Abstract
Background
Renin–angiotensin system inhibitors (RASis), including ACE inhibitors (ACEis) and angiotensin receptor blockers (ARBs), are commonly initiated in primary care for various clinical indications. However, the risk of hyperkalaemia and increased serum creatinine levels after initiation has not been well studied in Asian populations.
Methods
We conducted a retrospective cohort study of all patients aged 18 years and older who initiated RASi treatment at National University Polyclinics, a network of seven primary care clinics in Singapore, between 28 September 2020 and 31 June 2024. The study outcomes were the incidence of hyperkalaemia and elevated serum creatinine levels after RASi initiation. Multivariable regression was used to evaluate the covariates associated with these outcomes.
Results
The cohort comprised 9,926 patients whose baseline creatinine and post-initiation creatinine or potassium data were available. Within our cohort, 181 patients (1.8%) had hyperkalaemia, comprising 149 (1.5%) mild cases, 28 (0.3%) moderate cases and 4 (< 0.1%) severe cases; 249 (2.5%) patients had a major elevation in serum creatinine of 30% or more from baseline. Risk factors for hyperkalaemia include age, Indian or ‘Other’ ethnicity, increased baseline serum creatinine and increased baseline serum potassium. For major serum creatinine elevation, risk factors included ‘Other’ ethnicity, comorbid diabetes mellitus and concomitant diuretic use.
Conclusion
Hyperkalaemia and major serum creatinine elevation are uncommon adverse events following RASi initiation.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12875-026-03195-8.
Keywords: Renin-angiotensin system inhibitor, Hyperkalaemia, Adverse effect, Primary care, Asian population
Introduction
Renin–angiotensin system inhibitors (RASis), including angiotensin-converting enzyme inhibitors (ACEis) and angiotensin receptor blockers (ARBs), are essential for managing chronic conditions such as hypertension, diabetes mellitus, chronic kidney disease, ischaemic heart disease and heart failure [1, 2]. As RASi initiation may cause hyperkalaemia and elevated serum creatinine, international and local institutional guidelines recommend checking for serum potassium and creatinine within two to four weeks after RASi initiation and to reduce or stop RASi in case of moderate to severe hyperkalaemia or major serum creatinine elevation [1–3].
Hyperkalaemia after RASi initiation occurs due to impaired renal potassium excretion arising from reduced intraglomerular pressure and inhibition of aldosterone secretion [4, 5]. The incidence of hyperkalaemia following RASi initiation reported in earlier studies ranges from 1.2% to 10.8% [6, 7], but this data is of limited applicability in a primary care setting owing to inconsistent definitions of hyperkalaemia and disease-centric trial populations such as chronic kidney disease or heart failure [8]. More recent population-level studies revealed a lower range of hyperkalaemia risk following RASi initiation. In 2017, the Stockholm Creatinine Measurements (SCREAM) study reported a 5.6% incidence of mild hyperkalaemia (> 5.0 mmol/L) [9], whereas a UK study in 2021 reported a 3.4% risk for moderate or severe hyperkalaemia (> 5.5 mmol/L) [10]. The identified risk factors for hyperkalaemia in these two studies included older age, diabetes mellitus, congestive heart failure (CHF), chronic kidney disease (CKD) and concomitant medications such as nonsteroidal anti-inflammatory drugs (NSAIDs), mineralocorticoid antagonists (MRAs), and potassium supplements. The SCREAM study findings were used to derive a risk score comprising 6 risk factors (male sex, baseline serum potassium, baseline eGFR, diabetes mellitus, heart failure and concomitant potassium-sparing diuretic use) which was subsequently validated in another RASi initiation cohort.
Serum creatinine elevation is recognised as a potential adverse effect of antihypertensive treatment in general. Intensive blood pressure control in the landmark ACCORD-BP and SPRINT trials was associated with a greater risk of ‘incident CKD’ (defined as a decrease in eGFR of ≥ 30% to < 60 mL/min) regardless of antihypertensive class [11–13]. The decline in renal function is thought to be due to impaired renal autoregulation of intraglomerular pressure, particularly in patients with chronic hypertension and diabetes [14, 15]. RASis have a greater propensity for serum creatinine elevation due to decreased intraglomerular pressure from efferent arteriole vasodilation, especially in cases of volume depletion, heart failure or renal artery stenosis [16]. Prior studies reported a 1.7% to 3.4% risk of elevated creatinine of 30% or more from baseline following RASi initiation, with identified risk factors including age, increased comorbidity burden and polypharmacy [17, 18]. To date, international guidelines recommend an elevation of 30% of more from baseline as a threshold for modifying therapy but acknowledge that this threshold is based on expert opinion [2, 3].
Singapore, a highly developed and urbanised city-state in Southeast Asia, has a multiethnic population comprising approximately 76% Chinese, 15% Malay and 7% Indian ethnicities [19]. The use of RASis is anticipated to increase in the Singapore population, along with other Asian populations facing an increasing burden of cardiometabolic disease [20]. However, providers’ concerns over potential adverse effects may create barriers to initiating these potentially beneficial medications, of which hyperkalaemia and creatinine elevation are the most frequently cited concerns [21, 22]. Given that the risk of hyperkalaemia is influenced by dietary potassium intake [23], it is conceivable that ethnic differences may affect the risk of hyperkalaemia compared with the predominantly Western populations where prior studies were conducted. Another anecdotal barrier commonly encountered in our setting is the need for repeat serum potassium and creatinine testing within 2–4 weeks after RASi initiation, which may deter patients who are unable to return early for testing.
To our knowledge, no studies have examined the incidence and risk factors for hyperkalaemia and creatinine elevation following RASi initiation in an Asian population. To address this gap, we investigated these outcomes in a multiethnic Asian primary care population in Singapore. Estimating this risk more accurately and identifying risk factors will inform primary care providers who are considering initiating RASis for their patients. Given that the need for repeat testing within 2‒4 weeks is a barrier to initiating RASis in our setting, further stratification of these risks may reduce the need for post-initiation testing in low-risk patients, thereby increasing RASi utilisation where indicated.
Methods
Setting and participants
A retrospective cohort study was conducted among ambulatory patients with a first dispensation of RASis in National University Polyclinics, a network of seven primary care clinics serving Western Singapore, from 28 September 2020 to 31 June 2024.
Data extraction
Deidentified data were extracted from the electronic medical records of all patients who received ACEis or ARBs between 28 September 2020 and 31 June 2024 inclusive. The current version of the database was implemented between 28 September 2020 and 2 November 2021 across the seven clinics. An initiation event was identified by listing all ACEi or ARB dispensations per patient within the study period and selecting the earliest date of dispensation. Single-dose dispensations were excluded. Initiation events within 365 days of each clinic’s database implementation were excluded as a washout period to avoid continued prescriptions from the previous version of the electronic medical record. Data on age, sex, ethnicity, body mass index (BMI), baseline laboratory data before 365 days of initiation, comedications and the presence of comorbid diabetes mellitus or congestive heart failure were extracted. Ethnicity was classified into Chinese, Malay, Indian or ‘Others’ according to local convention. The comedications studied mirrored the comedications included in prior studies [9, 17, 18] and included potassium supplements, diuretics, beta-blockers, spironolactone (the only MRA available), NSAIDs, cyclooxygenase-2 inhibitors (COX-2is) and sodium‒glucose cotransporter-2 inhibitors (SGLT2is). Comorbidity data were extracted based on International Classification of Diseases 10th Revision (ICD-10) diagnostic codes. Informed consent was not required by institutional policies because of the exclusive use of deidentified retrospective clinical data. The study was exempt from ethical review by the institutional Domain Specific Review Board (DSRB Reference 2023/01018).
Inclusion and exclusion criteria
All patients aged 18 years and above for whom ACEis or ARBs were initiated were included. Patients were excluded if they did not have a baseline serum creatinine within 365 days before initiation or if they lacked serum potassium or creatinine measurements between 14 and 90 days after initiation.
Study outcomes
The outcomes included the incidence of hyperkalaemia (> 5.0 mmol/L) and major elevations in serum creatinine post-initiation, defined as an increase of more than 30% from baseline. Hyperkalaemia events were classified as mild (5.1 to 5.4 mmol/L), moderate (5.5 to 6.0 mmol/L) or severe (> 6.0 mmol/L) [24]. We also studied a combined outcome of either moderate or worse hyperkalaemia or major serum creatinine elevation as a safety threshold, which typically prompts providers to stop or reduce the dose of RASis. An additional exploratory outcome evaluated was a combined outcome of major serum creatinine elevation and a post-initiation estimated glomerular filtration rate (eGFR) < 60 mL/min.
Statistical analysis
Baseline eGFR was calculated from sex, serum creatinine and age data with The 2009 CKD-EPI equation in alignment with current institutional practice, with all patients assumed to be non-black [25]. Medications were considered concomitant if they were dispensed between 90 days before initiation and the date of post-initiation laboratory testing. Post-initiation serum potassium and creatinine levels were extracted within 14 to 90 days after RASi initiation. If there were multiple measurements, the earliest value within the window was selected. Multivariable logistic regression was used to calculate odds ratios (ORs) and 95% confidence intervals (CIs) for hyperkalaemia and significant creatinine elevation using covariates identified in previous studies, namely, age, sex, comorbid diabetes mellitus and congestive heart failure and baseline serum creatinine and potassium. While BMI and baseline urine albumin to creatinine ratio (uACR) data was extracted, they were not included in the final regression model due to substantial proportions of missing data but were included in a complete case analysis.
The following sensitivity analyses were performed to evaluate the robustness of the multivariable regression findings: testing for dose-response effects for the most common RASi, using post-initiation data within 14–30 days and testing varying methods of analysing the post-initiation serum potassium data, firstly by extracting the highest rather than earliest post-initiation serum potassium value within the date range and secondly by excluding patients with baseline hyperkalaemia (serum potassium ≥ 5.1 mmol/L).
All analyses were performed in R version 4.5.1 [26] with the following packages: tidyverse for data processing [27], nephro for eGFR calculations [28] and gtsummary for Table [29].
Results
Baseline and post-initiation testing
A total of 23,704 patients initiated RASi treatment between 28 September 2020 and 31 June 2024. Among these patients, 14,463 (61.0%) had a baseline serum creatinine measurement within 365 days of initiation. The final cohort comprised 9,926 (41.9%) patients with both baseline and post-initiation serum potassium or creatinine measurements taken within 14 to 90 days post-initiation (Fig. 1).
Fig. 1.
Data extraction
Baseline characteristics
Within the cohort (Table 1), The mean age was 64 years, and most patients were male (53.7%, n = 5,326) and of Chinese ethnicity (74.2%, n = 7365). The mean baseline serum creatinine was 80.9 µmol/L, 83.9% (n = 8,329) had a baseline eGFR of 60mL/min or greater and 95.1% (n = 9,443) had normal baseline serum potassium; 42.7% (n = 4,239) had diabetes mellitus and 0.6% (n = 58) had congestive heart failure. Among RASi initiations, 74.3% (n = 7372) were ARBs, with losartan being the most common. For comedications at RASi initiation, 12.6% (n = 1,255) were on beta-blockers and 5.2% (n = 520) were on SGLT2i; less than 2% of the cohort was on the other comedications studied. Baseline BMI and uACR data was missing for 21.8% (n = 2,172) and 23.5% (n = 2,337) of the cohort respectively.
Table 1.
Baseline characteristics of the final cohort and the patients excluded because of missing post-initiation data
| Final Cohort (n = 9926) |
Post-Initiation Data Missing (n = 4537) |
|||
|---|---|---|---|---|
| Age – mean (SD) | 63.9 (12.0) | 63.7 (12.9) | ||
| Sex – count (%) | ||||
| Male | 5326 | (53.7) | 2604 | (57.4) |
| Female | 4600 | (46.3) | 1933 | (42.6) |
| Ethnicity – count (%) | ||||
| Chinese | 7365 | (74.2) | 3094 | (68.2) |
| Malay | 1287 | (13.0) | 630 | (13.9) |
| Indian | 646 | (6.5) | 465 | (10.2) |
| Others | 628 | (6.3) | 348 | (7.7) |
| BMI in kg/m2 – mean (SD) | 26.9 (5.2) | 26.8 (5.4) | ||
| Missing | 2172 | (21.8) | 1935 | (42.6) |
| Type of RASi Initiated – count (%) | ||||
| ARB | 7372 | (74.3) | 3036 | (66.9) |
| Losartan | 4902 | (49.4) | 1499 | (33.0) |
| Valsartan | 1285 | (12.9) | 740 | (16.3) |
| Telmisartan | 1090 | (11.0) | 622 | (13.7) |
| Irbesartan | 73 | (0.7) | 127 | (2.8) |
| Candesartan | 22 | (0.2) | 48 | (1.1) |
| ACEi | 2554 | (25.7) | 1501 | (33.1) |
| Lisinopril | 1732 | (17.4) | 835 | (18.4) |
| Enalapril | 764 | (7.7) | 525 | (11.6) |
| Perindopril | 54 | (0.5) | 138 | (3.0) |
| Captopril | 4 | (< 0.1) | 3 | (0.1) |
| Baseline Serum Creatinine in μmol/L – mean (SD) | 80.9 (29.5) | 84.4 (34.5) | ||
| Baseline eGFR in mL/min – count (%) | ||||
| ≥ 60 | 8329 | (83.9) | 3754 | (82.7) |
| 45 – 59 | 1103 | (11.1) | 461 | (10.2) |
| 30 – 44 | 385 | (3.9) | 236 | (5.2) |
| < 30 | 109 | (1.1) | 86 | (1.9) |
| Baseline Serum Potassium in mmol/L – count (%) | ||||
| < 5.1 | 9443 | (95.1) | 4271 | (94.1) |
| ≥ 5.1 | 101 | (1.0) | 59 | (1.3) |
| Missing | 382 | (3.8) | 207 | (4.6) |
| Baseline Microalbuminuria by KDIGO Categories – count (%) | ||||
| A1 | 2139 | (21.5) | 1551 | (34.2) |
| A2 | 4697 | (47.3) | 1076 | (23.7) |
| A3 | 753 | (7.6) | 276 | (6.1) |
| Missing | 2337 | (23.5) | 1634 | (36.0) |
| Comorbidity – count (%) | ||||
| DM | 4239 | (42.7) | 2096 | (46.2) |
| CHF | 58 | (0.6) | 57 | (1.3) |
| Comedication – count (%) | ||||
| Potassium supplement | 97 | (1.0) | 0 | (0.0) |
| Spironolactone | 30 | (0.3) | 0 | (0.0) |
| Diuretic | 176 | (1.8) | 54 | (1.1) |
| Beta-blocker | 1255 | (12.6) | 0 | (0.0) |
| NSAID/COX-2i | 103 | (1.0) | 0 | (0.0) |
| SGLT2i | 520 | (5.2) | 0 | (0.0) |
SD indicates standard deviation, BMI, body mass index, RASi renin-angiotensin system inhibitor, ARB angiotensin receptor blocker, ACEi angiotensin-converting enzyme,inhibitor, eGFR estimated glomerular filtration rate, KDIGO Kidney Disease: Improving Global Outcomes, DM diabetes mellitus, CHF congestive heart failure, NSAID nonsteroidal anti-inflammatory drug, COX-2i cyclooxygenase-2 inhibitor, SGLT2i sodium-glucose cotransporter 2 inhibitor
Compared with patients excluded due to missing post-initiation data, the final cohort had a similar age, baseline renal function and baseline serum potassium, a lower proportion of those of Indian ethnicity (6.5% compared to 10.5%), a higher proportion of ARB initiation (74.3% compared to 66.9%), a higher proportion of patients with diabetes mellitus (42.7% compared to 46.2%) and a lower proportion of patients with comedications.
We additionally analysed data among patients with missing baseline data and found that while they were largely similar in terms of demographic and comorbidity characteristics, missing baseline serum creatinine data was almost accompanied by other missing laboratory data including serum potassium and uACR status which largely precluded further analysis (Table S1).
Rates of hyperkalaemia and elevated serum creatinine after RASi initiation
Among the 9,694 patients with post-initiation serum potassium measurements, 181 (1.8%) had hyperkalaemia of which 149 (1.5%) were mild, 28 (0.3%) were moderate and 4 (< 0.1%) were severe (Table 2). Among 9,892 patients with post-initiation serum creatinine measurements, 249 (2.5%) had a major elevation in serum creatinine, of which 122 (1.2%) also had a post-initiation eGFR < 60 mL/min, and 275 (2.7%) had the combined outcome of either moderate hyperkalaemia or major serum creatinine elevation (Table 2).
Table 2.
Adverse outcomes at 14–90 days after RASi initiation
| Outcome | Event Rate (%) | |
|---|---|---|
| Hyperkalaemia | 181 | (1.8) |
| Mild (5.1 to 5.4 mmol/L) | 149 | (1.5) |
| Moderate (5.5 to 6.0 mmol/L) | 28 | (0.3) |
| Severe (> 6.0 mmol/L) | 4 | (<0.1) |
| Serum creatinine elevation from baseline | 3323 | (33.6) |
| > 10% | 2344 | (23.7) |
| > 20% | 730 | (7.4) |
| > 30% | 249 | (2.5) |
| Serum creatinine elevation > 30% from baseline and post-initiation eGFR < 60 mL/min | 122 | (1.2) |
| Serum potassium ≥ 5.5 mmol/L or serum creatinine rise > 30% from baseline | 275 | (2.8) |
Risk factors for hyperkalaemia and major serum creatinine elevation of after RASi initiation
To examine the risk factors for hyperkalaemia and major serum creatinine elevation after RASi initiation, we conducted a multivariable regression analysis (Table 3). For hyperkalaemia, significant risk factors included age (OR per 10 years 1.27 [95% CI: 1.08–1.50]), Indian or ‘Other’ ethnicity (OR 2.26 [95% CI 1.34–3.67] and 2.31 [95% CI 1.30–3.90] respectively), increased baseline serum creatinine (OR per 10 µmol/L 1.01 [95% CI 1.01–1.01]) and increased baseline serum potassium (OR per 0.1 mmol/L 14.0 [95% CI 9.04–21.9]). When the odds ratio for hyperkalaemia was analysed against baseline serum potassium, the risk of hyperkalaemia was found to increase substantially beyond a baseline serum potassium of 4.5 mmol/L (Fig. 2).
Table 3.
Factors associated with hyperkalaemia and major serum creatinine elevation 14–90 days after RASi initiation
| Characteristic | Hyperkalaemia | Serum Creatinine Elevation > 30% from Baseline | ||
|---|---|---|---|---|
| Event Rate/Total (%) | Adjusted OR (95% CI) | Event Rate/Total (%) | Adjusted OR (95% CI) | |
| Age (per 10 years) | - | 1.27 (1.08 – 1.50) | - | 1.09 (0.97 – 1.23) |
| Sex | ||||
| Male | 80 / 5198 (1.5) | 1.00 | 108 / 5308 (2.0) | 1.00 |
| Female | 69 / 4496 (1.5) | 1.32 (0.92 – 1.89) | 141 / 4584 (3.1) | 1.05 (0.76 – 1.44) |
| Ethnicity | ||||
| Chinese | 88 / 7209 (1.2) | 1.00 | 178 / 7337 (2.4) | 1.00 |
| Malay | 18 / 1251 (1.4) | 1.02 (0.56 – 1.72) | 39 / 1284 (3.0) | 1.29 (0.88 – 1.84) |
| Indian | 24 / 626 (3.8) | 2.26 (1.34 – 3.67) | 9 / 643 (1.4) | 0.55 (0.26 – 1.04) |
| Others | 19 / 608 (3.1) | 2.31 (1.30 – 3.90) | 23 / 628 (3.7) | 1.71 (1.06 – 2.65) |
| Type of RASi | ||||
| ARB | 111 / 7229 (1.5) | 1.00 | 179 / 7353 (2.4) | 1.00 |
| ACEi | 38 / 2465 (1.5) | 0.89 (0.59 – 1.31) | 70 / 2539 (2.8) | 1.10 (0.82 – 1.45) |
| Baseline Serum Creatinine (per 10 μmol/L) | - | 1.01 (1.01 – 1.01) | - | 0.99 (0.98 – 0.99) |
| Baseline Serum Potassium (per 0.1 mmol/L) | - | 14.0 (9.04 – 21.9) | - | 0.61 (0.43 – 0.86) |
| Comorbidity | ||||
| DM | 85 / 4109 (2.1) | 1.19 (0.83 – 1.71) | 136 / 4219 (3.2) | 1.69 (1.29 – 2.21) |
| CHF | 1 / 55 (1.8) | 0.47 (0.02 – 2.61) | 1 / 58 (1.7) | 0.49 (0.03 – 2.46) |
| Comedication | ||||
| Potassium sup. | 0 / 97 (0.0) | - | 4 / 97 (4.1) | 0.83 (0.23 – 2.22) |
| Spironolactone | 0 / 29 (0.0) | - | 2 / 30 (6.7) | 3.26 (0.50 – 11.8) |
| Diuretic | 3 / 173 (1.7) | 0.46 (0.09 – 1.57) | 13 / 176 (7.4) | 3.47 (1.70 – 6.47) |
| Beta-blocker | 29 / 1245 (2.3) | 1.35 (0.86 – 2.06) | 31 / 1255 (2.5) | 0.94 (0.62 – 1.38) |
| NSAID/COX2i | 3 / 103 (2.9) | 3.35 (0.78 – 9.76) | 3 / 103 (2.9) | 1.11 (0.27 – 3.02) |
| SGLT2i | 12 / 516 (2.3) | 1.33 (0.66 – 2.48) | 14 / 520 (2.7) | 0.81 (0.44 – 1.39) |
RASi indicates renin-angiotensin system inhibitor, OR odds ratio, CI confidence interval, ARB angiotensin receptor blocker, ACEi angiotensin-converting enzyme inhibitor, DM diabetes mellitus, CHF congestive heart failurem, NSAID non-steroidal anti-inflammatory drug, COX-2 cyclooxygenase-2 inhibitor, SGLT2i sodium-glucose cotransporter 2 inhibitor
Bold text denotes statistically-significant results
Fig. 2.
Point estimate of odds ratio for hyperkalaemia against baseline serum potassium
For major serum creatinine elevation (Table 3), significant risk factors included ‘Other’ ethnicity (OR 1.71 [95% CI 1.06–2.65]), comorbid diabetes mellitus (OR 1.69 [95% CI 1.29–2.21]) and concomitant diuretic use (OR 3.47 [95% CI 1.70–6.47]). Elevated baseline serum creatinine and potassium were found to be protective (OR 0.99 [95% CI 0.98–0.99] and 0.64 [95% CI 0.43–0.86]).
When complete case analysis including baseline BMI and uACR was performed (Table 4), increased BMI was additionally found to be a protective factor against hyperkalaemia (OR 0.93 [95% CI 0.87–0.98]), while Indian ethnicity, increased baseline serum creatinine and potassium remained significant risk factors. For major creatinine elevation, diabetes mellitus and comorbid use of diuretics remained significant risk factors while increased baseline serum creatinine remained protective.
Table 4.
Complete case analysis of factors associated with hyperkalaemia and major serum creatinine elevation at 14-30 days after RASi initiation (n = 6102)
| Characteristic | Hyperkalaemia | Serum Creatinine Elevation > 30% from Baseline | ||
|---|---|---|---|---|
| Event Rate/Total (%) | Adjusted OR (95% CI) | Event Rate/Total (%) | Adjusted OR (95% CI) | |
| Age (per 10 years) | - | 1.16 (0.90 – 1.50) | - | 1.10 (0.94 – 1.29) |
| Sex | ||||
| Male | 40 / 3167 (1.3) | 1.00 | 74 / 3204 (2.3) | 1.00 |
| Female | 34 / 2863 (1.2) | 1.55 (0.90 – 2.66) | 84 / 2898 (2.9) | 0.73 (0.49 – 1.11) |
| Ethnicity | ||||
| Chinese | 49 / 4522 (1.1) | 1.00 | 113 / 4574 (2.5) | 1.00 |
| Malay | 7 / 770 (0.9) | 1.03 (0.41 – 2.28) | 27 / 777 (3.5) | 1.29 (0.80 – 2.02) |
| Indian | 11 / 386 (2.8) | 2.27 (1.02 – 4.64) | 7 / 393 (1.8) | 0.62 (0.26 – 1.27) |
| Others | 7 / 325 (2.0) | 1.79 (0.68 – 4.03) | 11 / 358 (3.1) | 1.30 (0.65 – 2.37) |
| BMI (per kg/m2) | - | 0.93 (0.87 – 0.98) | - | 1.02 (0.99 – 1.05) |
| Type of RASi | ||||
| ARB | 58 / 4610 (1.3) | 1.00 | 114 / 4651 (2.5) | 1.00 |
| ACEi | 16 / 1420 (1.1) | 0.89 (0.48 – 1.56) | 44 / 1451 (3.0) | 1.22 (0.85 – 1.73) |
| Baseline Serum Creatinine (per 10μmol/L) | - | 1.02 (1.01 – 1.02) | - | 0.98 (0.97 – 0.99) |
| Baseline Serum Potassium (per 0.1 mmol/L) | - | 13.5 (7.22 – 25.4) | - | 0.77 (0.49 – 1.21) |
| Baseline uACR (per 0.1 mg/mmol) | - | 1.01 (0.96 – 1.05) | - | 1.03 (0.98 – 1.06) |
| Comorbidity | ||||
| DM | 39 / 2619 (1.5) | 0.87 (0.51 – 1.47) | 84 / 2658 (3.2) | 1.55 (1.11 – 2.16) |
| CHF | 0 / 23 (0.0) | - | 0 / 25 (0.0) | - |
| Comedication | ||||
| Potassium sup. | 0 / 67 (0.0) | - | 1 / 67 (1.5) | 0.28 (0.01 – 1.45) |
| Spironolactone | 0 / 15 (0.0) | - | 1 / 15 (6.7) | 3.21 (0.17 – 16.9) |
| Diuretic | 1 / 93 (1.1) | 0.24 (0.01 – 1.72) | 7 / 94 (7.4) | 4.72 (1.85 – 10.4) |
| Beta-blocker | 20 / 803 (2.5) | 1.75 (0.97 – 3.05) | 17 / 808 (2.1) | 0.78 (0.44 – 1.29) |
| NSAID/COX2i | 2 / 71 (2.8) | 4.34 (0.67 – 15.7) | 2 / 71 (2.8) | 0.98 (0.16 – 3.20) |
| SGLT2i | 7 / 375 (1.9) | 1.58 (0.61 – 3.58) | 6 / 376 (1.6) | 0.44 (0.17 – 0.94) |
RASi indicates renin-angiotensin system inhibitor, OR odds ratio, CI confidence interval, ARB angiotensin receptor blocker, ACEi angiotensin-converting enzyme inhibitor, DM diabetes mellitus, CHF congestive heart failure, NSAID non-steroidal anti-inflammatory drug, COX-2 cyclooxygenase-2 inhibitor, SGLT2i sodium-glucose cotransporter 2 inhibitor
Bold text denotes statistically-significant results
Subgroups were analysed to determine if there was a dose-response effect for either outcome in patients initiated on losartan (Table S2) and lisinopril (Table S3), the most common ARB and ACEi respectively. Increased daily average initiation dose of both medications was associated with an increased risk of major serum creatinine elevation (OR per 10 mg of losartan 1.15 [95% CI 1.03–1.28], OR per 10 mg of lisinopril 1.06 [95% CI 1.01–1.12]) but not with an increased risk of hyperkalaemia (Table S2 and S3).
Shortening the inclusion period for post-initiation data from 14 to 90 days to 14 to 30 days to align with international guidelines on post-initiation testing [2] (Table S4) decreased the rate of hyperkalaemia from 1.8% to 1.5% and the rate of major creatinine elevation from 2.5% compared to 2.4%, but the multivariable regression findings remained largely unchanged (Table S5). Analysing various permutations of extracting post-initiation serum potassium results (Table S6) found that when the highest serum potassium within the 14-to-90-day range was extracted, the rate of hyperkalaemia increased from 1.8% to 2.3%, and when patients with baseline serum hyperkalaemia were excluded, the rate of hyperkalaemia decreased from 1.8% to 1.6%. Significant risk factors for hyperkalaemia did not vary between these permutations (Table S7).
Discussion
Concordance with testing guidelines before and after RASi initiation
Concordance with guidelines for post-initiation monitoring of serum creatinine and potassium was low in our cohort (41.9%), which is consistent with other studies where less than half of the patients initiated on RASis underwent post-initiation testing [30, 31]. This is contrary to international guidelines that recommend at least annual serum creatinine checks in patients with hypertension [1], diabetes mellitus [32], heart failure [33] or chronic kidney disease [3] and post RASi initiation testing of serum potassium and creatinine [2, 3]. The effect of such missing data is difficult to determine as a proportion of them are from patients transferring existing RASi prescriptions from other institutions to our clinics or ad hoc top-ups of medications normally supplied by other institutions; these cases should rightfully be excluded from analysis. It is common for Singapore patients to consult multiple primary care providers and for speciality clinics to transfer follow-up of stable chronic conditions to primary care providers where testing may have already been performed in other institutions or not clinically indicated. Patients with missing baseline creatinine data also tended to have missing baseline BMI, serum potassium and uACR data which are routinely measured only in patients consulting for chronic disease management, which supports our postulation that these encounters are less likely to be formal chronic disease consultations. Of course, patients may also not have post-initiation data due to poor adherence to follow-up. For these patients, poor concordance with monitoring guidelines suggests substantial barriers to patient adherence. Based on anecdotal evidence in our setting, we postulate that the inconvenience imposed on patients by the short interval between RASi initiation and testing is one such barrier.
Hyperkalaemia
At 1.8%, the rate of hyperkalaemia following RASi initiation in our cohort is one of the lowest reported in the literature within which the majority (1.5%) were mild and would typically be mitigated by reducing dietary potassium intake [4]. Only small minority (0.3%) of moderate or severe cases would necessitate adjusting RASi therapy.
We postulate that one reason for the low rate for hyperkalaemia in our study is lower dietary potassium intake. A prior survey on global dietary potassium intake [34] showed that Singapore had a relatively low mean potassium intake of 2.40 g/day, compared to 3.30 g/day in Western Europe, 2.85 g/day in Australasia and 2.67 g/day in North America.
Significant risk factors for hyperkalaemia were, in descending order of effect size, baseline serum potassium, Indian or ‘Others’ ethnicity, age and baseline serum creatinine. Baseline serum potassium and creatinine were also 2 of the 6 significant risk factors found in the SCREAM study [9], however diabetes mellitus, congestive heart failure, sex and use of mineralocorticoid receptor antagonists were not. The effect size of baseline serum potassium was far larger, while that of worsening baseline renal function was lower.
Differences in baseline characteristics might contribute to the difference in findings; our cohort is older and has a higher prevalence of diabetes mellitus and baseline eGFR < 60mL/min. A lower prevalence of congestive heart failure and use of comedications at the point of RASi initiation may have contributed to insufficient statistical power for these factors. In addition, our study included ethnicity data which also proved to be a significant modifier of risk. This may be due either to variation in dietary potassium intake among the different ethnicities studied or yet-undiscovered differences in mechanisms maintaining potassium homeostasis. We are unfortunately unable to analyse the composition of the ‘Others’ ethnic category which in the latest 2015 Singapore General Household Survey included all other groups including Filipino, Caucasian, Eurasian, Arab, Thai and other ethnicities [35].
Taken together, our findings suggest that not only is the risk of hyperkalaemia in our Singapore primary care cohort among the lowest in the literature, but also that its risk factors may also be different with baseline serum potassium as the strongest predictor and those of older age and Indian or ‘Others’ ethnicity having a higher risk.
Major serum creatinine elevation
Major elevations in serum creatinine were uncommon, with 2.5% of our cohort experiencing a serum creatinine increase of 30% or more from baseline which is in keeping with prior studies [17, 18]. Diabetes mellitus and the concomitant use of diuretics were found to be risk factors for major creatinine elevation both in our primary and complete case analysis, similar to those found in prior studies [17, 18]. We postulate that the underlying mechanism is that both potentiate the fall in intraglomerular pressure during RASi initiation; the former due to microvascular dysfunction from chronic hyperglycaemia and the latter due prerenal volume contraction. While those of ‘Others’ ethnicity had a higher risk of major serum creatinine elevation, similar to hyperkalaemia we were unable to analyse this further due to the heterogeneity of the ethnic groups in this category. Finally, in our subgroup analysis we noted that for both losartan and lisinopril which were the most common ARB and ACEi initiated, there was a dose-response relationship where increasing mean daily doses of either drug conferred a higher risk of major serum creatinine elevation.
We note that increased baseline serum potassium and serum creatinine was protective against major serum creatinine elevation, the later only mildly so. We interpret these findings with caution given that they were not significantly in the complete case analysis, although this may suggest that the pathophysiology behind major serum creatinine elevation is different from that of hyperkalaemia.
We note that within the 2.5% of patients with major serum creatinine elevation, less than half of them (1.2%) had a post-initiation eGFR falling below 60mL/min which is the threshold for clinically significant renal impairment. This raises an open question of whether all major serum creatinine elevations are equivalent regardless of post-initiation renal function, given that the 30% proportional threshold is more likely to be reached for patients with low baseline serum creatinine such as female patients with low muscle mass.
Finally, what is striking when both hyperkalaemia and major serum creatinine elevation analyses are compared is that there are almost no shared risk factors for both outcomes. This suggests that while they are commonly considered to be related adverse effects of RASi therapy, in they are influenced by different pathophysiological mechanisms and the risks of either have to be considered separately.
Implications on clinical practice
Our study’s findings suggest that there may be a role of risk-stratifying the need for serum potassium and creatinine testing post-RASi initiation rather than imposing a blanket requirement, which may improve uptake of RASis in patients who may benefit. In young, non-diabetic patients without use of diuretics, pre-existing renal impairment and a lower baseline serum potassium, our data serves to inform a risk-benefit discussion which may potentially allow a longer interval of lab testing post-RASi initiation, important in this population of patients who are frequently deterred by the inconvenience of repeat laboratory testing within 2–4 weeks.
Strengths and limitations
To the best of our knowledge, our study is the first large-scale cohort study on RASi initiation in an Asian population and the first to study both hyperkalaemia and elevated serum creatinine, which are the main adverse biochemical outcomes. Using data from a primary care cohort increases its applicability to providers in similar settings. Our use of baseline data within 365 days and a post-initiation data cut-off of 14–90 days was designed to align our study with current recommended practices and to isolate the effect of RASi initiation on our outcomes.
Our study has also uncovered a possible role of ethnicity in influencing the risk of such outcomes, particularly for hyperkalaemia which may be influenced by dietary habits. This may have contributed to our cohort having one of the lowest hyperkalaemia rates in the literature and should motivate further study to confirm this finding in larger local cohorts.
This study has several limitations. First, a considerable proportion of RASi initiation cases were excluded because of missing baseline or post-initiation data, which was due either to suboptimal concordance with monitoring guidelines or to patients who were initiated on RASis elsewhere and who continued their medications in our clinics and thus were not true initiations. We postulate that this may be partially attributed to our study period including the COVID-19 pandemic during which primary care attendances for chronic disease consultations decreased due to pandemic restrictions.
Second, our retrospective cohort study design meant that there were gaps in baseline and post-initiation data such that covariates such as baseline uACR and BMI could not be analysed in our main regression model, and that we were not able to determine if patients were adherent to the medications dispensed. Even though non-adherent cases who did not complete follow-up tests were excluded from analysis, non-adherent cases who remained adherent with follow-up tests may have led to an underestimation of the incidence of hyperkalaemia and major creatinine elevation.
Third, our cohort comprises primary care patients from only one of the three public healthcare clusters in our study; hence, patients with RASi initiation or laboratory testing in inpatient settings, specialist ambulatory clinics or other healthcare clusters could not be captured.
Fourth, there was a low prevalence of advanced CKD or heart failure in our cohort, as RASi initiation in our clinics mostly occurred in patients followed up for hypertension or diabetes mellitus who developed microalbuminuria. In contrast, in most patients with congestive heart failure or advanced CKD, RASis are initiated in the inpatient or speciality outpatient clinic setting; hence, our findings are unlikely to be applicable in settings with a greater prevalence of such comorbidities. Our classification of comorbidities based on ICD-10 diagnosis codes may have been an underestimation of their true prevalence in our cohort owing to inconsistent coding practices.
Areas for future research
Our findings suggest the need for larger-scale studies in Asian populations, including data from primary care, inpatient, and specialty clinic settings, to refine our preliminary findings about the risk of hyperkalaemia and major serum creatinine elevation post RASi initiation. Prior studies have developed risk scores for hyperkalaemia [9, 36, 37], which may be useful for guiding clinicians in monitoring after RASi initiation, but these scores have yet to be validated in an Asian population. Finally, while major serum creatinine elevation remains a safety signal post-RASi initiation and may preclude certain patients from fully utilising RASi, long-term studies should be performed to clarify if such major elevations correlate with hard cardiorenal adverse outcomes and whether baseline or post-initiation renal function further stratifies this risk. Further, such studies should evaluate whether the reduction or discontinuation of RASi in these patients has negative effects on the long-term management of their chronic conditions.
Conclusion
In our cohort of 9,926 patients with RASi initiation, 1.8% had hyperkalaemia, of which the majority had mild cases, and 2.5% had major serum creatinine elevation. Risk factors for hyperkalaemia include elevated baseline serum potassium, baseline serum creatinine and Indian or ‘Others’ ethnicity. Risk factors for major serum creatinine elevation include diabetes mellitus, diuretic use and ‘Others’ ethnicity.
Supplementary Information
Supplementary Material 1. Table S1: Baseline characteristics of the cohort including patients with missing baseline data. Table S2: Subgroup analysis of the effects of the losartan initiation dose on hyperkalaemia and major serum creatinine elevation at 14-90 days after RASi initiation (n = 4839). Table S3: Subgroup analysis of the effects of lisinopril initiation dose on hyperkalaemia and major serum creatinine elevation at 14-90 days after RASi initiation (n = 1677). Table S4: Adverse outcomes at 14-30 days after RASi initiation. Table S5: Subgroup analysis of factors associated with hyperkalaemia and major serum creatinine elevation at 14-30 days after RASi initiation (n = 7526). Table S6: Hyperkalaemia at 14-90 days after RASi initiation with various permutations of potassium data extraction methods. Table S7: Factors associated with hyperkalaemia 14-90 days after RASi initiation with various permutations of potassium data extraction methods.
Acknowledgements
We would like to thank Mr. Zachary Lim, Ms Joelle Lam, Ms Moxi Looi, Ms Nur Rasyidah Binte Azman and Ms. Chang Yang Yi (Family Medicine Development, National University Polyclinics) for their assistance with data extraction and Professor Jose M Valderas (Head of Department, Department of Family Medicine, National University Health System), Dr. Sky Koh (Lead, Renal Speciality Advisory Group, National University Polyclinics) and Dr. Hong Weizhen (Senior Consultant, Division of Nephrology, Department of Medicine, National University Hospital) for their editorial assistance and comments on the study design and statistics.
Authors’ contributions
MWL conceived the idea for the study. MWL designed the methodology and performed the statistical analysis, with AS, DO and SL contributing substantially to the study design. All authors were involved in critical discussions and interpretation of the findings. MWL drafted the manuscript, and all authors critically revised it for important intellectual content. All authors approved the final version of the manuscript.
Funding
No funding was required for this study.
Data availability
The dataset supporting this study is proprietary to National University Polyclinics and may be available upon reasonable request from the corresponding author in accordance with prevailing data protection protocols and subject to the necessary ethics approvals.
Declarations
Ethics approval and consent to participate
The study was exempt from ethical review by the institutional Domain Specific Review Board (DSRB Reference 2023/01018).
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Mancia G, Rosei EA, Azizi M, Burnier M, Clement DL, Coca A et al. ESC/ESH Guidelines for the management of arterial hypertension. 2018;39(33):3021–104. [DOI] [PubMed]
- 2.Cheung AK, Chang TI, Cushman WC, Furth SL, Hou FF, Ix JH, et al. KDIGO 2021 clinical practice guideline for the management of blood pressure in chronic kidney disease. Kidney Int. 2021;99(3):S1–87. [DOI] [PubMed] [Google Scholar]
- 3.National Institute for Health and Care Excellence (NICE). Chronic kidney disease: assessment and management (2021) NICE guideline NG203. London: National Institute for Health and Care Excellence (NICE); 2021. [Google Scholar]
- 4.Palmer BF. Managing hyperkalaemia caused by inhibitors of the Renin–Angiotensin–Aldosterone system. N Engl J Med. 2004;351(6):585–92. [DOI] [PubMed] [Google Scholar]
- 5.Raebel MA. Hyperkalemia associated with use of angiotensin-Converting enzyme inhibitors and angiotensin receptor blockers: hyperkalemia with ACEI and ARB. Cardiovasc Ther. 2012;30(3):e156–66. [DOI] [PubMed] [Google Scholar]
- 6.Pfeffer MA, Køber L, de Werf FV, Edwards S. Valsartan, Captopril, or both in myocardial infarction complicated by heart Failure, left ventricular Dysfunction, or both. N Engl J Med. 2003;349(20):1893–906. [DOI] [PubMed]
- 7.The SOLVDI. Effect of Enalapril on mortality and the development of heart failure in asymptomatic patients with reduced left ventricular ejection fractions. N Engl J Med. 1992;3(10):685–91. [DOI] [PubMed] [Google Scholar]
- 8.Weir MR, Rolfe M. Potassium homeostasis and Renin-Angiotensin-Aldosterone system inhibitors. Clin J Am Soc Nephrol. 2010;5(3):531–48. [DOI] [PubMed] [Google Scholar]
- 9.Bandak G, Sang Y, Gasparini A, Chang AR, Ballew SH, Evans M, et al. Hyperkalemia after initiating Renin–Angiotensin system blockade: the Stockholm creatinine measurements (SCREAM) project. J Am Heart Assoc. 2017;6(7):e005428. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Wetmore JB, Yan H, Horne L, Peng Y, Gilbertson DT. Risk of hyperkalemia from renin–angiotensin–aldosterone system inhibitors and factors associated with treatment discontinuities in a real-world population. Nephrol Dial Transpl. 2021;36(5):826–39. [DOI] [PubMed] [Google Scholar]
- 11.Beddhu S, Rocco MV, Toto R, Craven TE, Greene T, Bhatt U, et al. Effects of intensive systolic blood pressure control on kidney and cardiovascular outcomes in persons without kidney disease: A secondary analysis of a randomized trial. Ann Intern Med. 2017;19(6):375. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Hu AH, Chang TI. SPRINT—A Kidney-Centric narrative review: recent advances in hypertension. Hypertension. 2021;78(4):946–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Collard D, Brouwer TF, Peters RJG, Vogt L, Van Den Born BJH. Creatinine rise during blood pressure therapy and the risk of adverse clinical outcomes in patients with type 2 diabetes mellitus: A post hoc analysis of the ACCORD-BP randomized controlled trial. Hypertension. 2018;72(6):1337–44. [DOI] [PubMed] [Google Scholar]
- 14.Palmer BF. Renal dysfunction complicating the treatment of hypertension. N Engl J Med. 2002;347(16):1256–61. [DOI] [PubMed] [Google Scholar]
- 15.Kaplan NM, Palmer BF, Palmer BF. Impaired renal autoregulation: implications for the genesis of hypertension and hypertension-Induced renal injury. Am J Med Sci. 2001;321(6):388–400. [DOI] [PubMed] [Google Scholar]
- 16.Wright JT, Bakris G, Greene T, Agodoa LY, Appel LJ, Rostand SG. Effect of blood pressure Lowering and antihypertensive drug class on progression of hypertensive kidney disease: results from the AASK trial. J Am Med Assoc. 2002;288(19):2421–31. [DOI] [PubMed] [Google Scholar]
- 17.Schmidt M, Mansfield KE, Bhaskaran K, Nitsch D, Sørensen HT, Smeeth L et al. Serum creatinine elevation after renin-angiotensin system Blockade and long term cardiorenal risks: cohort study. BMJ. 2017;356:j791. [DOI] [PMC free article] [PubMed]
- 18.Fu EL, Trevisan M, Clase CM, Evans M, Lindholm B, Rotmans JI, et al. Association of acute increases in plasma creatinine after Renin-Angiotensin Blockade with subsequent outcomes. Clin J Am Soc Nephrol. 2019;14(9):1336–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.National Population and Talent Division, Group S. Prime Minister’s Office, Singapore Department of Statistics, Ministry of Home Affairs, Immigration & Checkpoints Authority, Ministry of Manpower. Population in Brief 2024 [Internet]. Government of Singapore; 2024. Available from: https://www.strategygroup.gov.sg/files/media-centre/publications/Population_in_Brief_2024.pdf
- 20.Goh RSJ, Chong B, Jayabaskaran J, Jauhari SM, Chan SP, Kueh MTW, et al. The burden of cardiovascular disease in Asia from 2025 to 2050: a forecast analysis for East Asia, South Asia, South-East Asia, central Asia, and high-income Asia Pacific regions. Lancet Reg Health - West Pac. 2024;49:101138. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Yildirim T, Arici M, Piskinpasa S, Aybal-Kutlugun A, Yilmaz R, Altun B, et al. 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. [DOI] [PubMed] [Google Scholar]
- 22.Shirazian S, Grant CD, Mujeeb S, Sharif S, Kumari P, Bhagat M, et al. Underprescription of Renin–Angiotensin system blockers in moderate to severe chronic kidney disease. Am J Med Sci. 2015;349(6):510–5. [DOI] [PubMed] [Google Scholar]
- 23.Clase CM, Carrero JJ, Ellison DH, Grams ME, Hemmelgarn BR, Jardine MJ, et al. Potassium homeostasis and management of dyskalemia in kidney diseases: conclusions from a kidney disease: improving global outcomes (KDIGO) controversies conference. Kidney Int. 2020;97(1):42–61. [DOI] [PubMed] [Google Scholar]
- 24.McDonagh TA, Metra M, Adamo M, Gardner RS, Baumbach A, Böhm M et al. 2021 ESC guidelines for the diagnosis and treatment of acute and chronic heart failure. Eur Heart J 2021 Sept 21;42(36):3599–726. [DOI] [PubMed]
- 25.Levey AS, Stevens LA, Schmid CH, Iii AFC, Feldman HI, Kusek JW et al. A New Equation to Estimate Glomerular Filtration Rate. Ann Intern Med. 2009;150(9):604–12. [DOI] [PMC free article] [PubMed]
- 26.R Core Team. R: A Language and Environment for Statistical Computing [Internet]. Vienna, Austria. 2024. Available from: https://www.R-project.org/
- 27.Wickham H, Averick M, Bryan J, Chang W, McGowan L, François R, et al. Welcome to the tidyverse. J Open Source Softw. 2019;4(43):1686. [Google Scholar]
- 28.Pattaro C, Fujii R, Herold J, nephro. Utilities for Nephrology [Internet]. 2023. Available from: https://CRAN.R-project.org/package=nephro
- 29.Sjoberg DD, Whiting K, Curry M, Lavery JA, Larmarange J. Reproducible summary tables with the Gtsummary package. R J. 2021;13(1):570. [Google Scholar]
- 30.Parikh RV, Nash DM, Brimble KS, Markle-Reid M, Tan TC, McArthur E, et al. Kidney function and potassium monitoring after initiation of Renin-Angiotensin-Aldosterone system Blockade therapy and outcomes in 2 North American populations. Circ Cardiovasc Qual Outcomes. 2020;13(9):e006415. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Schmidt M, Mansfield KE, Bhaskaran K, Nitsch D, Sørensen HT, Smeeth L, et al. Adherence to guidelines for creatinine and potassium monitoring and discontinuation following renin–angiotensin system blockade: a UK general practice-based cohort study. BMJ Open. 2017;7(1):e012818. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.American Diabetes Association. Standards of care in Diabetes—2023 abridged for primary care providers. Clin Diabetes. 2023;41(1):4–31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Al-Naher A, Wright D, Devonald MAJ, Pirmohamed M. Renal function monitoring in heart failure - what is the optimal frequency? A narrative review: frequency of renal monitoring in heart failure. Br J Clin Pharmacol. 2018;84(1):5–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Reddin C, Ferguson J, Murphy R, Clarke A, Judge C, Griffith V, et al. Global mean potassium intake: a systematic review and bayesian meta-analysis. Eur J Nutr. 2023;62(5):2027–37. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Singapore Department of Statistics. Resident Population of Other Ethnic Groups by Age Group, Ethnic Group and Sex (General Household Survey 2015). data.gov.sg; 2023 [cited 2025 Dec 20]. Available from: https://data.gov.sg/datasets/d_5d80697df3d471b7c4ae4d0a60030368/view
- 36.Johnson ES, Weinstein JR, Thorp ML, Platt RW, Petrik AF, Yang X, et al. Predicting the risk of hyperkalemia in patients with chronic kidney disease starting lisinopril: PREDICTING THE RISK OF HYPERKALEMIA IN CKD. Pharmacoepidemiol Drug Saf. 2010;19(3):266–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Ahuja TS, Freeman D Jr., Mahnken JD, Agraharkar M, Siddiqui M, Memon A. Predictors of the development of hyperkalemia in patients using Angiotensin-Converting enzyme inhibitors. Am J Nephrol. 2000;20(4):268–72. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary Material 1. Table S1: Baseline characteristics of the cohort including patients with missing baseline data. Table S2: Subgroup analysis of the effects of the losartan initiation dose on hyperkalaemia and major serum creatinine elevation at 14-90 days after RASi initiation (n = 4839). Table S3: Subgroup analysis of the effects of lisinopril initiation dose on hyperkalaemia and major serum creatinine elevation at 14-90 days after RASi initiation (n = 1677). Table S4: Adverse outcomes at 14-30 days after RASi initiation. Table S5: Subgroup analysis of factors associated with hyperkalaemia and major serum creatinine elevation at 14-30 days after RASi initiation (n = 7526). Table S6: Hyperkalaemia at 14-90 days after RASi initiation with various permutations of potassium data extraction methods. Table S7: Factors associated with hyperkalaemia 14-90 days after RASi initiation with various permutations of potassium data extraction methods.
Data Availability Statement
The dataset supporting this study is proprietary to National University Polyclinics and may be available upon reasonable request from the corresponding author in accordance with prevailing data protection protocols and subject to the necessary ethics approvals.