Is Ambulatory Blood Pressure Monitoring Required for Elderly Hemodialysis Patients during the Interdialytic Period? - Experience of a Tertiary Care Center in South India

Introduction

Hypertension (HT) is a common and challenging problem in patients on chronic hemodialysis (HD). HT in dialysis has several unique features. First, the prevalence is very high with 75% to 90% of patients on HD being hypertensive.[1] Second, young HD patients have higher average systolic blood pressure (SBP) than the elderly, perhaps because arterial stiffening occurs at a younger age, and the elderly HD patient has an increased risk for cardiomyopathy, unlike the general population in which SBP increases with age.[2]

In the general population, a linear relationship exists between blood pressure (BP) and mortality.[34] In contrast, a U-shaped curve or a reverse J-shaped curve of association of BP occurs with the survival of patients on dialysis. Patients with low SBPs (<110 mmHg) have higher mortality and no deleterious effects observed until SBP reached >180 mmHg. Thus, high BP in chronic dialysis is associated with adverse outcomes, low BP has an even stronger association with mortality.[5] The National Kidney Foundation's Kidney Disease Outcomes and Quality Initiative (KDOQI) 2005 guidelines recommended aiming for a predialysis BP of 140/90 mmHg and a postdialysis BP of 130/80 mmHg.[6] Blood pressure variability occurs over the dialysis session because of volume shift, and hence peridialytic BP recordings are inaccurate compared with interdialytic BP recordings.[78]

In a multicenter study using peridialytic BP recording, among 89% of patients on antihypertensive, only 30% had control of HT. Therefore, in the absence of accurate measurement of BP in the interdialytic period, a large number of patients will be falsely diagnosed with controlled HT.[9]

The relationship between elevated BP recorded in the interdialytic period and mortality is direct and linear as in the general population.[10] Compared with peridialytic and home BP recordings, 24-hour ambulatory BP displays closer associations with indices of target organ damage and is a stronger predictor of mortality.[811] Interdialytic BP measurement is superior for diagnosing masked HT[12] and nocturnal HT, which is associated with target organ damage and increased cardiovascular (CV) events.[813] Therefore, ambulatory blood pressure monitoring (ABPM) is considered the “gold standard” approach for the management of HT among patients on dialysis.[714]

Here, we analyzed the interdialytic BP pattern, nocturnal BP dipping status, masked uncontrolled hypertension (MUCH), and factors associated with all-cause mortality of elderly HD patients for whom 24-hour ABPM was done.

Materials and Methods

Data of elderly patients undergoing HD for whom 24-hour ABPM was done between November 2017 and December 2019 in a tertiary care hospital dialysis center were collected. Among 41 patients, only 37 patients were included for analysis because BP recordings were inadequate in two patients and two had withdrawn from the dialysis. Patients undergoing dialysis for more than 3 months and 8 to 12 hours/week were included for evaluation. Routine laboratory investigations taken before dialysis within 4 weeks of ABPM measurements and echocardiography done postdialysis were taken for analysis. Adequacy of dialysis was assessed by online Kt/V done during the week of ABPM measurement.

Results

Data of a total of 37 elderly dialysis patients in whom 24-hour ABPM was done were analyzed. Clinical, demographic, and lab investigations are given in Table 1. The mean age was 67.73 ± 3.36, and the majority of them were (75.7%) males. Diabetes was present in 67.6%, HT in 75.7%, and coronary artery disease (CAD), a common comorbid condition in our population, in 81.1%. The average duration of dialysis was 26.64 months, and most of them were adequately dialyzed (Kt/V = 1.27). Mean predialysis systolic and diastolic BP were 144.05 ± 10.91 mmHg and 76.27 ± 8.53 mmHg, respectively. The mean postdialysis SBP was 148.38 ± 13.02 and diastolic BP was 74.76 ± 7.38 mmHg as shown in Figure 1.

Close

Figure 1

Mean blood pressure characteristics of peridialytic and interdialytic ABPM (ambulatory blood pressure monitoring)

Export to PPT

ABPM mean SBP was 155.02 ± 20.96 mmHg and diastolic BP was 75.94 ± 11.63 mmHg with a mean arterial pressure (MAP) of 102.94 ± 13.69 mmHg. A total of 28 patients (75.7%) were known hypertensive taking antihypertensive drugs. Using ABPM, all 37 patients (100%) were found to have HT, thus unmasking HT in nine (24.3%). There is a high prevalence of uncontrolled HT in our cohort (n = 30, 81%). All the patients with uncontrolled ABPM SBP had a nondipping pattern of BP. Thus, 81% of elderly patients in our dialysis unit were nondippers. Nine patients with normal peridialytic BP were found to have uncontrolled HT by ABPM as well as nondippers of nocturnal BP (P = 0.000). Demographic, clinical, and laboratory factors influencing the nocturnal dipping status of BP and the various ABPM characteristics are given in Table 2. Age, gender, dialysis vintage, diabetic status, and adequacy of dialysis were not correlated with nocturnal dipping status. CAD was significantly associated with nondipper state (n = 27, 90%, P = 0.004). None of the patients with nondipper status had postdialytic diastolic BP more than 85 mmHg (P = 0.036). In the nondipping group, only 10% had ABPM active period/daytime uncontrolled diastolic BP (P = 0.034). Significant variables associated with the nondipping pattern of BP are shown in Figure 2.

Close

Figure 2

Significant association of variables with nondippers

Export to PPT

Patients with uncontrolled HT were noted to have significantly elevated ABPM active period: SBP (P = 0.0001), diastolic BP (P = 0.04), and MAP (P = 0.01), as well as passive period SBP (P = 0.011) and diastolic BP (P = 0.01) as shown in Figure 3. CAD was significantly associated with uncontrolled HT (n = 26, 87.3% P = 0.05). A detailed description of the characteristics between uncontrolled and controlled HT is given in Table 3. Of the 37 patients, nine (24.3%) died during the study period. Although patients who all died had uncontrolled HT, none of the patients with controlled HT had any mortality as shown in Table 4. CV events were the common cause of death in four patients (44.4%), sepsis in four (44.4%), and stroke in one (11.1%). No significant difference was noted between the diabetic and nondiabetic groups regarding control of blood pressure, dipping status, and mortality.

Close

Figure 3

Interdialytic ABPM (ambulatory blood pressure monitoring) systolic blood pressure (BP) – Uncontrolled BP versus controlled BP

Export to PPT

Discussion

There is a 10- to 20-fold higher risk for CV events, particularly sudden cardiac death and heart failure, in HD patients than in the general population.[15] HT is an important risk factor for CV disease and mortality in HD patients[16] besides other unique factors such as endothelial dysfunction, arterial stiffness, and increased Fibroblast Growth Factor-23 (FGF23).

The dialysis unit BPs neither predict target organ damage nor all-cause mortality in dialysis patients.[11] Significant variability in an individual's peridialytic BP measurements exists with poor reliability compared with interdialytic BP measurements.[2] This is partly due to the measurement technique[17] because interdialytic self-measured home measurements of BP predict better for left ventricular hypertrophy (LVH) than peridialytic BP measurement.[11] Other factors that dictate inaccurate peridialytic readings include the white coat effect and fluctuations in volume status.[2] The poor reliability of peridialytic BP recordings compared with interdialytic BP recordings by 44-hour ABPM was shown in a meta-analysis.[18] High ambulatory BP is associated closely with increased mortality among HD patients.[8] Meta-analyses of randomized trials established that BP lowering with the use of antihypertensive therapy improves clinical outcomes.[19]

In our observation of 37 elderly dialysis patients by 24-hour ABPM, all were found to have HT (100%), whereas 30 (80%) patients had uncontrolled HT. Nine patients (24.3%) who were normotensive by peridialytic BP measurement were all found to have uncontrolled HT, thus unmasking uncontrolled HT (P = 0.000). Masked HT has been proven to be associated with increased CV risk in the general population.[20] MUCH affects a large proportion of dialysis patients and is associated with increased pulse wave velocity. The prevalence of MUCH ranges from 43% to 75% as measured by ABPM in the interdialytic period. For patients on antihypertensive medications and with elevated predialysis, BPs are more likely to be MUCH.[21] Nondipping of nocturnal BP was observed in the majority of our patients (81%). This is similar to many of the studies showing nondipper status in HD patients in 60% to 80%.[2223] A study from China had shown that more than 90% of patients on dialysis were “nondippers” or “reverse dippers,” with more than half of the patients in the “reverse dipping” group.[24] Most of the patients with uncontrolled HT are nondippers (n = 26, 86.7%), whereas four patients (n = 4, 57.1%) with uncontrolled HT are dippers (P = 0.05). In a prospective study of 57 treated hypertensive HD patients, Amar et al. reported that elevated nocturnal SBP and elevated pulse pressure were independently associated with CV mortality.[8] Fluid retention has been proposed as the major factor contributing to nocturnal HT and nondipping BP patterns in chronic kidney disease (CKD).[25] Nocturnal HT is thought to compensate for sodium retention during the daytime and enhanced pressure natriuresis during the night. Baroreflex dysfunction resulting in impaired BP regulation could contribute to increased BP variability in CKD.[26] Nocturnal hypoxemia that occurs in sleep apnea, which is highly prevalent in dialysis patients, has been associated with nocturnal HT.[27]

An increase in short-term SBP variability was present with advancing CKD stages in a large cohort. This increased SBP variability may be involved in the sharp elevation of CV risk with worsening renal function.[28] We also observed that there is no significant difference in ABPM SD, percentage time elevation (PTE), and hypertension index (HI) between survivors and nonsurvivors.

Our analysis showed that patients with 24-h ABPM uncontrolled systolic HT were also found to have uncontrolled systolic pressures in active (P = 0.0001) and passive periods (P = 0.01) as well as MAP (P = 0.008). CAD occurred more in patients with uncontrolled HT (n = 26, 86.7%, P = 0.05). Also, there was a significant association between uncontrolled 24-hour ABPM SBP and mortality (100%) compared to patients with controlled 24-hour ABPM SBP (P = 0.05). This is consistent with other studies showing increased mortality with elevated interdialytic blood pressure.[8] Gender, diabetic status, dialysis vintage, or adequacy of dialysis had no significant impact on dipping status or mortality in our population. In a study by Agarwal et al., the prevalence of HT in elderly dialysis patients was 86%, and adequate control of HT was found in 30% of patients only. HT was not associated with gender or ethnicity, and risk factors associated with elevated BP included diabetes, old age, and an increased number of antihypertensive in contrast to our experience with no significant association of uncontrolled HT and diabetic status.[9]

Conclusion

The prevalence of uncontrolled HT with blunted circadian rhythm was high among elderly HD patients as detected by ABPM in the interdialytic period. Uncontrolled SBP as well as nocturnal nondipping BP pattern were significantly associated with CAD. MUCH measured by ABPM was not uncommon in patients with normal peridialytic BP. Uncontrolled HT in the interdialytic period had a significant impact on mortality. A prospective randomized-controlled trial with a large sample size will be much more useful.