Masked uncontrolled hypertension in patients with end-stage kidney disease on in-center hemodialysis

Abstract

Background

Masked uncontrolled hypertension (MUCH) is associated with an increasing risk of morbidity and mortality. Current literature on MUCH lacks data on patients with end-stage kidney disease (ESKD) on hemodialysis (HD). We aimed to investigate the prevalence, ambulatory blood pressure (BP) characteristics, and risk factors of MUCH in this population in a low-middle-income Asian country.

Methods

We conducted a simple random-sampling, cross-sectional study on patients with hypertension and ESKD on HD. The outcome was MUCH, which was confirmed if the participants, who were on 24-hour ambulatory BP monitoring, had at least 1 of the following criteria: (1) daytime mean ambulatory BP ≥ 135 mmHg (systolic) and/or ≥ 85 mmHg (diastolic); (2) night-time mean ambulatory BP ≥ 120 mmHg (systolic) and/or ≥ 70 mmHg (diastolic); or (3) 24-hour mean ambulatory BP ≥ 130 mmHg (systolic) and/or ≥ 80 mmHg (diastolic). Data were presented using descriptive statistics. We used logistic regression to explore the risk factors for MUCH, and the results were reported with odds ratio (OR) and 95% confidence interval (95% CI).

Results

Among 104 participants included for analysis (median age 54.5, 48.1% being female), MUCH was reported on 85 of them (prevalence of 81.7%, 95% CI 73.2–88.0%.). Non-dipping status of BP were found in 98.1% of the participants, with non-dippers dominating the normotension group (prevalence of 57.9%, 95% CI 36.3–76.9%) and reverse dippers dominating the MUCH group (prevalence of 65.9%, 95% CI 55.3–75.1%). We also identified some factors that were associated with a higher risk of MUCH, i.e., being current smoker (adjusted OR = 3.49, 95% CI 1.07 to 11.40), undergoing HD for > 48 months (adjusted OR = 5.69, 95% CI 1.48 to 21.81), taking > 3 antihypertensive medications (adjusted OR = 3.64, 95% CI 1.11 to 11.92), and requiring α2-adrenergic receptor agonists for BP control (adjusted OR = 6.31, 95% CI 1.12 to 35.62).

Conclusion

The prevalences of MUCH and non-dipping of BP (non-dipper and reverse dipper) were very high in patients with ESKD who were undergoing HD for a median duration of 6 years without ambulatory BP monitoring. To avoid extra cost and inconvenience, risk factors should be initially screened for MUCH before monitoring out-of-office BP for a confirmed diagnosis.

Background

Masked uncontrolled hypertension (MUCH) is a clinical condition characterized by normal office blood pressure (BP) readings but elevated out-of-office readings in patients with confirmed hypertension [1]. Thus, MUCH can only be detected through ambulatory or home BP monitoring. This poses significant diagnostic challenges as it often goes unnoticed during routine clinical assessments, leading to insufficient treatment and management [2]. Patients with chronic kidney disease (CKD) are particularly susceptible to MUCH due to the complex interplay of fluid retention, hormonal imbalances, and vascular changes inherent to their condition [3]. The presence of MUCH in CKD exacerbates cardiovascular risks, contributing to increased morbidity and mortality [4,5,6].

The challenges of MUCH are even more worrisome in end-stage kidney disease (ESKD) [6], the most severe form of CKD [7]. As many patients with ESKD require renal replacement therapy, e.g., dialysis, the management of hypertension in this population can be complicated due to the dynamic changes in fluid status and BP fluctuations associated with dialysis sessions [8]. This may increase the burden of MUCH due to its masked nature in clinical assessment. Despite the high prevalence of hypertension in the ESKD population, the issues with MUCH in patients on dialysis remain underexplored. Current literature lacks comprehensive data on the prevalence and risk factors of MUCH in this group, especially in patients undergoing hemodialysis (HD) in low-resource settings. Understanding this knowledge gap is vital, as accurate and timely detection and management of MUCH could significantly improve cardiovascular outcomes and overall prognosis for patients with ESKD on HD [9].

Given the growing need for research, we aimed to investigate the prevalence of MUCH and the ambulatory BP characteristics in patients with ESKD undergoing HD in a low-middle-income Asian country. To facilitate screening and detection in clinical practice, we also aimed to explore the factors associated with the risk of MUCH in this population.

Methods

Study design and participants

We conducted a simple random-sampling, cross-sectional study at the Department of Nephrology of Trung Vuong Hospital (Ho Chi Minh City, Vietnam) from December 1, 2022, to August 31, 2023. We included patients who: (1) were ≥ 18 years old; (2) had ESKD (estimated glomerular filtration rate, eGFR, < 15 mL/min/1.73 m² for ≥ 3 months); (3) were undergoing in-center HD (not including inpatient intermittent HD or continuous renal replacement therapy) for ≥ 3 months; (4) had a confirmed diagnosis of hypertension; (5) were receiving antihypertensive regimen; (6) had pre-dialysis office BP reading < 140 mmHg (systolic) and < 90 mmHg (diastolic); and (7) were eligible for ambulatory BP monitoring. We excluded those who: (1) currently had acute illness, e.g., infection, acute heart failure, acute stroke; (2) had a history of atrial fibrillation; and (3) had bilateral arteriovenous fistula.

We conducted this study in accordance with the Declaration of Helsinki and reported our findings using the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement (Supplementary File, Table S1). The study was approved by the Ethics Committee of Pham Ngoc Thach University of Medicine and Trung Vuong Hospital under approval number 1292/HDDD-BVTV. All participants gave their written informed consent prior to participation.

Outcome

The outcome of interest was MUCH, which was reported as frequency with percentage. Given that all participants had hypertension and controlled office BP, we adapted the approach of prior studies and defined MUCH based on the elevated out-of-office BP [1, 4, 10], as recommended by the European Society of Cardiology (ESC) and European Society of Hypertension (ESH) [11]. In this study setting, MUCH was confirmed if the participants, who were on 24-hour ambulatory BP monitoring, had at least 1 of the following criteria: (1) daytime mean ambulatory BP ≥ 135 mmHg (systolic) and/or ≥ 85 mmHg (diastolic); (2) night-time mean ambulatory BP ≥ 120 mmHg (systolic) and/or ≥ 70 mmHg (diastolic); or (3) 24-hour mean ambulatory BP ≥ 130 mmHg (systolic) and/or ≥ 80 mmHg (diastolic). Due to the limitation in nocturnal BP readings, home BP monitoring was not considered for MUCH screening and diagnosis in this study. For the measurement of office BP and out-of-office BP (ambulatory BP monitoring, on non-dialysis dates with readings recorded every 30 min), we followed the standard practice outlined by the American Heart Association and ESC/ESH guidelines [11, 12] (Supplementary File, Table S2). A valid ambulatory BP monitoring session was considered as having ≥ 70% usable BP recordings, with ≥ 7 night-time readings and ≥ 27 24-hour readings [11]. Repeated monitoring was conducted if the previous session had been invalid.

Covariates

We collected and investigated the following covariates: age (year; <65 or ≥ 65); sex (female or male); residential area (rural or urban); height (cm); weight (kg); body mass index (kg/m²); social history (smoking, alcohol); underlying comorbidities (type 1/type 2 diabetes, heart failure, hyperlipidemia, and anemia); primary cause of CKD (diabetes, hypertension, diabetes and hypertension, glomerulonephritis, cystic disease, other); history of myocardial infarction and stroke (ischemic or hemorrhagic); time since ESKD diagnosis (month); time since dialysis initiation (month); pre-dialysis systolic and diastolic BP (mmHg); dipping status of BP (yes or no; extreme dipper [night-to-day BP ratio ≤ 0.8], dipper [night-to-day BP ratio > 0.8 and ≤ 0.9], non-dipper [night-to-day BP ratio > 0.9 and ≤ 1.0], and reverse dipper [night-to-day BP ratio > 1.0]); pre-dialysis laboratory values: hemoglobin (g/L), blood urea nitrogen (mmol/L), serum creatinine (µmol/L), serum corrected total calcium (mmol/L), serum phosphate (mmol/L); antihypertensive medications (angiotensin-converting enzyme inhibitors, ACEi; angiotensin II receptor blockers, ARB; α2-adrenergic receptor agonists, α2ARA; β-adrenergic receptor antagonists, BB; calcium channel blocker, CCB; diuretics); and medications for CKD-associated complications (anemia, hyperkalemia, and mineral and bone disorder). All of these covariates were also used to investigate potential risk factors of MUCH, except for height, weight, systolic/diastolic BP, and blood urea nitrogen due to the risk of multicollinearity in the analysis.

Statistical analysis

The reported findings from a previous study indicated that the prevalence of MUCH in patients with ESKD on HD could vary from 43.4 to 74.5%, depending on how MUCH was defined [4]. Considering the similar settings between that study and ours, a minimum of 95 participants was needed to have a 95% level of confidence that the true prevalence of MUCH in our population would be within ± 10% of the estimated value. To avoid loss during ambulatory BP monitoring or missing data, the sample size was adjusted by an extra 15%, resulting in a total of 110 participants being enrolled.

All participants with no missing data were included for analysis. We summarized and presented frequency with percentage for categorical variables and median with interquartile range (IQR) for quantitative variables. We used logistic regression to explore the risk factors for MUCH. Findings were reported as odds ratio (OR) with the 95% confidence interval (95% CI). We conducted all statistical analyses using R (version 4.2.1, R Foundation for Statistical Computing, Vienna, Austria).

Results

We enrolled 110 participants after the initial screening. We later excluded 6 participants who withdrew from the study due to significant discomfort during ambulatory BP monitoring. Among 104 participants included for analysis (Fig. 1), MUCH was reported on 85 of them. The prevalence of MUCH in patients with ESKD on HD was estimated at 81.7%, 95% CI 73.2–88.0%.

Fig. 1

Flowchart of the participants. Abbreviations: ESKD, end-stage kidney disease; HD, hemodialysis; MUCH, masked uncontrolled hypertension; PD, peritoneal dialysis

Socio-demographic and anthropometric characteristics

The included participants had a median age of 54.5 (IQR 48–64.3), with 48.1% being female. Participants with MUCH were comparable to those with normotension in terms of residential area, height, weight, and body mass index (Table 1). Despite undergoing HD, large proportions of the participants continued their smoking and alcoholic habits, especially those with MUCH (48.2% being current smokers and 37.6% having excessive alcohol use). Most participants avoided or refused to answer questions related to major socioeconomic indicators, such as educational attainment, occupation, and income. Thus, we could not report nor investigate the associations among these factors and MUCH.

Table 1 Socio-demographic and anthropometric characteristics

Medical characteristics

The medical characteristics of the participants are summarized in Table 2. The prevalences of diabetes, hyperlipidemia, and anemia were comparable between participants with MUCH and normotension. Those with MUCH appeared to have higher proportions of major cardiovascular events, such as myocardial infarction (17.3%) or stroke (5.8%). While both groups underwent comparable durations of hypertension, participants with MUCH had a longer duration of ESKD and a longer period of HD (median of 6 years) compared with the normotension group (median of 2 years). Most CKD-related laboratory values, including hemoglobin, blood urea nitrogen, serum creatinine, serum corrected total calcium, and serum phosphate, were also comparable between the 2 groups (Table 2).

Table 2 Medical characteristics

Blood pressure characteristics

The characteristics of pre-dialysis office and out-of-office BP of the participants are presented in Table 3. Both the office systolic and diastolic BP of the 2 groups were similar. The median numbers of ambulatory night-time and 24-hour readings were 10 (IQR 8–11) and 32 (IQR 28–34), respectively. Participants with normotension had consistent ambulatory BP patterns throughout the day (mean ambulatory systolic BP < 120 mmHg and diastolic BP < 80 mmHg, with mean daytime BP higher than that of night-time readings). Those with MUCH had persistently high ambulatory BP throughout the day, especially systolic and night-time readings (mean of daytime systolic BP > 140 mmHg, mean of night-time and 24-hour systolic BP > 150 mmHg). Non-dipping status of BP was highly prevalent among the participants (> 80%), with non-dippers dominating the normotension group (prevalence of 57.9%, 95% CI 36.3–76.9%) and reverse dippers dominating the MUCH group (prevalence of 65.9%, 95% CI 55.3–75.1%).

Table 3 Blood pressure characteristics

Therapeutic characteristics

We summarized the therapeutic characteristics of the participants in Table 4. Participants with MUCH were reportedly taking more antihypertensive medications (55.3% taking > 3) compared with the other group (26.3% taking > 3). The most frequently indicated antihypertensives were CCB (84.7%, including amlodipine and lercanidipine) and ACEi/ARB (83.5%, including lisinopril, losartan, and telmisartan) in the MUCH group while most of the participants with normotension were given diuretics (78.9%, including furosemide and spironolactone) and CCB (63.2%, including amlodipine and lercarnidipine). Given the CKD-associated anemia, all participants received intravenous iron and erythropoiesis-stimulating agents. Apart from the routine HD, participants at high risk of hyperkalemia were primarily managed with loop diuretics, and only about 5% reported taking sodium–potassium exchange resins. Since mineral and bone disorders were not assessed routinely in most participants, only about 6% received phosphate-binding agents, activated vitamin D, and/or calcimimetics.

Table 4 Therapeutic characteristics

Risk factors for MUCH

The following pre-specified factors were associated with a significant risk of MUCH (Supplementary File, Table S3; Fig. 2): being current smoker (adjusted OR = 3.49, 95% CI 1.07 to 11.40), undergoing HD for > 48 months (adjusted OR = 5.69, 95% CI 1.48 to 21.81), and taking > 3 antihypertensive medications (adjusted OR = 3.64, 95% CI 1.11 to 11.92). We did not find any interactions among these 3 factors (all p_interaction >0.05). For screening purposes, we also found, a posteriori, that particular type of antihypertensive medications might be a predictor of MUCH. In specific, the risk of having MUCH was significantly high in participants who required α2ARA to control their BP (adjusted OR = 6.31, 95% CI 1.12 to 35.62).

Fig. 2

Risk factors of MUCH. Abbreviations: 95% CI, 95% confidence interval; HD, hemodialysis; MUCH, masked uncontrolled hypertension; OR, odds ratio. Odds ratios were adjusted for the presented covariates. Some pre-specified covariates were excluded due to small number of observations and unreliable estimatess

Discussion

Overall, we found a high prevalence of MUCH in patients with ESKD who were undergoing HD for a median duration of 6 years without ambulatory BP monitoring. Reverse dipping of BP was the predominant status in these patients. We also identified some factors that were associated with a higher risk of MUCH, i.e., being current smoker, undergoing HD for > 48 months, taking > 3 antihypertensive medications, and requiring α2ARA for BP control.

As the standardized criteria for MUCH diagnosis are not currently provided by any professional associations, we had to adapt the proposed criteria from prior studies and the recommendations from ESC/ESH guideline [1, 4, 10, 11]. All of these criteria used the ambulatory BP readings as the diagnostic method, except for 1 study that combined both ambulatory and home BP monitoring [10]. However, as the cut-offs for ambulatory BP readings were inconsistent, MUCH could be under-diagnosed or over-diagnosed, leading to inappropriate antihypertensive regimens. Hence, there is a need for a consensus or guideline recommendation of MUCH diagnostic criteria.

We detected MUCH in 81.7% of our study sample, which was notably higher than most previously reported findings [4, 13]. These discrepancies could be partially due to differences in the populations, clinical settings and diagnostic criteria of MUCH. Nevertheless, we believe that the current evidence, which implies the high prevalence of MUCH, can highlight the inadequacy of relying solely on office BP measurements for hypertension management and underscore the need for routine use of out-of-office BP monitoring methods to accurately identify and manage MUCH in patients with ESKD on HD. Future studies should focus on exploring the underlying mechanisms of MUCH in this population and developing targeted interventions to mitigate its adverse effects on patient outcomes.

In our study, several risk factors for MUCH in patients with ESKD undergoing HD were identified. Notably, smoking emerged as a significant risk factor, which is consistent with the literature linking smoking to increased BP and cardiovascular risk [14]. Additionally, patients who were undergoing HD for > 48 months were more likely to have had MUCH by the time of the diagnosis. This finding aligns with the pathophysiologic effects of maintenance HD on BP, such as volume overload, arterial stiffness, endothelial dysfunction, and use of erythropoietin-stimulating agents [15]. Another risk factor was the necessity of intense antihypertensive therapy, typically > 3 agents, which is also similar to previous finding [4]. Given the difficulties and inconvenience in diagnosing MUCH, these factors can be integrated into a questionnaire, which serves as a quick, convenient, and non-invasive screening tool for patients with ESKD on HD. If such screening tool is created, further research should verify its precision and applicability in clinical practice.

To the best of our knowledge, this is the first study to investigate MUCH in patients with ESKD on HD in an Asian country. Evidence from our findings can be used to estimate the burdens of MUCH on clinical outcomes and healthcare costs in similar low-resource settings. However, there are some limitations. First, our sample size was not large enough to further investigate the effects of other covariates on MUCH. Second, due to the cross-sectional study design, we could not assure causal inference nor follow the participants for clinical outcomes. Third, we could not assess medication concordance in our clinical setting. Finally, our diagnostic approach for MUCH was not standardized for patients with ESKD on HD and might not detect patients with MUCH accurately.

Conclusion

The prevalences of MUCH and non-dipping of BP (non-dipper and reverse dipper) were very high in patients with ESKD who were undergoing HD for a median duration of 6 years without ambulatory BP monitoring. To avoid extra cost and inconvenience, a combination of these factors (being current smoker, undergoing HD for > 48 months, taking > 3 antihypertensive medications, and requiring α2ARA for BP control) could be considered to initially screen for MUCH before monitoring out-of-office BP for a confirmed diagnosis. However, further research is still needed to verify these findings.