Introduction: Angiotensin-converting enzyme inhibitors (ACEis) and angiotensin receptor blockers (ARBs) are frequently discontinued in patients with chronic kidney disease (CKD). Documented adverse drug reactions (ADRs) in medical records may provide insight into the reasons for treatment discontinuation. Methods: In this retrospective cohort of US veterans from 2005 to 2019, we identified individuals with CKD and a current prescription for an ACEi or ARB (current user group) or a discontinued prescription within the preceding 5 years (discontinued group). Documented ADRs in structured datasets associated with an ACEi or ARB were categorized into 17 pre-specified groups. Logistic regression assessed associations of documented ADRs with treatment discontinuation. Results: There were 882,441 (73.0%) individuals in the current user group and 326,794 (27.0%) in the discontinued group. There were 26,434 documented ADRs, with at least one documented ADR in 7,520 (0.9%) current users and 9,569 (2.9%) of the discontinued group. ADR presence was associated with treatment discontinuation, aOR 4.16 (95% CI: 4.03, 4.29). The most common documented ADRs were cough (37.3%), angioedema (14.2%), and allergic reaction (10.4%). ADRs related to angioedema (aOR 3.81, 95% CI: 3.47, 4.17), hyperkalemia (aOR 2.03, 95% CI: 1.84, 2.24), peripheral edema (aOR 1.53, 95% CI: 1.33, 1.77), or acute kidney injury (aOR 1.32, 95% CI: 1.15, 1.51) were associated with treatment discontinuation. Conclusion: ADRs leading to drug discontinuation were infrequently documented. ADR types were differentially associated with treatment discontinuation. An understanding of which ADRs lead to treatment discontinuation provides an opportunity to address them at a healthcare system level.

Angiotensin-converting enzyme inhibitors (ACEis) and angiotensin receptor blockers (ARBs) have been recommended as first-line agents for the management of hypertension in patients with albuminuria since 2003 [1, 3]. They decrease albuminuria and are an important intervention to slow the progression of chronic kidney disease (CKD) and decrease the risk of cardiovascular events [4, 5]. Despite this, ACEis and ARBs are under-prescribed in patients with CKD, with one study showing only 66% of patients with CKD stages 3–4 receiving these treatments [6, 7]. They are also frequently discontinued, which is associated with an increased risk of death, cardiovascular events, and end-stage kidney disease [7, 10].

High rates of treatment discontinuation are likely in part due to adverse effects such as acute kidney injury (AKI), hyperkalemia, and angioedema, which can be treatment-limiting in some cases [11]. Documentation of adverse drug reactions (ADRs) in the patient’s medical record is an important method of preventing repeated adverse events by alerting clinicians at the time of prescription that a patient has a history of an ADR to that or a similar agent. However, ADRs to ACEis and ARBs remain under-documented in the electronic health record despite increases over the last 20 years [12].

To better understand the relationship between ACEi or ARB prescription patterns and documented ADRs, we aimed to determine if the presence of a documented ADR was associated with ACEi or ARB discontinuation, quantify the frequency of specific types of documented ADRs, and evaluate relationships between documented ADR types and the likelihood of continuing versus discontinuing treatment.

Data Source

Real-world clinical data about US veterans seen in the Veterans Affairs (VA) healthcare system from January 1, 2005, to December 31, 2019, were obtained from the VA Corporate Data Warehouse (CDW). Outpatient and inpatient data including demographics, comorbidities, laboratory values, and pharmacy records were accessed via the VA Informatics and Computing Infrastructure (VINCI). Comorbid medical conditions were identified using International Classifications of Disease, Clinical Modification, revisions 9 (ICD-9) and 10 (ICD-10) codes (online suppl. Table 1; for all online suppl. material, see https://doi.org/10.1159/000530988). Baseline characteristics were obtained within 1 year prior to cohort entry.

Study Design and Participants

We conducted a retrospective cohort study, evaluating a national sample of individuals with new onset CKD stages 3–4 in the VA system. The study was approved by the Institutional Review Boards at Baylor College of Medicine, the Michael E. DeBakey VA Medical Center, and the Tennessee Valley Healthcare System VA, and a waiver of informed consent was granted at both locations, approval number (H-47583).

We identified adults with CKD stages 3–4, defined as an estimated glomerular filtration rate (eGFR) of 15 to <60 mL/min/1.73 m2 calculated by the 2021 creatinine-based CKD Epidemiology Collaboration (CKD-EPI) equation [13]. Individuals were included if they had two outpatient eGFR measurements <60 mL/min/1.73 m2 at least 90 days apart with no intervening values ≥60 mL/min/1.73 m2. CKD stage was determined by the second eGFR value and defined by guideline-based thresholds [14]. The index date for this analysis was the date of the second eGFR value meeting these criteria. Participants were restricted to those with at least one primary care or general internal medicine encounter during the study window to identify active users of the VA system. Key exclusion criteria included receipt of chronic dialysis or kidney transplantation identified by diagnosis and procedure codes using a highly sensitive and specific algorithm that has been previously published (online suppl. Table 2) [15]. Those who had never received an ACEi or ARB in the VA system within the 5 years preceding the index date were excluded from the study.

ACEi or ARB Prescription Status

ACEi or ARB prescription status was identified from pharmacy records. Those prescribed an ACEi or ARB at the index date were classified as current users. Individuals who were not prescribed an ACEi or ARB at the index date but for whom one had previously been prescribed and discontinued prior to the index date were classified as discontinued. All participants met criteria for inclusion in one of these groups. Individuals who had been switched from one drug to another, for example, those who developed cough with an ACEi and were transitioned to an ARB, were categorized as either current users or discontinued based on whether they had an active prescription for an ACEi or ARB on the index date.

Documented ADRs

Documented ADRs in the medical record affiliated with an ACEi or ARB were captured. Two investigators categorized all ADRs into 17 categories [16]. Disagreements were resolved by consensus. Categories included commonly reported reactions to ACEi and ARBs, such as AKI, hyperkalemia, angioedema, and cough, as well as other system-based categories. General symptoms included reports of vague or systemic symptoms such as dizziness, chills, fatigue, or muscle aches. Endocrine/reproductive symptoms included symptoms such as decreased libido, erectile dysfunction, and gynecomastia. The “other” category included ADRs that did not clearly fit into one of the named categories, such as photosensitivity, alopecia, anemia, flushing, and lymphadenopathy, among others.

Statistical Analysis

Baseline characteristics were compared between groups using χ2 tests. Multivariable logistic regression identified factors associated with treatment discontinuation compared to current users. Models were adjusted for age; sex; race; CKD stage; congestive heart failure; diabetes mellitus; hypertension; cerebrovascular disease; peripheral vascular disease; coronary heart disease; and the use of thiazide or loop diuretics, potassium-sparing diuretics, beta blockers, calcium channel blockers, and statins. ADR categories were presented graphically as the proportion of participants with an ADR reporting that category. Multivariable logistic regression assessed the odds of treatment discontinuation versus continuation among individuals reporting each category of ADR. Models were adjusted for demographics; CKD stage; comorbidities; and prescription of thiazide or loop diuretics, potassium-sparing diuretics, beta blockers, calcium channel blockers, and statins. ADR categories were also evaluated among subgroups by drug class associated with the ADR (i.e., ACEi or ARB), CKD stage (3a, 3b, or 4), and diabetes mellitus status. For the analysis by CKD stage, Cochran-Armitage p values assessing the trend are reported. For the analysis by diabetes status, χ2p values comparing results between the two groups are reported. The data were constructed using SQL and SAS. Analyses were conducted using R.

Baseline Characteristics

Of 1,760,791 individuals with CKD stages 3–4, we excluded 551,556 who had not received an ACEi or ARB within 5 years prior to the index date. We included 1,209,235 participants, with 882,441 (73.0%) persons in the current user group and 326,794 (27.0%) in the discontinued group (Fig. 1). All evaluated baseline characteristics differed between groups, with the exception of comorbid congestive heart failure (Table 1). Individuals in the discontinued group were more likely to be aged ≥70 years old and less likely to have diabetes mellitus than the current user group. CKD stage 4 was more common in the discontinued group than the current user group (6.1% vs. 4.5%). The discontinued group was less likely to be on statins or other antihypertensive medications, including thiazide or loop diuretics, potassium-sparing diuretics, beta blockers, and calcium channel blockers.

Fig. 1.

Design of the cohort.

Fig. 1.

Design of the cohort.

Close modal
Table 1.

Characteristics by ACEi or ARB prescription status

Variable, N (%)Current user groupN = 882,441Discontinued groupN = 326,794p value
Age, years 
 <40 1,127 (0.1) 520 (0.2) <0.0001 
 40 to <50 9,860 (1.1) 3,731 (1.1) 
 50 to <60 77,928 (8.8) 25,710 (7.9) 
 60 to <70 242,255 (27.5) 77,334 (23.7) 
 ≥70 551,271 (62.5) 219,499 (67.2) 
Sex 
 Male 860,287 (97.5) 318,110 (97.3) <0.0001 
 Female 22,154 (2.5) 8,684 (2.7) 
Race 
 White 625,640 (70.9) 227,792 (69.7) <0.0001 
 Black or African American 142,875 (16.2) 52,386 (16.0) 
 Other 14,083 (1.6) 5,101 (1.6) 
 Unknown 99,843 (11.3) 41,515 (12.7) 
Comorbidities 
 CKD stage 
  3a 639,075 (72.4) 229,093 (70.1) <0.0001 
  3b 203,985 (23.1) 77,936 (23.9) 
  4 39,381 (4.5) 19,765 (6.1) 
 Congestive heart failure 146,690 (16.6) 54,287 (16.6) 0.89 
 Diabetes mellitus 
  Without complications 294,864 (33.4) 96,947 (29.7) <0.0001 
  With complications 170,565 (19.3) 57,111 (17.5) 
 Hypertension 
  Without complications 804,441 (91.2) 295,471 (90.4) <0.0001 
  With complications 25,368 (2.9) 10,099 (3.1) 
 Atherosclerotic cardiovascular disease 267,012 (30.3) 103,666 (31.7) <0.0001 
  Cerebrovascular disease 130,218 (14.8) 51,361 (15.7) <0.0001 
  Peripheral vascular disease 126,317 (14.3) 49,709 (15.2) <0.0001 
  Coronary heart disease 65,190 (7.4) 24,567 (7.5) 0.02 
Medications 
 Thiazide or loop diuretic 416,630 (47.2) 102,791 (31.5) <0.0001 
 Potassium-sparing diuretic 77,487 (8.8) 24,553 (7.5) <0.0001 
 Beta blocker 442,720 (50.2) 120,623 (36.9) <0.0001 
 Calcium channel blocker 291,607 (33.1) 84,677 (25.9) <0.0001 
 Statin 583,935 (66.2) 148,300 (45.4) <0.0001 
 SGLT2 inhibitor 2,245 (0.3) 500 (0.2) <0.0001 
ACEi or ARB use 
 ACEi 715,136 (81.0) 276,012 (84.5) <0.0001 
 ARB 167,305 (19.0) 50,782 (15.5) 
 At least one documented ADR 7,520 (0.9) 9,569 (2.9) <0.0001 
Variable, N (%)Current user groupN = 882,441Discontinued groupN = 326,794p value
Age, years 
 <40 1,127 (0.1) 520 (0.2) <0.0001 
 40 to <50 9,860 (1.1) 3,731 (1.1) 
 50 to <60 77,928 (8.8) 25,710 (7.9) 
 60 to <70 242,255 (27.5) 77,334 (23.7) 
 ≥70 551,271 (62.5) 219,499 (67.2) 
Sex 
 Male 860,287 (97.5) 318,110 (97.3) <0.0001 
 Female 22,154 (2.5) 8,684 (2.7) 
Race 
 White 625,640 (70.9) 227,792 (69.7) <0.0001 
 Black or African American 142,875 (16.2) 52,386 (16.0) 
 Other 14,083 (1.6) 5,101 (1.6) 
 Unknown 99,843 (11.3) 41,515 (12.7) 
Comorbidities 
 CKD stage 
  3a 639,075 (72.4) 229,093 (70.1) <0.0001 
  3b 203,985 (23.1) 77,936 (23.9) 
  4 39,381 (4.5) 19,765 (6.1) 
 Congestive heart failure 146,690 (16.6) 54,287 (16.6) 0.89 
 Diabetes mellitus 
  Without complications 294,864 (33.4) 96,947 (29.7) <0.0001 
  With complications 170,565 (19.3) 57,111 (17.5) 
 Hypertension 
  Without complications 804,441 (91.2) 295,471 (90.4) <0.0001 
  With complications 25,368 (2.9) 10,099 (3.1) 
 Atherosclerotic cardiovascular disease 267,012 (30.3) 103,666 (31.7) <0.0001 
  Cerebrovascular disease 130,218 (14.8) 51,361 (15.7) <0.0001 
  Peripheral vascular disease 126,317 (14.3) 49,709 (15.2) <0.0001 
  Coronary heart disease 65,190 (7.4) 24,567 (7.5) 0.02 
Medications 
 Thiazide or loop diuretic 416,630 (47.2) 102,791 (31.5) <0.0001 
 Potassium-sparing diuretic 77,487 (8.8) 24,553 (7.5) <0.0001 
 Beta blocker 442,720 (50.2) 120,623 (36.9) <0.0001 
 Calcium channel blocker 291,607 (33.1) 84,677 (25.9) <0.0001 
 Statin 583,935 (66.2) 148,300 (45.4) <0.0001 
 SGLT2 inhibitor 2,245 (0.3) 500 (0.2) <0.0001 
ACEi or ARB use 
 ACEi 715,136 (81.0) 276,012 (84.5) <0.0001 
 ARB 167,305 (19.0) 50,782 (15.5) 
 At least one documented ADR 7,520 (0.9) 9,569 (2.9) <0.0001 

Adverse Drug Reactions

There were a total of 26,434 documented ADRs in 17,089 participants, including 12,382 in the current user group and 14,052 in the discontinued group. At least one documented ADR was present in 7,520 (0.9%) individuals in the current user group and 9,569 (2.9%) of the discontinued group. A documented ADR was independently associated with higher odds of treatment discontinuation, aOR 4.16 (95% CI: 4.03, 4.29) (Fig. 2). Other factors associated with higher odds of treatment discontinuation included age ≥70 years, black or unreported race, more advanced stage of CKD, hypertension, cerebrovascular disease, peripheral vascular disease, and coronary heart disease. Concomitant use of statins or other antihypertensive agents was independently associated with lower odds of treatment discontinuation (Fig. 2).

Fig. 2.

Factors associated with ACEi or ARB discontinuation. Among current users and the discontinued groups, multivariable logistic regression evaluated associations between patient characteristics and ACEi or ARB discontinuation.

Fig. 2.

Factors associated with ACEi or ARB discontinuation. Among current users and the discontinued groups, multivariable logistic regression evaluated associations between patient characteristics and ACEi or ARB discontinuation.

Close modal

Categories of ADR

The most commonly reported type of ADR among the entire cohort was cough, which accounted for 37.3% of all documented ADRs (Fig. 3a). Angioedema and allergic reaction were the next most common, comprising 14.2% and 10.4% of all ADRs, respectively. Hyperkalemia accounted for 8.7% and AKI for 3.7% of all documented ADRs (Fig. 3a; online suppl. Table 3).

Fig. 3.

Proportion of documented ADRs in the total cohort (a) and based on ACEi or ARB prescription status (b). These data are also displayed in online supplementary Table 3. *p < 0.05 for the comparison.

Fig. 3.

Proportion of documented ADRs in the total cohort (a) and based on ACEi or ARB prescription status (b). These data are also displayed in online supplementary Table 3. *p < 0.05 for the comparison.

Close modal

More patients in the current user group than the discontinued group had a documented ADR related to cough, allergic reaction, general symptoms, gastrointestinal symptoms, neurologic symptoms, hypotension, cardiovascular symptoms, or endocrine/reproductive symptoms (Fig. 3b; online suppl. Table 3). The discontinued group was more likely to have a documented ADR related to angioedema, hyperkalemia, peripheral edema, or AKI. There was no difference between the two groups in the proportion of participants with dyspnea, psychiatric symptoms, or electrolyte disorders other than hyperkalemia.

In models adjusted for demographics, comorbidities, and concomitant medications, individuals with a documented ADR related to angioedema, hyperkalemia, peripheral edema, or AKI had higher odds of treatment discontinuation (Fig. 4). An ADR related to cough, allergic reaction, general symptoms, gastrointestinal symptoms, neurologic symptoms, cardiovascular symptoms, or endocrine/reproductive symptoms was associated with lower odds of treatment discontinuation (Fig. 4).

Fig. 4.

Associations between ACEi or ARB discontinuation with documented ADR type. A forest plot shows the odds of having a documented ADR in a particular category for those in the discontinued group relative to those in the current user group. Adjusted for age; sex; race; CKD stage; congestive heart failure; diabetes mellitus; hypertension; cerebrovascular disease; peripheral vascular disease; coronary heart disease; and the use of thiazide or loop diuretics, potassium-sparing diuretics, beta blockers, calcium channel blockers, and statins.

Fig. 4.

Associations between ACEi or ARB discontinuation with documented ADR type. A forest plot shows the odds of having a documented ADR in a particular category for those in the discontinued group relative to those in the current user group. Adjusted for age; sex; race; CKD stage; congestive heart failure; diabetes mellitus; hypertension; cerebrovascular disease; peripheral vascular disease; coronary heart disease; and the use of thiazide or loop diuretics, potassium-sparing diuretics, beta blockers, calcium channel blockers, and statins.

Close modal

Subgroup Analyses

ADRs associated with an ACEi showed similar results to ADRs to either an ACEi or ARB (online suppl. Fig. 1a, b). Among ADRs related to ACEis, angioedema, hyperkalemia, peripheral edema, and AKI were associated with higher odds of drug discontinuation (online suppl. Fig. 2). ADRs associated with an ARB were most commonly due to cough, general symptoms, allergic reactions, gastrointestinal symptoms, or angioedema (online suppl. Fig. 3a). A higher proportion of patients in the current user group had an ADR related to cough, general symptoms, gastrointestinal symptoms, and cardiovascular symptoms (online suppl. Fig. 3b). Among ADRs related to ARBs, angioedema and hyperkalemia were associated with higher odds of treatment discontinuation (online suppl. Fig. 4).

ADR types were also evaluated across subgroups by CKD stage and diabetes mellitus status. More advanced stages of CKD were associated with lower rates of ADRs for cough, angioedema, and general symptoms but higher rates of ADRs for hyperkalemia, AKI, hypotension, and unspecified reasons (Fig. 5a; online suppl. Table 4). Individuals with diabetes mellitus were less likely to have an ADR related to cough, angioedema, or allergic reaction and more likely to have an ADR for hyperkalemia or hypotension (Fig. 5b; online suppl. Table 5). There was no difference in documented ADRs for AKI based on diabetes status.

Fig. 5.

Proportion of documented ADRs by CKD stage (a) and diabetes status (b). These data are also displayed in online supplementary Tables 4 and 5. *p < 0.05 for the comparison. ACEi, angiotensin-converting enzyme inhibitor; ADR, adverse drug reaction; ARB, angiotensin receptor blocker; CKD, chronic kidney disease.

Fig. 5.

Proportion of documented ADRs by CKD stage (a) and diabetes status (b). These data are also displayed in online supplementary Tables 4 and 5. *p < 0.05 for the comparison. ACEi, angiotensin-converting enzyme inhibitor; ADR, adverse drug reaction; ARB, angiotensin receptor blocker; CKD, chronic kidney disease.

Close modal

In this analysis, we showed that among patients with CKD for whom treatment with an ACEi or ARB had been prescribed, 27.0% had discontinued treatment. Documented ADRs were uncommon, present in only 2.9% of the discontinued group for whom there was likely a reason that prompted the discontinuation. Nonetheless, a documented ADR was strongly associated with treatment discontinuation independent of demographic factors, comorbidities, and concomitant medications. The most commonly reported types of documented ADR were cough, angioedema, allergic reaction, and general symptoms. Of individuals who received an ACEi or ARB in the VA system, those experiencing angioedema, hyperkalemia, peripheral edema, and AKI were more likely to have treatment discontinued, while those experiencing cough, allergic reaction, general symptoms, gastrointestinal symptoms, neurologic symptoms, cardiovascular symptoms, or endocrine/reproductive symptoms were more likely to have treatment continued.

It has previously been shown that ACEi and ARB are underutilized in patients with CKD. One prior study of US veterans with CKD stages 3–4 showed that only 66% (interquartile range 62, 69) were receiving an ACEi or ARB within 12 months before or after the index primary care visit [7]. Another study showed that 63.4% of patients with CKD stages 3–5 were prescribed an ACEi or ARB after their last visit with nephrology, indicating that these patients had been receiving nephrology subspecialty care [11]. These differences in ascertainment are likely to contribute to the lower prescription rate found in our study, which evaluated active prescription at the time of CKD diagnosis to define current users. When considering the 551,556 individuals with no ACEi or ARB prescription within 5 years, only 50.1% of patients with CKD were current users of an ACEi or ARB, 18.6% had a previous ACEi or ARB discontinued, and 31.3% had not received a prescription for an ACEi or ARB. These results were similar to another study that showed that 49% of patients had an active ACEi or ARB prescription on the index date [17].

Patients with CKD are more likely to have ACEi or ARB discontinued than patients without CKD [18]. One study reported that 56% of patients with type 2 diabetes and incident CKD discontinued their ACEi or ARB within 1 year of initiation [19]. Prior studies have investigated reasons for underutilization of ACEis or ARBs in patients with CKD. One study showed that ACEi or ARB use was less likely in patients with a history of AKI, hyperkalemia, and lack of nephrology care but more likely in individuals with hypertension, proteinuria, or diabetes [17]. In another retrospective review of medical records, the most common reasons for non-initiation of an ACEi or ARB in patients with CKD stages 3–5 included no clear reason (37.4%), past or current hyperkalemia (13.8%), suspicion of renal artery stenosis (13.8%), acute deterioration of renal function (12.5%), hypotension (12.5%), and very low glomerular filtration rate (10.0%) [11]. Our study differs from these by evaluating the utility of the documented ADR structured data item, which may be more practical for assessing reasons for treatment discontinuation in larger datasets when individual chart review to ascertain all reasons for treatment discontinuation is not feasible. Our contemporary data build on our current knowledge about reasons for discontinuation.

Documented ADRs were uncommon, even among the discontinued group who likely had a clinical indication that prompted clinicians to stop or hold their ACEi or ARB. There are likely some individuals for whom treatment was held with the intention to restart at a later date, but clinician understanding and comfort may lead to differences between providers in willingness to restart treatment after an event like AKI or hyperkalemia. Although the presence of an ADR was independently associated with treatment discontinuation, because so few individuals had a documented ADR, this structured data element alone is likely insufficient for research purposes to accurately identify the reasons for treatment discontinuation. However, given that in other studies hyperkalemia and AKI were among the most common reasons for discontinuation [11], datasets with access to laboratory values may incorporate these elements to more comprehensively identify reasons for treatment discontinuation. It is also possible that reasons for discontinuation are reported in clinical encounter notes and may be detected with natural language processing.

The frequency of several types of ADR we presented were similar to a prior study of documented ADRs that identified cough in 41.0%, angioedema in 6.4%, and hyperkalemia in 4.4% of reported ADRs among inpatients [12]. However, documentation of ADRs likely differs from the actual rates of those adverse events in clinical practice. A severe adverse event such as angioedema may be more likely to be captured in a patient’s medical record than a non-life-threatening reaction such as peripheral edema and mild hypotension. Additionally, a reaction that is well known to clinicians such as cough with ACEis may be more likely to be documented as an ADR than a vague or unexpected symptom that is less clearly caused by the medication, such as anxiety. Because of this, our results examining only documented ADR data differ notably from a prior study that conducted more complete retrospective reviews of medical records to determine the causes of ACEi or ARB discontinuation among patients with CKD stages 3–5. This study cited reasons for ACEi or ARB discontinuation as hyperkalemia (66.6%), acute deterioration of renal function (17.6%), hypotension (5.9%), renal artery stenosis (3.9%), very low glomerular filtration rate (2.0%), angioedema (2.0%), and cough (2.0%) [11]. The differences between more comprehensively obtained adverse events and documented ADRs suggest that ADRs are likely not documented consistently across categories, notably with hyperkalemia and AKI being under-documented and cough and angioedema being more commonly documented.

CKD stage and diabetes mellitus status impacted the frequency of some types of documented ADRs. Individuals with more advanced CKD stages were more likely to have a documented ADR related to hyperkalemia, AKI, or hypotension, and patients with diabetes mellitus were more likely to have one related to hyperkalemia or hypotension. These findings are expected based on the physiology of CKD and diabetes and are likely largely attributable to increased adverse event rates rather than altered rates of ADR documentation.

Accurate assessment of reasons for ACEi or ARB discontinuation is critical to reduce the underutilization of ACEis and ARBs in patients with CKD. Within the time constraints of real-world clinical practice, a clearly documented and easily identifiable reason for drug discontinuation may assist frontline clinicians in quick decision-making about re-initiating treatment. In research, accurately identifying reasons for discontinuation may reveal novel pathways to address healthcare delivery and increase utilization of these medications to improve long-term kidney outcomes. Documenting ADRs may be an important component of this but would require much more consistent use to achieve these goals. System-level interventions to increase outpatient treatment reinitiation of ACEi or ARB after they were held in response to an AKI or episode of hyperkalemia may also improve utilization of these important agents in patients with CKD.

This study has several strengths. We investigated a large, national sample with comprehensive available clinical data. Few prior studies have described the prevalence of documented ADRs or explored the utility of this structured data item for identifying reasons for discontinuation of ACEis or ARBs. Our study also has important limitations. We did not review participant charts to comprehensively ascertain reasons for treatment discontinuation to compare to documented ADRs. Comorbidities were ascertained by diagnosis codes. Despite these, our real-world data from a large health care system build on existing knowledge about ADR documentation and its importance.

In conclusion, in this descriptive retrospective cohort study, we demonstrated in a national sample that although adverse reactions to ACEis and ARBs were uncommonly documented as an ADR structured data item, a documented ADR was strongly associated with ACEi or ARB treatment discontinuation. We reported the rates of specific types of documented ADRs and identified ADR categories that were associated with treatment discontinuation or continuation. Further efforts are needed to incorporate documented ADRs and other structured data items to more comprehensively identify reasons for ACEi or ARB discontinuation in patients who may benefit from these medications. Interventions or clear guidance for the reinitiation of ACEis or ARBs in appropriate patients after AKI or hyperkalemia may be a valuable approach to increase treatment utilization and improve kidney outcomes.

The opinions expressed reflect those of the authors and not necessarily those of the Department of Veterans Affairs, NIH, or the US government.

The study was approved by the Institutional Review Boards at Baylor College of Medicine, the Michael E. DeBakey VA Medical Center, and the Tennessee Valley Healthcare System VA, and a waiver of informed consent was granted at both locations, approval number [H-47583].

Dr. Virani has received honoraria from the American College of Cardiology in his role as the Associate Editor for Innovations, acc.org. Dr. Gregg serves as an Editorial Fellow for the Journal of the American Society of Nephrology. Dr. Matheny reported receiving consulting from a VA-DoD-NIH Pain Management Consortium. Outside the submitted work, Dr. Navaneethan reported receiving personal fees from ACI Clinical, AstraZeneca (Data Safety Monitoring board), Bayer, Boehringer Ingelheim/Eli Lilly and Co, GSK, Intercept and Vertex (Event Adjudication Committee), and Vifor; receiving grants from Keryx; and receiving research funding from the Department of Veterans Affairs Health Services Research and Development.

This work was supported by a Department of Veterans Affairs Health Service Research and Development Service Investigator Initiated Grants (IIR 16-072, IIR 19-069) and the Houston VA Health Services Research and Development Center for Innovations grant (CIN13-413). Support for VA/CMS data provided by the Department of Veterans Affairs, VA Health Services Research and Development Service, VA Information Resource Center (Project Numbers SDR 02-237 and 98-004). Dr. Gregg is supported by a VA Clinical Sciences Research and Development Career Development Award (IK2CX002368). SDN is also supported NIH/NHLBI K24 HL161414-01A). Dr. Navaneethan is also supported by Garabed Eknoyan MD Endowed Professorship. Dr. Virani is supported by research grants from the Department of Veterans Affairs, NIH, and the Tahir and Jooma Family. Dr. Matheny was supported by research grants from the Department of Veteran Affairs HSR&D IIR 16-072, 19-069, and SDR 18-194.

Conception and study design: L. Parker Gregg, Peter A. Richardson, Michael E. Matheny, Salim S. Virani, and Sankar D. Navaneethan. Data acquisition and analysis: L. Parker Gregg, Peter A. Richardson, Michael E. Matheny, Salim S. Virani, and Sankar D. Navaneethan. Interpretation of the data: L. Parker Gregg, Peter A. Richardson, Michael A. Herrera, Julia M. Akeroyd, Shehrezade A. Jafry, Glenn T. Gobbel, Sheena Wydermyer, Jennifer Arney, Adriana Hung, Michael E. Matheny, Salim S. Virani, and Sankar D. Navaneethan. Drafting or revising the manuscript: L. Parker Gregg, Peter A. Richardson, Michael A. Herrera, Julia M. Akeroyd, Shehrezade A. Jafry, Glenn T. Gobbel, Sheena Wydermyer, Jennifer Arney, Adriana Hung, Michael E. Matheny, Salim S. Virani, and Sankar D. Navaneethan. Approval to be published: L. Parker Gregg, Peter A. Richardson, Michael A. Herrera, Julia M. Akeroyd, Shehrezade A. Jafry, Glenn T. Gobbel, Sheena Wydermyer, Jennifer Arney, Adriana Hung, Michael E. Matheny, Salim S. Virani, and Sankar D. Navaneethan.

All data generated or analyzed during this study are included in this article and its online supplementary material files. Further inquiries can be directed to the corresponding author.

1.
Chobanian
AV
,
Bakris
GL
,
Black
HR
,
Cushman
WC
,
Green
LA
,
Izzo
JL
Jr
.
Seventh report of the joint national committee on prevention, detection, evaluation, and treatment of high blood pressure
.
Hypertension
.
Dec 2003
42
6
1206
52
.
2.
Taler
SJ
,
Agarwal
R
,
Bakris
GL
,
Flynn
JT
,
Nilsson
PM
,
Rahman
M
.
KDOQI US commentary on the 2012 KDIGO clinical practice guideline for management of blood pressure in CKD
.
Am J Kidney Dis
.
2013
;
62
(
2
):
201
13
.
3.
Kidney Disease Improving Global Outcomes KDIGO Blood Pressure Work Group
.
KDIGO 2021 clinical practice guideline for the management of blood pressure in chronic kidney disease
.
Kidney Int
.
2021
99
3S
S1
87
.
4.
Agodoa
LY
,
Appel
L
,
Bakris
GL
,
Beck
G
,
Bourgoignie
J
,
Briggs
JP
.
Effect of ramipril vs amlodipine on renal outcomes in hypertensive nephrosclerosis: a randomized controlled trial
.
JAMA
.
2001
;
285
(
21
):
2719
28
.
5.
Xie
X
,
Liu
Y
,
Perkovic
V
,
Li
X
,
Ninomiya
T
,
Hou
W
.
Renin-angiotensin system inhibitors and kidney and cardiovascular outcomes in patients with CKD: a bayesian network meta-analysis of randomized clinical trials
.
Am J Kidney Dis
.
May 2016
67
5
728
41
.
6.
Chu
CD
,
Powe
NR
,
McCulloch
CE
,
Banerjee
T
,
Crews
DC
,
Saran
R
.
Angiotensin-converting enzyme inhibitor or angiotensin receptor blocker use among hypertensive US adults with albuminuria
.
Hypertension
.
2021
;
77
(
1
):
94
102
.
7.
Navaneethan
SD
,
Akeroyd
JM
,
Ramsey
D
,
Ahmed
ST
,
Mishra
SR
,
Petersen
LA
.
Facility-level variations in kidney disease care among veterans with diabetes and CKD
.
Clin J Am Soc Nephrol
.
Dec 7 2018
13
12
1842
50
.
8.
Walther
CP
,
Winkelmayer
WC
,
Richardson
PA
,
Virani
SS
,
Navaneethan
SD
.
Renin-angiotensin system blocker discontinuation and adverse outcomes in chronic kidney disease
.
Nephrol Dial Transplant
.
2021
;
36
(
10
):
1893
9
.
9.
Leon
SJ
,
Whitlock
R
,
Rigatto
C
,
Komenda
P
,
Bohm
C
,
Sucha
E
.
Hyperkalemia-related discontinuation of renin-angiotensin-aldosterone system inhibitors and clinical outcomes in CKD: a population-based cohort study
.
Am J Kidney Dis
.
2022
;
80
(
2
):
164
73
. e1.
10.
Mathew
RO
,
Lo
KB
,
Tipparaju
P
,
Phelps
E
,
Sidhu
MS
,
Bangalore
S
.
Patterns of use and clinical outcomes with angiotensin-converting enzyme inhibitors and angiotensin receptor blockers in acute heart failure and changes in kidney function: an analysis of the veterans' health administrative database
.
Cardiorenal Med
.
2021
11
5–6
226
36
.
11.
Yildirim
T
,
Arici
M
,
Piskinpasa
S
,
Aybal-Kutlugun
A
,
Yilmaz
R
,
Altun
B
.
Major barriers against renin-angiotensin-aldosterone system blocker use in chronic kidney disease stages 3-5 in clinical practice: a safety concern
.
Ren Fail
.
2012
;
34
(
9
):
1095
9
.
12.
McConeghy
KW
,
Caffrey
AR
,
Morrill
HJ
,
Trivedi
AN
,
LaPlante
KL
.
Are non-allergic drug reactions commonly documented as medication “allergies”? A national cohort of Veterans’ admissions from 2000 to 2014
.
Pharmacoepidemiol Drug Saf
.
2017
;
26
(
4
):
472
6
.
13.
Inker
LA
,
Eneanya
ND
,
Coresh
J
,
Tighiouart
H
,
Wang
D
,
Sang
Y
.
New creatinine- and cystatin C-based equations to estimate GFR without race
.
N Engl J Med
.
2021
;
385
(
19
):
1737
49
.
14.
Stevens
PE
,
Levin
A
Kidney Disease: Improving Global Outcomes Chronic Kidney Disease Guideline Development Work Group Members
.
Evaluation and management of chronic kidney disease: synopsis of the kidney disease: improving global outcomes 2012 clinical practice guideline
.
Ann Intern Med
.
2013
;
158
(
11
):
825
30
.
15.
Norton
JM
,
Ali
K
,
Jurkovitz
CT
,
Kiryluk
K
,
Park
M
,
Kawamoto
K
.
Development and validation of a pragmatic electronic phenotype for CKD
.
Clin J Am Soc Nephrol
.
2019
;
14
(
9
):
1306
14
.
16.
Virani
SS
,
Akeroyd
JM
,
Ahmed
ST
,
Krittanawong
C
,
Martin
LA
,
Slagle
J
.
The use of structured data elements to identify ASCVD patients with statin-associated side effects: insights from the Department of Veterans Affairs
.
J Clin Lipidol
.
Sep-Oct 2019
13
5
797
803
. e1.
17.
McCoy
IE
,
Han
J
,
Montez-Rath
ME
,
Chertow
GM
.
Barriers to ACEI/ARB use in proteinuric chronic kidney disease: an observational study
.
Mayo Clin Proc
.
2021
;
96
(
8
):
2114
22
.
18.
Vaduganathan
M
,
Fonarow
GC
,
Greene
SJ
,
Devore
AD
,
Albert
NM
,
Duffy
CI
.
Treatment persistence of renin-angiotensin-aldosterone-system inhibitors over time in heart failure with reduced ejection fraction
.
J Card Fail
.
2022
;
28
(
2
):
191
201
.
19.
Fried
L
,
Schmedt
N
,
Folkerts
K
,
Bowrin
K
,
Raad
H
,
Batech
M
.
High unmet treatment needs in patients with chronic kidney disease and type 2 diabetes: real-world evidence from a US claims database
.
Nephrol Dial Transplant
.
2023
;
38
(
3
):
630
43
.