempo transcorrido desde diagnóstico de HPTS grave (p = 0,001) em um modelo ajustado para idade, colesterol sérico e pressão arterial (r2 ajustado = 0,545, p = 0,001). Conclusão: A perfusão subendocárdica foi menor em pacientes com DMO, refletindo o maior risco cardiovascular nesta população. Investigações adicionais são necessárias para definir se a paratireoidectomia precoce no curso da doença renal crônica poderia interferir neste risco. Resumo


Resumo
Palavras-chave: diálise renal; hiperparatideroidismo secundário; sistema cardiovascular.Even though associated with higher mortality in the general population, such vascular evaluation in patients on hemodialysis has not been extensively studied.Methods: In this cross-sectional study, hemodialysis patients were submitted to flow-mediated dilation, subendocardial viability ratio (SEVR) and ejection duration index assessment, in order to estimate the impact of BMD markers on vascular dysfunction.Results: A matched cohort of patients with (n = 16) and without (n = 11) severe secondary hyperparathyroidism (SHPT) was studied.Additionally, time spent under severe SHPT was also evaluated.Patients with severe SHPT had lower SEVR and higher ejection duration index, indicating higher cardiovascular risk.Lower SEVR was also associated to diastolic blood pressure (r = 0.435, p = 0.049), serum 25-Vitamin-D levels (r = 0.479, p = 0.028) and to more time spent under severe secondary hyperparathyroidism (SHPT), defined as time from PTH > 500pg/ml until parathyroidectomy surgery or end of the study (r = -0.642,p = 0.027).In stepwise multiple regression analysis between SEVR and independent variables, lower SEVR was independently associated to lower serum 25-Vitamin-D levels (p = 0.005), female sex (p = 0.012) and more time spent under severe SHPT (p = 0.001) in a model adjusted for age, serum cholesterol, and blood pressure (adjusted r² = 0.545, p = 0.001).Conclusion: Subendocardial perfusion was lower in patients with BMD, reflecting higher cardiovascular risk in this population.Whether early parathyroidectomy in the course of kidney disease could modify such results still deserves further investigation.

IntRoductIon
Deaths secondary to cardiovascular (CV) disease are proportionally higher in end-stage renal disease (ESRD). 1 Besides the underlying disease, this excess mortality can also be attributed to mineral and bone metabolism disorders (BMD) 2 and to a faster progression of atherosclerosis 3 and vascular calcification is a consequence of altered bone metabolism in this population. 4Non-invasive vascular structural and functional assessment has been used as surrogate marker of CV disease once it can detect early signs of atherosclerosis.A useful tool to assess such vascular alterations is the subendocardial viability ratio (SEVR), which has already been associated with cardiovascular events and early detection of individual CV risk. 5ven though SEVR has been studied in general population, few studies have addressed SEVR evaluation in patients with kidney disease.In this present study, we investigated the association between SEVR and BMD markers in a cohort of patients on hemodialysis.

Study population
This was a cross-sectional observational study that included 27 prevalent patients on maintenance hemodialysis (HD) for at least 12 months, age > 18 and < 70 years old, who agreed to undergo a noninvasive vascular assessment.Patients were drawn from HD unit at the Hospital das Clínicas, University of São Paulo, Brazil.Exclusion criteria were: presence of bilateral arteriovenous fistula for hemodialysis, Diabetes Mellitus, resistant hypertension, peripheral vascular diseases and refusal to give written consent.Demographic variables including history of coronary disease, heart failure and stroke were recorded.
Severe secondary hyperparathyroidism (SHPT) was defined as a PTH ≥ 500 pg/ml, confirmed by at least two measurements with an interval of 3 months or history of previous parathyroidectomy (PTX).The time on severe SHPT was calculated from the day of PTH ≥ 500 pg/ml until the day of PTX or end of study.

echocARdIogRAm
As part of routine clinical care in HD unit, transthoracic echocardiogram was performed yearly in all patients.Variables recorded from this routine annual assessment were left ventricular ejection fraction (EF), left ventricular mass index (LVMI), septum and posterior wall thickness.

EndothElial dEpEndEnt and indEpEndEnt vasodilation
On the opposite arm to the arteriovenous fistula for dialysis, flow-mediated dilation (FMD), which assess endothelial-dependent vasodilation, and vascular smooth muscle response to trinitrate vasodilators, or nitrate-dependent dilation (NDD), were evaluated after a two-day interdialytic interval in all patients.The brachial artery was assessed above the antecubital fossa.The diameter of the artery was verified by ultrasound machine (Sequoia Echocardiography System, version 6.0, Acuson, Siemens, Ca, USA) equipped with a multifrequency linear transducer (7-12 MHz) and coupled to a computer specifically programmed to record and analyze this type of data.

pulsE wavE analysis
Pulse Wave Analysis was performed by SphygmoCor System (AtCor Medical, New South Wales, Australia).This software provides the following indices, as shown in Figure 1: i-Augmentation index adjusted to a heart rate of 75 beats per minute (AIX@75): pulse pressure (PP) wave is a composite of the ongoing wave (from left ventricular systole) plus the reflected wave (from peripheral vascular bed).Augmentation pressure (AP) is the amount of aortic pressure increase secondary to this reflection.AIX@75 is an index of PP augmentation (AP/PP) at a standard heart rate of 75 beats per minute.iii-Subendocardial viability ratio (SEVR or Buckberg index).As shown in Figure 1, central aortic pulse wave allows estimation of the area under curve of both systolic and diastolic phases of the cardiac cycle, or systolic (SPTI) and diastolic time index (DPTI).As myocardial oxygen supply and demand occur during diastole and systole, respectively, the relationship DPTI/SPTI reflects the ischemia propensity of the subendocardial tissue.
The study was approved by the Research Ethics Boards of the University of São Paulo (#235.350)and all subjects provided written consent before participation.

StatiStical analySiS
Data are presented as mean SD or median (25,75 percentiles) according to normal or abnormal distribution, respectively.Categorical data are presented as N and percentage.Comparison between patients with and without severe SHPT was performed by the independent samples t test, Kruskal-Wallis, Pearson Chi Square or Fisher exact test when appropriate.Multivariable relationships between SEVR and independent variables by stepwise linear regression, with p < 0.05 to enter and p > 0.1 to remove.
Collinear variables were excluded from multiple regression modeling.As we are limited because of the small sample size, a low number of events per variable may prone our study to overfitting.For this reason we have identified highly correlated risk factors, and subsequently modeling the best subset of variables in a stepwise multivariable analysis.Statistical analyses were performed with the SPSS system 21.0 (SPSS Inc., Chicago, IL, USA) and with Graphpad prism 6 (CA, USA).A two-sided p value less than 0.05 was considered significant.

Results
We consecutively studied 27 patients from February to August 2014, whose characteristics are shown in Table 1.In general, patients were relatively younger than globally described ESRD population.All patients were receiving recommended dialysis doses, as verified by single-pool Kt/V.Most patients (78%) presented left ventricular mass index (LVMI) above normal range when considering > 88 g/m 2 for women and > 102 g/m 2 for men. 6For technical difficulties in performing PWA, only 21 patients underwent this specific vascular assessment.
The population was subdivided according to the presence of severe SHPT.As expected, patients with severe SHPT had longer dialysis vintage and had lower percentage of arteriovenous fistula.SEVR was lower and EDI was higher in patients with severe SHPT, indicating higher ischemia risk and prolonged mechanical systole, respectively.No other biochemical, demographic and echocardiography parameter were different between the two groups, except for PTH levels.Nine (43%) patients were submitted to PTX due to poor response to the available clinical treatment.
SEVR correlated with diastolic blood pressure (r = 0.435, p = 0.049) and serum 25-OH Vitamin-D (r = 0.479, p = 0.028), and a borderline significant trend with age (p = 0.090) and serum cholesterol (p = 0.068).By stepwise multivariable analysis to examine factors that contributed to lower SEVR indices, female sex, lower 25-vitamin D, and higher time spent on severe SHPT were independently associated with SEVR and together explained 54.5% of its variability (Table 2).The inclusion of the presence of an arteriovenous fistula in the model did not change previous results.
In order to better evaluate factors associated to worse SEVR, the population was subdivided according to SEVR higher or lower than median (141%), as shown in Table 3 cause for worse myocardial perfusion.To that end, we retrospectively identified the first day the patient presented serum PTH higher than 500 pg/ml up to the day of PTx or the day of the end of study.This period was higher in patients with SEVR ≤ 141% than in patients with SEVR > 141% [24 (12,   52) vs. 0 (0,2.2) months, respectively; p = 0.004].Figure 2 shows an inverse relationship between SEVR and time spent under severe SHPT (r = -0.642,p = 0.002).Severe SHPT (PTH > 500 pg/ml) was found in all but one patient with SEVR lower than median, and only in 2 patients with SEVR higher than median.In more details, regarding these two patients, one was already submitted to PTx and his PTH levels are 304pg/ml, and the other is currently on clinical treatment for SHPT (calcitriol) with PTH levels of 890 pg/ml.

dIscussIon
This study suggests that prolonged time spent under higher PTH levels is related to changes in non-invasive vascular indices assessed by pulse wave analysis, leading to a higher risk of subendocardial ischemia, which might not be reversed by PTX.In addition, we have demonstrated that female sex and patients with lower levels of 25 Vitamin-D are also subject to higher cardiovascular risk, assessed by SEVR.
Patients with CKD-MBD, especially those with SHPT, have an abnormally higher relative risk of death from CV diseases due to assortment of reasons: it causes valvular and arterial walls calcification, 7,8 promotes faster atherosclerosis progression, including coronary artery disease 9 and induces myocardial hypertrophy. 10CKD-MBD may worsen anemia by impairing bone marrow response to erythropoietin stimulating agentes. 11Altogether, such findings, associated to a uremic background, increase the propensity for cerebral and cardiac ischemic events.Ganesh at al found that PTH levels > 495 pg/ml are associated to 25% increase in fatal CV events. 12ew studies have addressed the evaluation of SEVR index in ESRD patients.Di Micco et al. showed that the reduction in this index could predict mortality in this population; 13 also, the creation of an arteriovenous fistula acutely reduces SEVR. 14,15In the present cohort, low SEVR was associated to lower 25(OH) Vitamin D levels, and higher cholesterol levels, factors already associated with higher CV risk.
Such finding highlights the importance of hypovitaminosis D in HD patients, once such deficiency was already described as a predictor of worse CV outcomes in patients with coronary artery disease 16 in the general population and also among patients on HD. 17 Another interesting factor consists in the higher risk for subendocardial ischemia in female patients: It could reflect the higher susceptibility to SHPT in comparison to their male counterparts 18 or the loss of the protective hormonal status due to premature menopause commonly observed in HD patients. 19n our data, there was a lower percentage of an arteriovenous fistula among patients with severe SHPT.However, the impact of an arteriovenous fistula in our findings is distinct from the previous study, as it is a marker of more prolonged time on hemodialysis instead of acute impact on SEVR.
Patients with severe SHPT presented a longer ejection duration index than those without severe SHPT.Such finding indicates a more prolonged systolic interval over diastolic time during the cardiac cycle and this is related to a reduced blood supply to the subendocardium tissue.This adverse change in timing of cardiac cycle components and presumed decrease in coronary artery perfusion cause an imbalance in the cardiac supply/demand ratio, which was verified by reduced SEVR in these patients, reflecting higher CV risk.
Our data have shown that lower SEVR was mostly likely associated to more time spent under severe SHPT, reflecting the higher CV risk resulting from this clinical condition.Even though PTX could bring PTH levels close to normal range and reduce mortality in these patients, 20,21 possibly it would promote survival advantage if done earlier in the course of the SHPT. 9herefore, the remaining question is why patients already submitted to PTX persist under high subendocardial ischemia risk.There are some possibilities that might explain such finding: first, our patients usually spend prolonged time from PTX indication until proper surgery is done, resulting in permanent damage to cardiovascular system; second, PTH was already identified as a central mediator of pathologic cardiac remodeling, by promoting intracellular calcium overload and, eventually, leading myocardial cells to apoptosis.Inflammation, which succeeds such process, might induce myocardial fibrosis. 10

Figure 1 .
Figure 1.Cardiovascular indices obtained from pulse wave analysis.