Intensive BP Control and eGFR Declines: Are These Events Due to Hemodynamic Effects and Are Changes Reversible?

Abstract

Purpose of review:

Acute declines in estimated glomerular filtration rate (eGFR) are often observed during intensive blood pressure (BP) lowering. This review focuses on identifying the various mechanisms of eGFR decline associated with intensive BP lowering and evaluates the evidence linking BP control with kidney and cardiovascular (CV) outcomes.

Recent findings:

In 2017, the American College of Cardiology and the American Heart Association (ACC/AHA) began recommending treatment of all individuals to a BP target of <130/80 mmHg. Since then, multiple post hoc analyses of BP trials have associated intensive BP lowering with acute declines in kidney function and acute kidney injury; whether these represent reversible changes in the kidney is still debated.

Summary:

There is ample evidence that intensive BP lowering is associated with declines in eGFR. The clinical implications of these events remain unclear. Individualizing the risks and benefits of intensive BP therapy continues to be warranted.

Introduction:

Hypertension affects approximately 1.1 billion people worldwide and has consistently been the leading risk factor for preventable deaths [1, 2]. Treatment of hypertension is important for reducing risk of adverse cardiovascular (CV) events, including myocardial infarction, stroke, heart failure, and death. Hypertension guidelines have evolved over time, initially with the Joint National Committee (JNC) 6 and 7 reports recommending a blood pressure (BP) target of < 140/90 mm Hg and JNC 8 liberating the BP goal to < 150/90 mm Hg for patients older than 60 years of age (Table 1) [3–5]. In 2017, the American College of Cardiology and the American Heart Association (ACC/AHA) released new guidelines lowering the clinic systolic BP goal to < 130 mmHg for all patients regardless of age [6••]. The most recent changes to hypertension guidelines have been driven mainly by the results of the landmark Systolic Blood Pressure Intervention Trial (SPRINT) published in 2015 [7••]. SPRINT demonstrated that intensive BP control (systolic BP < 120 mmHg) was superior to standard control (systolic BP < 140 mmHg) and reduced the risk for CV events by 25% and all-cause mortality by 27% among patients with high CV risk but without diabetes [7]. Given the significant CV benefits of intensive BP lowering, SPRINT was stopped prematurely after a median follow-up of 3.3 years.

Table 1.

The evolution of hypertension management guidelines in the United States

Year	Guideline	BP Target
1997	JNC 6	< 140/90 mmHg and “lower if tolerated”
2003	JNC 7	< 140/90 mmHg
		< 130/80 mmHg if CKD and diabetes
2014	JNC 8	< 140/90 mmHg if age < 60 years, CKD, or diabetes
		< 150/90 mmHg if age > 60 years
2017	ACC/AHA	< 130/80 mmHg

Abbreviations: BP, blood pressure; JNC, Joint National Committee; CKD, chronic kidney disease; ACC/AHA, American College of Cardiology and the American Heart Association

While intensive BP control to systolic BP <120 mmHg confers significant CV benefits, similar benefits to the kidney have not been observed. In SPRINT, higher rates of acute kidney injury (AKI) and incident chronic kidney disease (CKD) in the intensive-treatment group has sparked debate regarding the potential adverse effects of intensive BP lowering on kidney outcomes [7, 8•]. When BP is lowered with antihypertensive medications, it is not uncommon for serum creatinine to rise acutely to a magnitude greater than 0.3 mg/dL or 1.5-times from baseline, meeting the definition of AKI based on Kidney Disease Improving Global Outcomes (KDIGO) guidelines [9]. However, such acute changes in kidney function during the intensification of BP control are traditionally thought to be hemodynamic and reversible. This contrasts to potentially irreversible changes that may also occur when a patient experiences AKI while under intensive BP control which can lead to tubular necrosis and subsequent progression to CKD (Table 2). In this review, we discuss the pathophysiology and recent literature surrounding changes in kidney function during intensive BP treatment and the potential longer-term clinical implications of these changes.

Table 2.

Paradigm of acute changes in kidney function that may be hemodynamic and reversible vs potentially irreversible

	Hemodynamic and Reversible	Potentially Irreversible
Physiology	Decrease in intraglomerular pressure without ischemia or tubular injury	Decrease in intraglomerular pressure with ischemia and possible tubular injury
Pathology	No change	Possible tubular injury and tubulointerstitial fibrosis
Outcomes	May decrease risk for progression of kidney disease	Increase risk for progression of kidney disease
Predisposing factors	Intensive BP control, ACEI/ARB or SGLT-2 inhibitor use, HFrEF	Intensive BP control, ACEI/ARB or SGLT-2 inhibitor use, HFrEF in the setting of an individual with altered autoregulation (e.g. elderly, advanced CKD, diabetes) and volume depletion
Preventative measures	Unclear need for preventative measures; potential slower up-titration of anti-hypertensive agents	• Avoid hypovolemic states  • Hold anti-hypertensive medications in the setting of acute illnesses such as diarrhea  • Avoid concomitant use of nephrotoxic agents during intensive BP therapy

Abbreviations: BP, blood pressure; CV, cardiovascular; ACEI/ARB, angiotensin-converting enzyme inhibitor/angiotensin receptor blocker; SGLT-2, sodium glucose co-transporter 2, HFrEF, heart failure reduced ejection fraction; CKD, chronic kidney disease

Acute eGFR Declines During Intensive BP Lowering: Potentially Reversible Changes in Kidney Function

It is postulated that acute declines in eGFR during BP lowering are due to reversible intrarenal hemodynamic changes in the setting of autoregulation rather than structural injury to the kidney [10–12]. In the kidney, autoregulation in response to changes in BP occurs via two main mechanisms: the myogenic reflex and tubuloglomerular feedback, both of which help maintain renal blood flow and glomerular filtration rate (GFR) when changes in systemic BP occur [10]. When systemic BP drops, a decrease in stretch is sensed in the afferent arteriole, leading to a myogenic reflex that results in vasodilation and preservation of GFR. Tubuloglomerular feedback complements the myogenic reflex by sensing changes in distal sodium chloride delivery and triggering dilation of the afferent arteriole and constriction of the efferent arteriole, thus helping to maintain GFR in the setting of lower BP. These mechanisms of autoregulation maintain constant renal blood flow as the mean arterial pressure (MAP) varies between 80 to 180 mmHg [13], but acute declines in MAP below 80 mmHg can lead to a decline in GFR.

Chronic hypertension and CKD can lead to remodeling of the afferent and efferent arterioles, limiting their ability to effectively constrict and dilate. These pathologic changes in the kidney vasculature may lead to a decrease in autoregulatory capacity, whereby the kidney is only able to compensate when systemic BPs are higher [10]. When systemic BP is lowered acutely with antihypertensive treatment, the resultant decrease in eGFR could be a sequelae of the systemic BP falling below the lower threshold of patients’ impaired autoregulatory responses. For some patients, this initial eGFR decline can stabilize or resolve as BP control leads to remodeling of blood vessel structure and resetting of autoregulatory function [10].

Acute declines in eGFR during intensive BP treatment have been observed in patients with and without CKD. The Secondary Prevention of Small Subcortical Strokes (SPS3) trial [14] revealed that among patients with relatively preserved kidney function (mean eGFR 80±18.5 ml/min/1.73 m²), intensive BP therapy was associated with a greater reduction in eGFR and higher risk of rapid decline in kidney function (> 30% decline in eGFR) compared to usual BP therapy at 1-year follow-up [15]. Notably, rapid decline in eGFR was associated with higher risk for stroke, major vascular event, or the composite outcome (death, major vascular event, MI, or stroke) among patients randomized to the usual BP arm (systolic BP 130–149 mmHg) but not the intensive BP treatment arm (systolic BP < 130 mmHg) [15]. However, because the changes in eGFR were determined over a one-year period, it is unclear whether the observed declines in eGFR were due to the hemodynamic effects of BP treatment, interim AKI, or progression of CKD. Furthermore, observational analyses are subject to confounding, making it challenging to discern whether the higher risk for rapid declines in kidney function reflects an adverse outcome of intensive BP therapy or the inherent higher risk of individuals who experience this outcome.

Several recent studies have used urinary biomarkers to assess the implications of acute eGFR declines during intensive BP lowering. In a post hoc analysis of participants in SPRINT with CKD, eGFR declined by 10% in the intensive BP arm versus 3% in the standard BP arm at 1-year follow-up (p < 0.001) [16•]. However, participants receiving intensive BP treatment in SPRINT did not have increased levels of urinary biomarkers of tubular damage compared to the standard BP treatment group at 1-year after randomization [16, 17•]. Another study using data from a subgroup of the Action to Control Cardiovascular Risk in Diabetes BP (ACCORD-BP) trial revealed similar findings. Like SPRINT, the ACCORD-BP trial randomized patients to intensive (systolic BP < 120 mmHg) versus standard (systolic BP < 140 mmHg) BP treatment, but recruited patients with type 2 diabetes mellitus [18]. After mean two years of follow-up, eGFR was 11% lower among participants in the intensive BP arm compared to the standard BP arm (p < 0.001), but there was no difference in levels of kidney injury biomarkers between the two groups [19•]. However, these findings must be interpreted with caution. The lack of increase in biomarkers of tubular damage 1–2 years after intensive BP lowering does not necessarily mean that the eGFR decline occurring during hemodynamic fluctuations was completely benign, but supports that no continued ongoing injury detected by urinary biomarkers was occurring by 1-year after BP targets were achieved. Notably, there is no evidence that these acute eGFR declines due to intensive BP lowering attenuated the CV and mortality benefits of intensive BP control [20•].

Analyses of two completed BP trials of individuals with CKD at baseline, African American Study of Kidney Disease and Hypertension Study (AASK) [21] and Modification of Diet in Renal Disease (MDRD) [22] trials, have aimed to shed more light on the long-term effects of acute eGFR declines during intensive BP lowering [21, 22] in those with more advanced CKD. These two trials recruited patients with CKD who had lower baseline eGFRs than participants in recent BP trials (median eGFR 40 ml/min/1.73 m²) [23], and thus were at higher risk for ESKD. A post hoc study linked US Renal Data System (USRDS; the national United States ESKD registry) with participants of AASK and MDRD to extend the ascertainment of ESKD for up to two decades after randomization [24•]. The authors found that intensive BP treatment leading to declines in eGFR of < 20% was not associated with higher risk of ESKD compared to declines of < 5% in the usual BP arm, but a ≥ 20% eGFR decline was associated with higher risk of ESKD in both the intensive and usual BP arms [24]. Interestingly, participants who derived the most benefit from intensive BP control (from an ESKD or mortality standpoint) experienced at least a mild acute eGFR decline (5–20%) [24, 25]. While these findings were based on non-randomized group comparisons, the data support the hypothesis that acute declines in eGFR during intensification of BP treatment may signal some reduction of intraglomerular pressure that may be beneficial in the long-term. Larger eGFR declines with intensive BP lowering may lead to kidney injury and was associated with increased risk for progression to ESKD.

Acute eGFR Declines Following Renin-Angiotensin-Aldosterone (RAS) Inhibitor Initiation: Potentially Reversible Changes in Kidney Function

The hemodynamic effects of intensive BP lowering are particularly applicable to the initiation of RAS inhibitors. High glomerular filtration pressures from hypertension can lead to nephrosclerosis and progressive loss of kidney function [26]. An important goal of hypertension treatment is to decrease intraglomerular pressure to prevent long-term structural damage to the kidney. ACEIs and ARBs accomplish this through efferent arteriolar dilation, which can lead to hemodynamic-related declines in eGFR.

A rise in serum creatinine of up to 30% from baseline (equivalent to a 27% decline in eGFR) [27] is often thought to be acceptable during the initiation or intensification of anti-hypertensive therapy, especially in the setting of RAS inhibition [10, 11, 28•]. Some have suggested that this degree of initial eGFR decline may indicate the successful reduction in intraglomerular pressure that is necessary for long-term kidney protection [29–32]. However, it is important to note that while this 30% threshold is commonly cited and clinically accepted, the safety and appropriateness of this threshold has not been rigorously evaluated. This recommendation stems primarily from a systematic review of 12 randomized clinical trials reporting that acute increases in serum creatinine of up to 30% is not associated with long-term kidney injury, although the vast majority of persons included in this systematic review had relatively normal kidney function [28].

A post hoc analysis of the Reduction of Endpoints in Non-Insulin-Dependent Diabetes Mellitus with the Angiotensin II Antagonist Losartan (RENAAL) trial found that patients who had larger acute drops in eGFR when starting losartan experienced slower rates of long-term eGFR decline [11]. The inverse relationship between the degree of acute, initial eGFR fall and rate of long-term eGFR decline has also been observed among patients without diabetes [28, 29]. These studies concluded that the acute declines in eGFR induced by RAS inhibition reflect reversible and even desirable hemodynamic responses that may slow progression of CKD.

More recently, several studies have challenged this previously widely accepted theory that acute declines in eGFR after initiation of RAS therapy reflect purely benign, hemodynamic changes that confer long-term kidney protection. A population-based study found that larger rises in creatinine during ACEI or ARB therapy over one year were associated with increased risk for ESKD, even among patients who had creatinine rises below the 30% threshold [33]. However, this study was observational and it is unclear whether the acute rises in serum creatinine were directly related to RAS inhibition or a reflection of progressive CKD given the longer duration (one year) over which the eGFR changes were measured. Additionally, it is possible that individuals who experience acute rises in serum creatinine with RAS inhibition are inherently at higher risk and more susceptible to long-term declines in kidney function.

Analysis of data from the ADVANCE (Action in Diabetes and Vascular Disease: Preterax and Diamicron Modified Release Controlled Evaluation) trial showed that acute increases in serum creatinine of ≥ 20% after treatment with perindopril-indapamide was associated with increased risk for adverse kidney and CV events compared to serum creatinine increases of < 10%, but there was no heterogeneity in treatment effect of the randomized intervention (ACEI therapy) based on the degree of serum creatinine rise [34]. Another secondary analysis of two randomized-controlled trials of RAS inhibition (ONTARGET [Ongoing Telmisartan Alone and in Combination With Ramipril Global End Point] and TRANSCEND [Telmisartan Randomized Assessment Study in ACEI Intolerant Subjects with Cardiovascular Disease]) demonstrated that acute declines in eGFR of > 13% after initiation of RAS blockade were associated with worse kidney outcomes compared to eGFR declines of < 5% when eGFR changes were computed using eGFR at week 0 as the baseline [35]. In their primary analysis using week 2 or 8 eGFR (after RAS inhibition) as the baseline, this trend towards higher risk of adverse kidney outcomes was not observed.

The apparent discrepancies in the data surrounding the clinical significance and implications of eGFR declines during RAS initiation are difficult to reconcile. Differences in the baseline characteristics of study populations, such as the degree of proteinuria, may account for some of these discrepancies. For example, in the RENAAL study [36], the median urine albumin to creatinine ratio (UACR) was approximately 1,200 mg/g, while in the other trials (ONTARGET [37], TRANSCEND [38], and ADVANCE [39]) participants predominantly had UACR < 300 mg/g. It is possible that the benefits of acute hemodynamic declines in eGFR induced by RAS blockade may be more pronounced and applicable to patients with proteinuria, although there is insufficient data to inform this issue at this time. Importantly, the results of these post hoc studies should be interpreted with caution given that these trials were not designed to test the clinical significance of kidney function decline after RAS inhibition, and thus are prone to potential residual confounding.

Patients with heart failure with reduced ejection fraction (HFrEF) are particularly susceptible to the hemodynamic perturbations associated with RAS inhibitor therapy and yet have pressing indications for this class of medications for its cardioprotective properties. The inherent pathophysiology of heart failure (elevated cardiac filling pressures and decreased forward flow to the kidney) along with frequent use of diuretics may predispose these patients to a higher risk for eGFR declines during RAS initiation. A post hoc analysis of data from the Studies of Left Ventricular Dysfunction (SOLVD) trial found that randomization to enalapril (versus placebo) was associated with increased risk for acute early declines in eGFR (within 6 weeks of randomization) among patients with HFrEF, but this increase in risk did not persist after the first year [40]. Another study also using SOLVD trial data found that among patients with similar degrees of acute eGFR decline treated with either enalapril or placebo, enalapril was associated with lower risk of all-cause and CV mortality and heart failure hospitalizations over a 2.8-year median follow-up, and there was no evidence of effect modification by eGFR decline [41]. Based on these analyses in patients with HFrEF, acute declines in eGFR upon initiation of RAS inhibitors do not seem to increase risk for future adverse kidney events or diminish the CV benefits afforded by RAS therapy.

AKI and Structural Kidney Damage During Intensive BP Treatment

Patients who are under intensive BP control may be more susceptible to AKI and be at higher risk for irreversible structural injury that may over time lead to long-term kidney dysfunction [24, 42]. The higher rates of AKI among participants assigned to intensive BP treatment in SPRINT has in recent years ignited considerable discussion regarding the clinical significance of this finding. Of the 9,361 participants in SPRINT, 179 (3.8%) experienced AKI requiring hospitalization in the intensive BP arm and 109 (2.3%) in the standard BP arm (hazard ratio 1.64, 95% CI 1.3–2.1, p < 0.001) during the 3.3 year trial follow-up [43•]. Approximately 60% of AKI events were mild stage 1 (increase in serum creatinine ≥ 0.3 mg/dL or 1.5- to 2.0-fold from baseline) and over 90% of AKI events that occurred among participants in the intensive BP arm recovered (decrease in serum creatinine to within 20% of baseline) [43]. The most common cause of AKI in both arms of the trial was dehydration or intravascular volume depletion [43]. While the overall rates of AKI were relatively low in SPRINT, a study applying the SPRINT eligibility criteria to the 1999 to 2006 National Health and Nutrition Examination Survey (NHANES) found that implementation of intensive BP goals would result in 88,700 cases of AKI per year [44].

This has raised concerns because of the known significant morbidity and mortality associated with AKI. Patients experiencing AKI are at higher risk for a myriad of adverse outcomes, including prolonged hospitalization, increased risk for ICU transfer, discharge to long-term care facilities, and death [45, 46]. AKI is also a major risk factor for ESKD [47, 48], particularly for elderly patients with CKD [49, 50]. While most of the cases of AKI in SPRINT were mild, outside of clinical trials or in the setting of routine clinical monitoring, these events may be more frequent and of greater severity.

Subgroups at Higher-Risk for AKI With Intensive BP Treatment

Certain patient subgroups may be at higher risk for AKI while under intensive BP control. The elderly and those with advanced CKD (CKD stages 4 and 5) tend to have higher burden of vascular arteriosclerosis that may limit the kidneys’ autoregulatory capacity in the setting of intensive BP treatment, which could increase their risk for AKI and subsequent kidney ischemia and chronic damage [46, 51, 52]. In post hoc analyses of elderly participants in SPRINT, the CV and mortality benefits of intensive BP control were observed regardless of age [53, 54], but intensive (versus standard) BP control was associated with higher risk of AKI requiring hospitalization in participants ≥ 80 years old (HR 2.12; 95% CI 1.37–3.26) [54]. A retrospective analysis of SPRINT also found that intensive BP treatment in participants with eGFR < 45 ml/min/1.73 m² was associated with a 73% increased risk of AKI (compared to standard BP treatment) without reduction in the risk for the primary CV outcome, although there was no interaction between baseline eGFR and the randomized intervention in the primary trial [55].

SPRINT did not include patients with diabetes or urine protein excretion > 1 gram/day [7], but patients with diabetes also have high prevalence of micro- and macrovascular diseases that impair renal autoregulation and may increase risk for AKI [13, 56]. High concurrent rates of RAS inhibitor and sodium glucose co-transporter 2 (SGLT-2) inhibitor use in the diabetes population could further place these patients at risk for AKI with intensive BP treatment. However, it is unknown whether AKI that occurs in the setting of intensive BP treatment identifies patients inherently at higher-risk for adverse kidney events or signals the need to de-intensify BP control. Weighing the risks and benefits of intense BP control strategies in subgroups of patients who may be prone to AKI (independent of BP control) remains prudent.

Potential Long-Term Implications of AKI During Intensive BP Treatment

The potential transition of AKI to CKD and ESKD is of major clinical significance. Overall, during the 3-year follow-up of SPRINT participants, intensive BP treatment correlated with a higher incidence of AKI and greater degrees of kidney function decline over time, but was not associated with increased risk of ESKD regardless of presence of CKD at baseline. Among the 28% of participants with CKD at baseline (eGFR 20–59 ml/min/1.73 m²), the rates of change in eGFR were relatively small in both arms, with a 0.47 ml/min/1.73 m² per year decline in the intensive BP group and 0.32 ml/min/1.73 m² per year decline in the standard group (p< 0.03) [57•]. While eGFR decline was steeper in the intensive BP arm, the rate is similar to declines attributable to normal aging [58]. In SPRINT participants without CKD at baseline, intensive BP treatment resulted in a 2.6% increase in absolute risk for incident CKD at 3-year follow-up (defined by an eGFR decline ≥ 30% with a confirmed eGFR < 60 ml/min/1.73 m²) compared to standard BP treatment [8]. Among patients with and without CKD in SPRINT, larger declines in eGFR observed in the intensive BP arms over time did not attenuate the benefits that intensive BP control had on CV events and mortality [8, 57].

These findings from SPRINT are in agreement with that of similar studies performed using data from the ACCORD-BP trial, which included patients with diabetes. At mean follow-up of 4.6 years, the cumulative incidence of CKD (defined as 40% decline in eGFR to < 60 ml/min/1.73 m²) among ACCORD trial participants in the intensive arm was 3.7 per 100 person-years compared to 1.6 per 100 person-years in the standard arm (HR 2.3, 95% CI 1.89–2.76) [59]. Similar to the SPRINT findings, there was no difference in progression to ESKD between the intensive and standard BP arms [59].

While analyses of the SPRINT and ACCORD-BP trial did not reveal increased risk of ESKD despite higher incidences of AKI and CKD among participants in the intensive BP arms compared to standard BP arms, these observations must be interpreted in the setting of a couple of important caveats. Overall ESKD events were rare in both SPRINT and ACCORD-BP. These trials recruited participants with relatively high baseline eGFRs, with a mean of 71.7 ml/min/1.73 m² in SPRINT and 91.6 ml/min/1.73 m² in ACCORD-BP [7, 18]. The majority of patients in these trials did not have CKD at baseline and had minimal levels of proteinuria (median baseline albuminuria 0.04 g/g in SPRINT and 0.01 g/g in ACCORD-BP), and thus their risk for development of ESKD during the trial follow-up was low. Furthermore, follow-up for these trials were a median of 3.3 years in SPRINT and 4.7 years in ACCORD-BP, which in a population with relatively preserved kidney function, limits practical long-term assessment of ESKD.

Conclusions

Optimal BP control has been a subject of ongoing debate. Acute declines in eGFR during intensive BP treatment are commonly observed and usually attributed to intrarenal hemodynamic changes that may be reversible. However, greater declines in eGFR, especially in patients with altered intrarenal autoregulation, has been associated with increased risk of CKD and ESKD. Optimizing CV and mortality risk without exposing patients to long-term structural damage of the kidney can complicate the management of hypertension, especially in select populations where the data are less robust (e.g. elderly, advanced CKD, diabetes). While intensive BP control has been associated with higher rates of eGFR decline, this did not attenuate the benefits that intensive BP control had on CV events or mortality. Current hypertension management guidelines recommending BP goal < 130/80 mmHg are primarily driven by the results of SPRINT. As future studies provide new data informing clinicians of the risks and benefits of applying specific BP treatment goals to select patient populations, individualization of care remains prudent.