Bioanalytical assessment of plasma concentrations of angiotensin-converting enzyme II inhibitors and angiotensin receptor blockers: A pilot study among patients hospitalized with acute heart failure

Abstract

Background:

While angiotensin-converting enzyme II inhibitors (ACEI) and angiotensin II receptor blockers (ARB) improve chronic heart failure outcomes, their potential harms and benefits in acute heart failure (AHF) is less clear.

Study Question:

We explored the relationship between ACEI or ARB plasma concentrations among patients with AHF with in-hospital change in estimated glomerular filtration rate (eGFR).

Data Sources and Study Design:

August 2016-June 2017, patients with AHF prescribed an outpatient ACEI or ARB were enrolled before AHF treatment. All patients were given twice their home dose of diuretic intravenously and received clinical care at the discretion of the medical team. Of 61 patients in the parent study, saved plasma from 34 who were prescribed an outpatient ACEI or ARB was included in this sub-study.

Measures and Outcomes:

Liquid chromatography tandem mass spectrometry (LC/MS/MS) was performed to assess ACEI or ARB plasma concentrations prior to AHF treatment. Change in estimated eGFR was computed using the CKD-EPI equation, which adjusts for age, sex, and race; diuretic dose and enrollment eGFR were used to adjust for HF severity. Multiple linear regression adjusting for enrollment eGFR and diuretic dose was performed to examine the relationship between drug concentration (undetectable/low vs. in/above-range) and in-hospital change in eGFR.

Results:

Of 34 AHF patients, median age was 63 years (interquartile range, IQR, 58 to 78 years), 19 (55.9%) were female, median eGFR at enrollment was 55.6 ml/min (IQR 35.2 to 75.3 ml/min), and for 11 (32.4%) no ACEI or ARB was detectable in plasma. Medication concentrations in or above reference range were associated with in-hospital decrease in eGFR of 8.3 ml/min (95% confidence interval 15.3 to 1.3 ml/min decrease), after adjusting for enrollment eGFR and diuretic treatment.

Conclusions:

Bioanalytical assessment of medication levels may be useful to guide in-hospital ACEI and ARB therapy for patients with AHF.

Introduction

Clinical guidelines for heart failure (HF) with reduced ejection fraction (HFrEF) recommend continuing angiotensin-converting enzyme II inhibitors (ACEIs) and angiotensin receptor blockers (ARBs) during acute heart failure (AHF), although this is based level B evidence obtained overwhelmingly from HFrEF patients with stable renal function.¹ Evidence to support continuing ACEI therapy during AHF is limited, and some suggests it may aggravate renal dysfunction in AHF.^2,3

ACEI and ARB initiation are well known to cause a transient rise in creatinine, and their use can cause hypotension, worsening renal function, and fluctuations in potassium. All of these effects are associated with worse outcomes in AHF,^4–6 but the relationship between ACEI or ARB plasma concentrations and in-hospital or post-discharge outcomes after hospitalization for AHF is not known. Patients with AHF despite taking medications as prescribed may have more severe or worsening underlying HF, while non-adherent patients may improve by simply addressing adherence. Currently, adherent and non-adherent patients are difficult to distinguish and therefore are both treated with diuresis, even though they likely experience very different clinical effects. We hypothesized that among patients hospitalized with AHF higher ACEI or ARB plasma levels, indicative of higher medication adherence prior to hospitalization, would be associated with a decrease in estimated glomerular filtration rate (eGFR) during hospitalization compared to patients with undetectable or low ACEI or ARB plasma concentrations at the time of hospitalization.

Methods

This exploratory study was conducted as part of an observational study of diuretic resistance in AHF conducted at a tertiary care medical center from August 2016 to June 2017 (NCT02751242).⁷ Patients were eligible for the parent study if they had a clinical diagnosis of AHF in the emergency department (ED) based on clinical history, physical exam, chest radiography, and natriuretic peptide levels. Patients were excluded for: systolic blood pressure <90 mm Hg, IV diuretic administration before enrollment, allergy to furosemide or bumetanide, or any dialysis. All patients were given twice their home dose of diuretic intravenously and then standard clinical care at the discretion of the medical team during the remainder of their hospitalization. Of 61 patients in the parent study, saved plasma from the 34 who were prescribed an outpatient ACEI or ARB was used in this sub-study. The study protocol conforms to the ethical guidelines of the 1975 Declaration of Helsinki as reflected in a priori approval by the institution’s human research committee. Written informed consent was obtained from each patient or an acting surrogate.

Bioanalytical Assessment of Systemic Exposure of ACEI or ARB

Liquid chromatography tandem mass spectrometry (LC/MS/MS) was used to assess ACEI or ARB plasma concentrations prior to AHF treatment. This comprehensive laboratory test enabled detection of 277 analytes for 263 parent medications (Precera Bioscience, Inc., Franklin, TN),⁸ including 13 of the evidence-based medications recommended for patients with HFrEF.¹ Patients were prescribed benazepril, enalapril, lisinopril, losartan, or valsartan. LC/MS/MS results were categorized: undetected; quantifiable but below reference range; or quantifiable and within or above reference range.^9–11

LC/MS/MS was performed according to standard operating procedures adopted by Good Laboratory Practice (GLP), Clinical Laboratory Standards Institute (CLSI), The College of American Pathologists (CAP), and Clinical Laboratory Improvement Amendments (CLIA). Plasma samples were thawed, mixed, and transferred to 96-well plates, and an internal standard working solution was added. Protein precipitation was performed using Phenomenex Impact Protein Precipitation Plates, and eluate was transferred to a new plate and dried under Nitrogen. Reconstituted samples were processed using a Shimadzu Nexera X2 liquid chromatography system (Columbia, MD) fitted with a Phenomenex 2.1 × 50 mm, 1.7um C18 column (Torrence, CA). Sample analysis was performed on a Sciex 5500 Q-Trap Mass Spectrometer (Framingham, MA) with TurboV ion source. Data collection was performed with Sciex Analyst software, version 1.6.2, and data analysis was performed using Indigo BioAutomation Ascent software (Indianapolis, IN). Optimal grade methanol and acetonitrile were obtained from Fisher Scientific (Waltham, MA). Formic acid, ammonium acetate, ammonium formate, and water were LC/MS grade and obtained from Sigma-Aldrich (St. Louis, MO). Dimethylsulfoxide (DMSO) was obtained from Sigma-Aldrich. Ammonium hydroxide was obtained from Thermo Fisher Scientific. All analytical standards were obtained at the highest purity available. Stock solutions were prepared individually in DMSO, water, methanol, or acetonitrile, and then combined. Standard Curve and Quality Control samples were prepared in drug naïve human serum (Bioreclamation IVT, Westbury, NY). Assay linearity, precision, accuracy, and detection were assessed by adding varying amounts of each test drug to human serum. Each of the analytes assayed passed strict analytical validation criteria. All samples were run in a single batch, and samples were thawed a single time for analysis.

Statistical Methods

Summary statistics, including medians with 25^th and 75^th percentiles or frequencies, were computed. The primary outcome was in-hospital change in estimated glomerular filtration rate (eGFR). Serum creatinine was assessed at the time of presentation and 48-hours later, or, for two patients discharged <48-hours, prior to hospital discharge. EGFR was computed using the CKD-EPI equation, which adjusts for age, sex, and race.¹² Diuretic dose was total diuretic within 24 hours of arrival. Bumetanide doses were multiplied by 40 to compute an equivalent furosemide dose.¹³ Diuretic dose and enrollment eGFR were used to adjust for HF severity.

A scatterplot with a fitted polynomial and 95% confidence interval illustrates drug concentration category (undetectable, low-but-quantifiable, in-range-and-quantifiable, above-range-and-quantifiable) by in-hospital change in eGFR. Multiple linear regression was performed to examine the relationship between drug concentration (undetectable/low vs. in/above-range) and in-hospital change in eGFR. There was no evidence of interactions for enrollment creatinine or diuretic dose with in-hospital change in eGFR at a P-value threshold of 0.10. Computation of eGFR accounts for age, sex, and race; therefore, the final multiple linear regression model adjusted for enrollment eGFR and diuretic dose in the first 24 hours of arrival. Post-hoc analysis was done removing an outlier and by outpatient prescription for lisinopril and losartan. No adjustment was made for multiple analyses.¹⁴ All analyses were performed with Stata 14.2 (StataCorp, College Station, TX).

Results

Of 34 patients, 19 (55.9%) were prescribed an ACEI (17 were prescribed lisinopril) and 15 (44.1%) were prescribed an ARB (12 were prescribed losartan). Median age was 63 years (25^th percentile 58, 75^th percentile 78 years), 19 (55.9%) were female, 11 (32.4%) were Black, and 9 (32.1%) had an ejection fraction ≤40% (Table). Baseline median eGFR was 54.6 ml/min (25^th percentile 35.2, 75^th percentile 75.3 ml/min). ACEI or ARB was undetectable for 11 patients (32.4%). The proportion of patients with EF ≤40%, outpatient ACEI or ARB doses, enrollment renal function, diuretic dose, and worsening heart failure did not differ by in-hospital change in eGFR.

Table:

Clinical characteristics (N = 34)

Age, median (IQR), years	63 (58, 78)
Female, no. (%)	19 (55.9)
Black, no. (%)	11 (32.4)
EF≤40%, no. (%)	9 (32.1)
Comorbidities, no. (%)  Hypertension  Diabetes	32 (94.1) 16 (47.1)
Serum creatinine, median (IQR)	1.2 (0.9, 2.0)
EGFR, median (IQR), ml/min	54.6 (35.2, 75.3)
BNP, median (IQR), pg/ml	729.5 (374, 1130.5)
Worsening HF, no. (%)	25 (73.5)
Day 0 cumulative median diuretic dose, (IQR), mg	120 (80, 200)

Abbreviations: IQR, 25^th percentile, 75^th percentile; no., number; EF, ejection fraction; HF, heart failure

ACEI or ARB plasma concentration in- or above-reference range at the time of hospital presentation was associated with a decrease in eGFR during hospitalization of 8.3 ml/min (95% confidence interval, CI, 15.3 to 1.3 ml/min decrease, P=0.02) after adjusting for enrollment eGFR and diuretic dose, compared to patients with undetectable or low ACEI or ARB concentrations (Figure).

Figure.

In-hospital change in eGFR by ACEI/ARB drug concentration among patients hospitalized with acute heart failure

In post-hoc analysis, after excluding an outlier, ACEI or ARB plasma concentration in- or above-reference range at the time of hospital presentation was associated with a decrease in eGFR during hospitalization of 6.4 ml/min (95% CI −12.4 to −0.3). Among patient prescribed lisinopril (n=16), eGFR decreased 5.5 ml/min (95% CI 13.5 ml/min decrease to 2.5 ml/min increase). Among patients prescribed losartan (n = 12), eGFR decreased 12.5 ml/min (95% CI 42.2 ml/min to 5.3 ml/min decrease).

Discussion

This pilot study demonstrates that bioanalytical assessment of guideline-directed medications for HFrEF is feasible and found that higher ACEI or ARB plasma concentrations are associated with decrease of in-hospital eGFR. To our knowledge, this is the first such study among patients with AHF in a hospital setting. Optimal medication levels and ranges are not known for guideline-directed medical therapy for HFrEF, particularly in the setting of co-administered medications and acute illness. This exploratory study is an important step towards determining optimal ACE or ARB dosing AHF. This method can also identify inter- and intra-individual variability in drug absorption and metabolism or directly assess adherence to tailor therapy and medical care to increase medication effectiveness. Bioanalytical assessment of medication levels as a snapshot of adherence also overcomes limitations in recall and social desirability bias from self-reported adherence measures and insensitivity in medication refill data.

Because medication concentrations were assessed before administration of medication in the hospital, they may be a useful assessment of outpatient medication exposure. A significant proportion, 11 (32.4%), did not have detectable ACEI or ARB plasma concentrations, due either to medication non-adherence or a combination of medication half-life, dose, and time since last ingestion. These factors should all be considered when interpreting LC/MS/MS results.

Limitations

Although we controlled for age, sex, race, enrollment renal function, and HF severity, the long-term implications of an in-hospital eGFR decrease of 8 ml/min are unclear. Enrollment median eGFR of 55 ml/min suggests that patients may have entered the study with mild-to-moderate chronic kidney disease, which may increase the risk for decreasing in-hospital eGFR. Among patients with stable HF who start an ACEI or ARB and experience a rise in serum creatinine, therapy should be continued in most cases,¹⁵ and patients discharged with an ACEI or ARB after a critical illness and AKI have lower mortality.¹⁶ In the setting of AHF, however, acute changes in eGFR may herald developing cardiorenal syndrome. Difficulty distinguishing between transient change in eGFR and early AKI raises questions regarding ACEI or ARB administration, dosing, and escalation of AHF therapy.¹⁷ Further work is needed to determine whether higher ACEI or ARB concentrations indicate worsening HF despite adequate medical therapy or contribute to an acute decline in renal function.

In conclusion, bioanalytical assessment of ACEI and ARB concentrations may be a useful tool for guiding in-hospital medical therapy for AHF. Further investigation is needed to examine the relationship between ACEI and ARB levels with renal function and clinical outcomes among patients hospitalized with AHF.